I’m not going to discuss the writeup itself in depth — you can read it if you’re so inclined. But I’ll touch on my mindset with this sort of document while using it as an example of the value of fresh eyes on a project.

It’s 28 pages of issues with few compliments, so you might assume I’m negative on the game. But I could write this volume of text for any in-progress game. I think Poly Fighter, while early, is pretty cool.

If you make a long list of issues to fix, then fix them, you’ll probably end up with a game that’s good if not great. A long list of issues is daunting, and fixing them is way less fun than dreaming up new features. But that’s why game development is a job and not just a fun hobby. (Though it can be that as well)

Many great games don’t have exciting new features and aren’t uniquely inspired. They become great simply by virtue of tediously improving what needs it.

I would refer to 15 Years of Indie Dev In 4 Bits of Advice and this bit:

Redfall has a 57 on Opencritic. The exact same game with issues addressed would be a 77, maybe up to an 82. Pushing higher than that might involve more fundamental changes: ditching the animatic-styled cutscenes, making the unique character abilities more interesting, delivering on the Far Cry 2-style systems-driven open world that was teased but is nowhere in the shipped game. But creating a long list of issues to fix then tediously fixing them would have given the game a significant bump.

I view a writeup like this one for Poly Fighter as a list of issues to address. As such I don’t do “sandwich a complaint between two compliments.”

Too many negatives and no positives might be demoralizing. But a list of compliments, while motivating, isn’t actionable, and overly-kind feedback can be misleading. A common element of some of the most troubled projects is a “toxic positivity” mindset where negative feedback is both ignored and discouraged, or where the developers have an insular view of their game. (If nothing else, adding two compliments for every issue would make my document 84 pages long…)

“Toxic positivity” gets thrown around often now. Here’s Josh Sobel, a lead tech artist at Wildlight who was let go along with most of the staff immediately following release, describing Highguard:

“The future seemed bright. Everyone I knew who had any connection to the team or project had the same sentiments: ‘This is lightning in a bottle.’ ‘I trust this team wholeheartedly.’ ‘If there’s one project nobody in the industry is worried will fail, it’s yours.’ ‘This has mainstream hit written all over it.’ ‘There’s no way this will flop.’ ‘I could play this game all day,’” Sobel said.

I’ve played Highguard and I don’t think it’s a bad game. But I could make a long list of issues that could be improved or fixed, and that’s a released game, not an in-development one.

I don’t love the term “toxic positivity” as people apply it to teams without knowing anything about that team’s culture. Sometimes teams suppress negative feedback through a variety of cultural mechanisms, and reward the relentlessly positive as “team players.” But sometimes teams are just naturally optimistic and perhaps a smidge pollyannaish — that’s not “toxic positivity” that’s plain old positivity. But plain old positivity can still be a problem and result in the same conclusion.

As such I think following a “sandwich a negative between two positives” approach does the team a disservice, that spares their feelings at the risk of damaging their long-term success chance.

A last note about my Poly Fighter writeup is that, were I a new hire on the team, or a consultant rather than a total outsider, it would be much less talky, exclude many known issues, and cut down on the number of suggestions. Stating known issues can be valuable to underscore their importance, but I wouldn’t mention the end-of-match jingle, for example, if I knew it was a temporary asset.

Issue Blindness and Losing Perspective

Older games from the era before extensive playtesting often have glaring balance, understandability or control issues.

One reason for this is simple: while working on a game you eventually get used to the current state, in the same way someone becomes used to the particular smell of their house.

A game developer might have played their game for hundreds of hours. Not only are they used to enemy patterns, the controls, etc, but they designed those enemy patterns and controls. Game designers often vastly overestimate how understandable their systems are because they designed those systems and understand their underlying logic and motivation, whereas a player can only interpret that logic via what’s surfaced in game and can only guess at why that system exists at all.

As a developer you never have a legitimate new player experience on your own game, and to whatever extent you approximate it that quickly becomes a distant memory. Very few developers can put themselves into the shoes of a new player.

To use some specific examples here, players complain that the loot-finding stage of Highguard is dull. The developers understand the map layouts, the traversal and mining mechanics, and the relative strength of mined crystals and store-bought items vs loot found in treasure chests. So to them optimizing their paths to maximize preparation and economy is an interesting minigame. But to a new player there aren’t interesting tradeoffs because they don’t understand what those tradeoffs are.

2XKO has some odd control choices but after hundreds of hours they become muscle memory. The developers aren’t going to forget which button activates which move or accidentally get a super move instead of a special move because they pressed a button sequence too quickly.

I played the demo for Queen’s Domain, a fantastic-looking faux-retro Kings Field-like. The first enemies I ran into were insanely difficult — each of my hits did 10% damage to them and they did 30% damage to me, and I was outnumbered three-to-one.

The developers probably know that if you run and move in a certain way the enemies can’t hit you at all, and if anything think they’re too easy.

Low Prioritization of Avoidable Problems

Another element of blindness is voluntary: when developers know an issue exists but continually de-prioritize it because they know to avoid it.

To use a personal example, I’m working on a game with a 3D camera. The older implementation of the camera had some odd behavior: as the player rotated the camera down into the ground the camera would slide along the ground closer to the character, then as the player rotated the camera back up it would slide back along the ground instead of immediately moving upwards. It felt like the camera was magnetized to the ground. It didn’t honor the player’s input; the camera moved out when the player told it to move up. I’d planned to fix this issue eventually, but I didn’t consider it a big issue because I simply learned to avoid it. As the funny Doom tweet said “you control the buttons you press.” I stopped rotating the camera down into the ground so the resulting behavior stopped bothering me.

Coming back to the project after some time off I realized how incredibly annoying this issue actually is, and fixed it and a number of similar issues immediately.

The Burden of Historical Knowledge and Being Smart Enough to Rationalize Bad Decisions

A common game development failure pattern is when decisions are constantly revisited or undone — when flying in Anthem is added then removed then added again. But an opposite problem can also occur: a feature is tried and dismissed, and remains mentally categorized as a failed effort when the context (including office politics) around it changes.

It’s usually the case that when players make an obvious suggestion the developers already tried it or avoided it for good reason: the feature didn’t work out in practice or was too work-intensive. It’s rare that the single-most-obvious idea just never occurred to the team. But it’s not that rare that a feature that didn’t make sense at the time makes more sense now.

In that case knowledge of development history is a burden, and it can be clarifying to hear a naïve “just try this obvious thing” take.

Sometimes it’s useful to ask “what would a fresh player and absolute dummy (no offense) think?” An example of this is Nintendo’s insistence on displaying long lists of options as a scrolling single-line rather than a grid or radial-menu or matrix.

Presumably Nintendo has real reasons to stick to this design — they did some research that showed that players struggle with more complex UI, for example. But I suspect that those reasons are outdated, inapplicable and need revisiting. Maybe sticking to a single horizontal line makes sense when the UI only displays up to 15 icons, but breaks down when Echoes of Wisdom displays dozens of choices.

There’s a smart person at Nintendo who could make a sophisticated argument for this UI style, while any dummy can say “this single-line list is just too damn long!” But in this case the dummy is probably correct.

Similarly Nintendo’s sports games no longer have the story / RPG modes that people enjoyed. I assume there’s a good rationalization for that but that’s not the same as a good reason. Sometimes developers can outfox themselves and execute on a sophisticated philosophy that performs worse than “give the people what they want.”

Becoming Blind to Strengths

There are many ways to become blind to weaknesses in a game, or blind to fruitful potential avenues. It’s also possible to become blind to the strengths of an in-development game. I haven’t played it so this is an uninformed take, but my impression of Big Hops is that it has many features that don’t relate much to the core appeal. It has a Mario 64 move set and adds wall running, climbing, and swinging with a tongue — that’s more than enough for a full game. (It’s more than what Mario 64 offers) But it also has items (not just temporary powerups), an inventory and inventory management, swappable trinkets that provide different perks, NPCs, quests, and minigames where you use your tongue to solve Pipe Dream-style puzzles. Looking at these minigames I can’t help but wonder if the team, after hundreds of hours of play, grew too used to the core mechanics and worried that the game was too plain and needed more pizazz in the form of new features.

Game Developer Critique of In-Progress Games

In my first blog post, originally written for GameDeveloper.com in 2015 about Dark Souls, I explained my motivation this way:

This is inspired by Matt Zoller Seitz's Please, Critics, Write About the Filmmaking, the gist of which is that critics, rather than writing solely about how a film made them feel, should examine how the film made them feel that way.

I consider much of what I write craft critique.

Any player can write “I liked the music” or “the controls are unresponsive.” There’s nothing wrong with those observations but a game developer sharing them is functioning as a player, not as as a professional — like a director praising the work of another director with “I enjoyed the movie!”

I don’t expect you to read my 30-pages of Poly Fighter impressions (really it’s like 17 pages and some screenshots) but I’d highlight a few sections to underscore the difference between player and game developer feedback.

I don’t think many players would include drawing-order issues in their feedback. They might identify related symptoms: the action is harder to read than it should be, it can be hard to tell if a hit landed or where it landed, hits are less satisfying than they could be, some VFX can look wonky, etc. But they probably wouldn’t identify z-axis sorting issues as the culprit.

Lack of clarity around in-combat rules like air-juggling is another issue they might obliquely gesture at without identifying directly.

Craft Critique vs Other Type of Criticism

This sort of feedback isn’t a review. I often bristle when reviews contain suggestions. At that point the game has been released so (patches and live service games notwithstanding) that feedback isn’t actionable. Most game reviewers aren’t game developers so “here’s what I would have done” is silly. But a game developer giving suggestions on an in-progress game is potentially helpful.

Live Play Criticism

I would contrast this sort of criticism to criticism-as-content. I’ve watched videos where a game developer plays a game on Twitch or YouTube and shares criticism — but I’ve never seen a good one. That format isn’t conducive to useful, honest feedback.

Playing a game live there’s an incentive to just keep talking. If you keep asking play-testers for their impressions, after they’ve exhausted their initial feedback, they’ll often simply make things up to avoid appearing like they have nothing to say. In the same way a game developer doing running commentary has to fill silence with something, whether it’s pertinent or not.

The format discourages grace and consideration. You might play an RPG that introduces a new battle mechanic that’s initially hard to grasp — often RPGs are designed with the expectation that new systems are initially opaque but will become clearer over time. But if you’re playing a game live and doing running commentary as content there’s a strong incentive to declare that the tutorial is bad or that the game has “poor onboarding.”

My greatest objection to this sort of commentary is that it’s often ego-driven, where the developer providing the criticism is fundamentally motivated by the desire to show off and say “here’s how I would have done this better.”

This sort of criticism helps the person doing the play-through accumulate views and gives their audience something to chortle at, but doesn’t help the game’s developer at all. This sort of criticism delivers flexing, not real insights.

Arguably any criticism intended for any audience beyond the game’s developers is “content” — I’ve turned my Poly Fighter writeup into blog content. But I’d consider most live-plays “#content” — a pejorative.

Subscribe now

Mining New Hires for Impressions

Not many game developers are going to write something like my Poly Fighter writeup — at least not for free. This kind of craft critique is valuable but in scarce supply.

As a developer you can do playtests, betas and early access, but while that provides useful feedback it doesn’t provide this type of feedback.

I’ve heard that some companies solicit this sort of feedback when interviewing game design candidates; but I’ve also heard that game designers don’t like being asked for this sort of feedback. I’m not sure what the exact objections are but I imagine it’s awkward to be asked to critique a game you’re hoping to get hired on — a sort of “describe your greatest weakness” question that dissuades honest answers. It might also come off as spec work. (Valuable but unpaid real work.)

At conventions other game developers might play your games and offer feedback, but those impressions are usually based on a few minutes of play. And I’ll be honest here: when I’ve shown up to E3 industry-only hours, played a game and been asked for feedback, I’ve always been positive, even when I wasn’t positive on the game. I pretended not to notice the screen-tearing in Killer is Dead and I said the cinematic QTE car chase sequences in Spider-Man: Web of Shadows2 were “neat” when I found them dull. Conventions are fun atmospheres and players, including other game developers, shy away from being downers. If your game gets a tepid response at a show that’s a bad sign, but if it gets a good response at a show that’s probably a nothing sign.

You can ask game-developer friends to play your game and give feedback, but my impression is that that feedback, though it varies widely, is often quite succinct. And depending on the personality of your friend and the nature of that friendship it’s likely to be overly positive.

You can pay outside consultants for mock reviews and game evaluations. A mock review is when an ex-journalist writes a review as if they were still at a review website: “your game would get a 7.5 at Gamespot and here’s why.” A game evaluation is when they write “and here’s how it could get an 8.5 instead.” Mock reviews are valuable, but the people offering these review and evaluation services tend to be ex-journalists with no experience making games, so I don’t know why you’d take their advice on how to make yours. Game journalists consulting on a game are articulate gamers more than they’re game developers, even if they’re now technically in the game development business via consulting.

So while game developer feedback is valuable there aren’t many places to find it. Which brings me to new hires.

It can be awkward for new hires to be critical, but it’s less awkward than it is for job seekers. Asking a new hire to provide this feedback isn’t asking for a favor or spec work — it’s asking them to do their job. (As you define it)

Some people aren’t well-suited to offer this sort of feedback. They may be too polite, too narrow in interests, or have off-kilter opinions. But each new hire is at least a potential resource.

I don’t like to give advice as I hardly have life figured out. But I think it’s wise to treat new hires as a potentially valuable resource with a limited shelf-life — certain flaws fade to low-priority or invisibility within weeks or even days. Have them meet the team, get set up and familiarized with processes and software, go through all the HR rigamarole, then have them play the game and provide feedback before they’ve been subject to in-depth system and mechanics explanations, so that they’re getting an experience close to that of a new player. That gives you access to an expert, in-depth opinion.

And that opinion isn’t fire-and-forget, the way playtest sessions or outside consulting might be. I was once hired on a project and some of my first feedback was “the pathfinding stalls the entire game”; one of my first tasks was fixing the pathfinding stalls. Which makes sense — let the guy most bothered by that issue improve it till it meets his standards. Even if I wasn’t technical enough to fix the issue myself I could have been assigned to approve any fix. This is also a natural way to combat toxic negativity, by allowing people to be the change they want to see rather than encouraging fruitless griping.

• Monster Hunter Wilds is My Most Disappointing Game of the Year

• The Video Game Industry is Not Doomed but It Is a Mess

• A Eulogy for Games Media

• The Year I (Re)Discovered that Youtube is Bad

• Thoughts on The Game Awards

• The Rise of Indie Games

• Do I like The Hundred Line: Last Defense Academy?

Monster Hunter Wilds is My Most Disappointing Game of the Year

I normally wouldn’t identify a most disappointing game, but Monster Hunter Wilds is worth highlighting for its breadth of misses. I can’t think of any sequel with a higher rate of misguided decisions.

The back of the box features turned out to be meaningless. Weapon switching occasionally saves you a few seconds by letting you skip returning to your tent — that’s it.

It’s open world only in a strictly technical sense. It’s a game of zones stitched together with loading corridors; you can use these corridors to seamlessly transition between zones but thanks to fast travel you never should. I travelled between zones exactly once, just to confirm it was possible. Despite being an open world game it includes significantly less exploration and discovery than Monster Hunter Rise. You can auto-traverse each zone on an off-brand Chocobo and fast travel, meaning you never have to learn the landscape, which is poorly designed for on-foot travel anyway. And each zone is first introduced then traversed via automated and tedious walk-and-talk sequences, rather than letting the player explore and discover them.

It has few open world features like quests, meaningful points of interest or survival elements. The closest it gets is campsites which allow for fast travel — campsites that monsters will often destroy 30 seconds after they’ve been built, and that feel like vestigial remnants from a more ambitious game that never materialized.

The story is dire. Monster Hunter games typically have bare-bones functional stories that give the player a reason to hunt monsters. Wilds has that same plot but with ten times the volume, told through interminable on-rails segments and cutscenes where characters ramble about “Wyvern Milk.” Your character’s role in the story is that of a silent enforcer, like the modern himbo Link minus the charm. The central character is an annoying child, making the game one long escort quest.

That child’s character arc is one of the worst I can recall. He begins the story with empathy for monsters — can’t humans and monsters live in peace? The big turn — the emotional climax and growth moment — comes when he realizes that sometimes you just gotta murder monsters.

Technically it’s a mess. It performs too poorly for the visuals it outputs, and is an example of what’s wrong with modern game engines from a graphical perspective. Thanks to low resolution plus upscaling the image looks both pixelated and blurry. It relies heavily on upscaling, temporal AA and other smearing techniques to hide a low-quality base image and stochastic effects.

Here’s a screenshot I took where I tried to turn off all the upscaling / TAA / smearing to look at a raw full-resolution image — it looks terrible. The rim lighting and specular on the character in front of the forge is random pixels; far worse than the rim lighting in Breath of the Wild for Wii U. The cloth canopy above her has unmotivated alternating light and dark vertical stripes that smearing is supposed to hide. This particular scene runs at half the framerate of Helldivers 2 on my PC and looks half as good.

Worst of all, the core gameplay loop — hunt monster, use the spoils to craft better gear, hunt stronger monster — is completely busted. There’s no hunting at all, it’s effectively a boss rush game where you fast travel to arenas. The story serves up each region and monster, which you’re expected to beat the first time, so by the time the game opens up you’ve seen every area and beaten most enemies. The “real” game ends as soon as it begins; it seems designed for streamers to beat in 12 hours, swear they had a great time, then never touch again.

The monster fighting itself is the saving grace, but even that’s blemished. I enjoyed the wounds system at first, but the addition of a free-aim toggle turns the combat into something more like a Skyrim or Avowed, where you’re a strafing pseudo-first-person DPS machine.

2XKO is a game with major problems but it has the raw materials of a good game. It’s fixable with some commitment. There’s no fixing Wilds. The cracks are everywhere and they run deep. Players complained that the game is too easy, so in an update Capcom added a monster that can one-shot players from full health. That makes the game harder but no more enjoyable. The underlying systems forced Capcom’s hand: at any point during combat you can call your mount to scoop you up and make you effectively invincible while you heal up, so it’s almost impossible to die from small damage instances.

Wilds isn’t a terrible game — it’s certainly not the worst game of the year. But it might be the most misguided one. Nearly every change is for the worse.

The Video Game Industry is Not Doomed but It Is a Mess

Video game industry doom and gloom kicked into high gear when Matthew Ball released his State of Video Gaming in 2025 report. That report was turned into a video from Alanah Pearce titled “The games industry is screwed,” and that got turned into Eurogamer’s “GTA 6’s delay doesn’t mean the games industry’s in trouble - it’s already dead.”

That report drove much of the conversation but, even to those blissfully unaware of its content or derivatives, clearly there’s something not right in the world of video games. At this point I don’t think I can buy that layoffs are simply balancing COVID-era over-hiring. Microsoft’s gaming division is flailing, Amazon’s is flailing, Netflix is selling theirs off. Microsoft acquired Activision to make Call of Duty their Game Pass crown jewel just as COD lost its luster. Sony acquired Bungie as Destiny began a downswing and Marathon struggled with tepid reception and a plagiarism controversy. Rise of Hydra and Fairgames were delayed, Wonder Woman and Perfect Dark were cancelled. EA’s Black Panther was cancelled and after 2 or 4 years of development (depending on who you ask) had nothing to show but concept art.

That last one is emblematic of modern video game development. EA committed to a slow process that included two years of “incubation” — I understand what that means for tech companies but no idea what it means for video game developers — then cancelled the project because it was moving too slowly. Microsoft cancelled Everwild after working on the game for years despite not knowing what it was. As I titled a blog post, we’re struggling to make games.

From a worker standpoint the games industry is in a terrible state. My personal canary-keeled-over-in-the-coalmine is when graphics programmers are laid off, as that’s an important position without many qualified applicants. When associate producers are laid off that could be “trimming the fat” (no offense associate producers) but laying off graphics programmers says “we’re getting out of this business.”

Listing all the troubled consolidations, closures, cancellations, Saudi buyouts etc would be an entire post in itself.

That said video game companies are still making lots of money, and video games are still in high, if only steady, demand. From a consumer perspective there are dozens of great games releasing every year, so these production woes haven’t translated into worse quality on the whole.

The problem facing video game media (up next) is that there’s not much demand and little purpose being served — it’s largely obsolete. In the past year I’ve read so many “young male NEETs play video games all day” think pieces — games can’t be obsolete. But they are in turmoil.

A Eulogy for Games Media

I’ve toyed with writing something negative about games media in the past, like a “Top 5 Worst Articles of All Time” list, but it felt too mean and like kicking a horse when it’s down.

Gaming outlets have made mistakes but the fundamental problem is outside their control: the original service that games media provided — information on games — is no longer required, so outlets have to continually search for new angles to remain relevant. The last of those angles was guides but now Google search or AI-generated websites scrape and present those guides while bypassing their creators.

There will always be gaming websites — in that sense games media will never die. But many new outlets are based on goodwill accrued from a past era, while those past-era websites close up or shamble along with reduced staff and low-effort content. Aftermath is a website for those who miss Kotaku, and Easy Allies is an outlet for those nostalgic for Game Trailers. If you miss the old Giant Bomb maybe you’ll support the new Giant Bomb, which was sold to current employees by the former owners who didn’t see it as a worthwhile business.

In many cases these outlets aren’t looking to grow, only to subsist or slowly wither. They can’t pay forward their path to success of minting reputations.

The still-existing major websites frequently announce new layoffs and have stripped content down to bare essentials like Marvel movie listicles. I just checked Gamespot and one of their top stories today is “Marvel’s Spider-Man Magic: The Gathering Booster Box Is Nearly 50% Off At Amazon.”

That’s not to knock these authors — if I was paid $40 an article I’d also write a listicle or push a referral code.

Every time an outlet closes we hear “actually it was profitable and there’s still high demand” — but if so why don’t laid-off employees from Fanbyte or Waypoint make their own website and take that profit for themselves? (Waypoint has a streaming-centered spinoff but it’s another site based on fondness for the previous site)

I don’t think the demand is there. Old gaming magazines like EGM and Gamefan thrived on previews, but now video game publishers provide that preview material. Modern gaming websites have leaned heavily into cultural criticism but every geek culture consumer has a venue to share their punditry. We’re drowning in takes.

Now here’s where I say some nice things about games media.

I have a low-brow view of games journalism — I don’t expect or require it to deliver “important” content or topical cultural criticism.

Polygon once ran a piece titled “What Watch Dogs can teach us about the situation in Ferguson.” I have no interest in what video games journalists think Watch Dogs has to teach us on that subject, any more than I care about Gaza commentary from a vacuum cleaner magazine. It’s odd and a little unfair that games media writers are expected to serve as cultural commentators, as if consuming geek culture products gives one an expert understanding of world affairs.

(I’ll get to the nice part I swear)

It would be nice for games media to regularly produce deep analysis of video games and that’s well within their purview, but I assume that’s high-effort low-reward content.

So what I realistically expect is useful reviews that answer, indirectly, “would the reader like this game?” And — I’ve gotten to the nice part — gaming media delivers that.

I find game reviews far better than movie, book and especially TV reviews. Television reviews are typically based on a partial season. Movie and television reviews are highly susceptible to trends and to reviewing the idea of the work rather than the work itself. Someone invested in the post-Get Out movies-as-racial-social-commentary movement is reluctant to write that another black-trauma show is a bummer that doesn’t have “something to say” as much as something to repeat, and someone who resents that movement gives Sinners a bad review when their review makes clear they didn’t understand basic plot points.

Professional book reviews of “serious” works are often either the reviewer promoting a colleague or taking down an enemy. Science fiction book reviews are often more twitter-presence review than anything else: this author’s bio says they’re an Asian enby, has both Ukranian and Palestinian flag emojis, and they’re good at tweeting out sick burns, so they probably wrote one of the best books of the year. If you think this is cynical or me being “anti-woke” there’s an entire trilogy of books that won awards and esteem due mostly to funky pronoun use.1 Then there’s the Isabel Fall fiasco, where a who’s who of social media stars got “I Sexually Identify as an Attack Helicopter” pulled from publication for being offensive and not “#ownvoices” (irrelevant and false), made up that the author was a Nazi, then later admitted that they hadn’t read past the title of the story. Some of the top (cough) science fiction guild members, critics and writers (truly baffling!) don’t bother to read the work at all. Critics who judge based entirely on the identity of an author, and/or without reading the work, is a common scandal in genre and young adult fiction.

In that context video game reviews are pretty damn good!

One of the highest compliments I can pay to reviews is that I frequently read a positive one and conclude that I won’t like the game, or vice-versa, as it effectively describes the content and not just the author’s opinion of the content.

I could lodge some complaints about game reviews. A Demonschool review doesn’t need to start with “does it rise to the top of the class or should it have been held back?” and end with “it makes the grade but the tedious demon volcano2 final boss is hardly magma cum laude.” But — maybe this is me at my most basic bitch — a review serves a utilitarian purpose. If it gets the job done it’s fine if your Doom review isn’t Dostoevsky.

Some bristle at the idea of reviews as buyer’s guide. But that’s why they exist, historically, and you can include heady analysis in your review. You just don’t have to.

Written reviews are one of the few remaining domains of games media, and something games media does well. I’d rather read a smattering of written reviews than watch YouTube videos, and “content creators” often make for poor critics. Maybe everyone else has moved on to video reviews, video essays and podcasts, but I still enjoy Ye Olde Fashioned Reading.

The Year I (Re)Discovered that YouTube is Bad

I’ve played fighting games for decades but until researching 2XKO I had no exposure to fighting game Reddit groups or “content creators.” As such I didn’t have first-hand knowledge of how bad the “which YouTuber did you get your opinion from?” problem is.

It’s shocking how willing people are to repeat whatever their favorite YouTuber or streamer last said, as if it represents their own experience, sometimes at the expense of their own experience. I suppose when people use Chat GPT to write birthday wishes to friends it shouldn’t be shocking, but it is.

Nearly all the popular 2XKO talking points originate with a handful of influencers. If those influencers fixate on irrelevant issues the “fighting game community” does as well, and if those influencers ignore major issues so does that community.

Various fighting game subreddits argue whether Marvel Tokon “has sauce”, because an influencer claimed that the game is “sauceless.” When you read these conversations you’re quickly forced to conclude that the participants have no idea what “sauce” means — it’s just a word they heard and repeat.

I came back to Helldivers 2 after a break, did a few missions, and noticed the Automaton faction had a new enemy type, the War Strider. It seemed pretty cool but not all that threatening. I encountered a bunch and defeated them using the advanced technique of “shooting them a lot.”

When I checked the subreddit it was packed with complaints that the War Strider breaks the game, is too hard, is “bad game design” and so on.

And of course, there’s a YouTube video with 200k views making exactly these points.

One of the main points of the video is that the new enemy is harder than existing ones. Well no shit. Players have gotten better and gotten access to new equipment, so the new enemy exists to be harder. That’s the point. The old enemy took one shot to dismember but this one takes two. Yes exactly!

The video ends with the YouTuber complaining that the Helldivers 2 developers listen to the wrong people — the “glazers” — who want the game to be arbitrarily difficult. Many of his videos feature him dying then doing a shocked face and moaning that the game is too hard and poorly designed.

But his channel also has videos like “I Solo’d Max Difficulty with a Dance Pad.”

So the game is too hard, too cheap, too poorly designed, and caters too much to the hardcore — but the guy making this claim can solo (it’s a game designed for 4 players) the hardest difficulty missions with a gimmick peripheral.

This is exactly who developers should not listen to: someone who creates hammy artificial content and who plays the game in a niche way.

When I track down the oldest reddit threads from the War Strider’s introduction they’re more positive. Once the community absorbed the YouTube content the tone shifted to complaints about high difficulty, “bad design” etc.

Standard game development wisdom is that players are often right about problems (if not the underlying problem then at least that one exists) but wrong about their solutions.

This is wisdom I’ve never taken issue with; it’s just been plainly true for decades. But now I’m not so sure. It held true because players complained about problems they experienced, but now many players complain about problems they experience only vicariously.

What I came to realize this year is that when Arc System Works processes their survey data from the Marvel Tokon: Fighting Souls beta weekend, the 10,000 responses they get aren’t 10,000 summations of lived experience. A number of those responses — probably a large number — are effectively the same handful of content-creator opinions repeated over and over, sometimes at the behest of those creators encouraging their viewers to fill out the survey a certain way.

Which defeats the purpose of the survey.

Until this year I hadn’t realized to what extent YouTube drives — or rather substitutes for — player impressions. That’s what I mean when I say that YouTube is bad.

That said this account of a cat that does flips is pretty cool.

Brief Thoughts on The Game Awards

Gaming awards shows face an intractable fundamental problem: voters can’t be expected to play all the nominated games, let alone all the potential nominees. The Oscars can’t get voters to watch five 2-hour films, there’s no chance a voter for gaming awards is going to spend hundreds of hours playing all the games.

As such there’s a low ceiling to how legitimate gaming awards can be. That’s why I don’t get too worked up over who wins or that the show is more trailers than awards.

Some argue that industry awards like GDC or DICE should be the prestige awards, as they’re voted on by game developers. But these awards are worse in that games media members play more games than game developers. Voting for the GDC Awards is a mix of popularity contest and reputation-based, where voters cast votes for the most impressive-seeming games, the few can’t-miss games they’ve played, or for the smaller mobile and indie games they actually had time for.

The GDC 2024 game awards list is just sad, with the same games nominated in every category, when those categories are supposed to reward specific elements of craft.

The best you can hope for from an individual site is a list of favorites, not a list of bests, and for a broader more definitive list the best you can do is close to what The Game Awards does: opening the voting up to a variety of gaming media outlets.

Second — this one is aimed mostly at other developers — much of the discourse around Expedition 33 is silly. I do roll my eyes a bit when the same game wins its 6th award, but it’s hard to argue that those awards are undeserved.

As an indie developer you’re supposed to adopt a “rising tide lifts all boats” mindset and cheer when another indie developer does well. But I forgive (and sometimes embody) resentment and jealousy. It’s ok to be petty, within reason, as long as you’re self-aware enough to not frame that resentment as rational objection. The Expedition 33 team didn’t cheat, the accolades were earned, and to their credit that team hasn’t adopted the too-common “finally a group of white people has fixed the wretched JRPG genre, a genre we last touched in 1994” attitude. In short, don’t hate them because they’re beautiful.

It rankles some that a $10+ million game from a publisher counts as an indie game, but the “indie” moniker is never going to make sense. Hollywood was ruled by five major studios that owned backlots and equipment, signed exclusive contracts, and faced no foreign competition — in that context it’s easy to identify a major publisher. There are hundreds of video game publishers from all over the world, with each separated from the next by 10% net revenue.

Subscribe now

The Rise of Indie Games

A huge boon to indie games is the room for lower-priced games. When games were $60 and physically-distributed a budget game might be $50 or $40. These days indie game developers have access to digital storefronts, powerful low-priced tools, and plenty of room beneath $70 and $80 titles. $5, $10, $20 and $40 are all viable price points on separate tiers. Budget games are a great deal when for one Mario Kart World you can get a $40 prestige indie game, a $20 game, a $10 game and take a flyer on two $5 titles. Games like Helldivers 2 also use this pricing headroom to great effect.

Another boon to indie games is the general lack of productivity and disarray in the AAA space.

A third factor is that AAA game studios have voluntarily ceded popular genres to indie games. There are few AAA Minecraft-likes, survival games, “friendslop” games or multiplayer horror games.

In a leaked internal email Phil Spencer wrote:

Over the past 5-7 years, the AAA publishers have tried to use production scale as their new moat. Very few companies can afford to spend the $200M an Activision or Take 2 spend to put a title like Call of Duty or Red Dead Redemption on the shelf.

AAA studios compete largely on production value, which means they shy away from gameplay-centric genres where players accept or embrace cheaper productions and jank.

When AAA companies do make a Minecraft-style or survival game they tend to turn the dial from mechanics, systems and openness towards polish — Dragon Quest Builders is more constrained than Minecraft, but with more plot, structure and better monster graphics, since that’s what major studios are good at and geared for.

It’s tempting to think that, instead of making one 300-person game, AAA studios should make ten 30-person games, or one-hundred 3-person ones. But major studios aren’t good at and aren’t built for that. In a 3-person team there’s no room for a dedicated cutscene artist or rigger, let alone an associate producer. Major studios avoid jack-of-all-trade types, when those are the types 3-person teams often need. Instead they’re full of specialists and processes that don’t scale down.

So it would be hard. But it also might not be wise.

Every 15-20 years there’s a breakout small-budget horror film: Halloween, The Blair Witch Project or Paranormal Activity. The current state of the game industry allows for a Blair Witch Project every month, along with a number of other games that fly more under-the-radar but still do well.

Which is great!

But this year there are 20,000 releases on Steam alone. With that in mind a Blair Witch a month is much less impressive. From that perspective the landscape is worse than when we’d get monthly “Indiepocalypse” think pieces.

Even among people with industry knowledge and access to numbers there’s tendency to see six breakout indie games and assume that indie games are doing numbers across the board — just make one and rake in the cash. But the numbers say that the average Steam release (depending somewhat on how you define “average”) is a flop.

It’s a hard market to predict. I used to get NPD numbers and in those days (though it was a very different market) AAA sales were predictable based on publisher, platforms, genre, etc. There was an occasional outlier — Prototype springs to mind as a game that sold way over my expectations. But most games did about what you’d expect once you were familiar with the data, and I assume that’s still true today to a lesser extent. (Especially lesser for multiplayer GAAS games)

Indie game sales are hard to predict and the indicators are less clear. There was a time when an indie game with high production value — a Darkest Dungeon — was easy to peg as a winner. It just looks well-made and expensive.

But then there were games like Slay the Spire. It doesn’t look bad — much of the animation is nice — but it doesn’t look great or particularly expensive. These days we have our Vampire Survivors and Megabonks and various “slops.” The best I can do with these titles is look at enough of them and their sales numbers to develop an instinct for which games will sell. Essentially acting as a human LLM, making predictions based on training data minus any explainable underlying reasoning. In Megabonk you can ride a skateboard so it’s kind of memey and it’s more real 3D than isometric, and the navigation looks fast and physics-based and that’s a thing people like…I guess? There’s lots of animation and stuff happening and a quick pace so that’s good — but much of it doesn’t look great but maybe that makes it more charming? It’s a lot more gut-instinct and subconscious understanding of trends than noting that Darkest Dungeon looks expensive.

Megabonk has 40,000 Steam reviews but I’d believe 4,000 reviews or even 400. Looking at bullet hell survivor-likes on Steam, Deadly Days: Roadtrip looks memey and systems-heavy but has 400 reviews. I’d believe 4,000 or even 40,000. Halls of Torment doesn’t look particularly better than or different from many other games but has 20,000 reviews. Of course it helps to actually play these games and see if they’re fun, especially when long-term progression design is key for the genre, but still. It’s unpredictable and the craft competency that AAA studios bring to the table matters less in that space.

I don’t know that AAA devs can compete in that arena or that they’d want to.

That said I think it would be wise for AAA games to compete on mechanics more, rather than ceding entire mechanics-heavy genres to lower budget games. To bring up Monster Hunter Wilds again, it’s an open world game with none of the systems, complexity or appeal that more modest open world games offer; an open world game where you auto-walk from one checkpoint to the next. Why not a Monster Hunter meets Valheim? Kill a monster, craft a tent, kill a monster, craft an anvil, forge your first bone weapon, build a cabin, etc. Especially when Capcom has two Monster Hunter branches to play with.

The Forest has 170,000 Steam reviews. The closest major-publisher game is…Grounded?

There are plenty of AAA games with a bevy of real mechanics, but there are plenty of games that lean heavily into dull conversation trees, playing back prefabricated sequences and canned animations, navigating your character from one map marker to the next, doing light procedures in lieu of solving puzzles, and performing rote hacking and pipe dream minigames. Your Metroid Prime 4s and Star Wars Outlaws could be well-served by nudging the dial towards mechanics.

Do I like The Hundred Line: Last Defense Academy?

A common refrain from Expedition 33 skeptics is that it’s not doing something much different from other JRPGs. The Hundred Line isn’t a JRPG but it has similar elements, and has been namechecked by the Expedition 33 team, so I’ll compare the two a bit.

Last Defense Academy is very tropey. Characters are defined by a single trait. There’s a guy who hates himself and every line is a variation on “I’m a miserable insect not fit to eat dirt.” 5% of the time these characters reveal more depth: the impossibly emo girl obsessed with Killing Games had a rough childhood, and the creepy torturous psychopath can be a loyal friend. But 95% of the time the cast is rabidly one-note.

Tonally the game is all over the place and very “anime.” In the midst of a scene where characters discuss how to prevent more of their close friends from dying, my character hears two girls mention they’re sleeping in the same room for safety and gets horny, then thinks “what am I doing I’ve got to focus!” What are you doing dude? (I pose that question to the writers as well)

It’s packed with repetitive dialog and plot restacking, with flashbacks to events from minutes earlier. (Some of that is due to the structure of the game, which allows you to travel to any point on divergent timelines at any time, necessitating reminders of the current context)

I put these things up front to make this point: I’d rather play a Shin Megami Tensei IV than an Expedition 33, but the claim that Expedition 33 is doing basically the same thing as anime-style games is weapons-grade copium. We can identify clear distinctions between 33 and these games without straying into “white people have fixed the JRPG” discourse.

Moving on.

The most interesting aspect of Last Defense Academy, from a design standpoint, is that it’s a choose-your-own-adventure game (with some other elements) with no real choices until you beat the game once then unlock alternate timelines. That first playthrough is probably 20+ hours, during which it’s a choose-your-own-adventure game with no choices. I don’t hate it but I find it absolutely bizarre.

There’s a second element of structure that I’ve wanted to write about since first playing Virtue’s Last Reward, a visual novel from some of the same creators as Last Defense Academy and the sequel to 999: Nine Hours, Nine Persons, Nine Doors. (For my money one of the best DS games and one of the best visual novels)

Some choose-your-own-adventure games present multiple paths assuming that most players will only play the game once or maybe twice — games like Dispatch and Detroit: Become Human. Those paths exist to provide interesting choices, different outcomes (or the illusion of different outcomes) and to spark conversation. 999 and previous visual novels, including the foundational “Sound Novel” games, expect the player to run through short scenarios many times. The player makes one set of choices, gets one ending, plays through again with different choices and a different ending, then stops when they’ve seen the best ending or simply had enough. These games feel like wandering through a possibility space.

This is a good video on Sound Novels (a term for early visual novels) and particularly on their structure. Some of those early games don’t have paths as much as they have parallel tracks the player can swap between to create a number of similar-but-distinct stories.

Virtue’s Last Reward is the first game I played that was neither repeatedly wandering through a possibility space nor a Dispatch-style one-and-done.

Virtue’s Last Reward is a map game.

It presents the player with a map of branching choices with the expectation that players will make every combination of choices and explore the entire map. Detroit: Become Human has a similar map, as do a number of Asian FMV games like Five Hearts Under One Roof. (Someone should write about these games, there are a lot of them and they’re pretty popular!) But I don’t think Quantic Dream expects players to explore the whole map.

Virtue’s Last Reward expects you to explore the whole map. That’s what the game is fundamentally about. (Last Defense Academy expects you to explore the map but I suspect the creators realize the game is very long and that most players won’t explore everything)

I don’t think I like this style of map game.

I like the idea of wandering through a possibility space of unknown complexity and boundaries. When the map is so explicit, and you’re expected to explore all of it, the game becomes about exhaustively searching a well-defined finite space.

If you choose to save one character over the other you’ll later backtrack and save the other character instead to explore that branch. You’ll eventually make every possible choice, so there’s no sense of forging a unique adventure or that the choices really matter.

I lost steam on Last Defense Academy when I completed a few distinct paths and the game became about backtracking to revisit choices. At that point I’m not making choices that I want to make, I’m making choices I’m compelled to make for the sake of filling out the map.

Specific to Last Defense Academy is that many of these branches relate little to the main plot, using the same cast of characters in a different story and different genre like horror or romance. The ending of the first route dangled many questions that the other routes barely address, and as I’m playing through slowly I’m forgetting what the main plot even is beyond the broadest details. In one branch someone is dead, in another they’re alive, in one they’re a friend, in another they’re a love interest or evil. It’s hard to stay invested or even follow when there’s a plot for every occasion. It doesn’t help that many of the alternate branches are very tropey.

I find it oddly compelling and I’m still playing it. But I feel the same way I did when I first played Virtue’s Last Reward: that this overarching structure is a mistake. I prefer both a) wandering an unknown space and b) creating my own one-off adventure, to filling out a map. Filling out the map is more rote task than adventure.

Do I like The Hundred Line: Last Defense Academy? I’m 50 hours in and I’m not sure — ask me this time next year.

I believe it serves a valuable function against omnipresent and overrated concepts, and also it’s just my nature. As such I planned on writing a pointed takedown of “fulfilling the player fantasy” as a game design concept.

Before I write pieces I sometimes pose related questions on social media as part of my research. I don’t have a high social media profile (not a complaint) so I don’t typically get many responses. This time, however, I got quite a few. I’m going to incorporate it throughout, but the high-level takeaway is that people are more skeptical of “player fantasies” than I’d assumed - the median response was along the lines of “it can be a useful tool but I’m well aware of various limitations.”

I’m still going to evaluate the concept of “player fantasies” in detail, but I don’t feel the need to push back as hard against dubious conventional wisdom, because I don’t know that the importance of “fulfilling the player fantasy” is the conventional wisdom. (As I edit this I realize I went pretty hard, but with a softer tone at least)

When I write I try to either present a new angle, or, arrogant as this sounds, to produce the best version of an existing argument.

I hope for some people this level of examination of “player fantasies” is new, and that for others it serves as a useful compilation or helps them crystallize existing thoughts.

I’ll end with some musings on the nature of public game design discourse.

Player Fantasies Defined

No matter how I define a “player fantasy” someone will quibble - one of the weaknesses of the term is that it’s vague and overlaps with other concepts. But my best attempt is that players come into a game, or at an aspect of the game, with a certain fantasy, and as a developer you need to deliver on that fantasy.

I already see potential disagreement: “player fantasy is about team alignment and deciding on a vision, not on delivering on a vision that players already hold.”

Is the player fantasy invented or discovered?

It depends on who you ask. Which already sets off my Spider-Sense. Maybe the best way to define what “fulfilling the player fantasy” is is with the clichéd example: “if you’re making a pirate game the player should feel like a pirate.”

The Best Case for Player Fantasy as a Useful Concept

There’s one place where I think player fantasy is a valuable concept with few caveats: licensed IP and established properties, especially those that focus on escapism, fantasy fulfillment and power fantasy in particular.

A likely buyer for an Incredible Hulk game is someone who likes the Hulk and wants to do Hulk things: jump high, get angry and smash. The line between “fantasy” and “expectation” can be blurry but in this case using “fantasy” is appropriate.

It’s also easy to point to examples of licensed game not fulfilling the player fantasy to their detriment. In the Suicide Squad game you use guns more than powers. I expect to use powers in a superhero game, but also I want to use powers - that’s the fantasy.

It’s unfair to judge 007 First Light but I’ve seen plenty of reservations that, by focusing on a young and inexperienced Bond, it’s not delivering the fantasy. Presumably Bond players want to drive fancy cars, employ cool gadgets, sleep with and then callously discard hot women, etc. Maybe the game will be a hit because it offers a refreshing change of pace, but if it’s not I expect “it’s not the Bond I wanted” to run rampant.

Games that focus on wish-fulfillment, like dating sims, also seem like good candidates for centering player fantasies.

Outside of licensed games and wish-fulfillment, games with a well-established lineage are also good fits for player fantasy as a useful concept. Though, having said that, it’s hard to think of too many examples. I suppose Halo players want the “fantasy” of being Master Chief but neither ODST nor Reach prominently feature Master Chief. The Assassin’s Creed games are all over the place in terms of setting and mechanics, and have largely dropped the connecting tissue that makes a game Assassin’s Creed. God of War reinvented itself by shaking up the “fantasy.”

You could say that the “player fantasy” of Mario Kart is you drive karts around tracks, but while you could say that you probably shouldn’t, as that’s a dressed-up way of saying “it’s a kart racing game.” Discussing Mario Kart that way doesn’t seem useful. Players expect some consistency and continuity in a franchise, and for a genre entry to deliver the basic premise of that genre, but that doesn’t strike me as “fantasy.”

The Weaknesses of Player Fantasy as a Useful Concept

Overlap with other terms and concepts

The usefulness of any jargon or concept is limited if it lacks a shared, well-understood definition. As research I’ve watched many videos and read various blog posts and papers; player fantasy isn’t well-defined and often overlaps with other, more useful terms.

Fantasy vs Expectation

“Fantasy” implies desired, not just expected, but in practice “fantasy” is often used where “expectation” also works. As one commenter put it, player fantasy is often used to mean “making sure the gameplay can cash the checks that the IP/art/narrative are writing.”

In cases where both “fantasy” and “expectation” work I prefer “expectation” for one important reason: expectations are something you help set, where a fantasy seemingly originates with the player.

As a practical example I’d look at how Riot Games uses the term “player fantasy.”

Riot uses this term a lot - doing research on 2XKO and running across it frequently is probably why I’m writing this now. League of Legends players, in turn, discuss “player fantasies” frequently, mimicking the language of Riot employees.

Riot has struggled with certain League of Legends characters and classes, in balancing the “player fantasy” vs other values like ability to counterplay. The fantasy of the Assassin class is that you can delete another player quickly, but in a multiplayer game a character who can easily delete another character is problematic.

Balancing the “assassin fantasy” vs other players having fun is a difficult problem, but also a self-inflicted one, that began by naming a class “Assassin.” DOTA2, another MOBA, doesn’t have an assassin class. (Even though it has three characters with “Assassin” in the name…it’s complicated)

Riot tried to rename Assassins to “Slayers.” They also renamed ADC - “Attack Damage Carry” - to “Marskman”, presumably for similar reasons: the “carry” role expects to carry the game, to “1v9”, which is an inherently unenjoyable fantasy to inflict on the other 9 players.

The character Yorick, often considered the most problematic LoL character, has similar issues with naming and expectations. He was advertised as half Gravedigger, a role that carries almost no gameplay implications (he hits people with a shovel), and half Necromancer, which implies a lot. It’s hard to deliver a satisfying Gravedigger / Necromancer hybrid as the pairing is so lopsided. You could call this “failing to deliver the fantasy”, but I’d call it improper expectation setting.

Fantasy vs Experience

“Experience”, as in “the desired player experience”, is another term that gets mixed up with “fantasy.” People seamlessly transition from “you should carefully consider what experience you’re offering” to “and that’s why you need to fulfill the player fantasy”, as if the former implies the latter.

As I mentioned up top, there’s a fundamental confusion over whether the fantasy is the experience you’re offering or the pre-existing notion you’re trying to fulfill.

This video is an example of this sort of fundamental confusion.

When it comes to game design, one of the most essential skills is the ability to comprehend the specific fantasy that you’re trying to create.

If I want the player to really feel like Batman that’s a fantasy I’m fulfilling. If I’m making Sonic the Hedgehog I don’t think I’m fulfilling a fantasy - I don’t think many people fantasize about being a hedgehog or even know what a hedgehog fantasy could mean. In particular I don’t think hedgehogs are blue, have attitude, or run fast. The actual hedgehog fantasy, if it exists, is probably more like the game Shelter - “experience the wild as a badger sheltering your cubs from harm.”

The video goes on to use The Witcher as a positive example of creating a specific fantasy. But The Witcher isn’t a fantasy CD Projekt RED created, it’s a licensed game.

The video claims that, due to CD Projekt RED’s care, players really “feel like the titular role”, and praises the one-liners Geralt delivers as being true to the character. I find this very difficult to follow. I assume that most Witcher players haven’t read the books. How can players feel like the Witcher when this is their first exposure to him? For all they know one-liners are dramatically out-of-character!

This is like claiming “when you play Prototype you really feel like Alex Mercer.” I have no idea who that is!

The term “fantasy” is more confusing than useful here. We can just say “Geralt is a compelling, well-written character and his one-liners lend him a distinct personality” or “the game does a good job of translating the book characters.”

This video uses “theme”, “fantasy”, “emotive fantasy”, “feeling like [the character]”, atmosphere, setting, and various elements of craft in such loose and overlapping ways that I don’t know what to think. Even though the video is fundamentally about fantasy I think you could make a better video by stripping the word out entirely.

Not every video needs to be straightforward advice, but this video feels like advice, but at the end I have no idea what that advice is.

I would compare this to Sakurai videos, which present a workmanlike and practical approach to game development. Not that intellectualizing topics is wrong, but sometimes it’s more obfuscating than enlightening.

The Pirate Problem

In my GameDeveloper piece “When Theme and Mechanics Collide” I took on ludonarrative dissonance. (I republished it on substack with that section edited out as it was confusing)

I had two main issues with the term:

1. Most uses of “ludonarrative dissonance” could be replaced by a phrase like “tonal incoherence” with no loss.

2. In practice people used the term mostly to criticize Uncharted and Bioshock, and the Bioshock criticism wasn’t particularly strong. “Nathan Drake is affable in cutscenes but murders people without a second thought in gameplay” is a fair criticism but it’s very narrow.

I would make similar claims about “player fantasy.” Much of the time it could be replaced with more well-defined terms. And in practice it’s much narrower than it first appears.

I typically make outlines first, and my outline included this:

People have “fantasies” (aka expectations) when you give them something familiar: generic pirate, generic wizard, generic ninja, etc. (I would note that almost every example of player fantasy is pirate, wizard or ninja!)

After I posted to social media asking for thoughts on player fantasy, the first two replies I read that mentioned specific roles both used pirate. The first example I found online, referencing Jesse Schell, also used pirate as the example of “making a player feel like” the thing.

I would half-jokingly suggest that “fulfilling the player fantasy” is less a broad design philosophy and more a strategy for making pirate games. Less jokingly I would point out that it’s harder than it first appears to come up with a good fill-in-the-blank entry for “really makes you feel like ____”

Pirate, sure. Ninja. Ok. Wizard. Sure. Go on.

Samurai? We already said Ninja but I’ll allow it. Cleric? Fighter? Now we’re just naming other D&D classes.

Space Marine? The space marines in Halo, Aliens and Warhammer 40k are all very different. Bounty Hunter? Like a Samus or a Boba Fett? Jedi, Smuggler, etc - those are all Star Wars (licensed IP), and to me a weakness of Star Wars is that it keeps recycling “iconic” character types - the Han Solo type and the Boba Fett type.

I’m not sure that even the easy answers are slam dunks. We expect ninjas to look the part, do flips, and use ninja weapons. But a ninja game could be a sneaking / stealth game like Mark of the Ninja or a balls-to-the-wall action game. In Shinobi 3 you ride a surfboard - is that part of the ninja fantasy? Maybe that’s a subgenre, see also: Surf Ninjas.

Galford from Samurai Shodown is a San Francisco sherriff’s son, sailor and ninja, who fights with a dog companion. (He’s really hitting all the archetypes) Earthquake is a fat-guy ninja from Texas. Is that a player fantasy? Maybe that’s also a subgenre; see also: Beverly Hills Ninja. The Ninja Warriors are robot ninjas, and in Ninja Warriors Again one has stretchy arms and another is a wrestler.

Ninja is a broader concept than it first appears. Importantly, you could consider stealth and assassination as part of the core fantasy, or ignore them completely.

I suspect the reason people use “makes you feel like a pirate” so often is that they’ve heard other people say it and it’s one of the few archetypes that really works, if you understand we mean a TV / movie / cartoon pirate. Cowboy - again a movie / TV kind - also works well. But surprisingly few things do. It’s tempting to slot in “spy”, for example, but as a guiding principle “we want the player to feel like a spy” is too broad and varied. The Mission Impossible spies wore disguises they’d rip off as part of the big reveal, while James Bond rarely disguises himself and famously introduces himself with his full name. I had a friend who joined the NSA after college and I don’t think her job was to put on a leather catsuit and sneak through air vents.

Encouraging Genericism

If you want the player to feel like a pirate you’re probably not offering an Our Flag Means Death about a nobleman who chooses the pirate life. Or even Pirates of Dark Water. One responder on social media shared that on Destiny 2 they used player fantasies as a way to quickly get the team on the same page for Seasons of Plunder. If your goal is quick team alignment then using “let’s do pirates” only works if you use a shared understanding of pirates, and ignore the team members who possess oddly specific pirate factoids.

Here’s some advice I found online regarding player fantasies and authoring characters:

I find a useful exercise is to suppress your initial impulse to heavily define a protagonist’s character, and strip it back to the minimum with the following questions in mind:

What features of this character directly support the player fantasy? How much definition is needed for the player to feel they are this type of character? The messy detective’s office and the 1930s decor, perhaps, but do they need to know they were once a longshoreman, owned a dog called Frederick, and their mother’s name was Martha? Do those details function as useful character-defining quirks, or are they distracting fluff or even worse do they act as limitations?”

To be fair I’m stripping out some context here. The post is about players inhabiting blank-slate characters. Adding details can make a player feel disconnected from the character if those details don’t align with the player, and it’s arguably harder to inhabit a character when that character has very specific traits. Characters like Lara Croft, Samus and Doomguy have been weakened with the addition of ill-fitting backstory. (Though the “ill-fitting” part is key there)

But it’s hard not to read this as “make it more generic!” Does Columbo need to mention his wife?

Here’s another quote from this post:

You want players to feel they’re playing Indiana Jones, or D’Artagnan, or Sam Spade — or a character which feels very close to those templates.

You can only make a character who feels like Sam Spade because someone else invented Sam Spade. Don’t you want to invent the next Sam Spade?

I’m not sure that many players like “a Sam Spade type character” anyway. Over time it’s easy for popular characters and archetypes to become genericized in the public mind. There are hundreds of Alien knockoffs (in my Tubi watch history alone!1) - most knockoffs feature a snarling, chitinous, gooey creature - the Xenomorph has become generic. But Alien inspired those knockoffs because the specifics made it memorable and interesting.

When I revisit old works that I recall as generic I’m often surprised by how specific they are. Every fantasy novel has Orcs and Goblins and magic rings, but The Lord of the Rings doesn’t read like vanilla fantasy. We may remember It as a story about an evil clown, but the evil clown parts aren’t even the most striking parts of the story.

When you offer up a “Han Solo type character” or an “Indiana Jones type character” you’re probably offering up a genericized version, without the specifics that made those character stand out in the first place.

You can end up with someone like Jet Brody from Fracture. Here’s a wikipedia blurb:

1UP.com criticized Fracture for taking multiple features from other shooters such as Halo, Gears of War and Resistance: Fall of Man. The site described character Jet Brody as “perhaps the single most derivative character in the modern age of gaming, a direct combination of the heroes from Resistance/Mass Effect/Gears of War/Too Human. If the narrative weren’t so staunchly heavy-handed, I’d put my money on Brody being a literal parody of shooter-genre character tropes.” Game Informer continued this theme by ranking Brody first on its list of “the top 10 worst character names.”

There are many blank slate video game characters, but it’s hard to think of too many successful “type of guy” characters. Mario doesn’t have an elaborate backstory or complex personality but he’s not a pre-existing type of guy. Sonic isn’t a type of guy - instead he defined a type of guy, the animal mascot character with attitude, which begot characters like Bubsy and Awesome Possum.

Samus is a bounty hunter but she’s not a Boba Fett-like. I guess Lara Croft is half James Bond and half Indiana Jones, but that’s a mix of two guys at least, and Lara Croft isn’t a guy, something the original games and advertising made very clear. (wink wink)

Joel from The Last of Us is a dad-from-The-Road type of guy, but I don’t think that’s really a type of guy. The best I can come up with is Nathan Drake, who kinda is Indiana-Jones-at-home. But even then Drake has specific character traits and relationships - if he’s an Indiana Jones type he’s at least a specific instance of that type.

Here are the results of a most iconic video game characters poll.2 Very few of these are “a Sam Spade type.” These are, if anything, characters that established a mold rather than popped out of one.

It’s Hard to Escape Power Fantasy

Player fantasies don’t have to be power fantasies but it’s hard to escape the gravity of power fantasy, as the fantasy angle is so straightforward. Having cool superpowers is a fantasy in a way that struggling to drive a truck through the snow is not.

This video does an admirable job of trying to make “player fantasy” cover all sorts of fantasies:

1. Power

2. Aspirational (Guitar Hero)

3. My best life (Animal Crossing)

4. Being there (Breath of the Wild)

5. Transgressive (GTA)

6. “What would I do?” (Historical shooters, State of Decay)

7. Emotional (Ico)

This is a good attempt but I don’t find it convincing.

“I want to be a super hero” and “I want to be a rock star” are straightforward fantasies. “I want to inhabit a cool space” strikes me as an extremely vague fantasy. “What would I do?” is an interesting thought experiment or narrative hook but I don’t know that I’d term it a fantasy, nor do I see how it applies to WW2 shooters. “What would I do?” clearly does apply to choose-your-own-adventure games, but “these games deliver a what-would-I-do player fantasy” is circuitous vs “it’s a compelling narrative that unlike a book or movie responds to your choices.” If you’re making a choose-your-own adventure game offering “what would I do” choices is baked into the genre - that players make choices explains the appeal but provides little guidance to developers.

Maybe TCG Card Shop Simulator and other mundane-task games offer a best-life fantasy - you’re your own boss, you can order the products and arrange your store how you like, you presumably make enough money to pay rent, etc. “Best life” may describe the appeal of the game but again as a guiding principle it seems weak. And how does it apply to Viscera Cleanup Detail?

“The player should feel like a pirate”, as cliché as it is, is a reasonable guiding principle - include treasure maps, don’t include heat-ray vision. Animal Crossing is a “my best life” fantasy but it begins with your character in massive debt.

You can describe Ico or its appeal as delivering an emotional fantasy but “wanting to experience emotions” is an awfully vague fantasy. You’re better off describing Ico by the specific emotions and atmosphere it evokes: loneliness, connection, trust, dependence, wistfulness, loss, vulnerability, etc. (We won’t even broach thorny questions like how applicable is “fantasy” to sadness, loss, horror, etc?)

If I state, as a guiding principle, that Ico seeks to invoke a sense of wistfulness, otherworldliness or magical realism, implementation details like the use of heavy bloom and unintelligible language make sense - using that more specific language provides guidance that “fantasy” doesn’t.

The use of “fantasy” here doesn’t strike me as terribly appropriate or useful.

Where do Player Fantasies Come From? On Psychographic Profiles

Let’s assume that you fulfill a fantasy rather than create it - you’re making a pirate game and want players to “feel like a pirate” based on their existing idea of pirates.

How do you know what that idea is?

The most well-known gaming psychographic profile is Timmy, Johnny, Spike from Magic: The Gathering. Those profiles seem reasonable as you can readily observe those types of players - some players really do just like big creatures. I find attempts to create psychographic profiles across video games as a whole much less successful. According to the MDA Framework paper there are 8 kinds of fun and 8 types of fun-enjoyers - why 8, and why those 8?

Another use of psychographic profiles is to invent a target customer: the Wii Fit mom who wants to stay in shape while engaging the family. That sounds like a plausible customer, but Wii Fit style games have fallen off hard. Maybe what we thought of as Wii Fit mom was actually a fad-following mom.

You could invent the Warhammer 40K: Kinect Dance mom - the mom who loves Warhammer and wants a family-friendly activity to do with her kids. The doesn’t sound as plausible as the Wii Fit mom but she could exist - ultimately we’re making people up, which limits how useful this exercise is.

When you say “we’re not delivering on the ninja player fantasy”, if you aren’t doing player surveys and market research, what you’re really saying is “I made up a guy who doesn’t like what we’re doing with ninjas.” When you say “our ninja game needs stealth to fulfill the player fantasy” you’re saying “I made up a guy who agrees with me that ninja games need stealth.”

When you discuss the player fantasy like this you’re discussing your assumption of the player fantasy - you’re making up a guy and assigning them a viewpoint. (That, most likely, suspiciously mirrors your own)

Saying “players expect stealth in a ninja game” is often just saying “I expect stealth in a ninja game”, with added obfuscation and an appeal to an invented authority. If you think a ninja game needs stealth argue it on the merits, rather than making up a guy who agrees with you.

Dispatch is a superhero game where you do mostly mundane tasks. From a player fantasy angle that seems wrong - one could imagine a producer saying “we need open-world 3D action parts where you punch bad guys in your mech suit - it’s a superhero game!” But maybe a more mundane simulator and workplace comedy is itself a compelling “fantasy.” Or maybe you don’t have to deliver the superhero fantasy because you’re delivering on the fantasy of dating hot superheroines. It’s easy to make up a guy who accuses Dispatch of failing to deliver their desired fantasy, and easy to make up a guy who praises it for delivering their fantasy. If you’re creative enough you can make up a guy for any occasion!

It Just Doesn’t Ring True

Little of “player fantasy” talk relates to how I personally approach games.

My interest in genres like action or SRPG is mostly mechanical. There are certain settings, character types and atmospheres I enjoy but I’m not looking to inhabit a specific role, adventure in a particular type of world, or fulfill a set fantasy. I enjoy the “being there” fantasy of the green grass and blue skies of Breath of the Wild, but I also enjoy dropping into a Helldivers 2 hell world. When I play a Platinum-style action game my “fantasy” is that the game is a good one of those kinds of games. I’d like an SRPG to have nice graphics and music and interesting system mechanics, a variety of characters with cool abilities, the right level of difficulty and required strategy, etc etc - but that’s hardly a “fantasy.”

I play both fighting games and MOBAs, which are often discussed via player fantasy rhetoric. But I choose characters in those games based on a combination of their specific mechanics, look and theme, “fun factor”, etc. I don’t go into a MOBA thinking “I want to play as a wizard who controls the battlefield” or into a fighting game as “I want to play as the rushdown character.”

I’m not looking to play as a wizard, and if you give me a wizard I still don’t expect something specific like huge damage, damage over time or crowd control. If the wizard is tricky and teleports around that’s fine, if they throw huge fireballs that’s also fine. Honestly if they hit guys with a big stick that’s also fine if it’s fun. In DOTA2 I play a lot of Ogre Magi, who is ostensibly a wizard but I play him by walking up to people and bonking them with his club. Despite being a mage, Ogre Magi is canonically a fucking moron, which is, to me, a main part of his appeal. I don’t know that a real dumb wizard that hits people with a club fulfills a player fantasy but it works for me.

The player fantasy rhetoric feels foreign to me - it’s not how I approach games. And, personal approach aside, it doesn’t seem widely applicable.

What is the fantasy of Expedition 33? To live in a world where all your friends die on schedule, well before their time? To lead an expedition that (early spoilers) immediately goes horribly wrong? To encounter and kill an evil mime then put on his bloody garments, presumably because you’ve suffered a full psychotic break?

If anything, as I noted in a previous blog, a major strength of Expedition 33 is that it’s not familiar, predictable or archetypal.

What is the player fantasy of Blue Prince? Ball x Pit? Is the fantasy of Peak to try to climb mountains but really suck at it? Of course Peak offers some genre expectations - it’s janky and has proximity chat. But you have to stretch very far to come up with a “player fantasy” for Peak that isn’t incredibly broad like “the fantasy of having fun with friends.”

What is the fantasy of Sonic the Hedgehog? To live out life as a hedgehog? Again I don’t think hedgehogs run fast. Maybe the fantasy there is to let players go fast, but “going fast” is an incredibly broad fantasy, and Sonic games feature sections of slow platforming.

Maybe I’m being too literal with “fantasy.” In a quote below someone suggests that “fantasy” may mean “the goal of making the act of playing a role feel like what the player imagines it should feel like.” But I still don’t see how that works for Sonic the Hedgehog, or Expedition 33, or even for The Witcher. (As again, I have no idea who that is)

Is This Useful?

The ultimate question I’d pose about “player fantasies” is how often is this the most useful rhetoric?

I’d refer to George Orwells’s Six Rules for Writers, or just “use the best word for the job.” Player fantasy doesn’t strike me as being the best tool for the job very often.

To take a more cynical angle, if you’re making a pirate game you should make the player really feel like a pirate. Sure.

But what if I’m not making a pirate game?

Player fantasy rhetoric isn’t literally limited to pirate games, but efforts to expand much beyond that are messy. My read is that some people are enamored with “fantasy” as a sort of theory-of-everything, and try to justify using it in a variety of contexts where it doesn’t work, or at least doesn’t work as well as other framings.

“When they play Ball x Pit they should really feel like The Cogitator, one of the playable characters in Ball x Pit!” Why would we talk this way?

Social Media Quotes

The dominant sentiment, in my eyes, was that player fantasy framing is a useful but limited tool, particularly in that it trades in the familiar. Multiple people observed that it might be a better tool for positioning, marketing and original concept creation than for day-to-day game design.

Here are some specific quotes - I’m not attributing them because I don’t think it was clear that I was soliciting them for a writeup.

I think it’s useful as a navigational point, but being chained to the idea of what players expect leaves less space to surprise or challenge them.

I’m leery of the term. It evokes this idea that a designer is just ‘discovering’ what the player wanted all along, but I think most of the time designers are creating new experiences the players never quite knew they wanted.

my bio on the other place was once “creating elaborate excuses for inherently enjoyable activities”, so, like, I am definitely sympathetic to the idea that a thing games can do is provide a way to do a thing a player wanted to do before they saw the game

This one is strongly worded (fine by me):

I think fulfilling player fantasies is a bad and infantilizing way of referring to a more serious goal which is basically ludonarrative consonance in more academic terms. basically the goal of making the act of playing a role feel like what the player imagines it should feel like. This doesn’t have to be anyone’s actual fantasy and it’s really easy to get wires crossed with it and things like power fantasy and wish fulfillment which can be but are not necessarily related

The idea that a player fantasy doesn’t have to be anyone’s actual fantasy seems true, but speaks to the confusing nature of the term. “Infantilizing” is pointed but I can see it, especially where player fantasy intersects with power fantasy.

How we Talk About These Kinds of Topics

Before writing this my assumption was that most people held the idea of “player fantasy” in high esteem. Of course social media feedback isn’t scientific or representative, but I think I overestimated how much cachet the concept has.

Which prompts me to ask: why?

When I encounter the topic it’s often in blog posts that lean towards advice, thought leadership or sharing the secrets of game development success. Sometimes it’s in videos, papers or presentations.

It’s rarely in venues that invite or even allow discussion. I think the reason I overestimated the esteem people hold for “player fantasies” was, for lack of a better term, homefield advantage. There are many spaces where people sell their ideas but precious few where they discuss them.

If you watch and disagree with a live GDC presentation it’s difficult to stand up during the limited question time and challenge the speaker. It can come off as gauche and confrontational. It’s difficult extemporaneous speaking, where the person objecting has to quickly think up and summarize off-the-cuff arguments, against someone who has prepared their positions in advance. And I’m not sure that I’d want the end of GDC presentations to frequently feature arguments over the preceding material.

If you watch the presentation in video form later, or read a PDF, there’s no venue for discussion at all.

I suspect some would point to the The Polaris Game Design retreat as a positive example of game design discussion - designers attend this retreat and work in teams to produce papers. But there’s no mechanism for discussion of those papers once they’re released - they’re mostly fire-and-forget.

When Clint Hocking wrote his take on ludonarrative dissonance in Bioshock many of the commenters differed and poked holes. When I wrote for Gamasutra people sometimes disagreed or offered up alternate views in comments, and I disagreed in comments as well. (Maybe too often!)

GameDeveloper, the Gamasutra successor, doesn’t allow comments. That Hocking blog post, were it written today, might be on a forum that doesn’t allow comments, or exist in hacked-up form on Twitter, or simply not exist at all.

I don’t know what the solution is. As GDC rebrands to the Festival of Gaming one thing it promises is more debate. I can see this working, but I can also see it going horribly wrong, where industry “personalities” retread tired topics. Debate entails pitting people against each other, which isn’t earnest discussion. What I want is people willing to disagree and discuss those differences, not people compelled to disagree via format.

The social media comments I received were surprisingly good. Useful, thoughtful and thought-provoking. I wish we could have these sorts of discussions more often!

So often public-facing gaming discussion is repetitive discourse, cycling through the same tired topics. And often that discussion is characterized by culture-war positioning and developers circling the wagons against “gamers.” Difficulty, accessibility and localization discussions are all so thoroughly dominated by culture war concerns that I don’t bother to read them anymore.

My sense is that developers discuss game design and development among their co-workers and immediate social circle, and come to social media to polemicize and fight. The first half of that makes sense but the second half is unfortunate and avoidable.

I have many saved PDFs or bookmarked Youtube videos or even saved HTML pages (literally right-click “Save as…” on the page) on development but so many of them are old. And when I recall meaningful game development discussions those are also old. Some of that is a function of I’m older and have less need for PDFs on rendering techniques or for discussions on ludonarrative dissonance. But it’s hard to avoid the sense that older discussions featured more vitality, more back and forth, and less sales pitches and preaching to the choir. Much of the praise I found for “player fantasies” was Linked-In speak (a good portion literally on Linked In) - someone trying to sell me on the concept, on themselves as a thought leader, or on consulting services.

Maybe that’s a broad complaint about the arc of social media. But my fantasy, if you will, is to encounter useful, thoughtful, diverse viewpoints and discussions more often - “discourse” without the need for quotes.

But I often cover control issues. I’ve touched on controls in fighting games a few times, including in Quarter Circle Forward: Quite Cool and Fun, which is entirely about them.

I’ve played games with control schemes similar to 2XKO, and I’ve played fighting games since the 90s. (Actually I enjoyed the oft-disparaged Street Fighter 1, a late 80s game)

I’ve read reports and player impressions, watched videos and match footage. But 2XKO is invite-only and I haven’t been invited.

I think the game has issues, particularly in the controls. I don’t see it breaking through to a larger audience - it’s going to be a game for the usual fighting game players and a smattering of League players, most of whom will drop it quickly.

I’m going to explain why I think that. I’m not big on making predictions, so I prefer to think of this as expounding on concerns. But I’ll take my lumps if this ends up being way off.

2XKO’s “Lack of Vision”

2XKO’s mechanics significantly changed from “alpha lab 1” to “alpha lab 2.” Those changes weren’t well-received, at points lacked obvious motivation, and introduced new problems, leading to accusations that the game’s creators “lack vision.”

“Lacks vision” is vague; I would avoid that phrase. But the game lacks a central mechanical conceit, particularly one that relates to the League of Legends brand.

These are, as far as I can tell, 2XKO’s major selling points:

1. Has League of Legends branding

2. Is free to play

3. Is “accessible” via a simplified control scheme (we’ll get into this)

These dovetail nicely - together they make 2XKO enticing to try out. But they aren’t a mechanical vision and don’t answer “why would I stick with this?”

X-Men: Children of the Atom, Capcom’s first superhero fighting game, immediately stood out as bigger, bolder and more dynamic than previous games. Instead of throwing small fireballs characters shot out 6-hit beams. They could run, high jump, and use screen-filling super moves. Children of the Atom is not Street Fighter with X-Men characters swapped in, it’s mechanically a superhero fighting game.

Killer Instinct has long, easy-to-perform combos, but those combos can be “broken”, so players remain active on defense. That doesn’t have much to do with the plot of the game - Killer Instinct is about a fighting tournament hosted by a corporation, that a werewolf and a skeleton attend for some reason. Thematically it offers generic 90s edge - the character Orchid has a fatality where she shows you her boobs then you die of a heart attack. (If the other player is also Orchid they get jealous and pout - laudable attention to detail, if questionable gender politics) The game has a mechanical hook - the combo system, a Mortal Kombat lite attitude, and even a technical hook - it’s an Ultra 64 game with pre-rendered sprites and a dynamic camera. But none of those hooks are related.

Then there’s Teenage Mutant Ninja Turtles: Tournament Fighters, a highly competent Street Fighter 2 knockoff. It has no novel mechanical elements and the gameplay isn’t related to the Turtles. You don’t collect ooze or eat pizza to power up your moves, it’s Street Fighter with turtle graphics.

That’s the sort of game Riot is making with 2XKO. It has a grab bag of mechanics borrowed from other games, in hopes of offering a best-of-breed combination. Nothing about League of Legends informs the gameplay of 2XKO other than that it stars League characters. It doesn’t have, for example, a system where characters gain XP and level up over the course of a match, they way they do in the MOBA. Nor does it have a mechanic like combo-breakers, novel but separate from the theme.

2XKO is like many games created by a generation that grew up on games. Those creators often have a strong grasp on genre conventions, are aware of and avoid mistakes of the past. Someone creating a 16-bit style JRPG in 2025 probably isn’t going to give characters the same painfully slow walking speed they had in 1993. But creators with rich knowledge of past efforts can be constrained by that knowledge, creating endless numbers of Metroidvanias, throwback RPGs and point-and-click adventure games content to sand off the rough edges from prior eras.

2XKO aspires to pick and present the best mechanics from other fighting games, with a League of Legends skin on top.

Does that qualify as a vision?

I find unique gimmicks overrated - plenty of great games lack a unique selling point. But the lack of any mechanical differentiator does seem risky.1

Casual-Friendly Means (and only means) Removing Motion Inputs

2XKO has been billed as casual-friendly from the start. This is the first text on the official website:

2XKO is a free-to-play fighting game with explosive 2v2 gameplay and fast, intuitive controls.

(Side note: I don’t think “fast controls” means anything)

This is a long and complicated post, but here’s the short version: every time Riot says 2XKO is easy to learn, intuitive, welcoming, etc, what they mean is that it doesn’t use special move motions like the traditional quarter-circle-forward fireball. And that’s all that they mean, mechanically speaking.2

And here’s the rub: maybe removing special move motions can make a game easier to learn and play. But when you remove those motions you have to replace them with something, and the replacement scheme that 2XKO employs is harder to learn than motion inputs and compares poorly to other “modern” control schemes. I can’t think of any fighting game control scheme, traditional or otherwise, that’s less intuitive. (I was tempted to say Primal Rage but I don’t think that’s right)

What Casual Fighting Game Players Want

Potential fighting game players want the things players always want: nice presentation, good sound, etc.

Beyond that baseline, fighting game (and MOBA) players want characters that appeal in both form and function. This is a traditional strength of Street Fighter, which offers characters from an electrified green werewolf to an Indian guy with stretchy limbs to a sumo wrestler to a sexy lady, with wildly different playstyles. (This is a weakness of City of the Wolves, in which most of the characters play similarly)

Casual players want games that can plausibly be learned without hours of extra-game research. The genre has almost uniformly terrible tutorials, which present a rapid-fire litany of techniques with no chance to absorb lessons, and go zero-to-sixty from “press up to jump” to 30 hit combos.

This is a video, not an interactive tutorial, but it’s typical of how fighting game tutorials are structured. I can understand this video by relating it to mechanics from existing games: “super assemble” is DHC from the Marvel vs Capcom and “assemble smash” is Street Fighter 6’s Drive Impact combined with Street Fighter 3’s Universal Overhead. If you haven’t played those games though…good luck.

The biggest barrier to casual adoption of fighting games is one that plagues arcade games and their derivatives: lack of content. A home racing game can feature dozens of tracks, hundreds of cars, a career mode and open world events, while an authentic arcade racer is typically a handful of tracks and cars with no persistent game modes. This problem has existed since the NES days, when ports like Rygar and Strider dramatically differed from arcade versions in part to expand content.

Fighting games are incredibly fun to play locally with friends. Online play is an acquired taste. What many casual players want out of fighting games is meatier single-player modes that are more about unlocks and progression and less about one-off fights; your Soul Calibur 2 Weapon Master mode or the various RPG-ish modes in more recent Mortal Kombats. But significant single-player content is rare. There’s a huge gap in the content offered by NBA 2K vs NBA Jam, but Street Fighter 5 offers basically the same content that Street Fighter Alpha 2 did in arcades in 1996. (It’s difficult for fighting games to offer meaningful unlocks, in part for balance reasons and in part because those unlocks are often sold as DLC)

What Fighting Game Developers Think Casual Fighting Game Players Want

The removal of motion inputs. The end.

There’s a near pervasive belief among fighting game makers that motion inputs erect a steep barrier to entry and that removing them - while making no other gameplay or content nods to casual players - makes a game casual-friendly.

2XKO is launching with 10 characters, a low number especially in a tag fighter. 2XKO doesn’t have many weird guys - your Dhalsims or Twelves - but also lacks your obvious standard-issue Ryu / Ken / Kyo / Hanzo / Ryo / Haohmaru / Sub Zero / Jago / Leonardo. So it doesn’t score well in terms of characters players can vibe with.

In a 2024 Kinda Funny interview a Riot dev said tutorials weren’t something they were thinking about yet, so I assume they won’t be a quantum leap forward for the genre. What exists currently is a pretty classic fighting game tutorial. (As a reminder, that’s bad)

2XKO includes a complex array of mechanics and systems, and as a tag fighter it’s harder to learn and to visually digest than a 1-on-1 game. I expect the game to present the same learning woes as other fighting games.

Riot has been tight-lipped about content - it’s not clear if the game will include even a basic story mode, let alone something approaching the career modes found in sports games. It has missions like “use a super move in a match 5 times” and a battle pass, so it does have unlocks, but doesn’t appear to offer significant single player modes outside of staples like training and versus.

2XKO has been billed as a casual-friendly fighting game, and the free-to-start model is financially friendly. But in terms of gameplay and content I can’t see anything that makes it casual-friendly other than the removal of motion inputs. It doesn’t address any of the common pain points that have dogged fighting games for decades.

The Age Old Motion Inputs Debate That We’ll Sidestep

25 years ago motion inputs were the standard, so theoretical musing was the only way to explore alternatives. These days plenty of games ship with alternate control schemes, so there’s no need for heady intellectual debate. Instead we can look at the particulars.

What I’ve come to belatedly realize is that the motion input debate pits a specific well-understood control scheme - Street Fighter’s control scheme - against an unspecified theoretical replacement.

It’s reasonable to discuss the pros and cons of motion inputs. But a comparison requires two subjects. The pertinent question, for 2XKO or any game with a “modern” control scheme, is not “is removing motion inputs good?” It’s “is our replacement control scheme better than what we’re replacing?”

For 2XKO the answer is “no.”

Controls Deep Dive

I’m not going to, but I could end now with “the prosecution rests.”

There’s a lot here - a lot of moves, terminology and mechanics. I don’t see how anyone could look at this and think it represents an easy-to-learn fighting game for casual players.

There’s dashing, running, high and low parries, “chain dashing”, “retreating guard”, tagging, “handshake tagging.” There are 4 types of assists - regular, charged, super and push. There are no concessions here in terms of simplifying mechanics for ease of use and understanding.

What’s a little less obvious is how fussy and difficult-to-operate these control are. There’s double taps, holding buttons, and pressing multiple buttons at once - the sort of control complexity that removing motion inputs is supposed to alleviate.

Here are the controls on a game pad.

This is an eight button game. Some of these buttons are macros: the parry button is a shortcut for low + high and dash is a shortcut for medium + high. So maybe it’s a 6 button game? (This is a picture from the official website and it has 8 buttons, don’t blame me!) But on a pad only the bottom-left and top-right pairs of face buttons are natural to press with your thumb. To press low + medium you have to reorient your thumb, or have a thick thumb tip. Pressing the left and right face buttons together, without accidentally hitting another button, is even harder. So while the macro buttons are technically optional some of them are practically necessary. This is similar to City of the Wolves, which is technically a 5-button game but is real pain to play without additional macros.

Blue Sky Special Move Inputs

3D action games frequently include special moves and don’t use motion inputs since “quarter circle forward” is confusing in 3D space, and doesn’t work well on analog sticks. Those games usually use a bumper as a modifier key to the face buttons - hold the right bumper and press one of four face buttons to do one of four special moves.

Using a bumper as a modifier key is my first instinct. If you want more than four moves you can use the other bumper, both at once, etc. If that’s still not enough…well at that point I don’t think you’re making a casual-friendly fighting game.

I’m surprised this scheme isn’t used more often. I’m not sure why it isn’t; maybe because it’s natural on a gamepad but doesn’t map well to sticks. (Street Fighter 6 does use roughly this setup for the “assist” button)

Riot is in a tough spot there. Casual gamers mostly own pads, but fighting game fanatics often use sticks or leverless controllers. Nobody complains that Smash Brothers doesn’t work well on sticks, but fighting game players expect a game that looks like a traditional fighting game to work on them. League of Legends is a PC game so Riot wants to accommodate keyboard players as well.

For whatever reason let’s say modifier keys are out. Another obvious approach is combining a special move button with different directions. This gives you 5 different special moves if you restrict to cardinal directions, and 9 if you include diagonals.

This is how Marvel Tokon works and how Street Fighter 6 works - and Smash Brothers as well, basically.

2XKO Special Move Inputs

2XKO offers two different special move buttons, “S1” and “S2”, that combine with directions for a total of 18 available inputs.

The decades-old motion input debate has centered on arguments like “a 360 motion is harder than a fireball motion, so the 360 can be more powerful.” I sidestepped these arguments earlier because 2XKO’s biggest issue isn’t one that comes up often in these debates.

The biggest problem with 2XKO’s control scheme is that it employs no systemic logic or organizational principles. It can’t be learned as much as memorized.

Systemically Organized Special Moves

Street Fighter 2 has a systemic, and therefore learnable, control scheme. The top row of buttons is punches and the bottom kicks - makes sense as your legs are beneath your arms. Attacks on the left are weaker, faster, and have less range, and attacks on the right are stronger, slower and have more range.

In the Street Fighter series there’s a clear hierarchy of moves. Pressing an attack button gives you a normal attack. Pressing a direction and a button sometimes gives you a “command normal” - these moves tend to have extra properties like built-in movement. (Chun-Li’s aerial head stomp is an early example) Then you have special moves, which require a motion plus a button, do block damage, and can cancel normal moves.

The special moves follow the same rules as the basic button layout: moves that use your upper body are done with punch, lower-body moves use kick. Using a left button produces a move lesser in some sense - weak punch produces a slow fireball and heavy punch a fast fireball.

Consistent use establishes patterns, both in games and across them. Quarter-circle-forward + punch throws a projectile in Street Fighter 2, in Marvel vs Capcom, and in SNK’s King of Fighters. A move that’s functionally a Dragon Punch but uses the feet is a Dragon Punch motion plus kick - that’s how you do Gen and Adon’s uppercuts in Street Fighter Alpha 2, or Joe Higashi’s Tiger Kick.

Some of these rules are clear enough to recognize on a conscious level. Some of them are less clear but still follow a subconsciously perceptible pattern. The hurricane kick uses a backwards fireball motion. Moves that look like hurricane kicks, where characters spin around a vertical axis, tend to use backwards motions. Fei Long’s anti-air special, a rising spinning kick, is done with a reverse dragon punch motion plus kick. That’s a combination of three separate rules (or guidelines at least): anti-air moves use dragon punch motions, kicks use kick buttons, and spinning moves use backwards motions.

This system breaks down on the edges - when Blanka rolls up into a ball that’s not clearly an upper or lower body move. But for the most part this is systemic and learnable, or at least falls into repeated patterns.

It’s enough for players to grasp these patterns on an intuitive level. I recently played the demo for Cairn, a mountain-climbing game in which you control all four limbs without explicitly selecting which one you’re controlling. After a few minutes it felt natural, even though I can’t describe exactly how it works. (I guess you control whichever limb is supporting the least weight)

Special Moves as a Series of One-Offs

Now let’s take 2XKO’s control scheme. There are three attacks - light, medium and heavy. This is already less descriptive than Street Fighter in that “weak attack” conveys less than “weak kick.” Tekken and Mortal Kombat also use more descriptive button names - left/right punch/kick and high/low punch/kick.

2XKO uses two special move buttons, “S1” and “S2”. There’s no indication, based on name, why a move would go on one button or the other.

Ideally we’d name these buttons to differentiate their use. Maybe one button is punch and one kick, or “offensive special” and “defensive special.” DNF Duel uses two special attack buttons, “skill” and “MP skill” - MP skills use MP and are more special. (The “skill” button is arguably more of a third attack button than a second special move button)

Naming the buttons enforces organization, but it’s also limiting, or at least demands creative thinking. What if you name the buttons “offense” and “defense” but can only think of one defensive move for Jinx? Or name them “punch” and “kick” but one character uses guns and one swings a giant axe?

If we’re giving up on meaningful names there are still ways to enforce order. We can create broad rules like “anti-air moves are always down + S2” or “if a character has a projectile, it’s always S1, and if a character has a second projectile, it’s always towards + S1.” Marvel Tokon, DNF Duel and Smash Brothers all do this to some degree - Up B in Smash is your vertical recovery move.

Characters could also follow individual logic - “Jinx has two guns, with each assigned to one button” or “Blitzcrank’s S1 is long range moves, and S2 is short range moves.”

Unfortunately if 2XKO employs such logic it’s not clearly evident. Braum has a move where he enters a defensive stance, it’s back + S2. Vi also has a defensive stance, but it’s down + S1. Darius and Blitzcrank both have long range grabs that pull the opponent in, and those are both S1, so that’s something. Illaoi also has a long range grab, it’s down/back + S1. A Riot developer reportedly suggested that Ekko displays some internal logic - his S1 controls his Timewinder projectile and his S2 does…other stuff. But towards + S1 is a rushing attack that doesn’t use Timewinder.

Illaoi follows some internal logic in that S2 plants tentacles and S1 uses them. But some of her S1 moves slide forward and hit the enemy, and some of her S2 moves do as well. So while there’s a clear functional difference between the buttons there isn’t a clear visual one.

Jinx has three guns. Two of the guns are mapped to S2 and one to S1. S1 also does non-gun things. Pressing any down direction and S2 gives you a move that follows a global “upward moves are done with down directions” pattern. But on S1 shooting upwards is back + S1, not down. You can also get an upward-angled move by pressing forward + S1 then holding the button. So “upward moves are done with down directions” is a loose pattern at best.

The forward + S1 move, when held, looks nearly identical to Ekko’s Timewinder Toss - in both cases the characters use one hand to throw a lingering projectile into the air. But Ekko’s is done with down + S1, not forward.

This lack of systemic rules extends to how individual moves work. Here’s an example of a Blitzcrank move:

Forward translating spinning strike. Hits both in front and behind Blitzcrank. Each hit charges a bit of Steam for Blitzcrank. The startup of Spinning Turbine can be special cancelled into either Prompt Disposal or Trash Compactor while retaining the forward momentum from Spinning Turbine.

This description uses the terminology “special cancel.” In Street Fighter if a move can be “special cancelled” it can be cancelled into any special move - you can do crouching medium kick → fireball, crouching medium kick → dragon punch, or crouching medium kick → hurricane kick. In Street Fighter the “special cancel” is a systemic concept.

This move can be “special cancelled” but only into two specials, not all of them. Correction: Trash Compactor is an Ultimate move. So this move can be cancelled into one other special move or into an Ultimate move. It can’t be cancelled into other special or super moves.

In Street Fighter every move that can be cancelled into a special can be cancelled into a super, but the converse isn’t always true - some moves can be cancelled only by supers. This is another systemic rule: super moves are a ladder rung above special moves so they can do anything a special move can do and more.

This Blitzcrank move can be “special cancelled” but just by one special move. It can’t be super-cancelled, even though super moves exist a tier above special moves. But it can be Ultimate-cancelled.

This is a series of one-off rules phrased to look like they exhibit systemic logic.

Subscribe now

Special Moves as a Series of Fussy One-Offs

The knock against motion inputs is that they’re difficult. But there are multiple axes of difficulty. Traditional motion inputs may be hard to perform but learning what to perform isn’t difficult.

In 2XKO learning what to perform means memorizing all the moves. Pressing a direction and a button isn’t hard, but remembering which direction and button to press isn’t easy when they don’t follow patterns. But I have to grant that pressing a direction and a button is, operationally, easier than pressing three directions in succession followed by a button.

A significant number of the special moves aren’t simply a direction and a button though. There are many “hold ok” moves and other additional input complexities.

Here is Blitzcrank’s S1, Rocket Grab:

Long range hit grab that pulls the opponent in on hit and on block. When Blitzcrank has a bar of Steam, they will shock the opponent on hit. Hold S1 to charge Steam. Pressing forward and S1 will Rocket Grab even when held. If Blitzcrank successfully hits the opponent on the ground, this can be cancelled into Power Fist.

As best I understand this, pressing S1 does Rocket Grab. Holding S1 charges up Steam, which is a totally different move mapped to the same button. Pressing forward and S1 does a Rocket Grab even if you hold the button.

Hmm.

I assume in some cases players were trying to do a Rocket Grab but holding the button down for too long and getting the Steam Charge instead, and this alternate version avoids that issue. But in that case why not make Forward + S1 Rocket Grab and S1 Steam Charge, or S1 Rocket Grab and back + S1 Steam Charge? Why do we have 2 moves mapped across 3 inputs?

Here’s Ekko’s Chronostrike:

Traveling strike. Creates an Afterimage. Hold for Volatile Afterimage. Hold forward to ensure Ekko does not rewind.

This is another move where holding the button gives you a variation and holding forward gives you another variation that’s apparently a sort of impossible-to-screw-up version. Presumably some players were rewinding when they didn’t want to, so this move has an optional additional input that prevents those future inputs being taking effect? Honestly at this point I’m a bit lost.

This is arguably practical - the developers observe that players repeatedly flub a handful of moves and add workarounds. But I think you have to concede that something has gone wrong when your game that advertises an easy-to-operate, intuitive control scheme adds alternate impossible-to-flub versions of moves because players are getting tripped up. Maybe this is less error-prone once you master it but it’s definitely less understandable.

Control Wrap Up

In game development these two things are often both true:

• Something is obviously flawed and just not-it, and it feels like there has to be a better way

• Nobody on the team can think of a better way

In my experience this is especially common with control schemes.

The 2XKO control scheme to me is just not it. It’s a game with 6 to 8 buttons but only 3 normal attacks. It requires simultaneous presses of buttons that aren’t aligned. It uses analog triggers for moves that are done often and need precise timing. (As opposed to say Street Fighter 6 using trigger for Drive Impact) We have two special move buttons that we can’t think of names for, and we also struggle to come up with any rules for their use, across characters or even just for one character in isolation.

I made this chart (before Blitzcrank was revealed) of which direction-plus-button combinations characters use. Knowing nothing else about the control scheme this should give one pause. Glancing at some of these characters there’s an apparent “no diagonals” rule - but that’s not a rule. Only three of the eight characters use a lot of both buttons. Yasuo uses neutral + S2 and back + S2 (the neutral is implied in the graphic ) - that’s it for S2. He doesn’t use back + S1, so Windwall, his back + S2, could go there. That leaves neutral + S2 as the only move S2 move. But if you’re willing to use diagonals, and we are for some characters, just stick it on a diagonal. Or on up. That move is a projectile so we could also, radical suggestion, make it a fireball motion. “How do we make Yasuo throw a fireball?” feels like a problem solved in 1991. But in 2025 there’s effectively a dedicated fireball button.

Our intuitive control scheme includes holding buttons to get move variations, and sometimes holding a button produces an entirely different move. For some moves holding a direction gives you an alternate, less-error-prone version of the same move, I guess?

There are many contextual follow-ups, so the controls are highly state-dependent. Those follow-ups also lack clear patterns.

And there’s the more subjective hand-feel problem: command normals, special moves, tags and assists are all done via direction + button - they’re all effectively command normals, without textural variation.

On top of all that, a significant swath of players actively resent the removal of motion inputs. They think it results in a “baby game”, removes the satisfaction of left and right hands working together, adds extra buttons, overloads the right hand, doesn’t work well on stick, etc etc. A sizable number of players assume that a game that only offers “modern controls” isn’t for them. (I’m one of those players) The developers of Invincible originally said it wouldn’t include motion inputs then acquiesced following outcry.

This is the sentiment you’re up against when you replace motion inputs with a questionable substitute. Some players find this alternate control scheme difficult or unintuitive, and others just don’t like it. I didn’t cherry pick these comments, they aren’t outliers. On any forum discussing 2XKO or fighting game control schemes they’re common.

I’ve recently touched on the belief that developers need to build systems before evaluating them. That’s sometimes true, and it’s often true of controls. But controls are tricky enough to pose a trap: what if you implement a control scheme and it’s merely ok? Control schemes take refinement to go from good to great, so maybe the control scheme, while flawed, “shows promise” - not bad enough to give up on, but not good enough to be confident in. I suspect this is a trap 2XKO has fallen into, and a trap many developers fall into when they’re ideologically enamored with a feature. Personally I would downvote even trying out a system of two special move buttons where we can’t name the buttons or describe what they do - to me that indicates more thought is required at the conceptual stage.

The End of the Experiment

We keep hearing that games without motion inputs are more attractive to players but those games keep coming out and not attracting more players. 2XKO is based on a huge cross-media IP, is made by one of the richest gaming companies, and is a solidly traditional fighting game other than that it lacks motion inputs. It’s as good a test subject as can exist in real world conditions.

If the player base ends up being fighting game stalwarts and a handful of League players can we admit that removing motion inputs has limited value?

In 1999 the idea that motion inputs were limiting the appeal of fighting games was an interesting and untested hypothesis. In 2025 it’s well-tested and it just doesn’t appear true. I wrote about this in 2018, when I noted that Fantasy Strike, a fighting game without motion inputs, was one of crowdfunding platform Fig’s biggest flops. Nothing has changed since then - there’s been no significant success story. At best you can argue that Street Fighter 6’s modern controls provide a nice onboarding option for casual players. But Street Fighter 6 makes motion inputs optional rather than removing them and offers a huge (for the genre) amount of single-player content.

Maybe motion inputs do put a ceiling on appeal versus some ideal alternative. Maybe Hollow Knight’s bug theme caps its appeal - that’s probably true! But that’s not a valuable observation unless you can suggest a better alternative.

When we remove motion inputs we have to replace them with something. In 2XKO that something is left wanting.

In that post I wrote:

Video game business advice is typically paired with “of course you have to make a great game”, but with precious little focus on how to do that.

The standout element of Expedition 33 is not an innovative business strategy, choice of genre, or the result of market analysis and research. It’s that they made a great game.

A game that I wouldn’t make.

Of course I wouldn’t make Expedition 33 for many reasons: it wouldn’t occur to me, I’m busy with other things, it would take resources I lack. I’m not French. But the pertinent reason is that on paper Expedition 33 didn’t strike me as a great game. It’s the kind of design I shy away from.

This post is about what I learned from Expedition 33; more exactly, what I learned from being wrong about it.

“The Real Time Elements in Expedition 33 are Bad Design” - Me Six Months Ago

The reveal of Expedition 33’s battle system, in which players can completely negate damage via well-timed button presses, worried me. My gut reaction was that it probably wouldn’t work - that it would be unbalanced and dangerously destabilizing.

Before Expedition 33 released I’d been batting around my own idea for a tactics / strategy RPG with timed button presses for active defense or extra damage. (Think Fire Emblem x Project X Zone)

I hadn’t put much thought into the extra damage element - “adjust to taste” seemed good enough. But I’d put serious thought into how damage reduction should work. Here’s how I first approached it:

If a well-timed button press reduces damage by 10% the feature may as well not exist. 10% damage reduction feels irrelevant, and it doesn’t make sense to spend development time on an irrelevant feature. So low numbers like 10% are out.

If you’re going to reduce damage by 30% or 40% you may as well do 50% - it’s a nice round number and easy to understand.

Reductions like 60% or 70% seem odd for the same reason 30% and 40% do - might as well just say 50%. Extensive playtesting could reveal that, due to the particulars of your damage formula, encounter length and rate, consumables, etc, the sweet spot is a number like 33% or 62%, but I’d never start with those numbers.

My conclusion is that there are only three damage reduction numbers worth considering up front: 50%, 100% and almost 100%.

50% sounds very safe and “fair.” A 50% reduction is noticeable and worth doing, but it’s not overwhelmingly powerful - it would be a mechanic without being the mechanic. 99% is basically 100% but in my mind less exploitable - if combat grinds on forever the player has to eventually heal or switch strategies or do something.

100% reduction strikes me as dangerous. If the player is great at parrying and parries negate all damage then nothing else in the combat system matters. Defensive buffs are pointless when you’re already taking zero damage. Exploiting the elemental weaknesses of enemies that can’t hurt you is purely a time-saver.

At the highest level my concern is this: what if the player fights the final boss and can parry all of their attacks, so the boss poses no threat? That sounds lame as hell.

This is how I’d thought about my own (theoretical) game, but figured this applied to Expedition 33 as well. That it would end up being a unenjoyable degenerate case: a game that offers many systems, only one of which matters.

The Perils of Personality

So anyway that was wrong.

My initial thoughts had some validity - the real-time elements rendering other battle mechanics meaningless was surely a potential problem.

For some players the game is too easy. On the flip side, because active defense is so strong enemy damage is high to compensate, which means if you’re bad at dodging and parrying the game can be too hard. One of the most downloaded mods relaxes the timing on dodges and parries.

I’ve seen plenty of players point out these sorts of problems, so they’re real to some extent. But some percentage of players will complain about anything and the game was well-received. I think I have to concede that my fears were largely unfounded. If anything, some skill interactions and bugs could trivialize combat far more than parrying and dodging could.

I also think I have to concede that my initial negative assessment of the battle system largely comes down to that’s the type of person I am.

In terms of game design (and maybe just life in general) I lean reserved - I’m not terribly spontaneous or adventurous. In life that means coffee places quickly learn my order since I get the same thing every day. In game development that means I’m the type of person willing to say that an idea sounds bad and we probably shouldn’t try it; which is, I think, a valuable type of person to have, but an entire company composed of me would be miserable.

I’m wary of imbalances and exploits. Exploits can ruin a multiplayer game for everyone. In single player games they’re less important, but players can still ruin a game for themselves by relying on an exploit or oversight, then blame the developer. It’s tempting to retort with “you control the buttons you press”, but some players will choose to press buttons that break the game, then bemoan the lack of challenge.

I lean towards mechanically subtle and sophisticated rather than big and bold. In Dark Souls you can follow a parry with a context-sensitive, hugely damaging attack. In Sekiro a parry depletes an enemy’s poise, which you have to do multiple times before you can deathblow them. There’s an even more restrained version of that though: do we need the deathblow at all?

In Street Fighter 3 there’s no poise meter and no deathblow mechanic. A parry negates damage and recovers quickly, which allows you to damage the opponent, but it’s up to you to maximize the damage based on the scenario.

One could imagine a version of Sekiro where breaking the enemy’s poise makes them briefly reel and it’s up to the player to determine and perform the maximum damage conversion - to open the biggest can of whup-ass. That’s a more subdued version of a death blow mechanic - less binary, more skill-intensive, more technical and finnicky, but less splashy than a canned animation where you slice a guy’s head in half.

Expedition 33’s total damage negation on parries and dodges is a dramatic feature - very splashy, in-your-face, strictly binary. The kind of feature I’m not naturally drawn to.

A Battle System Re-Analysis

When Expedition 33 released to acclaim I challenged myself: instead of poking holes in the system can I “steelman” it - can I make the argument that it’s a good system, and poke holes in my previous hole-poking?

One of my main objections was that if players can parry consistently then other aspects of the game barely matter. If the game is horribly broken and players can mash the parry button to parry every attack then yes, the battle system is fatally flawed, but that would be a bug, not a design problem.

We should design this battle system such that most players can’t parry enemies consistently right off the bat. Maybe the first time they fight a new enemy type they parry 15% of the time, then the next encounter is 20%, and so on.

If the threat of failing parries is real - and it should be for most players if we balance the game properly - then defensive buffs matter, and using offensive options to finish fights faster also matters. This is especially true due to how consumables work. In older Final Fantasy games you can load up on hundreds of potions and heal to full at the end of a battle - the damage doesn’t stick. In Expedition 33 consumables have a low fixed count and only refill at checkpoints (Estus Flask style), so using potions is attrition. Reducing damage by finishing fights faster or via defensive buffs lowers that attrition.

With practice maybe players can parry an enemy 100% of the time. But is that a problem? In most RPGs you can grind for experience and gold to make fights easier - repeatedly fighting an enemy and gaining proficiency is also “grinding” but it’s the player increasing their prowess rather than the characters. If anything that seems preferable - instead of accumulating meaningless video game stats the player improves a real life skill. Not a particularly valuable or transferrable real life skill, but a real life skill nonetheless.

Earlier I wrote:

What if the player fights the final boss and they can parry all of their moves, so the boss poses no threat? That sounds lame as hell.

That could be lame as hell. But also it could be cool as hell. If the boss is throwing out long attack strings and multi-hitting moves and the player, through practice, can deflect all of them: good for them. “What if the player gets good at the video game?” That seems ok.

Most single-player games can be trivialized with enough practice and skill.

Subscribe now

On Important and Loud Features

At this point my objections to the system have fallen away. But I haven’t made an affirmative case - total damage negation might not be problematic, but why might it be preferable to say 50% damage reduction?

I think the answer is that, while I’m not a big believer in “unique selling points”, games need to stand out. They need to answer the question “how is this better or different than other games?”, or at the most basic level, “why buy this?”

Real-time blocks and parries negating all damage is a standout feature. Were the damage reduction more conservative the real-time elements would be a tertiary feature - the same tertiary feature that’s appeared in games for at least 30 years.

I often think about features in terms of tier and amplitude: how important a feature is and how loud that feature is.

Many games feature inventory management but in Resident Evil games inventory management is a central feature - a tier 1 feature.

Loud features are highly visible. They tend to be important features but I think these are different concepts. Street Fighter 6’s Drive Impact is the most visible, splashy, back of the box feature, but for skilled players it’s less integral than Drive Rush. Early Guilty Gear and Tekken entries featured instant-kill moves and ten-strings, respectively, both of which were back-of-the-box features but mostly irrelevant after a few hours of play.

Loud features are the features that set a game apart at first glance, and important features set it apart in the long run.

In Expedition 33 the parries and dodges are both important and loud. They appear to be a defining element of the game, and they are.

Active defense in turn-based games is hardly new. The Last Remnant has it, Sea of Stars has it. Super Mario RPG has it. What’s new about Expedition 33 is that well-timed button presses are the central battle mechanic, not a secondary one.

In Sea of Stars a well-timed block reduces damage (halves maybe?) but doesn’t negate it entirely. Well-timed attacks and blocks don’t look and feel impactful, backed by only modest sound and VFX work, which communicates a lesser importance, regardless of numbers. (The notable exception here, at least in the demo, is the boomerang weapon, which will satisfyingly bounce many times with well-timed presses)

The Sea of Stars battle system has competing elements with more visual and seemingly mechanical weight. You can interrupt planned enemy-attacks by breaking locks floating above them, and can collect energy orbs to power up your attacks.

Well-timed button presses in Sea of Stars aren’t the system, they’re just a system. I’ve seen suggestions that Expedition 33 would be a better game - more well-balanced - if dodges and parries were less powerful. Even if that’s true - which is down to taste - the game would have less to hang its hat on in that case.

High-amplitude features are memorable and serve to differentiate. Sandfall Interactive has copped to doing some light trolling with some features, like the platforming, which is arguably memorable in a bad way. But players can not only tolerate but enjoy memorably bad sections, as long as those sections don’t stretch on for too long. See the poison swamp levels in FromSoft games.1 The infamous Turbo Tunnel in Battletoads is remembered fondly, and the game probably would be as well if everything past that point weren’t equally maddening; the last 90% of Battletoads is all metaphorical poison swamp.

Raising the amplitude of features can mean higher highs and lower lows. That’s preferable to consistent blandness, and in some cases low lows don’t particularly detract.

In terms of practical production, the more central and obvious a features is the more a developer is incentivized to make it good. The well-timed button presses in Sea of Stars aren’t terribly important and they aren’t particularly satisfying - I don’t think that’s a coincidence. In Expedition 33 they’re more satisfying, and they need to be, as they’re the central mechanic. The puzzles in Silent Hill: Shattered Memories are easy to the point of barely being puzzles, but that’s ok because it’s not a puzzle game, just a game with light puzzle elements. The puzzles in The Witness have to be good for the game to work, because it’s a puzzle game.

In game development it’s much easier to polish visible features. In my experience it’s common to add clarifying UI, sound and visual effects to a system2 - to make the system and how it works more obvious - and immediately spot improvements or problems. Sometimes after making the effects of a system more obvious you realize the system didn’t work at all! In many older RPGs certain stats don’t work as intended or don’t do anything.

What I Learned From Expedition 33

What I learned from Expedition 33 is not “RPGs are more engaging with real-time elements” - I’m not a fan of blanket game design rules or turning food for thought into lessons learned.

I did gain appreciation for “crank up the waveform on features.” Raising the volume on features isn’t a new tool, but it’s one I might reach for quicker. I probably underestimate to what degree loud features can make games more fun and more marketable. I can recall a specific example of this: I was talking to a Starbucks employee about a game I was working on where you play as various animals. He asked about the craziest creature you could be, prodding me to say something like “giant fire-breathing dragon”, but my answer was more like “turtle.” He wanted to hear a splashy selling point, while I was concerned with more mundane matters like that a dragon might be too big to fit through a door.

So there’s that.

But the most important thing I learned was about myself.

There are good reasons to be wary of Expedition 33’s combat design. But my wariness was less rational than gut reaction - I just enjoy spotting potential problems and downsides more than I enjoy spotting potential opportunities and upsides. My biggest takeaway is that I could stand to be less like that.

Much of the criticism I see of Expedition 33, especially of the battle system, comes from those who are proudly old-school, conservative about game design, or who are prone to pessimistic over-intellectualization. Less thoughtful critique than “bah humbug.”

To those people I’d suggest the exercise I tried: instead of dwelling on the flaws of a system try focusing on the upsides. Delve into how it could go right, in terms of attracting players or providing a standout experience, rather than how it could go wrong. Instead of brainstorming ways it could fail brainstorm ways around those potential failures. You don’t have to Stepford Wife yourself into blithely accepting any feature proposal, but at least try on that way of thinking.

More broadly, game developers can be stuck in their ways for not-entirely-rational reasons. The designer who rejects special move motions out of hand, or rejects “modern” controls out of hand. The developer opposed to any randomness or asymmetry in games, not because it’s less fun or strategic but for dogged ideological reasons they can only explain via essay. The guy who accepts anything as long as it satisfies the “rule of cool”, or the guy who rejects fun ideas because “we don’t use the word ‘fun’ here.” Having a point of view is valuable, but that can cross over into stubbornly clinging to a shortcoming.

These learnings can be convincing in the moment, but are often fleeting or contradictory. Clair Obscur benefits from compactness, but Baldur’s Gate 3 benefits from being stuffed with content. These are both valid observations but taken together present no neat lesson.

In this post I’m going to cover the difficulty of developing meaningful takeaways from the success of Expedition 33, something I admittedly enjoy doing. But while I don’t think there’s anything to learn from the success of games like Flappy Bird I do see legitimate takeaways from Expedition 33, so I’ll propose a few of those. In the follow-up post, tentatively titled What I Learned from Expedition 33, I’ll discuss a very specific way that it adjusted my attitude on game design.

What We Can Supposedly Learn from Expedition 33

I’ll start with a post from the popular GameDiscoverCo newsletter, Clair Obscur & Drive Beyond Horizons: subgenre + execution = hit!

This post identifies three reasons Expedition 33 succeeded:

Clair Obscur picked a counter-intuitive & undersupplied subgenre: there’s been few AA+ turn-based RPGs made by Western devs, except a handful of retro titles like Sea Of Stars.
…

Its ‘look and feel’ are closer to the Western mainstream than Asian competitors

…

The game ‘threaded the needle’ for global appeal, esp. in China

That last point - that it did well across the globe - is less a reason for success and more the definition of success: the game did well because it sold well everywhere.

There’s no explanation for why it did well globally or especially well in China. Was the Chinese voice acting great? (Does it even have Chinese voice acting?) Did the game have a Chinese publisher that did…something? Was it designed with Chinese audiences in mind, or do the themes and gameplay naturally appeal to Chinese players for some reason?

On the point about Western look and feel: I do think there’s something to the art style of Expedition 33 that I’ll get to later, but I don’t think it’s as simple as that “anime aesthetic” is a turn-off. Genshin Impact has made 6 billion dollars!

On that last point: “Clair Obscur picked a counter-intuitive & undersupplied subgenre.”

Is JRPG an undersupplied subgenre?

The claim here seems to be that JRPG is a genre (though surely JRPG is a subgenre of RPG) and that the subgenre is “AA+ turn-based RPGs made by Western devs [excluding retro titles].” This seems like a cheat to me - when you get that specific nearly every subgenre is thin on content. There are dozens of super hero movies but only three about a guy who’s ant-sized - is the ant-sized hero an undersupplied subgenre?

At the highest level, “subgenre + execution = hit” strikes me as meaningless. Every game can be placed into a subgenre, which leaves us with “execution = hit.” Which…sure.

The section on Expedition 33 ends on this:

under-developed subgenre + great narrative & art direction + lean, focused team + smart gameplay execution = big hit.

Many top-sellers aren’t in under-developed subgenres. Assassin's Creed Shadows, Oblivion Remastered, MLB The Show 25, Call of Duty: Black Ops 6, etc etc.

Maybe the lesson here is that great execution in an under-developed subgenre equals big hit, and great execution in a well-developed genre also equals big hit. In that case the lesson is “make a great game.”

It may seem like I’m picking on this post, but this is what most business analysis of Expedition 33 reads like, and more broadly what game industry business analysis usually reads like.

The Big Problem with Business Lessons

Video game business advice is typically paired with “of course you have to make a great game”, but with precious little focus on how to do that. This type of advice is aimed at people who don’t have a craftsman’s eye and don’t sweat the details.

As a budding director Sinners might broaden your horizons about the use of music in film. It could introduce you to a new shot composition, or something technical like a lens type. These aren’t “lessons” exactly but they’re elements of craft you can carry forward in your own work.

Film execs, on the other hand, want to learn lessons like “vampires are in!” Sony thought the lesson of Venom’s success was “people like Spider-Man anti-heroes” so they made Kraven and Morbius. (A vampire movie and a Spider-Man anti-hero, how could it go wrong?)

These high-level strategy lessons are the easiest to summarize in executive-friendly bullet points, easiest to digest, easiest to act on, and don’t require much product knowledge. “You need great narrative and great art direction and smart gameplay” is hard advice to act on if you don’t know much about video games or what makes them good, but “battle passes are in” is something anyone can understand and act on.

As such it’s tempting to treat “and you have to make a great game” as a lesser concern. But the video game audience is discerning. That’s been one of Microsoft’s constant woes. Games like Halo: Infinite, Redfall, Starfield and Crackdown 3 are on-trend and check the right genre / feature boxes, but just aren’t that good. “The people crave Left 4 Dead games” might be true but the 17th best Left 4 Dead-style game is a non-starter.

The Wisdom in Ignoring Existing Lessons

One of the most common pieces of business advice is that you should make a game in a lucrative genre - either a genre that’s currently big on Steam (simulation, for example), or a genre that appears underserved. I don’t think JRPG fits either of these criteria. The writeup I cited above alluded to this by calling the sub-genre “counter-intuitive.”

Another common wisdom is that making a game at the intersection of genres is a bad idea, because the potential player-base will be the intersection of those genre’s fans rather than the union. JRPG fans might be turned off by Souls-like elements, and Souls-enjoyers might balk at JRPG elements. By mixing genres you also run the risk of not delivering what players expect, or by setting confusing expectations.

This advice sounds like a warped version of the useful observation that making a game that offers disparate play styles is tough to pull off. When these games succeed it’s often because they’re more than the sum of their parts - neither the combat nor town-building in Actraiser are best-of-breed but the game works as a whole. It’s easy to see where this can go wrong, when a game simply is the sum of its parts, or is dragged down by the lesser elements.

Alternating genres isn’t the same as merging genres - Expedition 33 doesn’t alternate between JRPG and Souls-like, it offers one primary mode with elements of both. And the notion that it’s wise to avoid offering disparate genres seems outdated to me anyway, given the number of games in form of “farming plus dungeon crawling”, “base building plus exploration”, etc. Grand Theft Auto is alternately a shooting game, a driving game and various mini-games. (Though they’re all presented as part of the same 3D world rather than as independent modes, I suppose)

A final bit of conventional wisdom that Expedition 33 bucks is that it doesn’t have a “unique selling point.” In retrospect you can claim that the French nature gives it a certain je ne sais quoi, but 2 years ago “the unique selling point is that it’s French” would have drawn laughs. It has real-time combat elements but so do Mario RPG, Sea of Stars, Lost Odyssey, etc. You can get into the mechanical weeds - instead of a stack of items you have a fixed number of refillable consumables - but that’s smart design, not a back-of-the-box feature. There’s no game that has its exact combination of art style, music, story, etc, but that’s true of all games. That’s not a unique selling point in the same vein as Cuphead’s presentation.

Lessons Learned as Biases Confirmed

I’ve read much analysis of why Kamala Harris lost to Donald Trump and in nearly every case the purported takeaways are the result of motivated reasoning to reinforce existing world views.

Some Democratic pundits and consultants blame “the groups” for Harris’ loss. They don’t blame Harris’ policies because they agree with those policies, or Harris’ advisors because those advisors are part of their social circle.

Critics of Gen Z blame Gen Z voters. (Or non-voters) Those who dislike boomers blame boomers. Those invested in identity issues fixate on how Democrats slid with Hispanics or Asians, or blame white male voters for lurching right. “Manosphere” trackers blame the manosphere and the lack of a “liberal Joe Rogan”, those invested in Palestinian rights blame Biden’s Gaza policy. Centrists believe the campaign should have appealed more to centrists, when appealing to centrist voters was the main strategy.

Few of these people learn lessons that contradict their existing beliefs.

People who think the game industry is guilty of gatekeeping believe that “hire people off of Reddit” is a lesson to learn from Expedition 33. Those looking for funding for their own “counter-intuitive” genre entry see Expedition 33 as evidence that we should fund those games. The guy who rants about games designed by committee credits the success to a singular vision, while the one who rants about bloated teams praises Sandfall’s catlike agility. The person who dislikes the typical JRPG credits the game with breaking away from that formula, and the person who likes the JRPG credits the game with borrowing elements of that formula. Expedition 33 has been held up as both an example of western developers besting Japanese ones, and by comparison as an exemplar of the slide of western developers.

Of all this sort of analysis (both political and video game) I’d ask the following: is it a lesson learned if there was no change in belief, and the lesson is simply “I was right all along”?

A notion reinforced still has value, but most people are eager to reinforce their existing notions, so that value is limited.

Subscribe now

On The Incredible Productivity of 30 (plus or minus 382) People

Perhaps the biggest topic of Expedition 33 discussion is the push and pull over the claim that Expedition 33 was made by (a core team of) 30 people.

On one extreme some argue that since the core team is 30 people that the game was “made” by 30 people. On the other extreme are those who think the four Oboe players in the orchestra count as game developers and “made” the game as much as anyone else.

To me this is like debating whether craft services workers (the people who set up lunch spreads on film sets) count as “filmmakers.” It’s a purely definitional argument that film folks don’t debate because a) they all know what craft services does, and b) films use real metrics like budget and shooting days, not head count.

Video game analysis often uses poor metrics because those are the only metrics available. We use concurrent user counts or Twitch viewers in lieu of sales because publishers are stingy with sales data. Business analysis is focused on Steam because Steam metrics are easier to come by.

Head count is a metric we use to assess developers because we aren’t privy to real metrics like budget.

Ubisoft owns motion capture studios, so in an Ubisoft game the people who provide motion capture assistance are part of the team. Sandfall Interactive rented1 a motion capture studio, so the workers at that studio don’t count as “core developers.” But in both cases the games use motion capture that came out of a budget.

Paying a “core team member” takes money. So does paying a contractor or an outsourcer to do the same work, buying assets, renting out motion capture studios and so on.

In his video on Expedition 33, Youtuber Skillup asked if the budget was under $10 million and didn’t receive an answer. Clearly the budget wasn’t Spider-Man 2’s $300 million but “how did they do this on that budget?” isn’t a good question when we don’t know that budget.

Game Developers Can Be Very Productive

Given the amount of doom and gloom in the game industry (which I contributed to in my last blog post) it’s easy to adopt a broad “everyone just sucks at making games now” attitude. But game developers can be very productive in the right environment.

Revenge of the Savage Planet2 was also made by about 30 people. It got a bit buried (in part because it released two weeks after Expedition 33) but it’s well-made and arguably bigger in scope.

I’m not sure how many people work at Bokeh Studios, who recently released Slitterhead, but it seems same ballpark at least. When I fired up the demo I expected to be met with jank, but it’s a polished experience from the start.

Neither of these games captured the public like Expedition 33, but Sandfall isn’t lapping these teams in terms of raw productivity. Clearly Sandfall overachieved but I don’t think they’ve discovered a double XP hack.

Instead I’d propose the following: at some game studios an engineer might spend half their time in meetings, Slack, emails, etc, while the other half is spent on work that’s ultimately discarded. At another studio an engineer can spend their whole day working on content that appears in the shipped game. In those scenarios a “10x” engineer (or more) is very plausible.

Some companies don’t value productivity, don’t work to create environments where it can flourish, or expect and then receive low productivity. Or effectively sabotage production by re-litigating old decisions, re-doing work, reversing course only to reverse again, etc.

That some companies don’t value productivity sounds like a bold claim. But take Microsoft’s use of contract workers on games like Halo. That’s a decision to save money at the cost of lower productivity and quality.

Or take Double Fine Productions. Their top-down design philosophy often sees workers spinning their wheels waiting for directives from Tim Shafer. They realized this was a problem enough to hire someone to fix it, then fired him in part because they weren’t committed to fixing it3. Maybe this approach works for them, and with a more bottom-up approach they’d end up making generic action platformers rather than memorable ones. But it’s an approach that ranks productivity below other factors.

I don’t know much about Expedition 33’s production process, despite looking, so I don’t have a concrete takeaway here. But I’d note that the early demo of Expedition 33 looks a lot like the finished game. Visually it’s rough, with placeholder graphics and constant camera twitching. The plot and locations are different. But the bones are there.

According to James Gunn the biggest problem with the movie industry is that “people are making movies without a finished screenplay.” Even though most in Hollywood would agree that you should finish a script before filming starts.

Everyone knows that making a prototype that demonstrates the game then building that out into a full game is a good process. But maybe a lesson here is “actually do that though”, because so many games don’t follow that process. It’s hard not to side-eye games that, after 4 to 5 years, are still “struggling to find direction.”

Game developers can be very productive in the right environment. Three years into a project and still debating what sort of game you’re making is decidedly not that environment.

What We Can Learn

I enjoy poking holes in “what we can learn” analysis, but I do think there are real things to be learned from Expedition 33.

An Optimist’s Approach to Engine Use

Expedition 33 is an Unreal Engine 5 game, but lots of games use Unreal Engine 5.

It’s possible that Sandfall used Unreal Engine in novel ways, but according to this writeup their use was pretty standard: they used blueprints to wire up logic and Niagara for VFX because those are the standard built-in Unreal tools for visual scripting and VFX.

A couple uses stand out as minorly notable. One is that they used Sequencer (Unreal’s cinematic authoring tool) for skill animations. That’s a choice not all studios would make, instead authoring all the pieces (VFX, animations etc) separately and combining them via callbacks or animation events rather than editing them on a shared timeline.

The second notable use of Unreal features is the use of Metahumans, including for facial capture. I’m sure some teams, especially a few years ago, would have written the system off as not production-ready, too limiting, or too foreign.

Eternal Strands is an Unreal Engine game made by a medium-sized team - its character conversations use static portraits. Prince of Persia: The Lost Crown does likewise.

Cutscenes in Redfall are often animatics - motion-comic style presentations where static images slide across the screen. Even playback of found footage is delivered this way, which is jarring since you’d expect found footage to reflect in-game graphics and animation.

Maybe Metahumans and metahuman face capture weren’t quite ready when these games were made, or maybe the teams were reluctant to use it. (Or maybe it’s harder and more expensive than I think.) You could call animatic use an artistic decision but I think that’s often cope, to be frank.

Animatics usually look cheap. Animated models and face capture add a lot of apparent production value for relatively low-cost, if you’re capturing faces with an iPhone and not fancy equipment.

More than specific feature use, though, it’s clear the team adores Unreal Engine. It’s easy, especially with experience, to see the flaws in game engines, and easy to blame engines and tools for production woes.

This team was willing to be early adopters, to use what Unreal gave them, and didn’t suffer from “we have to build it ourselves” syndrome. Some teams would outsmart themselves - convince themselves that they should write their own rendering or material system or input handling due to real or imagined flaws in the default offerings, then spend months or years on hard-to-maintain replacements that aren’t any better.

Sometimes a more enthusiastic (some might say naïve) approach works well because when people aren’t aware of limitations they aren’t bound by them.

Sincere Storytelling

The tale of Expedition 33 is more story than lore. It doesn’t begin, as so many games do, with “ten-thousand years ago….” Of course there’s backstory but it’s delivered as events unfold in the present.

That story is a sincere drama. These days characters are often written as if they’re aware that they’re on-camera in a CW show, delivering lines to please an audience in constant, passive-aggressive 4th-wall breaking.

There’s a certain vulnerability in earnestness, as it reveals real thoughts and emotions, rather than safely burying them under layers of ironic detachment. And earnestness just seems uncool at the moment. Thor: Ragnarok allows Thor to feel sad about the destruction of his civilization for two seconds (I checked) before that’s undercut by a joke, and then his entire planet blows up as the punchline. This is a movie disinterested in questions like “how would the characters react in this situation, were it unfolding for real? (I enjoyed the comedy parts but the serious parts don’t work at all)

Expedition 33’s premise isn’t realistic, but given the premise the reactions feel real. Some people are resigned to their fates, while others are miserable, philosophical, or find cause for celebration.

Appropriate Art Style

Much has been made of Expedition 33’s “realistic” art style. I don’t know that I’d call it realistic exactly but I get what that means: you can use assets like Quixel scans (off-the-shelf rock textures) or “Forest Pack 3” with only minor modifications.

I’m wary of broad assertions like “gamers prefer realistic graphics” but I do think “realism” works for the game in a way other art styles might not.

I’ve found myself growing a bit tired of “stylized graphics” and the use of the term “stylized.” You’d imagine “stylized” implies a novel style, but often it indicates a lack of style - more often than not a “stylized” game is attempting to look like World of Warcraft-style Blizzard or the median stylized videogame.

Some stylized looks can be distracting or come off poorly in play. Animating “on the twos” Spider-Verse style, just doesn’t make sense to me in a medium that’s always valued framerate.

More importantly, I think the neutral art style of Expedition 33 plays into one of the game’s biggest strengths: it presents itself as an unknown quantity. You don’t know what you’re getting until you play it.

So much fiction today is IP based, or feels IP-based even when it’s not. Or, barring that, slots neatly into established conventions.

Take a look at one screenshot from The Hundred Line: Last Defense Academy and you know what you’re getting: a cast of wacky one-note anime tropes. At least you think you know what you’re getting, and right or wrong that feeling can nag you for much of the game, even if in the end the game subverts expectations.

In an anime-looking game you expect anime archetypes and plots, whatever that means to you exactly, even if anime isn’t a genre like romance or horror. Gothic Souls-like games, Gears of War knockoffs, etc, also come with certain expectations. Many “adult” games look the same - they use the same Daz3D assets - and many of them play the same4: Ren'Py visual novels with maybe some light sim elements.

Expedition 33’s look doesn’t confer many expectations. That works well with an original plot, and with a genre rooted in exploration and discovery. That’s very different from a game like Atelier Ryza, where despite being unfamiliar with the series I have expectations regarding everything from plot points to character design to voice acting style, based on a few screenshots.

Expedition 33’s combination of plot and art style reminds me of movies like Krull that we don’t get much of anymore. Krull is sci-fi and fantasy but it’s not speeder bikes, blasters, orcs and dragons. It’s its own thing in terms of plot, characters, costumes, locations and sets. It feels, if not unique, at least distinct. Clair Obscur's lack of presentational tells lends it the similar quality of a story unfolding before you, without the broad strokes known in advance.

Hiring for Talent vs Experience

I’m not keen on “hire off of Reddit” but the game industry leans too heavily into credentialism and experience.

Many of the worst hires in my past, both in gaming and in broader tech, share that we were looking for someone with specific experience that our team lacked. We needed a network programmer because none of us were network programmers, for example. It’s hard to evaluate candidates when your team is unfamiliar with the material, and this sort of vaguely desperate hiring can mean bumping a 7 candidate to an 8.5.

Western AAA studios prize experience with other western AAA studios, and often prize genre-specific experience as well. In theory someone who worked on a AAA game has proven chops. In practice it's much trickier; on a large team there are more places to hide and coast by. There’s an idea that someone who’s done the work is vetted - they can use source control, they aren’t an HR nightmare, etc. But as we’ve seen nightmare employees are sometimes actively protected.

The Western AAA MMO space ossified almost immediately, recycling the same members with previous MMO experience. World of Warcraft ate their lunch because, it turns out, being good at making games is more important than having experience making MMOs.

Of course there are certain jobs where prior experience is key. I’ve done short term contracts for porting or game performance - with three weeks to determine why a game runs poorly you can’t hire someone who has to learn performance optimization from scratch.

But if you’re planning on five years of work it’s fine to hire an artist who only knows Blender when you’re a Maya shop. The better artist will, in the long run, be a better hire than the artist who’s better at Maya, and the “long run” is probably months or weeks, not years. It’s fine to hire people with no FPS experience to work on your FPS, assuming your team isn’t totally clueless about the genre. In some cases hires with outsider perspectives may be more creative and less bound by convention.

The hiring stories behind Expedition 33 make me think of Roger Corman hiring James Cameron after seeing his short film Xenogenesis. You can find it online - it’s not good exactly but it demonstrates talent. Galaxy of Terror, a Corman film that Cameron did production design on, isn’t a good movie but it’s a good-looking movie, all things considered. Cameron then directed Piranha II: The Spawning. Then The Terminator.

At any point in this chain someone could have passed on Cameron because he worked on hastily-made low-budget productions, didn’t have “real” movie experience, or because the movies were bad.

Many in the game industry would blanch at hiring someone from gaming’s Galaxy of Terror, even if that candidate’s talents were evident to those willing to look. There’s a flip side to that: Expedition 33 didn’t hire many experienced AAA devs presumably due to cost and related factors, but many AAA devs would pass on the chance to work on an Expedition 33 if offered, because it’s not the sort of work experience the industry values.

In Conclusion

I don’t normally provide executive summaries but when in Rome…

• There’s a ceiling on how valuable business lessons can be, as these lessons ignore craft, when craft is a (the?) main reason the game did well

• It’s human nature to purport to learn what one already knows

• Expedition 33 didn’t hew closely to existing business advice - make of that what you will

• These guys fucking love Unreal Engine, which is a testament to Unreal Engine but also to the power of enthusiasm

• Expedition 33 delivers a sincere good story well told, something many games don’t attempt

• In story and presentation it doesn’t immediately invoke a been-there, done-that feeling

• Galaxy of Terror has an all-time great poster

In a follow-up post I’ll discuss how Expedition 33 made me confront and change an attitude I held about game design.

Until then…

This is far more literal: we struggle to ship games within reasonable budgets and timeframes. When those games do ship - if they do ship - they often fail to reflect those resources in both craft and audience appeal. This is a lament over modern video game production woes.

Dragon Age: The Veilguard released ten years after Inquisition. Recently-shuttered Monolith Productions last released a game in 2017, and wasn’t far along on Wonder Woman. Suicide Squad: Kill the Justice League was intended to capitalize on the Suicide Squad film - not The Suicide Squad from 2021, but Suicide Squad from 2016. I played Ubisoft’s Skull and Bones at an E3, I believe in 2018. It released in 2024.

Naughty Dog’s last game was The Last of Us Part II in 2020. Their next game is reported to release in 2027 at the earliest. Sony’s Bend Studio last released a game in 2019. Media Molecule’s most recent game was 2020’s Dreams. As far as I know they have no announced next project - I had to google to confirm they’re still in business. Bluepoint Games released Demon’s Souls in 2020 and has been quiet since.

Sony closed Concord developer Firewalk after one game. At the same time it closed Neon Koi, which never released a game. Deviation Games also closed before it released anything. Large swaths of Sony’s developers have effectively skipped a generation. (Before you object that Sony is a Japanese company, the gaming division has been run out of Europe as of late)

Western Woes

Video game development is volatile world-over, but this lack of productivity isn’t evenly distributed. Capcom is firing on all cylinders, releasing collections, remasters, remakes, sequels and new games on a regular cadence. Konami is back. When Nintendo takes a long time on games like Breath of the Wild or Tears of the Kingdom they do well critically and commercially - it’s evident where the time and money goes. Sega can make 8 Yakuza games in the time it takes Bioware to make one Dragon Age - that’s not hyperbole.

We regularly see successful games from upstart Korean or Chinese companies, or ones that have raised their profile and quality bar. Lies of P, Stellar Blade, Black Myth: Wukong, The First Berserker: Khazan, inZOI, Palworld, etc.

Concord cost over $200 million to make. It’s an arena shooter with 16 characters and 12 maps. It has no story mode and no set pieces. Overwatch 2 is a sequel that made a few rules tweaks, cut all the promised single-player content, and launched with a whopping three new characters.

Compare those to Marvel Rivals.

Riot Game’s League of Legends fighting game, 2XKO, just announced that it’s launching with 10 characters. It’s been in development since at least 2019. Fatal Fury: City of the Wolves started development in 2022, and just released with 17 characters.

When games take this much time and money to release (if they release at all) the pitch for a western-made game becomes untenable.

It’s Bad Management

Bad management is a well-explored topic by Jason Schreier and others, so I don’t feel the need to cover it extensively.

One aspect of bad management I’ll bring up, though, is the number of games and teams that, due to management, have almost no chance of success from the get go.

If there was ever a game begging for a “bigger, better, more badass” sequel it was Dragon Age: Inquisition. It significantly outperformed Dragon Age 2, marking a new direction for the series that could be milked (non-pejorative) for a few entries.

Instead it took Bioware a decade (granted, not all of it full production) to create a follow-up, and that follow-up almost certainly sold worse than a straightforward sequel would have.

Management didn’t plan on taking 10 years from the outset - that time was the result of a winding, fraught development. But, according to reports, a more straightforward Inquisition sequel was quickly ruled out or never seriously considered. Instead of a low-risk, medium-to-high reward approach, they opted for a high-risk approach. I didn’t write “high-risk, high-reward” - there’s no reason to believe their approach was particularly high-reward.

Sony’s approach to Concord was similarly high-risk dubious-reward. Say what you will about the quality of the game but an arena-shooter that cost $200+ million to make and includes no secondary monetization has a steep road to success. And, without impugning the quality of the game too much, it doesn’t scream $200 million.

Sometimes companies are set up to fail. This is particularly true of new companies that jump directly to “triple-I” or AAA production by hiring veteran, high-profile leadership teams attractive to investors.

In practice this means committing millions of dollars in salaries to people with job titles like director, creative director, narrative director, creative visionary, head of technology, and so on. Those jobs, when done well, can make a video game better, but they aren’t required to make a video game, and often they can’t make a video game. The people in those roles, if they can do hands-on development, have often graduated to management positions where they aren’t expected to.

These teams face a huge problem: on day one they have an empty Unreal Engine project file, on day 1500 they need the project file of a finished game, and they can’t make tangible progress towards that goal.

Prytania Media is one such example. This company spun up 4 different game development studios, using tens of millions of dollars of investment funds. All four of those studios have now closed without releasing a single game between them.

Fang & Claw used $3 million in seed money to hire a Studio Head, an Executive Producer, a Creative Director, an Art Director and a Technical Director / CTO.

Possibility Space’s first hires were for Visual Director, IP Director, Narrative Systems Design Director and Technical Director. I have no idea what an “IP Director” is supposed to do at a company with no IP. I’ve never heard “Visual Director” used as a job title in games. There’s a ceiling on how much meaningful work these roles can perform at a newly-formed company, regardless of how talented or industrious those hires may be, and the founders who spun up 4 studios at once were necessarily hands-off.

Similarly, while researching this post I came across a video from a Youtuber who was hired by Deviation Games in 2021 to serve as their community manager, and was laid off in 2023. Deviation Games never released a game and never announced a game, which really limits the scope of what a community manager can do!

There’s a raft of top-heavy, veteran-leadership companies funded 4-6 years ago that will never release a modest hit, or in many cases any game at all.

It’s Not Just Bad Management

Managers get paid to make the important decisions, so in that sense every failure is due to bad management. But sometimes “bad management” is an excuse, or at least, a woefully incomplete explanation.

Video game companies employ many managers, so one to blame is always in arm’s reach. Criticism of management is a bit inert when “management” can mean the publisher CEO, the studio head, the Creative Director or the Lead Writer.

When a team succeeds the credit doesn’t lie solely with management. So when a team fails it’s fair, at least some of the time, to spread the blame around a bit.

Perhaps it’s a gauche to write this now, when Company X just announced layoffs. But there’s no right time to write this, if that’s the criteria.

Some of You Have Odd Ideas about What Executives Do

Sometimes the CEO forces multiplayer games-as-a-service onto a game, only to change their mind three years later. Sometimes the CEO tells the team that his nephew loves Ninja Turtles so now your WW1 shooter needs Ninja Turtles. But C-suite members typically aren’t determining game content, authoring systems, designing characters, or even managing development schedules and budgets.

Many people blame “Sony meddling” for Concord’s1 failure but there’s no indication that Sony forced the team to make something they didn’t want to. Hermen Hulst probably didn’t design the characters or insist on the fearless-draft-style ranked rules.

No two people on any team have the exact same vision for the finished game, but the team seemed generally happy with what they produced, including the oft-criticized roster. An executive might issue an edict like “our game’s characters should reflect a diverse array of players” but they aren’t drawing the concept art.

I’ve seen some devs from the team unhappy with how the roster evolved over development, but the original and final designs are more similar than different.

The Saints Row team seemed happy with the direction of their latest game, pitching it as a more enlightened effort that would better connect with modern gamers. The team behind Dragon Age: The Veilguard seemed pleased with both the action focus and its more young adult tone.

We should blame “management” for approving Concord’s budget and for allowing the game to release without a secondary monetization strategy in place. (Yes, I’m basically saying executives should have forced MTX into the game, to give the game at least a chance to make money, though in this case it wouldn’t have mattered) It’s reasonable to guess that management was behind investing in CGI story vignettes, though that’s just a guess. But management didn’t write the off-brand Guardians of the Galaxy dialog or create Daw.

We tend to overestimate the amount meddling C-level positions do with game concepts. And it’s not uncommon for teams, not executives, to drive long development times and budgets.

Here’s Greg Zeschuk on BioWare working under EA.

BioWare co-founder Greg Zeschuk has challenged the common perception of EA as a controlling overlord forcing its studios to shoe-horn unwanted elements into its games, saying what EA really provides is freedom and resources - so developers can make their own mistakes.

…

"We had complete creative control over a lot of it; some fans didn't like some of it and some of it was experimental, quite frankly."

Sometimes “bad management” is giving the team too much freedom. Monolith Productions’ Wonder Woman game was reportedly struggling to incorporate the Nemesis System, first for enemies and then for allies. Neither use makes sense to me - the Nemesis System is a creative way to make randomly-generated, otherwise-repetitive enemies more memorable. A way of conjuring up worthwhile antagonists out of a white-noise sea of enemies. It’s an odd fit when you have named, well-established antagonists like Cheetah and Ares. Using it for allies means introducing a set of undistinguished allies - a solution in search of a problem.

Maybe Warner Brothers should have put their foot down and said “this doesn’t even make sense on paper, ditch the Nemesis System.” But had they done that both gamers and developers would have wailed - they have this great patented system that they aren’t even using!

Sometimes executives are too controlling, and other times not controlling enough. Those are both true at times. But that verges into tautological complaint. Maybe EA’s Patrick Söderlund should have dictated design elements of Anthem rather than letting the team work it out. But many would call that “bad management” - dictatorial meddling from someone only loosely involved in the game’s development. Letting the team figure it out for themselves - giving them “the rope to hang themselves”, as described by Zeschuk - is what most people would call good management, and what most game developers want out of an executive team.

Subscribe now

Problematic Game Development Attitudes

Many of our development woes are at least part attitudinal - the result of stubbornness, dubious conventional wisdom or adherence to outdated norms.

“Making Games is Hard” or “Every Shipped Game is a Miracle”

Game developers telling the public - Amazon delivery drivers, single moms working multiple part-time jobs, etc - that their jobs are hard has always rubbed me the wrong way. It’s the job, it’s (sometimes) fun and engaging, and at least in the US it pays well. And if it’s hard…so what?

Being a line cook is hard. You make $17 an hour in NYC, it’s hot, and you get cut and burned a lot. As a cook half your dishes can’t “fail to materialize” if you want to stay employed.

I hear of both movies and games that every release is a miracle. Movies can spend decades in development hell as scripts, directors and actors come and go. But once a movie starts filming - in game terms once it enters full production - it usually releases. Rob Morrow, the lead actor on The Island of Doctor Moreau, quit on the second day of filming. Marlon Brando didn’t show up and director Richard Stanley was fired on day three. The movie still came out.

Comparing films to games has limited use, but movie-making works the way game-making is advertised to work but often doesn’t. Getting a movie to the point where it’s ready to film is a lengthy but low-cost endeavor. “Development hell” means paying one guy $40k to rewrite a script. Development hell in games means paying a full team millions of dollars for work halfway between prototyping and production.

I’m starting to think this is a chicken and egg problem. The longer games take to make, and the more development cycles are cancelled or rebooted, the more the job of game development shifts from shipping a game to filling a seat. You show up, do some good work - though there’s a good chance that work will be discarded, so who cares really? - and maybe a game eventually ships.

I’ve worked on multiple games that were cancelled for reasons that felt beyond my control, so I get it. But the more we believe that every shipped game is a miracle the more it’s true. Does the Yakuza studio think every shipped game is a miracle? I’m doubtful. Their approach seems more much workmanlike: every day you wake up and make the donuts.

On the Power of Iteration

“Making games is iterative” is true, but that’s morphed into the more dubious proposition that simply spending time is good. So many stories of development hell involve years of iteration in service of “finding the fun”, without much fun actually being found.

Sometimes iterating accomplishes little because you get stuck honing in on a local maxima. I see this in 2XKO. The second “alpha lab” was marginally better (or, according to some, marginally worse) than the first. The team is willing to experiment on the edges but not so much at the core - willing to tweak the controls but unwilling to rework them, for example.

2XKO, like many games that undergo protracted development, also faces a Duke Nukem Forever problem, where development is outpaced by the industry as a whole. I thought 2XKO was visually appealing when it was first revealed, but since that reveal we’ve seen DNF Duel, Guilty Gear Strive, Street Fighter 6 and Fatal Fury: City of the Wolves, among others. 2XKO looks worse than all of those games, and crucially it looks older than all those games.2

I found a 2XKO interview where the director says “we’re sharing this game very early - a lot earlier than you typically would do.” That interview is from August 2024. The game was first revealed in October 2019.

Reading between the lines of that interview it’s clear that some players found the controls too complicated, when simple controls were touted as a main selling point. I’m sure they did some smaller-scale playtesting but the first large scale reality check came 5 years into development.

It’s not uncommon for games to take so long to make that they and a sequel could have released in that same timeframe, had they been more schedule-conscious. Even if you believe in “finding the fun”, focus testing, “user validation”, market research, etc, I think you have to concede that there’s nothing quite like feedback from the buying public. See the glowing Death Stranding 2 previews that just hit, with their emphasis on the way the game iterates on the first.

“When it’s done” or “a delayed game is eventually good” has morphed into an embrace of indefinite timeframes. “A delayed game is eventually good” was never true, and it’s less true today than ever. Delaying a game to add some final polish can push a game from good to great, sure, but eschewing a schedule can result in games that take 8 years to make and still release under baked.

The most common sentiment I see from dev teams working through development hell is “we’re cooking.” It’s possible that Monolith’s Wonder Woman game was the most slowly-cooking-but-ultimately-delicious action game stew ever created. But it’s also possible - and probably more likely - that with another 3 or 4 years it was going to be a pretty good action game with an upper sales ceiling below God of War, based on a movie property that was last hot in 2017. To mix cooking metaphors, while they may have been cooking the juice wasn’t worth the squeeze.

Management drives the schedule, but lack of urgency is hardly unique to management.

Ignoring Sentiment and Spite-Driven Development

Developers of games like Concord and Assassin’s Creed Shadows are in genuinely tough spots. Complaints about the casts of those games often come from flat-out racists. There’s a good chance that someone complaining about “bad writing” in Dragon Age is reflexively anti-woke.

But the writing in Dragon Age probably is too sanitized, young adult, and anachronistic. The roster of Concord isn’t great - that’s a real problem. (The roster of AC: Shadows, and of Star Wars Outlaws, are fine - confusing!)

Battlefield 5 was mired in controversy about historical accuracy and the inclusion of women, but culture war fighting aside, the first trailer looked more like a modern-day alt-history game that a WW2 one. Many complaints were motivated by regressive gender attitudes - but also, it’s a strange trailer! I think it’s safe to admit that now without fretting over giving emotional aid and comfort to anti-woke types.

Ignoring reasonable feedback because it smells unreasonable is understandable, but it can lead to situations like Saints Row where seemingly everyone but the developers saw danger ahead. Sometimes developers cross into spite-driven development, doubling down on misguided decisions, or producing community management messaging that effectively sabotages their game.

(I had some examples here but I removed them as it’s touchy subject and I don’t like contributing to culture war bullshit. Use your imaginations!)

Good Old Fashioned Exceptionalism

As I covered in “Elden Ring Discourse Emergency Dispatch”, when the Japanese game industry appeared to be struggling there was no shortage of western analysis.

I attended multiple GDC talks where western devs eagerly diagnosed the problems with Japanese game development: close-mindedness, not playing western games like GTA, lack of technical prowess, outdated game design, to even how source control was organized.

Now that Western game devs are struggling, however, there’s little explanation beyond “bad management.” Problems that aren’t written off as bad management are either ignored or attributed to the game industry at large.

There are people saying “American games just suck” but they’re Youtubers, not developers. I don’t think we need devs to say “American games just suck” but “maybe we’ve fallen into some bad habits” seems appropriate.

Game Development is Getting Away from Us

Some time ago I posted the following on Twitter:

Feel like AAA studios are increasingly creating well-made bad games. Games where the programmers and artists and sound people and cutscene editors etc all did their jobs well, but the game itself has a fundamentally flawed concept, a mismatch between the IP and the execution, etc

This got 800 retweets and 6,000 likes. I have 300 Twitter followers so this broke containment in a big way.

I wrote this with the gaming equivalent of The Rise of Skywalker in mind: everyone doing a good job except those paid the most.

Today this rings less true to me, in both movies and games. Sony’s superhero films like Kraven don’t have a good concept or script, they can’t maintain a reasonable budget, and for whatever reason they can’t match lines to lips or do ADR well either. It’s not that everyone on those movies is unskilled, but the production has crossed some threshold. Marvel’s Secret Invasion series has plotting and conceptual issues, but it also has basic directing, editing and camera-work issues. Much of this is “bad management” but it’s not all management. Ahsoka has bad makeup and costuming. (Or perhaps, the lighting and shooting angles aren’t doing the makeup and costumes any favors)

So many movies and TV shows have impossible-to-follow night scenes, and it’s not because executives forbid the use of lights for night shoots. Directors and DPs have convinced themselves that uniformly dark and dull scenes look good.

I see shades of that in how Dragon Age: The Veilguard’s writing is less Game of Thrones and more Legends and Lattes. From inside the industry there’s a push towards young adult writing, though there’s no indication that players prefer it. (If anything, the opposite3)

When I watch Star Trek: Section 31 my thought isn’t that everyone except the bosses did a great job. My thought is that the ability to make a good movie has just gotten away from them. The western game industry isn’t at that point yet, but it’s moving there.

I don’t know what the solutions are, but acknowledging problems would be a good start. There’s an understandable reluctance to give an inch on anything culture-war related, but these days everything is culture war related.

We should take cues from developers who have timelines and budgets more under control, and those consistently delivering at a high level. Over time our professional conferences and information-sharing have become more about advertorial content and influencers building personal brands. I frequently reference conference presentations but they tend to be older, even though the pace of change should render them obsolete. But the typical modern GDC presentation is “Community Clubhouse Developer Summit: Breaking the Productivity Trap: Reimagining Game Development in the AI Era (Presented by Community Clubhouse)”

Perhaps rethink what game development is. Less a beleaguered creative expedition into the game design wilderness and more a group of professionals doing a well-understood job.

Instead of paying lip-service to prototyping and failing fast either abandon those strategies as unworkable in practice or do them for real. There’s a huge disconnect between how we talk up those strategies and how we execute on them. Failing fast beats failing slowly and painfully, sure, but there’s little discussion of practical, non-theoretical processes that enable it. Why didn’t Naughty Dog fail fast with The Last of Us Online? “Fail fast” is often more meme than methodology.

Of course it’s management that sets schedules and budgets - or refuses to set them. Management determines production approach. But we collectively determine best practices, even if we can’t impose them on our bosses. And many of us are bosses, if not the boss.

Hearing this made me smile as I’ve felt the same for a while. It was also surprising, as my friend appeared to be a proponent of prototyping. It made me wonder how many developers, including those seemingly fully on board with the “you never know until you try” mindset, have unvoiced reservations.

In this post I’m going to voice those reservations.

The short version is that, in practice, prototyping often doesn’t produce the results it’s supposed to, and presents rarely-acknowledged but common problems.

Prototyping Defined

Prototyping, for purposes of this post, is a long-term development methodology. A cultural commitment to trying ideas out before running with them, throughout much of a project’s life.

Quickly building a prototype, at a game jam for example, isn’t the “prototyping” I’m discussing today. That’s building one thing, on a short schedule, with a clear deliverable and end goal in mind.

The prototyping I’m discussing today is, in practice, deferred decision-making. It’s a pseudo-scientific transposition of A/B testing onto game design: instead of making decisions up front, based on experience and judgement, try out the possibilities and choose what works best. (This sounds great on paper!) It’s “you never know till you try (so try everything).” It’s meandering, often without concrete goals or deadlines, through a nearly infinite possibility space.

It’s how you end up with games that are little more than a basic concept after years of development.

The Theory of Prototyping

The Platonic Ideal of prototyping looks something like this:

• It’s done with a small team, so while it takes time it’s not expensive.

• In fact, it saves lots of money by validating ideas and avoiding costly failed production paths.

• It’s a scientific process. You try out rival ideas, the team picks the objectively best one, then you repeat to build out a game composed of best-possible decisions. It’s game design via natural selection.

Unfortunately Platonic Ideals exist only in the aether, not in real life. In real life you do the above and still release a Concord.

Part 1: Prototyping Pitfalls in Practice

Lack of Overall Direction

One of the most common themes of failed games is a lack of shared vision among team members. This is a common complaint even on games that turn out fine, because wrangling even medium-sized teams is hard, but “we didn’t know what game we were making” is an especially acute and common failure of disastrous developments.

It’s hard to maintain a shared vision, period, as each team member imagines their own version of the game. But it’s especially hard to maintain a shared vision with a “you never know till you try” mindset and a refusal to commit. You can’t share a vision when no vision exists and you’re “finding the fun” along the way.

Don’t Rely on “Plan B”

When writing proposals or design docs, you might have some doubts about which spec is the right one to proceed with. ….

It’s tempting to approve of both options and leave the actual decision to someone else. That’s why so many proposals and design docs say “we could do this”, and the leave both doors open. But with proposals in particular, you’re better off ditching Plan B entirely. …

In the proposal phase, it’s important to stand by your idea and show you’ve really thought it through.

According to Sakurai we shouldn’t propose a Plan B. But prolonged prototyping consists of regularly proposing a Plan B, and often a Plan C and D - and then spending time and money implementing each of them.

You Can’t Evaluate Ideas, Only Implementations

Let’s imagine you’re faced with a question like “is our combat system better with or without stamina?”. Your approach is to try out both options and pick whichever is more fun1 - let’s say you pick the version without stamina.

Was that a better idea or was it just made better?

Perhaps all the animations you used (surely you’re not authoring a whole new set of animations?) looked nimble, like they shouldn’t require stamina. Maybe the stamina values were tweaked poorly - stamina use was too onerous or generous to the point of irrelevance. Maybe your enemy AI, at that point in your project, allowed enemies to attack all at once without turn-taking, and the addition of stamina made combat unfairly difficult.

To truly compare these systems you need new animations, new enemy AI and balance tweaks. Which is a lot of work! But without that work you’re invested time into engineering a half-assed comparison that reaches a foregone conclusion.

Now let’s tweak this scenario slightly. Some team members are jazzed about the stamina-free combat system so they work on that, somewhat separately from the team doing the regular combat system.

Now when you compare these two systems you’re comparing not just two different ideas, and not just two different implementations, but two different implementations from two different groups.

Maybe one group is more detail-oriented than the other, or puts more effort into game feel, sound FX and VFX. Maybe one has an animation background and the other doesn’t, or is more familiar with 3D action games.

It’s difficult, even for experienced professionals, to try out two combat systems and say “this one has better sound FX and better VFX and better overall game feel and more satisfying hit reactions but all of those are due to the specific execution - the one that feels worse in every way is fundamentally more sound.”

If they could say that - if they could tell that the less-fun implementation has more long-term potential - what did building and comparing the two different versions accomplish?

Choosing Better Ideas - The Meaning of “Better”

When I wrote about Silent Hill: Homecoming I made the point that it had the “best” combat in the series, but that made for one of the worst games, because the game was too combat-centric.

Dark Souls has basic combat. I played the entire game using a one-handed axe. I never used the heavy attack, and the light attack string uses basically the same animation for each swing. Effectively I used one attack the entire game! In a combat-centric game, like a fighting game or a Devil May Cry, that would be the sign of a terrible combat system. But while Dark Souls has a lot of combat it’s not a combat game in the same way Devil May Cry is. It’s not a “spectacle fighter” focused on the player-character’s capabilities.

If you were to compare the combat of Dark Souls and Devil May Cry in a vacuum you would probably decide that Devil May Cry has the “better combat.” But that conclusion is meaningless without context. In an actual A/B test you test something with clear metrics - do users click on the “download now” button more often if the button is red or blue? “Which combat system is more fun?” doesn’t have any clear evaluation metrics. It’s entirely subjective and it depends on the rest of the game.

Games aren’t collections of individually great mechanics. Many great games have mechanics that would be lousy in other games. A Mario game using Uncharted-style traversal would be awful, and vice-versa. Red Dead Redemption 2 is about the overall fidelity of the experience - not control fidelity but more authenticity. So opening a drawer or skinning a creature uses a realistic animation rather than a “gamey” one. If Mario played a slow animation every time he grabbed a coin it would be unbearable. Many great games have mechanics that are arguably lousy in their own game. Mashing to run in GTA. Driving a car in GTA. Shooting in GTA.

If you’re making a small game centered on a single mechanic asking “is this fun in isolation?” is useful. But in larger games answering “is this fun?” can be pointless or misleading.

The Importance of Content

A last point I’ll make here is with an excerpt from this piece from Daniel Cook of Spy Fox:

Content heavy games like RPG’s, Adventure games, some MMOGs and heavily scripted FPSs are difficult to prototype. The problem here is that these titles are actually highly evolved versions of a core game mechanic.

A RPG or MMOG is generally a single or multiplayer turn-based combat system with a whole bunch of content and meta-game systems layered on top. You can prototype the combat system, but the game’s competitive advantage is often “everything else.”

This seems far more true today that it was in 2005, even for genres that are traditionally mechanics-centered. Street Fighter 6 shipped with an entire open-world mode as a way to add more content to the game. Many AAA games rest on three pillars: combat, traversal, and exploration / puzzle solving. In many of those games the latter two are simplified or trivialized to better usher the player through content.

“Is this idea more fun than this other one?” often isn’t a good question to pose, it’s not easy to answer, and for many games, even if you can find an answer that answer isn’t too relevant.

The Production Realities of Prototyping

In theory prototyping might take time, but the team is small and the prototype can be unpolished, so it’s cheap.

In reality for prototyping to be effective senior personnel have to be involved. So while the team doing the prototyping is small it’s not all that cheap. And eating up the time of senior personnel has large opportunity cost.

In practice you often have a boss evaluating your prototyping efforts. Your boss may lack the ability to imagine untextured cubes as God of War: Ragnarok, so they expect a certain level of production polish. Which again isn’t cheap.

Sometimes instead of a boss you have a publisher, or a potential publisher or investor. I try to keep tabs on what publishers look for in pitches, and increasingly what they want are games that are far along and have great production value.

This is a video of a Bayonetta prototype. If you brought this to a publisher today they’d probably tell you to get lost. Or at least, ask you to make the game look like a final product on your own dime, then talk to them again.

Now let’s talk timetables.

We often discuss prototyping as if “prototype until you figure it out” is a short production phase that sits on the far left of the production timeline. But at many places prototyping is a production ethos, not a pre-production ethos. At those places it’s a constant productivity tax over the life of a project.

Another common way to think about prototyping is that you can prototype for as long as it takes before switching into costly full production mode. In effect it doesn’t count against the real schedule and budget.

But even Blizzard, which popularized the “when it’s done” mindset, regularly kicks under baked projects out the door now. Overwatch 2 released missing its main promised feature.

If you take too long the genre you’re targeting may become saturated. A fad or ethos you’re capitalizing on could be played out. You may hit the Duke Nukem Forever problem where games as a whole outpace your production, forcing you to constantly rework; you start off making a battle royal, then switch to a hero shooter, then switch to an extraction shooter. If you’re working on a sequel delaying it for too long risks the franchise losing momentum.

There are costs to a long production beyond monetary. And, at some point someone can (and probably will) demand recognizable revenue. That you can prototype virtually forever, with little downside, is more pipe dream than reality.

Many modern games take a long time to come out but still seem rushed. That goes far beyond prototyping - sometimes executives decide to turn the single-player RPG into a multi-player shooter, then reverse course a year later. But a common issue with these games is that the developers - not the executives, the developers - are indecisive about what they want to ship, sometimes delaying bedrock decisions for years under the guise of exploration.

From Jason Schreier’s report, “How Bioware’s Anthem went Wrong”

It’s a story of a video game that was in development for nearly seven years but didn’t enter production until the final 18 months, thanks to big narrative reboots, major design overhauls, and a leadership team said to be unable to provide a consistent vision and unwilling to listen to feedback.

…

Early iterations of flying—which, developers say, was removed from and re-added to Anthem several times—were more like gliding, and members of the Anthem team say it was tough to get the system feeling all that fun. Every time they changed the traversal, it meant changing the world design accordingly, flattening and stretching terrain to accommodate the latest movement style.

…

The most common anecdote relayed to me by current and former BioWare employees was this: A group of developers are in a meeting. They’re debating some creative decision, like the mechanics of flying or the lore behind the Scar alien race. Some people disagree on the fundamentals. And then, rather than someone stepping up and making a decision about how to proceed, the meeting would end with no real verdict, leaving everything in flux. “That would just happen over and over,” said one Anthem developer. “Stuff would take a year or two to figure out because no one really wanted to make a call on it.”

Anthem was a game about dressing up like Iron Man to fight baddies, but it took them years to decide if the Iron Man armor should fly! You can spin that as exploring whether flying was fun but at some point that’s just punting on answering “what game are we making?”

The Myth of Potential Energy

I suspect we all know people (including sometimes ourselves) who fall into the trap of planning without doing. The person who plans to get in shape, asks for advice on which exercises to do and which protein shakes to drink, but never does consistent exercise. I’m not talking about procrastinators who do no work at all, but people who substitute preparation for progress.

Something I see a lot in people who claim they want to make games — but who aren’t actually making them — is the idea that they can build up potential energy that will eventually transform into kinetic energy. Instead of working on their game they read game design books, which gives them greater game design insights, so when they do work on the game it will be that much better! Instead of messing around in Blender they watch Youtube videos on how to use Blender.

Consuming these things is fine, if you’re watching on your phone on the bus or during your lunch break. But for many people consuming this content is worse than doing nothing, since it tricks their brain into thinking they’re making progress. Food that fills them up but has no nutritional value.

Prolonged prototyping often rests on the idea that you’re building up potential energy. That you’re “laying the groundwork” and “answering outstanding questions”, so that when you begin work “for real” you’ll hit the ground running.

In reality you might stumble out the blocks when you realize the separate features you prototyped don’t play together nicely. That the game had “30 seconds of fun” in prototype form but each 30 seconds is the same as every other, so while it’s fun for 30 seconds it’s not fun for 30 hours. There are plenty of issues that only rear their head during production, even when a vertical slice is supposed to guard against that.

Laying the groundworks is important, but sometimes that we’re laying the groundwork is a white lie we tell ourselves.

The Personal Realities of Prototyping

It’s hard to stay excited about a project that remains exploratory for too long, or when it’s safe to assume that your work will be largely discarded. That creates turnover, it’s bad for projects, and I suspect it’s bad for the industry as a whole.

This Case of the Missing Cost-Benefit Analysis

“You never know till you try” is seductive as it has some element of truth: sometimes you try out a great-sounding idea and it turns out lousy, and sometimes (though, in my experience, not very often) you try out a lousy-sounding idea and it turns out great.

Making decisions up front has an obvious downside: you might make the wrong call. But a “try it out and see” approach has major downsides as well. Often missing is any discussion of those downsides, and a cost-benefit analysis.

I’d propose the following questions:

• How often does trying out multiple ideas and choosing the best one produce a better result than choosing without trying?

  • In other words, how reliable is your initial intuition?

  • Or, how quickly do you recognize and pivot away from bad choices?

• How much time and money does the process of trying out ideas take?

  • Would that time and money be better spent, on average, on execution rather than exploration?

In short, which has higher return: exploring multiple ideas, or executing on one?

One of the most common pieces of game development wisdom is “ideas are nothing, execution is everything.” In that case the answer to the above question is obvious. If we believe that ideas are nothing - are worthless - why would we advocate for a process that spends a great deal of time and budget choosing between them? (This is not a trick question, feel free to try to answer it)

I don’t entirely agree with “ideas are nothing” but execution is important. Missing out on the best ideas is a real danger, but so is short-changing those ideas (good or otherwise) with substandard execution.

I suspect Anthem would have been a better game had they committed to flying early. And would also have been a better game had they axed flying early.

Here’s another Sakurai video, “Don’t Put Decisions Off”, expressing similar sentiments as the previous one.

Each decision represents a fork in the road that will send your team down a long production path, so one wrong choice could lead to many hours of wasted work.

That’s truly a massive responsibility.

Thus, I can understand the desire to be careful, get second opinions, and figure things out as you go, but there often isn’t time for that kind of hesitation.

A quick, firm decision can end up helping more of your team that you might imagine. Every minute they have to wait is a minute the issue remains unresolved.

Intermission

I suspect many will get to this point and think “he keeps saying that prototyping is bad, but what he means is that prototyping done poorly is bad.”

Most developers reading this have probably taken part in stand-up meetings that devolve into two people arguing or getting lost in a tangent while everyone else loses interest. That’s not supposed to happen, but it happens.

The question is how often does it happen? If it happens 5% of the time you take the bad with the good. If those kinds of unproductive meetings happen 50% of the time you have a problem. If those kinds of unproductive meetings are the norm across the industry then we should stop using Scrum, no matter how great it sounds on paper.

What motivated me to write this piece is that it’s so rare for people to acknowledge when prototyping goes wrong, or that it can go wrong at all.

When in practice it goes wrong pretty often.

Prototyping: Origins

My friend who told me he’d grown weary of prototyping mentioned that he remembered it as popularized by GDC. I’ve done a little digging into that time period, and I’d like to highlight one representative presentation.

Game prototyping is all the rage these days. Chaim Gingold and I gave a fun lecture at the 2006 Game Developers Conference titled, Advanced Prototyping. The material is based on our experience prototyping games, technology, and user interfaces for Spore and at the Indie Game Jam over the past 4 years. Here's the abstract:

There are two points I’d make about this presentation and similar presentations from around that time period.

First, much of it goes against this hard-to-summarize thread by Daniel Cook of Spry Fox:

Why does 'finding the fun' fail so often as a design strategy?

#gamedesign

New designers are told to prototype and 'find the fun'. But the naive version of this is a garbage tactic that mostly results in poorly thought out prototypes that are never going to converge on gameplay.

The prototyping talks from around that time period lean heavily on “finding the fun” of individual features. They advance, in essence, the “bad version” from Cook’s thread: make the system, play the system, see if you’ve “found the fun” and if you co-workers (who are in this context little more surrogate players) are jazzed about it.

More generally, these presentations fall into many of the pitfalls I covered in the first section. The idea that games can be successfully composed of individually-selected features, chosen based on which are most fun in the moment. The conflation of trying out ideas with trying out executions. A commitment to a constant productivity tax, where Plans A B and C are regularly implemented, to aid in making even minor decisions. A lack of appreciation for the time cost of prototyping, and how that time could be spent elsewhere.

This particular presentation was well-received at the time because Hecker was working on Spore. Here’s a particularly amusing writeup of the prototyping talk, from Gamespy

Great games like Spore don't spontaneously appear from a design document. Here's how it happens...

Chris Hecker and Chaim Gingold are outspoken vanguards of wild, experimental approaches to game design and development. Is it any coincidence that these two designer-programmers are both working on Spore, Will Wright's experimental new Maxis title? This morning when the pair spoke about the processes they use to build up a breakthrough title like Spore, people were mobbed outside the door waiting to get in.

This writeup about how great and groundbreaking Spore is, thanks in part to heavy reliance on prototyping, was written in 2006. Two years before Spore released.

Spore still regularly shows up on most disappointing games of all time lists.

Spore relied heavily on prototyping individual features and minigames and that’s what it plays like - as per Cook it never converged on gameplay. It’s hard to even call Spore a collection of individually-fun elements that don’t converge, because most individual elements aren’t particularly fun on their own. The strengths of the game are the high-level concept (going from single-celled organism to a race of space explorers) and the creature creator.2 But adjusting the morphology of creatures - the most fun part of the creator - is only loosely connected to the gameplay. If you want to make a fast creature you don’t give it a long flexible spine and a dozen millipede-style legs, you give it a foot with a high “speed” stat.

So the second point here is an oldie but goodie: the proof of the pudding is in the tasting. Spore’s production methodology didn’t produce great results.

Part 2: The Prototyping Alternative

So if prototyping is bad (at least, more often than we’d like to admit) what’s the alternative?

I propose something like this:

1. Decide what game to make

2. Make the game

Step 1 might involve building prototypes, but in weeks or months, not years. It’s not the A/B/C/D testing of competing features or meandering through an infinite possibility space. And it’s not - this may be controversial - “finding the fun.” (Or hoping to stumble over it) You can hone the fun, or “follow the fun” - double down on the parts that are working best. But you should know what the fun of your game is supposed to be from the start. Otherwise what are you doing?

Step 2: make the game. This is not “doggedly move forward and stick to your original plan no matter how poorly it’s going.” Move forward in sane manner. Still start with rough versions and blockouts. Adjust to taste - if something is going poorly ditch or rework it.

What I’m proposing, more than anything, is an attitude shift from “this is exploratory work that’s probably throwaway” to “this will probably end up in the final game in some form.”

In practice these are less different than they sound. If you want to answer a question like “should this function be on left or right click?” you could try out both and gather team feedback. But you could also just choose one and swap down the road if needed. Instead of doing both up front and committing to a productivity tax, that tax is only paid when things go awry. (Though that tax may be higher, I suppose)

Anthem Flying and When #@&! Gets Real

The Anthem team explored different flying variations and eventually cut it entirely because it didn’t feel fun and complicated the production. That may sound like an example of prototyping done right: the team avoided a costly dead-end path.

But of course the game shipped with flying.

The leadership team’s most recent decision had been to remove flying entirely, but they needed to impress Söderlund, and flying was the only mechanic they’d built that made Anthem stand out from other games, so they eventually decided to put it back. This re-implementation of flying took place over a weekend, according to two people who worked on the game, and it wasn’t quite clear whether they were doing it permanently or just as a show for Söderlund. “We were like, ‘Well that’s not in the game, are we adding it for real?’” said one developer. “They were like, ‘We’ll see.’”

…

Then, according to two people who were in the room, Patrick Söderlund was stunned.

“He turns around and goes, ‘That was fucking awesome, show it to me again,’” said one person who was there. “He was like, ‘That was amazing. It’s exactly what I wanted.’”

Here’s a quote about flying from the ArsTechnica Anthem review

Almost everything about Anthem's flight system is awesome. The sheer act of lifting off looks, sounds, and feels great, no matter how many times you do it. There's a timing chain for the required jump-then-boost combo, coupled with a light-and-sound reaction of blasting off. The combined effect implies enough torque to make your real-life head rock back instinctually. Once you're airborne, the default speed is slow-and-maneuverable for newbies (aided all the more by a useful "hover in place" button), while pressing "forward" delivers a juicy amount of controllable velocity.

Flying wasn’t fun so they cut it, added it back in to impress their boss, then under the pressure of “make it work or we’re toast”, made it fun. That should scare prototyping enthusiasts.

The process that was supposed to help choose the correct feature set and avoid bad decisions - a process that took a lot of time and money - produced worse results than “it’s an Iron Man game so of course it should have flying - why are we even having this conversation?”

So what changed that made flying fun?

Sound, look and feel aren’t system features that show off in a rough prototype, they’re production polish. A hover button is something you might introduce to fill a need as the game comes together, when you notice that’s it’s hard to land or shoot precisely while flying. So one answer is the team wasn’t able to imagine cubes as God of War: Ragnarok - they weren’t able to project their prototype version of flying forward into a more polished more satisfying version.

Another answer is that there’s just a fundamental difference between “exploring” and working for real.

When Mark Darrah joined the project in the fall of 2017, he began pushing the Anthem team toward one goal: Ship the game.

“The good thing about Mark is that he would just wrangle everybody and make decisions,” said one former BioWare developer. “That was the thing that the team lacked—nobody was making decisions. It was deciding by panel. They’d almost get to a decision and then somebody would go ‘But what about this?’ We were stagnant, not moving anywhere.”

“He started saying basically, ‘Just try to finish what you’ve started,’” said a second developer. “The hard part about that was that there were still a lot of things to figure out. There were still a lot of tools to build to be able to ship the game we were making. It was very, very scary because of how little time there was left.”

“Finish what you’ve started” highlights the mindset difference between “this is exploratory work” and “this will end up in the shipped game”, as well as the importance of execution.

Earlier I posed the question: is it better to spend effort executing on the (possibly) wrong idea or to spend that effort choosing the right idea? But in this case execution revealed that the” wrong” idea was the right one.

Knowing Without Trying

Most games consist of familiar parts arranged in familiar fashion.

Many games are sequels. IP-based games typically don’t let the demands of IP define a radical new genre; instead the IP is married to an existing genre. Star Wars: Outlaws and Avatar: Frontiers of Pandora both take the form of familiar Ubisoft games.

New games are often “spiritual successors” or “inspired by” or “x meets y” - “cozy farming sim meets dungeon crawler.” Most games aren’t interested in big innovations, and it was a lot easier to be innovative in 1979 when there were only a few dozen games in existence.

If you’re trying to decide on an Estus-Flask style health system vs health packs vs regenerating health there are plenty of examples of games that use each of those, to varying degrees of success. The same is true for tactics games that use action points vs a fixed number of actions, hex vs square grids, various cover and line of sight rules, etc.

I recently saw a video about Vlambeer prototyping Luftrausers, as an example of how prototyping can be useful. Quickly building a prototype is the good sort of prototyping so I have no objections, but I could tell you without prototyping that Luftrausers would work because I’ve played a game called Asteroids. Not to mention Two Tigers, which is a very similar game. Blasting enemies while controlling a ship via Newtonian thrust has worked since Spacewar!

Something that’s always irked me about the game industry is how disinterested we are in our own prior art. Maybe it’s because the industry, being technology-based, prizes newness. Maybe it’s that high turnover leads to loss of institutional knowledge and historical memory. But for whatever reason our collective memory of video games often seems limited to hits from the past 5 or 10 years.

An example I’ve written about before is the difference between loot games that drop items and loot games that drop currencies. (Or that make it easy to convert items to currency and back) The former is almost universally more fun than the latter, so whenever I see a game doing the latter I want to ask: why?

Watching Sakurai videos it’s hard not to notice that he has a broad and deep knowledge of games. I suspect that’s why he has the confidence to be decisive and why it works out - I’m sure he has good instincts and intuition, but he also has a wealth of knowledge of prior art.

When I consider game ideas I always ask myself: did other games do this, did it work, and why or why not? Particularly useful is identifying that you’re repeating bad decisions from past games.

In the first Project X-Zone you have one primary resource that’s used for super moves, defensive moves, and minor buffs. Since these all draw from the same pool and super moves are way better than the minor buffs using those buffs is nearly always a mistake; playing optimally means ignoring the buffs system entirely.

In Project X-Zone 2 super moves and buffs use two different resources and don’t compete. Lesson learned.

In the PS2 God of War magic uses mana and a separate rage meter governs how often you can engage Extra Angry Mode, for presumably similar reasons. If magic and Rage Mode shared a meter it would be hard to pace the Rage Mode activations, hard to balance spell effectiveness and cost vs Rage, and players might feel bad for using magic since it would delay Rage Mode. (Perhaps indefinitely) You can see a similar dynamic in Street Fighter 6, which has a super meter and a separate Drive Gauge rather than one universal resource bar.

If you’re working on a game that involves special moves backed by a resource and pondering what would work best there are hundreds of previous examples to call upon. You don’t have to puzzle it out on your own. There are also dozens of examples of games that use cooldowns instead of resource pools. Just yesterday I played demos for Dynasty Warriors Origins and SWORD ART ONLINE Fractured Daydream. Both map special abilities to right bumper + face button. In Dynasty Warriors those abilities use resource points and in SOA they have independent cooldowns. I think those each make sense for their respective games, based on what those games are trying to accomplish.

I’m not trying to imply that most developers ignorantly stumble through this sort of analysis. But I think we collectively undervalue bringing prior art to bear.

I’ll end this section with a quote from Xalavier Nelson of Strange Scaffold:

I do think the overall principles of having, if you’ve got a time and a budget and a thing that you’re bringing to life, especially if you know what you’re building, and it’s based on any prior precedent, you can ship that thing - it’s ok - that’s possible. I was told a lot, early in my career, that that’s not possible, I was told you can’t know what game you’re making until you making. Which is fine as a philosophical concept - it doesn’t not keep you in business for very long though.

I’m not going to claim that Nelson has read this blog and agrees with my points. But I agree with his. You can know what you’re making up front rather than finding out along the way, and to some degree you can know if the game will work, because it’s likely composed of familiar pieces. Which, I would note, is less true of Strange Scaffold’s games than of the vast majority.

The Conclusion

This final quote comes from Ken Levine by way of Jason Schreier. It took me quite some time to find it but I’m glad I did, as provides a nice summary:

During a panel discussion a few years ago, Levine explained the final act of his process. “In almost every game I’ve ever worked on, you realize you’re running out of time, and then you make the game,” he said. ”You sort of dick around for years, and then you’re like, ‘Oh my god, we’re almost out of time,’ and it forces you to make these decisions.”

This is rarely admitted to but I don’t think it’s particularly uncommon.

Levine’s Judas looks an awful lot like Bioshock, by the way, which is common in these sorts of cases. After a prolonged prototyping phase Mass Effect: Andromeda was essentially a lesser version of previous Mass Effect games3. Not much in Anthem stands out as incredibly ambitious or experimental. When games run out of time and are forced to make decisions that often involves regressing to the mean.

I’ve done a fair amount of Anthem-bashing, but a point I’d make again is that had the Anthem team spent all that time actually making Anthem it would probably have been pretty great!

That’s the tradeoff we don’t often talk about. If they had made decisions quickly some might have been wrong. But they would have had more time to work on the best possible execution of those ideas, rather than 18 months to produce a flawed version of the theoretically best ideas. (And “theoretically” is doing a lot of work here) And boy, the lack of execution in the final game is very apparent, with horrible loading times, a mathematically flawed loot-scaling system, etc. The first-person home base (the rest of the game is third person) is bizarre in concept and just doesn’t work in practice. With more time at least someone could have cranked up the walking speed to make it bearable.

Studios like Capcom and Ryu Ga Gotoku put out quality games on a regular schedule, while western-made games increasingly struggle with bloated schedules and budgets. That goes way beyond a commitment to prototyping, but I think it’s fair to say that those more productive studios indulge way less in dicking around. Ryu Ga Gotoku released Like a Dragon, a turn-based Yakuza game, 2 years after the brawler Judgment - there simply was no time to dick around.

This post is not advice. I could never tell anyone, site unseen, that they’re prototyping too much or in the wrong way. But what I’m fairly certain of is that we, as an industry, aren’t entirely honest with ourselves about how often the “we dicked around and now we’re out of time” scenario plays out. About how often prototyping is a cover for indecision and a way of giving ourselves permission to dick around. And about how often prototyping provides worse results than advertised while being a constant drag on productivity.

Subscribe now

Screenwriter William Goldman famously wrote:

Nobody knows anything...... Not one person in the entire motion picture field knows for a certainty what's going to work. Every time out it's a guess and, if you're lucky, an educated one.

In the modern video game industry, by contrast, everybody knows everything. There are innumerable Youtubers and podcasters, circuit speakers and self-styled sales and marketing experts offering instant opinions on every topic.

A particular formula I’ve come to loathe is “What We Can Learn from Flappy Bird” - the idea that runaway successes and failures result in easily-digested takeaways like “keep it simple stupid.”

I don’t think there was much to learn from the success of Flappy Bird, as why it blew up remains a mystery. And I don’t think there’s much to learn from the failure of Concord.

Part 1: Broad Thoughts on Concord

Why Concord Failed - I Don’t Know (and Neither Do You)

Concord had some obvious problems and a tepid initial reception, with many immediately projecting it to flop. So in some sense everyone knows why it failed: misguided emphasis on story, unappealing characters, (apparent) been-there-done-that gameplay in a saturated genre, price, etc.

But the level of flop - a belly flop into the Grand Canyon that missed the water below entirely - is surprising1. Out of line with the quality of the game, in the same way that Flappy Bird’s success was out of line with its quality.

Oppenheimer is a 3-hour historical drama and Barbie is a girl-power movie based on dolls, and the two have little in common beyond release date. But somehow they became the merged cultural phenomenon Barbenheimer, with Oppenheimer making a billion dollars worldwide despite being the sort of talky drama that foreign markets shy away from.

Barbenheimer benefitted from a positive snowball. Though detractors would be loathe to admit it, Concord suffered from a negative snowball. It became a cultural phenomenon like Barbenheimer, but as a must-skip rather than must-see.

The movie Ishtar was once regularly called one of the worst movies ever made. It bombed at the box office and was dogged by reports of troubled production. As a bloated expensive flop it was easy to pile onto, and became a poster child for common sins of the movie industry.

These days Ishtar lies mostly forgotten because, in truth, it’s merely a mediocre movie, not terrible enough to be memorable. Removed from the mania of its time, with critics tripping over each other to write the most scathing review, it’s just your typical kinda-bad movie. This can go the other way as well - The Shipping News was once lauded as one of the greatest novels. I’m not sure if anyone seriously believes that today, and I’d be a little surprised if the majority of people reading this have even heard of The Shipping News.

Everyone “knows” that E.T. for Atari is one of the worst games ever made, but the vast majority of people who “know” this have never played it. I like to point out that not only is E.T. not the worst game ever made, it’s not even the worst Atari game of 1982 based on a Steven Spielberg movie and designed by Howard Scott Warshaw: that would be Raiders of the Lost Ark.2

SwordQuest: FireWorld is an ugly, unfun, and totally incomprehensible 2600 game. At least E.T. has a nice rendition of the E.T. theme. But for whatever reason SwordQuest: FireWorld isn’t a game people love to hate.

Overreacting people hate to hear “calm down you’re overreacting” - it just makes them madder! I see that effect with Concord - someone claims it’s one of the worst games ever made and a $200 million flop, and at the suggestion they might be emotionally overinvested, revise upwards to the single worst game ever made and a $400 million flop.

Animal Crossing: New Horizons became a game people loved to love, with a success that probably outpaced its merits. It became Barbenheimer or Hawk Tuah Girl. Concord quickly became a game that people loved to hate; it became Ishtar or Bean Dad. It has real problems, but anyone claiming the magnitude of failure is entirely on merits (or lack thereof) is willfully deluding themselves.

How Much Did Concord Fail?

I’ve seen people claim that Concord is the biggest AAA failure of all time.

Here’s the question I’d pose in response: is it even the biggest AAA failure of this year?

Warner Brothers lost $200 million on Suicide Squad: Kill the Justice League. That’s a huge monetary loss, and a huge hit to the Arkham brand, a billion dollar franchise. And to Rocksteady’s reputation, a developer that regularly made Best-Of developer lists.

Suicide Squad: Kill the Justice League posed a huge opportunity cost - the game took forever to make and that’s time the team could have spent on another mainline Arkham game or more-successful spinoff.

Concord wasn’t an established brand and Firewalk wasn’t an established developer. There’s some damage to Sony’s reputation but Firewalk is not one of Sony’s marquee developers. The studio was formed specifically to make Concord so any opportunity cost is much more academic - Sony could have used that money on other ventures, but Concord isn’t clearly taking the spot of another game. “With that money they could have made 10 Soul Sacrifice sequels” isn’t a real point, Sony wasn’t going to make those games either way.

I’m skeptical of claims that Concord lost significantly more than $200 million. But even if it did, Suicide Squad seems like the bigger bust this year, let alone of all time. Suicide Squad (and, to a lesser extent, Gotham Knights) has significantly damaged a key franchise from a premier developer; Concord was a one-and-done flop from a new studio.

I’ve read “estimates” (aka people making up numbers from thin air) that Concord took 5 years and $100 million, then 8 years and $250 million, then 8 years and $400 million, then 10 years and $800 million. I laughed out loud typing out that last one, as that would put it at four times the cost of some of the most expensive games ever made.

It’s a game with a small set of levels and a small set of (admittedly expensive) character models - if it cost as much as massive open world games something went very wrong. I’m not going to dwell on this because I don’t know the budget, but nobody “reporting” on the cost knows either. The number continually changes (in particular, it grows with each retelling) and is never sourced.

For various reasons people desperately want this game to be a historical failure, and that requires the cost to be very high.

If Concord was somehow one of the most expensive games ever made I’m not sure what the resulting analysis could be beyond “don’t spend $2 million on each crate in a warehouse level.” Maybe AAA games should cost less to make - that’s a reasonable topic to broach. But AAA games of Concord’s scope don’t cost that much. Spider-Man 2 cost a lot but it’s a lot of game - you can see where that money went. If Concord had an obscenely out-of-proportion budget that says little about AAA gaming and more that something went horribly wrong.

Part 2: A Survey of Nonsense

The Case of the Incredibly Long Credits

Analysis like this is aimed squarely at people who want to believe and retweet whatever confirms their biases, but since I’m not that person I watched (well, skimmed) the credits.

The credits video is over an hour long, true. But the credits scroll incredibly slowly and the main credits end at around 9 minutes when they start listing production babies.

The rest of the credits lists basically everyone even tangentially involved in Concord through Sony - Sony’s legal team, the people who update and maintain the PSN store, the people who submit games to the ESRB, etc. These were not full-time employees who worked on Concord for a decade - many of them just sat in the same office as someone who “worked” on Concord for an hour by sending an email. The credits also print out the full text of some open source licenses.

The credits do contain many senior / lead / principle titles and the names sound like people who live on the US West Coast rather than Myanmar, so they don’t come cheap, but the credits are an hour long only in a purely technical sense. And the formula that translates each minute of credits to $5.7 million of budget is certainly new to me.

“Culture Killed Concord”

Here’s a video from someone who - I’m going to be blunt here - simply has no idea what they’re talking about.

The idea that “culture killed Concord” is based on a chain of plausible inferences. But it’s easy to generate explanations that sound plausible enough but aren’t true.

The logic goes like this: we know that the game didn’t sell well. Therefore we can surmise that players didn’t like the game. Therefore we can surmise that playtesters didn’t like the game, and that the developers ignored playtest feedback or didn’t run playtests at all. Therefore we can surmise that the team was arrogant and had a toxic environment, which further implies that top talent left, since top talent flees toxic environments. Etc etc.

The first link in this chain - the game didn’t sell which means players didn’t like it - is immediately suspect. Maybe players disliked what they thought the game was, based on poor marketing and rollout. The game’s biggest haters, after all, are the people who haven’t played it and aren’t familiar with the basic systems. Potential players thought the game lacked appeal, sure, but appeal is largely a function of marketing. Maybe the team rejected feedback - or maybe they were too open to bad feedback.

What characterizes this video is a total lack of legitimate information. The logic deployed in this video could be leveraged against any game that underperformed, starting at low sales and ending up at the conclusion that the team was arrogant and toxic. The argument here is, in essence, is that any game that underperforms indicates a culture problem, and that’s simply not true.

Hi-Fi Rush lost money - is that a culture problem? Does Remedy have a culture problem? Redfall bombed but Arkane’s previous games did well - am I to believe a huge cultural shift occurred during Redfall’s development? Did a similar cultural shift at CD Projekt Red explain the poor release state of Cyberpunk? But then that culture reverted to old form as the game was fixed up?

This video is one of many I’ve seen to the tune of “I’ve worked in the video game industry and here’s what happened!” But the authors of these videos don’t know anything, and they aren’t even making particularly solid suppositions.

Part 3: Hindsight Analysis and All-or-None Evaluation

In “Do Games Need Hooks” I wrote about the danger of hindsight analysis.

A major issue with hook analysis is that it’s mostly post-facto.…If you believe that games need hooks to sell and a game sells well you might skip over “does this game have a hook?” and right to “this game must have a hook, so what is it?” Conversely if a game does poorly it must be due to a weak hook, and your job is to identify that weakness.

We know Overwatch sold well and is therefore a good game, and we know that Concord sold poorly and is therefore a bad game, so it’s easy to say that Overwatch has great character design while Concord’s is lousy.

I’ve seen multiple videos comparing these two character specifically, where people tangentially involved in the game industry argue that the left figure is awesome while the right one is terrible.

There’s one huge flaw with this analysis: I have eyes.

The left figure is, to me, frankly, goofy-looking. It looks like a side character you’d see in a mid 2000s 4Kids Anime titled “Sky Warriors: Beast Fighterz.” It’s got the comically large pauldrons that so many Blizzard characters have, unmotivated lion theming3 (though he’s shaped more like an ape), and a kewl aesthetic. He looks a lot like this “Battle Armor He-Man” toy I once owned.

Concord has character design issues in both conception and execution. I have a visceral negative reaction (for reasons I don’t consciously understand) to the blue lipstick. The character’s outfit looks blobby and inflated. Multiple Concord characters have odd tubing and their materials are often indistinct - not clearly metal or plastic or rubber or cloth. But the right character, with her football player appearance, is more clearly a tank than Reinhardt, the left character. Reinhardt apparently uses a hammer and an energy barrier - from the artwork above you get none of that.

I’ve seen so much analysis in the form of “when you look at this one picture of Reindhart you instantly know what his deal is” and that’s simply not true.

Video game analysis tends towards all or nothing. Some sites break scoring down into categories like “sound” and “graphics”, but usually the best games get high marks across the board, and bad games get low marks everywhere.

It’s easy - and a little lazy - to dwell on the negative and minimize the positive in a game that does poorly, and do the reverse for a game that does well. Most of the analysis I’ve seen of Concord works backwards from “we know the game is terrible”, and pegs every individual feature as terrible. But that the game didn’t sell doesn’t make every component terrible. There are some games where nearly every component is terrible - War Gods, for example. (Sorry War Gods, the truth hurts) But reality is often more complicated. The first game with cutting edge deferred rendering was the Shrek Xbox game of all things, with its 49 Metacritic.

Concord is a good-looking game. I have some quibbles with the character models (mostly in the materials) but they are undeniably well-done. The designs are a mixed bag and fairly weak overall, but that doesn’t mean the technical art is poor. People who’ve actually played the game tend to like the stacking buffs mechanic - I don’t think that mechanic is splashy enough to be a selling point, but existing in a game that failed doesn’t make it a failed mechanic.

I read on Japanse gamedev Twitter that the gem system in Capcom’s Marvel Super Heroes, in which gems fall from the sky for either player to scoop up, was inspired by the “Life Ball” in an NES Muscle wrestling game.

If you’ve watched Sakurai videos you’ve noticed he frequently references a variety of old, often fairly obscure games as inspiration. Picking out the good parts of bad games is how you end up with more diverse inspiration, and acknowledging the bad parts of good ones is how you improve rather than replicate. These are important skills for a game industry professional.

Part 4: What We Can Learn from Concord’s Failure

Live service games are fraught with danger, but also HoYoverse made $4 billion dollars.

“Make more Astrobots and less Concords” sounds nice, but if a knock on the industry is that it chases trends too often “make more Astrobots” is itself a variation on “do what was popular ten minutes ago.”

Smaller teams working on projects sometimes produce an Astrobot. But Concord had a pretty limited scope, with a modest number of maps and modes. Sometimes a small budget balloons into a big one. Sometimes a small, veteran team working on a modestly scoped project makes a Hi-Fi Rush then is shut down. Or they’re a Doublefine making a string of perfectly fine games. Or a FuRyu Corporation. (Google it)

I’m sure there are things Sony can learn from Concord. The fact that so many people immediately projected it to flop while Sony seemed somewhat4 unaware is concerning - there must be takeaways in terms of product evaluation, feedback collection, marketing strategies, etc.

But “here’s what the game industry can learn from Concord” is mostly the domain of engagement farmers, the people who charge for game consulting phone calls, and the guys making angry youtube videos aimed at low-information followers.

When a Flappy Bird blows up the lessons are “keep it simple, stupid”, “less is more” and “players crave simplicity.” When a Baldur’s Gate 3 blows up the lessons are more is more and that players can handle and crave complexity.

If the new Dragon Age does poorly people will point to the simplification of RPG mechanics as a culprit, but if it does well the lesson will be that people like that simplification. This sort of analysis is just stating what seems most obvious in hindsight, as part of a supposed larger trend that may or may not exist.

When Helldivers 2 blew up the devs were praised for sticking to their vision and their “a game for everyone is a game for no-one” philosophy. Now that Concord has flopped is the new takeaway “your vision might be suspect so run it by a group of 13-year-olds”?

A likely result of Concord’s failure is that Sony will rely more on focus testing, on “design by committee”, go through more “market validation” steps, involve more outside consultants - rely more on a formalized process that produces safer titles. A process that produces fewer Concords but also fewer Helldivers. A process that most gamers claim to dislike.

Executing on a unique creative vision is good, unless that vision is bad. Is that a useful lesson?

Maybe there isn’t always a pat takeaway. Life is not a string of Aesop’s Fables.

Maybe the lesson is they can’t all be winners. That sounds trite, but it’s no more trite than “the problem is it didn’t appeal to players.” Yeah - no shit!

If you’ve read my previous blogs you know I’m a fan of in-depth analysis - I spent hours running into rocks in Stellar Blade to understand the character controller. But understanding the details of a particular feature is very different from understanding what will sell. In that respect, as William Goldman wrote, nobody knows anything.

Subscribe now

Post-release I’m revisiting the game to discuss various design elements. The main takeaway: there’s a better game in here buried under a pile of just too much stuff.

Combat

My first post was about the tension of Midnight Suns being both a superhero game and a card game. Playing it extensively I see that tension in action, though the card game element hurts audience appeal more than it does gameplay.

Sometimes in Midnight Suns I play card called “Blast” and Iron Man blasts a guy. Makes sense. But sometimes I play a card called “Spider Sense” and draw two cards while reducing the cost of “heroic” cards by 1 - something that has no relation to Spider-Man sensing danger and doesn’t translate well from card game rules to in-game fiction.

Midnight Sun’s combat is at its best when you’re doing superhero things - grabbing a guy and slamming him into another guy, kicking a crate at some thugs to knock them over, etc. Where it’s less successful, for me, is when it leans too heavily towards card game mechanics and away from super-heroism.

Ghost Rider has a card that forces you to discard your hand, potentially discarding cards you wanted. That downside becomes an upside if you wanted to dump useless cards and draw new ones. This design has interesting tradeoffs and appeals to game designer brains. But it has nothing to do with Ghost Rider running people over with a flaming car. In theory it’s satisfying to craft a deck that does card gamey things like lowering casting costs and drawing cards, but to me that feels less heroic and less satisfying than just bonking a guy on the head with a crate.

In missions where you have to retrieve items you defeat enemies who drop “retrieve item” cards. What does it mean for an enemy to carry a card? Yu-Gi-Oh! is about card battles, so its characters wielding cards makes sense. But in the in-game fiction of Midnight Suns the characters are punching and shooting lasers, not playing cards.

Some defeated enemies drop a “Focusing Beacon” card. Canonically the enemy carried this beacon, and when you play the card your character deploys it. But sometimes an enemy drops a card like “Delay Truck” that’s just a card. It doesn’t represent a wheel lock. It’s a piece of rules with no in-fiction analog. The enemy was carrying the vague capacity to delay a truck, which was somehow transferred to your characters?

As I said in my first post on Midnight Suns, these are jigsaw puzzle pieces that don’t fit together perfectly. Not a deal-breaker but not ideal.

Combat Minimalism

You can only (normally) move one character per turn, but you can move them anywhere, so there’s no fussing with action points or movement grids. Maps are small and you only bring three heroes into battle. There’s no cover, no miss chance and no fog of war. Combat is minimalistic and puzzly - it’s far more Into the Breach than X-Com.

That’s fine - the battle is fun! It feels mobile-game-esque, but not in a bad way. But it’s a bit limiting. Positioning is very important; a large part of the game is knocking enemies into other enemies, knocking them into exploding barrels or electrified equipment, darting between nearby enemies, etc. But as the player you don’t have much control over your character’s movement, and between rounds enemies can move anywhere. So positioning is important without many ways to influence it.

In some missions you protect an object from enemy attacks, but the game has few protection mechanics. Enemies can move anywhere and target anything, so you can’t take a defensive formation or erect cover; protecting an object mostly means killing the enemies quickly.

As far as I’ve seen there are no missions where you infiltrate a position, find and escort some VIPs, flush out enemies, or have a running firefight. Instead of uncovering new enemies as you explore a map they simply run onto the map from out of bounds each turn. That can get tedious.

But, overall, the combat is pretty fun! The big problem with the minimalism, that I’ll get into later, is that the rest of the game is incredibly maximalist.

Plot and Narrative Setup: When the Least Interesting Character is You

I’m not going critique the writing or plot beats, but I want to touch on the narrative setup, as it greatly impacts the overall design.

Instead of playing as an existing Marvel character you play as “The Hunter”, a chosen one with amnesia, a decision I find outdated and misguided.

Genre stories often have an audience point-of view-character - a story set in the future might include a character from our time who wakes up from cryo-sleep. That gives us someone to relate to, and gives writers an excuse to deliver exposition, since that character, like the audience, needs to be filled in. Sometimes a character is meant to represent a target demographic - Wesley in Star Trek: TNG and Lucas on Sea Quest: DSV were both plucky smart teens, meant to appeal to teen male viewers.

The Hunter seems birthed from the mindset that players need an audience surrogate to help get them on board. It feels unconfident, in the same way the first X-Men movies clad its characters in black Matrix-like outfits instead of their real costumes. That arguably made sense at the time, but we’re in 2024 now. There have been 33 Marvel movies and 250 hours of Marvel film and TV content. Movie Wolverine can now wear a real Wolverine outfit and players can play as Wolverine rather than a “you” character. They don’t need a stand-in, and they really don’t need one who wanders around asking “so Blade…what’s your story? You a vampire or something?”

Comic books stories are often bad, but they aren’t often laden with exposition and backstory. If anything superhero comics have the opposite problem: they need a fight every 22 pages. Midnight Suns is full of exposition about things most players already know or could easily find out.

It’s - and I mean this in a derogatory way - a very video game setup. Amnesiac protagonist needs everything explained to them by way of “tell me about…” dialogue trees.

It’s hard to overstate how much of a non-character The Hunter is, in both personality and mechanics, especially put against the established Marvel characters.

Many of The Hunter’s conversations simply deliver exposition. Some conversations reveal the personalities of the Marvel characters, but The Hunter is taciturn and unspecific. You can game these conversations for both darkside / lightside points and for relationship points, which encourages min-maxing over role-playing, so The Hunter’s personality changes to match whatever’s optimal. But even if you wanted to roleplay the dialogue choices are lots of “that sounds tough”, “I don’t envy your position”, “tell me more” and so on. My Hunter has all the personality of a sheet of cardboard.

I’ve seen many complaints of “bad writing”, and I don’t entirely disagree with them, but making the protagonist a clichéd JRPG hero is doing narrative on hard mode. Even great writers would be hard-pressed to write compelling dialogue given constraints like “the plot must unfold very slowly as the hero languidly recovers from amnesia” and “the hero has to ask every character what the deal is with their powers.”

Mechanically this character is also a miss for me. Magik is pretty cool - you create portals then knock enemies through them. Captain Marvel can power up into Binary and has ways to taunt enemies and generate shields so she’s a bit of a bruiser / tank. Nico’s gimmick is she can’t cast the same spell twice, so her cards are based around random effects, often determined at draw time.

The Hunter character has vague “light” and “dark” powers which translates into healing, self-damage, and a lot of miscellaneous. (I assume in part to remain flexible since you have to take her on story missions)

When Ghost Rider knocks a guy into another guy, something many characters can do, he does it with a flaming chain, so it has some character-specific pizazz. My Hunter also knocks enemies around, but she does it by bonking them with her swords. When writing this I wasn’t sure if they were swords or tonfas, I had to look it up. She’s not a sword character in same way Ghost Rider is a chain character or Wolverine is a claw goblin, she just has swords she uses sporadically.

To put in bluntly my Hunter feels like an uninspired “OC” - a concept that doesn’t translate into a strong mechanical identity or personality.

Often in video game development you get stuck in something you know isn’t ideal. A frequent culprit is control schemes - you want the player to be able to sprint, jump, dodge and attack all at the same time, but you also have functions like block and parry, and you just can’t make it work with the available buttons. You have a nagging feeling that there’s a better solution but you never find it.

I think I understand why The Hunter is the central character. It allows for an original story crafted to fit the game without baggage. They have a large pool of cards, their powers are more broad, and you can tweak them with lightside / darkside points, so they have high mechanical flexibility - important since they fill one of your precious three slots in story missions. You can customize aspects like skin tone and gender, since they have no canonical appearance.

Allowing the player to play as any character, or to switch between them freely, would require a lot of work, and forcing the player into one specific existing character would be limiting. So I don’t know what the better solutions is, having pondered it for all of 20 minutes. But a better one has to exist - maybe have the player swap between different characters at different parts of the story, like the story modes in Mortal Kombat. I get why you play as The Hunter but I don’t like it, because again, The Hunter is the least interesting character.

Just let me play as Blade!

“It’s a Marvel tactics game that you play with cards” and “you don’t play as your favorite Marvel character, you play as a new guy we invented” both strike me as the dev team skating uphill.

Bloat, or, This Game Would Be Better if It Let Me Play It

So far it may sound like I’m lukewarm on Midnight Suns, but calling out issues without overstating them is important. I don’t like The Hunter as the protagonist but I wouldn’t care too much were the rest of the game fun. And the battle is fun.

Missions have a satisfying level of difficulty (I turned it up when that option became available) and are lightweight enough to prompt a “just one more mission” feeling. If the rest of the game was similarly lightweight relationship management it would be easy to recommend.

But that’s not what the rest of the game is.

The Quest for Content

Many older games in this genre, like Military Madness or Fire Emblem, were virtually all battles. You skipped from one mission to the next with some light exposition between. Other games had light exploration, crafting, outfitting or customization between fights. In Shining Force you explored an RPG-style map with towns and an overworld. In X-COM: UFO Defense you autopsied aliens and performed research. But in my mind nearly all those old games were primarily battle games.

Full-priced games in all genres have become meatier. At the same time, individual battles in tactics games are often slimmed down for accessibility and modernization reasons. Compare GBA-era Fire Emblem games, with their waves and waves of reinforcements, to the newest ones. This leaves developers in a bind - they need to add more “content” while slimming down the primary content.

Between battles in Fire Emblem: Three Houses you spend action points on different activities, set goals for and instruct students, and engage in conversations and diversions like fishing. That’s all engaging content, but apparently not enough content. So there are also chores. Scattered glowing orbs to happen across and lost items to find and return. Each month, to play optimally, you run through the entire monastery, searching every room and corridor, spending much time while extracting precious little fun.

Fire Emblem: Engage has even more home-base chores. There’s food to cook, locations to visit, training, mock battles, choosing farm animals, picking up their products. Almost none of these chores pose satisfying decisions, mechanics or aesthetics. They’re rote tasks - don’t forget to stop and feed the weird mascot character, because feeding them gives you a currency.

Midnight Suns is, unfortunately, one of the worst offenders in the misguided quest for content. It has bite-sized battles but a 40-hour critical path, ballooning to 80+ for completionists. So much of that time is busywork. Collecting respawning glowing orbs of different currencies. Checking the library each day for new book passages that might contain “Arcane Knowledge.” There are reagents to collect. Some story bits are gated by these - to attend mystical meetings you first have to deliver 10 mushrooms. Petting your dog does something tangible- I forget what - so every day stop and pet the dog. The gameplay in these sections is move around and press A at a prompt.

There are other activities I can’t honestly call chores, but ultimately are a chore. Most missions give you Gamma Coils, which you can open for new cards. That means jogging (there’s no running indoors) to the forge and watching a brief cutscene to open your spoils. At the forge you also queue up new research, which you do each day as research takes only a day to complete. Once you’ve opened a Gamma Coil and have new cards you now jog to the training area to combine them. In the training area you can also spar against training partners. These aren’t one-on-one battles; you select an opponent then get a still image listing your rewards. If you got new Intel Caches (another currency) you jog to the War Room and spend them on autonomous side missions. The Abbey, your home base, quickly becomes a laborious 3D menu.

The Abbey grounds have a Metroid-style setup in which you unlock new powers to access new areas. The grounds hold not only mushrooms and weeds to collect but journal entries from a half dozen different characters. There are also capital-M Mysteries strewn about - puzzles to solve that offer rewards or lore. The second time I discovered a mystery and realized that mysteries were a formalized game system it seemed neat. But to solve that mystery, the Mystery of Who My Father Is?, I just walked around and pressed A at prompts. When the mystery was solved the answer to “who is my father?” was “let’s talk more later.”

There are chests containing random cosmetics. To open them you need special keys, another currency. You can hang out with characters to build relationship points, or spend yet another currency, Compliment Points, to butter them up. There are various clubs: a spooky kids club, shop class, a book club. At the spooky kids club you recover your past memories - I did two or three and didn’t recover any memories of note, but top men were working on it.

Besides story missions there are side missions, and they all give you some resource you want. I suppose you could skip them, but combining cards is a primary deck-building mechanic, and to combine them you need currency. So I suspect many players won’t skip them.

The ratio of missions to everything else is way off, and the ratio of story missions - the more authored ones that involve plot movement, scripted moments, and memorable villains - is even lower. I played Midnight Suns for 15 hours and did maybe 3 or 4 story missions, which did almost nothing to advance the plot.

Please God, Just Let Me Play The Video Game

I’m not one of those people who gets angry about games that aren’t “real” games, or who thinks that “mechanics are everything.” I’m using “play” very loosely here: “let me play the game” means “let me engage with the interesting parts.” In Resident Evil that’s shooting zombies, but also includes watching cutscenes, because the cutscenes are entertaining. Playing a walking sim means trudging around and listening to audio logs, and “playing” a visual novel means reading. That’s fine - that’s the content you signed up for.

What isn’t fine is playing a Mega Man spiritual successor that’s constantly interrupted by 10-minute voiced animatics.1

The problem with Midnight Suns is I spend so much time not playing the game.

There’s too much talking for my tastes, or rather, too much fat to the talking - heroes explaining their powers and backstory, plot-centric conversations that don’t advance the story, slice of life moments that don’t reveal much about the characters. But I don’t dislike social sim elements, and I don’t mind casually conversing with Blade about his crush on Captain Marvel. Unlike some critics I don’t loathe the writing and I don’t dread social interactions.

What I dread are the chores.

Every day I wake up and do chores. Then do one mission - if it’s a side mission it might take 10 minutes. Then it’s evening and I do more chores. Then go to bed, wake up, and begin chores again. There’s so much fat that the non-chores - content like mysteries that should be engaging - become chores. If the game was battles, revealing conversations, and every 4 hours you solve a new Mystery those mysteries would be a pretty cool change of pace. But it’s hard to appreciate individual ingredients on a 47-topping pizza - in practice the mysteries are just part of the pile.

When I Stopped Playing

The last time I played I did a battle challenge. It was fun, for the 6 minutes it took to complete. On completion I unlocked a new navigation ability and a new area to explore. I wandered around collecting plants and came across an old church and a Mystery contained within. Strewn about the church were journal pages. Here’s a page from my own diary. Here’s a page of Wanda’s journal. Here’s a page of The Caretaker’s journal. Here’s a page of Agatha’s journal. (These people love journaling!) Here’s a page from Shaw’s journal. That was the first time I’d heard of Shaw - he’s also big into journaling. Also, here’s a painting lying on the ground for some reason.

I came across a mystical barrier - interacting with it played a small cutscene that said go talk to Doctor Strange. That meant I had to go back to the Abbey, have a conversation, then presumably come back here. Instead I decided to explore onward. I collected more mushrooms. I came across another mystery, which I hurried past. There were spots to use my new navigation power on to open routes or uncover hidden items.

Finally I reached the end of this winding path. And I was left thinking:

1. I still have to talk to Doctor Strange to advance this mystery. It’s not clear what the mystery I’m investigating even is exactly, so I’m not invested in it.

2. There are many spots on the map I can (and should) now revisit with my new traversal ability. Traversal abilities aren’t swinging around like Spider-Man or doing super jumps like the Hulk - both my abilities effectively open doors. Exploring the Abbey grounds isn’t like exploring in Super Metroid; it’s not mechanically engaging. You just jog around.

3. When I first entered this new area I had a choice to go north or south, and all of this was north. I haven’t gone south, so that’s more to explore.

4. I’ve been at this for 20 minutes, doing the rest is going to take another 20 minutes, and none of this is fun. It’s not battle, it’s not conversations, it’s not strategically engaging and it’s not fun for the fingers. It’s collecting mushrooms and reading lore.

5. When I’m finally done with all this my character will go to bed then wake up to more chores.

So I turned it off.

I’d Like an Abridged Version

The answer to “how could the developers fix up their game?” is nearly always “spend more money.” The Final Fantasy 16 demo left me lukewarm. It doesn’t have enough RPG elements to work as an RPG, it doesn’t have the stats and build variety to work as an action-RPG, and the combat’s too simple to work as a character action game. Improving that would require money - multiple playable characters with different movesets, different weapon types, maybe a full paper-doll equipment system.

The frustrating thing about Midnight Suns is that it’s a game that could be improved simply by removing elements.

The combat design is rooted in minimalism or “subtractive design” - when it doubt keep features out. The rest of the game is the opposite - it’s as if they held an all-hands meeting and said “everyone go around and throw out an idea for Abbey activities and we won’t say no to a single one of them.”

If they had cut some of those features and focused more on the rest (more combat variety, more plot momentum, etc) the game would be better. If they had cut some of those features and spent no additional resources on the rest it would still be better! The ratio of meaningful content to obligatory-feeling uninteresting content is just too high.

Maybe a 25-hour version wouldn’t sell as well, as it wouldn’t deliver the “value”-per-dollar that gamers demand. (Though presumably it would cost less to make) I don’t know - I’m not a sales and marketing expert. But a distilled version would be so much more enjoyable, and it’s sitting right there.

A game that’s half battles and half slice-of-life “Blade organizes a book club and gives a report on Little Women” may not be for everyone. But my criticism of the game isn’t the writing or the social aspect.

My criticism is this: I know devs want to be featured on those “can you pet the dog?” social media accounts, so I get that I can pet the dog. But why do I have to pet the dog, every day I see him? Midnight Suns has too much “content” by way of gristle that adds to the weight of a cut.

Subscribe now

as well as

I was originally planning on writing about dodging in Stellar Blade, but the dodging in it works fine, at least technically.

So instead I’m doing an overall evaluation of the demo, with a focus on control issues.

Evaluating Work-in-progress Builds

When joining an existing project one of first things I do is play the current build and give feedback on what needs improvement. I’m approaching the Stellar Blade demo the same way I would a work-in-progress build, though that evaluation would be a list rather than an essay.

I hope this is useful to game developers who haven’t played Stellar Blade, or those less versed in that style of game. For non-developers I hope this provides insight into the nuts-and-bolts of game development. Many people have a poor grasp of what game developers, especially designers, actually do, and even game developers often have a idealized view of design - for example the GDC “Game Design Challenge” is basically “the game idea challenge” and has little do with real game design. Day-to-day game design isn’t one big idea (though a splashy big idea can be helpful) - it’s continuous small ideas and improvements.

This isn’t a review of the game or the demo, it’s “here are some things that ideally would be fixed”, in enough detail to specifically identify the problems. Many reviews highlight “unwieldy” platforming, for example, without saying much beyond that. That’s fine for a review, but telling a team of developers tasked with movement and platforming that their implementation is “unwieldy” isn’t helpful. That’s the kind of feedback a CEO or playtester would provide, not the feedback a senior dev should provide.

I’m providing the sort of feedback a senior dev should (IMO) provide. It may come off as nitpicky, but being detail-oriented is part of the job. When working on a game I sometimes zoom in on a portion of the screen to diagnose a problem shader, run in super slow-mo to track down a one-frame animation glitch, or run at a capped low framerate to find time-dependent physics problems.

That said, offering detailed, specific (if in this case, unsolicited) feedback is different from being a dick. I’ve watched a few Twitch streams where game developers play a game and provide real-time evaluation, and in my experience those are both useless and off-putting. That format invites snap judgements and stream of consciousness babbling. And they often contain an undercurrent of “I could do this better than these idiots.” (See Jonathan Blow playing Breath of the Wild or Elden Ring, for example, if you enjoy wasting your own time)

There’s no implication here that the devs lack talent or that I could do better.

Part 1: Combat Controls

Dodging

Technically the dodging in Stellar Blade works fine. In the Lies of P demo your dodge would frequently get hung up on enemy characters, and if you queued up a dodge along with a direction the game ignored the queued direction. As far as I can tell there’s nothing like that here.

That said, the dodge seems almost totally useless by design. It has no invincibility to speak of, and it doesn’t move very fast or far either. I don’t mind the lack of i-frames - modern combat games sometimes lean too heavily on abusing invincibility windows. I like the idea of a dodge that avoids an attack with movement rather than with i-frames - something akin to Mega Man’s slide, which is quick and goes under some attacks but isn’t invulnerable. But the Stellar Blade dodge is slow and doesn’t move very far.

The balance between dodging and parrying is heavily weighted towards parrying. Parrying too early results in a block, and parrying successfully reduces enemy balance. (A stun meter, basically) If you mistime a dodge you get hit. If you succeed with a dodge…well you can’t really succeed because the dodge is so bad. The game really wants you to parry. The dodge seems only useful for avoiding unblockable attacks with a perfect dodge. Some enemies use color-coded attacks that you respond to using the dodge button, but those are more mini-QTE events than dodges.

You can dodge during most of an attack animation but not all of it - it’s more permissive than games like Monster Hunter and Dark Souls where dodges can only cancel recovery frames, but less permissive than games like Bayonetta where you can dodge at nearly any time. In Stellar Blade the rule of thumb is that you can’t dodge during the hitbox-active portion of the attack - the part where your sword can make contact and when the slashing VFX plays. (The most committed part of the animation) If you press dodge during this part of an attack the dodge is queued for later.

In practice it feels intuitive enough. I have no problem with it. That said, the game can feel unresponsive, and the fact that dodges during attacks are sometimes queued for later probably contributes to that.

General Unresponsiveness

The Stellar Blade demo has a lot of apparent input lag. It’s unclear how much of this is genuine input lag, where the engine takes multiple frames to respond to input, and how much it’s due to slow animation transitions.

Here’s a video of me trying to dodge attacks against a tutorial enemy.

Watching this video it’s hard to tell that I’m even trying to dodge, because the dodge animation takes a long time to kick in and isn’t easily recognizable. Sometimes my character slides to the side without animation - in those cases I think the dodge input has been processed and I’m technically starting a dodge, but the animation transition is slow or delayed. Other times my character plays some animation but it’s hard to tell if it’s a dodge animation or something else.

(Maybe dodging just doesn’t work at all against this tutorial enemy - but outside of this enemy it feels similarly unresponsive)

Blocking / parrying is similar. There’s a long delay between when you press block and when you character plays the block animation, and it’s unclear when in this transition you’re technically blocking. I think you count as blocking pretty quickly, so the blocking just looks slow - the animation transition lags behind the state change. You can successfully parry right before an attack lands, so at least parrying isn’t actually slow.

On the other hand, at least in the demo, attacking is genuinely unresponsive.

When you tap attack the move fires on button release rather than button press, which induces some lag as releasing takes extra time. Attacking on release makes sense for moves you hold down to charge, but these aren’t charged attacks.

Later in the game you unlock the ability to charge up basic attacks, so maybe that’s why attacks fire on button release. Typically games handle this sort of overlap (pressing a button does one thing, holding it does another) by playing the on-press animation then transitioning to the on-hold animation when needed. I believe that’s how launchers work in Spider-Man, for example: when you press attack Spider-Man starts his normal attack animation then as you hold he transitions into the launcher animation. For whatever reason here the animation doesn’t even begin to fire until you let go of the button.

Attacks also fire if you press and hold the button for long enough, but that length of time is very long. I assume that timeout also relates to the later ability to charge attacks. You’d never want to activate a standard attack this way.

(Note that this video is taken from the boss mission that unlocks after you beat the main demo area - maybe the controls in the starting area are a bit different since fewer abilities are unlocked)

Even when you tap the button as quickly as possible there’s noticeable delay on attacks. In Street Fighter 6 if I tap jab the reaction is effectively instant - I’m sure there’s technically a delay of a couple frames but it reads as instant. Tapping the button as lightly as possible in Stellar Blade still doesn’t produce anything close to an instant attack. Some people are more perceptive of this than others - to me the two games are a night and day difference.

So the general unresponsiveness of the game is a combination of issues:

1. There’s some level of engine-wide input lag

2. Sometimes your character technically enters a state like blocking or dodging but the animation doesn’t reflect that state change for a while, so it looks like your character is in the previous state (the developers have given interviews to this effect)

3. Attacks happen on button release and animate in slowly. It’s not clear if in this case you’ve technically swapped into attacking state and the animation is lagging behind, or if the state change itself is slow. It doesn’t matter much, since being in attack state isn’t particularly meaningful.

4. Dodges or blocks during certain parts of attack animations are deferred by design, to make the animations looks nicer and/or to introduce more commitment to attacking. This is a perfectly fine design choice but blends into the other issues.

Inconsistent Input Buffering

The simple explanation of an input buffer is it allows players to input moves a bit early and still have them activate, rather than forcing the player into frame-perfect inputs. Adon in Street Fighter Alpha 2 can combo two crouching strongs.1 This isn’t a “combo string” or a “chain combo” or anything fancy - you just press strong once, then press it again. If you press it too late the second attack activates but doesn’t combo. If you press it too early - while the first attack is still animating - the second attack won’t activate. Precise timing is required.

With an input buffer pressing the second attack a little early would queue it up and fire it off when it’s available. Input buffers are the norm in 3D fighting games and in action/adventure games; you don’t have to worry about tight button timing in Elden Ring. The input buffer in that game is so generous that it can cause problems like accidentally rolling twice or consuming two potions.

There are many variations on input buffers, including some that aren’t actual buffers at all, but the key point is that they make player’s lives easier by honoring their intent. The game does what the player tried to make it do, even if their timing wasn’t exact.

I’ve read complaints from players that Stellar Blade sometimes ignores their input. Some of this is surely due to the general low responsiveness, but the input buffer is also an issue. The input buffering in Stellar Blade feels ad-hoc rather than systemic, not always reflecting player intent.

First, here are the things that work how I’d expect:

• During an attack string you can queue up the next attack in a combo

• During an attack string you can queue up a dodge or block and it will fire off when available

• You can queue up a dodge during a dodge (and a block / parry during a dodge as well I think - I left it out of my notes)

But other things don’t work as expected:

• You can’t queue up an attack or jump during a dodge. So if you press dodge and then try to attack out of the dodge nothing happens if you attacked too early.

• You can’t queue up a jump after an attack

• You can’t queue up an attack after a jump - if you press jump and weak attack in quick succession you get a jump followed by nothing, rather than a jump followed by a quick airborne attack

• While landing from a jump you can queue an attack but not a dodge

Most input buffering systems have some special cases, but those special cases try to be invisible and exist to make the system work as expected. Stellar Blade’s input buffering feels like a collection of visible special cases. Sometimes you can queue up either an attack or a dodge, sometimes you can queue up an attack but not a dodge, and sometimes vice-versa. Jumping doesn’t seem to participate in the input buffer at all.

All the things that don’t work can feel like eaten inputs - you jump, land and dodge and the game ignores your dodge input, because for some reason dodge inputs aren’t queued in that scenario. Many players will interpret that as “I pressed dodge but nothing happened.”

Subscribe now

Part 2: Movement and Traversal

I wasn’t expecting to have so many issues with movement - character-action games tend to have simple collision geometry and limited or QTE-style movement options, so there’s a low limit on what can go wrong in them. Stellar Blade has complex geometry and movement options, with both systemic and QTE-style traversal, and the end result is ambitious but falls a bit short on execution.

Lack of Fine Movement Control

This video illustrates problems with fine movement control.

Reversing direction while running usually produces a sliding stop that keeps you moving along the same line, but sometimes you get a wider looping turn. (Presumably because while reversing direction your thumb strays a bit - skill issue)

It’s difficult to accurately move short distances when reversing direction. (Second section of video) If you quickly press the stick fully in a direction Eve does an immediate run-then-stop, which covers a fair distance. If you gently press the stick then quickly release she moves less far, but still drifts forward after you let go of the stick.

During the demo I sometimes had trouble lining up with treasure chests, keypads, doors, etc. Not a lot of trouble, but enough to be slightly annoying. The third section of the video shows one source of this problem - if you try to turn 90 degrees while your shoulder’s against a wall you can’t help but move laterally as Eve pivots on one foot rather than from her center point.

The last section of the video illustrates ledge stickiness gone wrong. If you’re near a ledge the game will helpfully make you grab it when you jump, but this help is too aggressive and kicks in even if you’re moving away from the ledge.

On their own none of these are significant problems, but together (along with other issues I’ll get to) they add up. I’ve seen reviews call out the platforming “unwieldy” and I can see why - you don’t have fine control of your character. When capturing video for this blog I noticed many places where I accidentally climbed a ledge or got momentarily hung up on an object. This is the kind of control I’d expect in a more “cinematic”, animation-driven, root-motion driven game, rather than in an action/platforming game.

A consistent issue with Stellar Blade’s movement and traversal is that the systems and level layouts don’t always mesh. Trouble lining up with interactive objects could be fixed by giving the player finer control, but it could also be fixed by making those objects more permissive in terms of required distance and angle. The more animation-centric approach isn’t necessarily bad, it just doesn’t always play nicely with level layout.

Something not in the video is that you can’t control your jump height (at least, not much?) by releasing the jump button early, and the jump steering isn’t great. The demo has a few parked cars that I tried jumping on and over - landing in the spot you want isn’t easy. Sometimes I came up short, jumped too far, or landed where I wanted then slid off. Again that would be fine in some games, but less so in this context.

Movement Inconsistencies

In the first part of this video I’m running alongside a bus while gently pressing into it. I tested this because a) I noticed that Eve sometimes plays a one-handed wall-touching animation while running next to a surface and I wanted to understand when that happened, and b) I noticed that running alongside this bus wasn’t smooth. Sometimes while running I got hung up on the bus and stopped moving until I adjusted the stick. (Presumably because the camera rotated a bit and now I was pressing too towards the bus) Sometimes Eve rapidly jittered while moving, getting stuck for a frame or two at a time. Other times I overshot the end up the bus and rounded the corner by mistake. Running alongside this bus feels rough.

I was never able to reproduce the one-handed glide consistently.

In the second part of the video I try to jump on the bus. This works the opposite of how I’d expect - if I push the stick towards the bus and jump I don’t vault onto it, but if I release the stick entirely then jump Eve grabs on and pulls herself up. The bus is tilted slightly towards the camera - maybe that has something to do with this behavior. But again, it’s unexpected.

In the third section of the video I tried to reproduce another neat animation, where Eve stops and pushes a wall with two hands. I’m sure there’s some logic to when it plays but it feels pretty random - maybe getting a running start into the wall helps a bit but I’m not sure. Intuitively this animation should play when you push directly into the wall - when the stick direction is facing the normal of the wall. In practice who knows.

This is mostly a cosmetic issue (at least in the demo) but it’s another example of inconsistency. But it’s not purely cosmetic - that animation fixes Eve in place rather than doing the normal collide-and-slide.

The last section of the video highlights issues with what I call the hard-landing animation. When you land from a jump Eve plays an animation that looks like a hard landing, but this animation typically doesn’t lock out other animations. It’s a nice cosmetic animation that sells the feeling of landing without affecting the gameplay.

Except sometimes the hard-landing animation isn’t just cosmetic, and does lock out other animations. In the video jumping into a wall forced a version of the hard-landing animation that blocked other animations. Normally you can repeatedly jump by mashing the button, but when pressing into this wall you can’t jump until the landing animation plays out.

In all these cases the game either isn’t doing what I think it should be doing based on my inputs, or what it’s doing is less predictable than I’d like.

Unclear Interactive Surfaces

In the first part of this video I notice a platform, try to jump onto it, and discover it’s surrounded by invisible walls and isn’t interactive.

In the second part of the video I test to see if I can grab a ledge. I can’t - the ledge is purely cosmetic. The lighting in this scene misleadingly highlights the ledge and draws the eye. You can also see other movement issues in this part of the video - while trying to line myself up for the jump I accidentally climb a small pillar, but while running to make the jump I don’t climb a similar pillar I bump into.

Some games have a “what you see is what you get” philosophy to geometry. In Super Mario Brothers or Dark Souls if a surface looks like real geometry it probably is - it has collision matching the shape and you can walk on it, wall jump off it, etc. On the other side of the spectrum sit most Platinum Games offerings, where complex geometry is mostly for show and blocked off by simple invisible walls. In Bayonetta if you’re in an alley with parked cars and overturned tables and brownstone door entrances you’re probably actually in a simple rectangular solid, and all those adornments are just graphics. You can debate these approaches but they’re both learnable.

Stellar Blade is less learnable. Sometimes ledges and platforms are purely cosmetic, but in other places oddly shaped junk piles have matching complex collision geometry. You can clamber up yellow-painted ledges but you can also clamber up normal unmarked ledges - sometimes. Sometimes those ledges have climbable collision geometry and other times they’re just graphics.

Complex Geometry Issues

This last video illustrates some issues with how the character controller and complex geometry interact.

In the first half of the video I jump at a ledge but don’t grab onto it. At a slightly different spot it works. Once on this junk pile Eve behaves erratically - sometimes she floats in mid-air for a while then plays a hard-landing animation. Running into the nearly vertical wall behaves poorly.

In the second half of the video I’m trying to jump around a corner from a junk pile to a walkway. (To be clear you don’t have to do this to progress, I chose to do it) As you can see the video looks awkward and unpredictable. Sometimes I float, other times I’m forced into hard-landing animations for no real reason. I fall into holes, bounce weirdly off surfaces, and get hung up on small outcroppings.

Traversal Issues Wrap Up

There are odd things going on with the character controller and with the collision geometry, but the problems arise more from how these things don’t play nicely together.

The traversal in Stellar Blade is both systemic and QTE-style. When you interact with yellow painted ledges the game uses your input as a broad suggestion to play appropriate animations, similar to games like Uncharted. If ledges aren’t lined up properly Eve will automatically do a diagonal jump. For ledges far apart she’ll do two hops. I’d call this “cinematic” traversal - the designers painted ledges with a special tool and the game plays carefully authored animations for pre-planned scenarios.

But the game also supports systemic traversal. You can vault onto and over surfaces that presumably aren’t specially marked. There are lots of nice traversal animations - Eve can hoist herself up, do a spinning jump, a fancy flip, etc. But interacting with the environment outside of the yellow-painted routes is often awkward as the character controller doesn’t handle complex surfaces well. Or maybe the surfaces are simply too complex for their own good, full of odd angles, jutting overhangs and sharp crevasses.

The end result is that Stellar Blade appears to support free-form systems-based traversal, but trying to use the systemic traversal systems on non-planned routes is less satisfying than it should be. In Breath of the Wild I can climb anything and for the most part it works pretty well; in Stellar Blade I can traverse a lot of things but I don’t want to, because it’s not fun.

It sometimes annoyed me in Bayonetta when an area looked complex but invisible walls shaped it into a cube or circular arena. In Stellar Blade areas look complex and are complex, but I construct my own invisible walls around them - “I can probably climb that junk pile but I’m not going to.”

Maybe this isn’t a huge problem - it’s a combat-focused game after all, not a rock-traversal simulator. But these are the things I’d call out if I was reviewing the work of a traversal team. (And also, apparently the final game has a fair amount of traversal that’s gotten mixed reactions)

Part 3: Odds and Ends

The first two sections were all things I’d bring up if I was brought onto a project midway through development. This last section includes less detailed observations, as well as a some more subjective and personal thoughts I wouldn’t include in that sort of review.

Combat

I was a bit disappointed by the combat system in a few ways. I’d read that the combat was halfway between a Sekiro and a Granblue Fantasy: Relink, which sounds appealing, but I don’t get that impression from playing it. The game has combo strings but these aren’t true combos and there isn’t much juggling or air combos.

Your attacks don’t induce hit stun against most enemies in most situations, so the combat is reactive - you wait for the enemy to attack then parry or do the appropriate color-coded response. But later in the game you apparently unlock more moves that hit stun reliably. So some players interpret the game as mostly reactive, and others complain that bosses can barely fight back. Either way, I was hoping for more of a fighting-game cadence - something between the “hit the enemy twice” of a Souls game and the “air juggle the floating enemy 200 times” of a button mashing ARPG. Something more like Tekken or Street Fighter or even Streets of Rage 4 where proper defense is important but you can unleash moderately fancy combos with the right setups. I can’t blame the game for my mismatched expectations - but I still had them.

I prefer when you play the game rather than what I call “playing the UI.” Avoiding the shadow of a giant meteor attack is playing the game; avoiding a giant red circle on the ground is what I’d call playing the UI - more MMO than action game. (Admittedly this is a blurry distinction)

In Stellar Blade you press towards and dodge for blue attacks, away and dodge for purple attacks, and perfect dodge out of yellow attacks. These indicators are similar to attack indicators in more recent God of War games or Gotham Knights. I prefer “you can do a special dodge through any enemy attack that looks like a forward lunge” to “you can do a special dodge for blue attacks”, as the former demands observation and learning. The latter is little too Simon Says, and a little too lock and key2, for my tastes.

My main complaints with combat, other than responsiveness issues, is that it wasn’t quite what I was looking for - which again, is more a me issue.

Plot and Theme

Maybe this is because Helldivers 2 and Starship Troopers are on my mind, but the opening of the game struck me as a farce. Fat ships deploy tiny shields that uselessly cover only their very front. When easily-shot-down drop pods manage to reach the surface they malfunction, and when they do work women without armor or ranged weapons emerge to die.

It’s not fair to judge the plot from the demo, but it didn’t grab me, either in narrative or theme. Astral Chain has a mundane plot but a cool theme, with anime space cops and a stylish, fitting UI. I have no idea what the plot of Lies of P is but it presented a moody, potentially interesting world. Playing Stellar Blade I wasn’t invested in the mystery of the Naytiba and assumed it would play out along an expected path.

Horndoggedness

I’m not particularly concerned with sexuality and horniness in games - I’m not wringing my hands while watching long video essays about problematic sexualization. (I own Rumble Roses for PS2 - no regrets) That said, while it’s a horny game it’s not a very sexy one - in particular many of the outfits are gaudy. Eve’s combat poses, particular in the legs, look ungraceful - a Chun-Li stance from someone unused to heels. Purely on aesthetics, without any moral calculus, it’s more X-Blades than Nier. It did not spark passion in me the same way some of my Tubi recently-viewed list does.

Culture War Nonsense

“This is the game that’s saving gaming” is, I get it, a silly culture war battle cry, not a serious statement. But it’s a ridiculous statement even given that. Stellar Blade embraces dominant trends, with yellow-painted ledges, a constantly-commentating Navi-style helper character, Detective Vision (every game must have that, it’s the law), familiar skill trees and upgrades and side quests, block pushing and other more-procedure-than-puzzles puzzles3. It’s a well-executed but familiar template, not a breath of fresh air. That’s fine - I’m looking forward to Earth Defense Force 6. But the idea that it’s a harbinger of change is frankly absurd. If other games take cues from Stellar Blade it will probably in the form of something modest like more elaborate combo strings than what’s typically in a Sekiro-like.

And Now, The Conclusion

I went into this expecting to write mostly about dodging. (That’s the gimmick of this series, after all) On a technical level I’m pleasantly surprised by how dodge is implemented.

The input buffering is noticeably ad-hoc but most games don’t totally nail input buffering, and while Stellar Blade’s isn’t ideal it’s not bad.

The apparent input lag - the mix of real engine-level input-lag, slow animation transitions, unclear state changes, and actions on button release rather than press - is bad. Some people might not notice it the way some people don’t notice frame rate, but like improving frame rate, improving input lag (which was reportedly done in the final version, though I’ve heard mixed impressions) would probably make the experience better for everyone.

What stands out to me most, surprisingly, is how much I noticed control quirks - once I started noticing them I couldn’t stop. (I have more footage of traversal issues I’m leaving out) One reason “waggle controls” lacked staying power was the lack of control fidelity - getting the wrong move in Mad World 5% of the time is incredibly frustrating, even if 5% sounds low. Too often in the Stellar Blade demo I found my character stopping too slowly, climbing something I didn’t want to climb, jumping too far, moving to the side when I wanted to move straight ahead, etc. In a game like Uncharted that would be fine, but I draw a distinction between a cinematic adventure game and an action game with more manual platforming. It’s not awful and traversal isn’t the star of the show, so it’s hardly a deal breaker. It’s not dodging in Callisto Protocol, which for me genuinely did sour the experience as whole. It’s just not ideal.

Again this isn’t a review - the point here isn’t to deduct points or “prove” that the game is bad. It could use work but every shipped game could use work. These are simply the small craft issues that stand out to me, especially from the perspective of game feel. Making the best game you can involves working through thousands of small issues like these as best you can.

(And sometimes, knowing when to ignore these issues entirely and move on, but that’s a topic for another time)

It’s rare for AAA games to have noteworthy high-level design.

They have low and mid-level design. Level designers lay out spaces, combat designers come up with new systems and movesets. Someone designs enemies and bosses and guns and swords. Someone makes a skill tree. But at a high level many AAA games play like other AAA games, mixing the same design elements in slightly different ways.

I’ll be comparing Helldivers 2 to Suicide Squad: Kill the Justice League a few times throughout this. Suicide Squad is a good example of a game I’d say has no real high-level design. Combat has some wrinkles like shield harvesting (a clunky version of Doom’s Glory Kills) and melee attacks that transition into gun juggles - someone had to design those, sure. Someone worked hard on elaborate areas like the Batman Museum and the individual traversal mechanics. But the high level vision for the game is “it’s an open-world mission-based GAAS looter-shooter - get off my back.”

Suicide Squad had 4 different director positions. (Maybe that was part of the problem!) I’m sure the directors had plenty to do in terms of managing production, but I suspect if you polled a group of open world looter-shooter players and averaged the results you’d get the design of Suicide Squad, more or less. It’s “travel to a location and defend some points for 30 seconds” missions with superhero themes on top. The toughest high-level design challenge it poses is how do you reconcile a looter-shooter with superheroes? That’s a nut they didn’t even try to crack - it’s really not a superhero game at all, it just uses superhero skins and names. In that sense it’s well behind the much-maligned Avengers.

I don’t think you could poll a group of players and end up with the design of Helldivers 2. It makes particular choices rather than a series of default choices - which isn’t that common in AAA or even AA games.

Someone on Batman: Arkham Asylum invented Detective Vision. As a AAA designer working today your job isn’t to invent your own unique element similar in spirit to Detective Vision that will become an industry-standard feature moving forward; instead your job is to copy Detective Vision, give it a new name, and invent some in-fiction justification for why it appears in your game. Lara Croft has strong instincts so she has “instinct vision.” In the upcoming Wolverine game Wolverine, with his strong animalistic senses, has Smell Vision. (Probably not the name they’ll go with) In the comics Wolverine uses his senses to investigate, sure. But let’s not kid ourselves - if Wolverine didn’t have super-senses they’d come up with another rationalization for including a vision mode. Spider-Man also has Detective Vision in his games, because reasons.

That’s how AAA design has trended for some time: borrow “tried and true”, “industry standard” ideas and combine them in not-particularly-interesting ways. And often water down the borrowed ideas. Ubisoft’s take on Breath of the Wild, Immortals Fenyx Rising, is the game dialed back from 10 to 7. Wherever Breath of the Wild made an interesting choice Immortals replaced it with a more conservative one.

The big choices in AAA games tend to be in narrative, technology, art and budget, not in design. I suspect that’s one reason big publishers are slow to jump on trends like Minecraft and survival-style games - those are genres that trade in game design, not face-scanned models and motion capture.

I don’t demand innovation - it’s fine that some games are comfort food, just more of the same but executed well. But it’s refreshing to play a game with genuinely new ideas, or at least new combinations of old ones. It’s nice to play a game where people with job titles like “creative lead” did something creative design-wise.

Which brings us to Helldivers 2.

The Design of Helldivers 2

The other day I read an interview about Dragon’s Dogma 2 and the design of fast-travel options. It struck me how rare it is to hear game creators talk about game design specifics. Not in a hand-wavy way like “we want the player to feel powerful emotions and a sense of purpose”, and not about low-level mechanics like “we want players to move constantly in combat which is why enemies throw grenades.” But more “this is the specific design choice we’re making to support our overall philosophy.”

Helldivers 2 is a game brimming with design choices that support the overall thrust of the game. I keep returning to one thought: if this were a game with a higher budget and higher sales expectations it would look very different; the interesting decisions and rough edges would be sanded off.

Friendly Fire

Helldivers 2 has friendly fire damage and that damage is high. It adds unpredictability but also frustration and “unfairness”, two things AAA games are keen to avoid. A mid-budget game can get away with “if you don’t like the friendly fire play another game, we find it funny”, but games with higher budgets eschew that approach.

There are “rational” arguments for toning down friendly fire. I’m sure someone could engineer a playtest session that “proved” that friendly fire is a bad idea, or have a “player experience engineer” well versed in “player psychology” make a presentation about how friendly fire “interrupts flow states”, “penalizes players for factors outside their control” and introduces “friction-inducing stressors.” And the higher the budget the more likely this is to happen, in the same way that increasing budget increases the likelihood of a side character saying “watch out! Those bugs spit acid!”

Stratagems

Stratagems are special abilities in your loadout - to use one (on controller by default) you hold left bumper and enter a code via the directional pad.

To enter the code you either have to take your thumb off the left analog stick, rendering you unable to move, or you have to use your right hand to enter the dpad code (awkward!), or do something even more exotic like take over the left stick with your right thumb.

This is non-ergonomic and “inaccessible” by design. It’s error prone, and if you enter the code wrong you have to start over. Sometimes in the heat of battle you enter the last bit of the code wrong, don’t realize it, and throw a grenade at your feet thinking it’s a stratagem.

Some designers believe that control difficulty is an invalid avenue to explore - that any control mechanism that takes effort isn’t part of the game but rather gets in the way of the game. They’d resent that the complexity of the codes is a balancing mechanism; stratagems that are either more powerful or intended for use outside of combat have more complex codes.

Then there’s the crowd that believes that controls have been solved per-genre and should be “standardized”, so Helldivers 2 should just do whatever Call of Duty does. And if that’s not possible because the game has different features those features should be changed to be compatible with an “industry standard” control scheme.

There’s little chance this system would survive user testing at a AAA studio. Most testers would frequently mess up the stratagem inputs, which would be flagged as a problem, and when the game shipped you’d hold a button to open a radial menu then select one of 8 possible stratagems from a wheel.

General Obtuseness and Reliance on Diegetic Elements

I died multiple times in the tutorial when I was supposed to crawl prone past turrets. One time I went prone too late, another time I dove then immediately stood back up. That’s not a problem but it is an early indication of a design tenant throughout: you’re allowed to fail.

The game is unfriendly, or at least seemingly ambivalent.

As I wrote in On Unfriendly Games

The designers of friendly games are helicopter parents. They show their love and affection by constantly intervening on your behalf. They are benevolent gods, and while they aren’t standing over your shoulder they help you through their angels of algorithms and floating arrows.

The designer of an unfriendly game is a Deist god - they set the world in motion then go drink a margarita. They show their love and affection by saying “I have faith that you can do it on your own little buddy.”

Helldivers 2 expects you to muddle through things. How useful equipment and stratagems are, or even what they do, is determined mostly through experimentation. Map objectives are often confusing on first encounter - the terminal tells you to “adjust radar dish” but it’s not clear whether to turn if left or right or by how much, or even where the radar dish is exactly. Some objectives have you turn multiple valves - one player will turn a valve to the on position, then another player will come by and mistakenly turn it back to off.

The other day I played a mission where 2 members of our team died a bunch then quit in the first couple minutes. At one point I also died and spent time observing my remaining teammate - he crawled on his stomach and crouch walked around, killing robot enemies one at a time with a pistol. The game explains nothing about aggro mechanics, lines of sight, noise, whether you’re less visible prone or crouched, etc. You could play the game for dozens of hours and never know that stealth is even an option.

Enemies have weak points but they aren’t glowing and obvious (well, ok the robot weak points do glow somewhat). How weak those weak points are exists on a spectrum - a thick frontal shield takes no damage, legs take some damage, back takes a lot of damage. The interaction of weapons and armor is left unexplained - again you just have to try things out in real-world conditions and see for yourself.

To destroy a robot factory you can throw a grenade through a vent, but you can also throw a grenade through the door as a robot walks out. That’s just something you might try or luck into. Some non-explosive weapons can destroy robot factories if you bank a shot into a vent by bouncing it off the metal covering - again something you might not realize even after dozens of hours.

I often say that in unfriendly games the hand of the designer is less evident, but what I mean is that the guiding hand is less evident. There’s no shortage of evident design in Helldivers 2 but there’s little sense the designers are helping you along. There’s not much indication that they set fresh play-testers down in front of the game then freaked out when they struggled.

Suicide Squad: Kill the Justice League constantly tells you what to do. Here’s an example of the Green Lantern fight. The top-right says “destroy construct defenses” - during this part of the fight Green Lantern summons “adds” that you have to destroy. A similar message sits in your face at the center of the screen: “destroy Lantern’s constructs to remove his shield.” The game designer is telling you exactly what to do. Green Lantern also has dialogue hinting at what to do - that’s three different pieces of onscreen text telling you how to beat the encounter. Character barks in this fight also function as hints - members of the Squad shout out (roughly paraphrased) “his shields are down, blast him” or “destroy his summons!”

This is a game deathly afraid of players not getting it. Many AAA games are designed with fear rather than confidence - what if a player gets stuck and then they hate the game and then they return it or don’t buy the sequel?

In Helldivers 2 you can defeat a tank by throwing a few grenades on top of it but there’s no immediate way to know that - maybe the top of the tank looks vulnerable but it’s pretty hard to tell through the constant stream of gunfire melting your face off. Meanwhile Suicide Squad gives you 3 or 4 layers of redundant help at once - text on the right that says the same thing as text in the middle that says the same thing as a character bark.

Presumably in Suicide Squad players weren’t picking up on the natural and diegetic elements - it wasn’t clear when or why Green Lantern’s shield went down, which made the fight dull and frustrating. So they added more onscreen text and barks. But then the text covered the shield graphics and the barks masked the sound of the shield taking damage and going down, so the game became less readable, so they added even more in-your-face hints to compensate.

The result is that despite being a faced-paced shooter the game often revolves around following on-screen text prompts like in an MMO. And there’s so much text and bark hints that they amount to noise.

Commitment to the Bit

Helldivers 2 is a game with vision. But vision is nothing without commitment to it.

AAA games have a strong pressure to revert to the norm. Later Dead Rising games eased off on the real-time elements. Similarly Pikmin games eased off on time pressure with future installments, while Resident Evil moved away from limited ink ribbons for saving.

A few weeks ago a story circulated about how 343 Industries had pitched a Halo game similar to Helldivers. But even if the pitch really was like Helldivers it’s hard to imagine the finished product being similar.

Helldivers 2 commits to the premise with the mechanics highlighted earlier, but also in more subtle ways. When your character dies they don’t respawn, they get replaced by a different character. If all players die and fail to extract missions still count as completed as long as the objectives were done. The game has minimal instruction, and the in-game tutorial (which is canon for your character) doesn’t prepare you at all for what’s to come. All of these tie into the same theme of disposable soldiers.

Loading screen hints, in-game orders, informercials that play on your ship’s tv screens - all of them represent commitment to the bit.

An example of strong vision with weak commitment is Watch Dogs: Legion. On paper the premise is intriguing and promises all sorts of possibilities. Maybe your entire squad dies except for one elderly grandma, a humorous near-loss condition. In theory two players could have radically different squads based on who they recruit.

In practice I suspect most players identify a few core team members early who cover the bases then stick with them the rest of the game. While you can play as anyone much of that time your character is controlling a Spider Bot or other gadgets, so who you’re playing is immaterial. The elderly characters aren’t particularly different from other characters - they animate differently but still ride around on motorcycles and capably engage in firefights.

There’s a permadeath mode that forces you into new operatives, but it seems like an afterthought, with players frequently complaining that they lose characters to glitches or control issues.

The pitch is much more interesting than the execution.

Perhaps the all-time example of a strong premise the developers shied away from is Spore.

This is how early Spore is described:

In the extremely early versions that I toyed around with, I was able to make creatures that shifted under their own weight,” explained mflux. “Creatures that exploited the length of their arms or legs for greater reach. Creatures that behave and move true to how they were built. A short bunny-creature would definitely be out-run by the long-legged dragon-giraffe. That was very neat, and it implied several exciting possibilities in gameplay…The strategy that earlier prototypes implied went beyond placement of parts. The length of limbs or spine felt like it mattered. If you had a forward-heavy animal with legs placed in the back, it would run poorly as it tries (and fails) to counteract its own weight.”

And this is how it ended up:

Of course, the editor enables an infinite number of visually distinct creatures, but the gameplay effects of the creature parts are unfortunately quite discrete. The feet components each carry with them a canned set of attributes – for example, Stubbtoe gives “Sprint 2,” “Dance 1,” and “Speed 2” – regardless of the position of the foot, the length of the attached limb, or the shape of the body. Thus, the attributes of each creature is simply a summation of all the named body parts, and although the procedural animation guarantees that a many-limbed creature will walk convincingly, the player’s creativity in designing the creature’s shape has no impact on actual gameplay.

I would go further: calling it “procedural animation” is a bit of a misnomer. The animation in Spore is authored, not procedurally generated, such that creatures animate similarly regardless of limb number, length or shape.

The combat and other interactions are mostly stats-based - fighting an enemy is like MMO combat.

A group of devs working on Spore were concerned that players could create creatures that weren’t effective - that would have difficulty manipulating objects or feeding themselves or navigating terrain. Rather than let the chips fall where they may, shrug and say “that’s nature” (or: “don’t design your creature that way then”), the solution was to make that sort of failure impossible by reducing creatures to their core stats.

Meanwhile in Helldivers 2 sometimes you try to use a terminal and there’s so much glare from an overhead lamp that you can’t read the screen. That’s a fundamentally different, and refreshing, design philosophy.

Why Game Design is Hard

Most graphics programmers have a book shelf that looks something like this:

You can learn graphics programming from books. Not just from books - you need to experiment and do hands-on work. But if you read GPU Gems (and Jose Canseco’s Juiced…) you’re off to a good start, because graphics knowledge is highly transferable. Whether you’re working on a racing game or a 3D platformer or a first person shooter they all need to draw and shade triangles, they all use a z-buffer and transform coordinates from object to world to view space.

I’ve never read a game design book cover-to-cover and I don’t think that’s particularly rare among people who didn’t go to game design school. Of the books I’m familiar with some, like The Art of Game Design: A Book of Lenses, seem pretty good. Others, like Rules of Play: Game Design Fundamentals, seem pointless or actively harmful. But all of these books share the same problem: at best they can “help you learn how to learn” rather than teach you specific things, because games are just too broad. Video games encompass 12-hour AAA cinematic action-adventure games, 100-hour open-world RPGs, 3 hour visual novels and infinitely repayable falling block puzzle games.

Doing design work in a new genre often involves narrowly applicable specifics. For example here’s Masahrio Sakurai’s Eight Hit Stop Techniques video.

I particularly like this bit about the direction of hit-stop-shake:

When I encountered this video I’d just finished working through some character and camera shake issues. My first thought was to shake a character based on the direction of the incoming hit - so if they’re hit with a horizontal swipe shake them left to right and if they’re hit with a vertical one shake them up and down. But in practice that makes the character clip into and detach from the ground, which our gamer brains interpret as a technical failing. Similarly it’s sometimes best to constrain camera shake to two dimensions, so that it moves up/down and left/right but not forward/back.

You can spend years in games and never broach this topic at all.

If you’ve only worked on FPS games your first impulse in a platformer might be to use a capsule collider, only to find that produces unseemly results at platform edges. If you’ve never worked on an action or fighting game you may not have a solid idea of how to set up an input buffer or even that you need one.

It’s easy to step into a new domain thinking “how hard could this be?” only to find out it’s actually pretty hard, and that the available information is spotty.

Which brings me to the humble damage formula. At first glance it’s simple enough: you bonk an enemy and do some damage. Create an “attack” and “defense” stat, yada yada yada, job done. But we yada yada’d over the complicated bit.

The Setup

My goal here isn’t to find the “best” damage formula. Best is highly subjective and dependent on other aspects of the game. Instead the goal here is to examine different damage formulas, discuss their pros and cons, which types of games they’re suited for, and to develop a framework for evaluating them.

The formulas could be for an RPG (Dragon Quest), action-adventure game (God of War), or a tactics game (Fire Emblem). Any game where damage is based on a formula involving something like an “attack” and “defense” stat.

Formula 1: Simple Subtraction

Damage = Attack - Defense

Many older games, including Fire Emblem, use this formula or a basic variation.

This is a great formula for Fire Emblem. It’s easy for players to understand and calculate. Even though newer Fire Emblem games display a comprehensive attack preview, ease of calculation is still important as it allows players to plan many moves ahead.

That said, this is a formula with many potential pitfalls.

Imagine we’re a designer working on Elden Ring, editing a big Excel spreadsheet, and we get to the cell for the defense of a later boss. What number do we type in here? Note that if enemy defense is higher than player attack the enemy takes no damage.

Fire Emblem games typically don’t have full equipment loadouts; characters have weapons and maybe one accessory, so the range of values is limited. The games are fairly linear and have only minor character customization. At any point in a game the designers have a good idea of what the player’s squad looks like, and can pick appropriate values for enemy stats.

But in Elden Ring the player could be using a greatsword or a dagger. Characters have many gear slots with huge variety for each slot. Not only is Elden Ring not linear, but we want players to encounter areas and enemies in the “wrong” order. So what number do we type in for boss defense? 10? 50? 500?

In Fire Emblem the player controls an entire squad, so if a fast-but-weak character can’t damage an enemy the player can use a different character. In Elden Ring the player controls a single character; if that character can’t damage a boss that’s a catastrophic failure.

Designing with Users in Mind

A common trap devs fall into is forgetting the user. Or really, the two users: the player and the other team members. If you’re a tech artist creating an Unreal Material the environmental artists have to drag textures into slots and move sliders around. If you’re authoring a damage formula designers have to type numbers into an Excel spreadsheet that feeds that formula.

One of the best tips I’ve picked up is if you’re creating anything that other team members will use, pick field names and descriptions that make sense from their perspective, not from yours. So, for example, if you’re making a car racing game make the “top speed” kilometers-per-hour and not “units per second.” If you’re making a platforming game maybe that field should be “units per second”, where one unit is the width of one Goomba.

I have a fog formula that uses math stuff like natural log and my “fog density” field is often a number like “0.0032” - totally incomprehensible, even to me. I wish I had chosen to present it as “distance at which an object becomes totally obscured by fog.”

Sometimes a system may not be a good fit for the users, even if you try to present it in easily-understood terms. If you tried to use the Fire Emblem damage formula in Elden Ring I suspect this is what would happen:

1. Designers would type numbers into spreadsheets then run into cases where the player can’t damage enemies at all

2. They would try to fix those numbers, only to find them broken again with a different stat and item build or when they ran through the game in different order

3. After going through that a couple times designers would give every enemy low defense because that’s less error-prone, and to make a tough enemy they’d just crank up their HP

Further Objections to Simple Subtraction

If you have just enough attack to do 1 damage, adding 1 more attack doubles your damage output. If you already do 10 damage adding 1 more attack is only a 10% increase.

For Fire Emblem I think is mostly a theoretical objection. “Time to kill” as a metric doesn’t really make sense in this context. Fire Emblem isn’t about having one character continuously hit another character. Instead Fire Emblem is about breakpoints - if an enemy has 10 HP, your archer does 2 damage and your knight does 7 that leaves you one damage short. In that case adding 1 point of attack to either of your characters hits that same threshold, even if adding damage to the archer theoretically changes time to kill more dramatically.

Going from 1 to 2 damage is a huge increase, but going from zero to still zero (because your attack is lower than enemy defense) does nothing at all. So sometimes adding attack to a low damage unit is very valuable, but other times it’s totally valueless.

There are similar issues with defense: the less damage a character already takes the more valuable defense is. If a unit only takes 2 damage a hit, adding one point of defense doubles their survivability. High-defense characters exist to sit on the front lines and take multiple hits, so “time to kill” is more relevant here.

But Fire Emblem units with high defense tend to have low total stats, including speed and magic resistance, so if “defense is overpowered if you already have high defense” is true in some sense it’s rarely an issue in practice. It is something to keep in mind, though, as it will come up again with later formulas.

Simple Subtraction Wrap Up

It works pretty well for Fire Emblem. It’s very understandable for players. More generally it works well for more constrained games like NES-era RPGs where designers can realistically hand craft and plot out character and enemy stats. But it’s unforgiving. If you get the numbers wrong defense either doesn’t do much or players can’t do damage, and the more complex a game is the more likely you are to hit these scenarios.

Formula 2: Percent Reduction

Damage = Attack * (100-Defense)/100

Here defense ranges from 0 to 100 and represents a percentage damage reduction. So if the character has 60 defense that indicates a 60% damage reduction - for every 100 points of damage received that character takes 40.

Game design elements generally aren’t good or bad but depend on other design elements - that said I would avoid this formula in most cases.

It has some strong points. It’s easy to understand. In some sense it’s forgiving. When I need to fill in my Elden Ring enemy spreadsheet I can give Smorgleborg: The Meandering a physical defense of 70 (70% reduction) and feel good about it - that feels safe. Unless I enter values close to 100 to make him invincible to physical damage there’s no catastrophic failure case.

But there are major problems with this formula, particularly for player characters in games with highly customizable characters.

One issue is that small numbers feel negligible. As a player if I equip a shield that has 2 defense I don’t take 2 less damage per hit, I take 2% less damage. At the start of the game if rat hits me for 8 damage and I equip a 2 defense shield I might still take 8 damage due to rounding - the shield does nothing at all.

So maybe a starter shield should have 10 defense. Does that mean a more advanced shield should have 20 defense? But what if I also have a chest piece, helmet, boots, etc?

This formula has the pesky “defense is better with already-high defense” problem from earlier. Going from 0 to 1 defense with this formula does effectively nothing, but going from 98 to 99 halves the damage you take. In Fire Emblem this was a theoretical problem for many reasons, including that the definition of “already-high defense” depends on enemy attack. But here this is a real problem - as your defense approaches 100 it becomes much more valuable.

This is the “what number do I type into the spreadsheet” problem, but instead of the problematic spreadsheet being enemy stats it’s player equipment and ability stats. If I’m staring at the Excel entry for “metal shield” what do I enter? I need room for “wooden shield” and for “stout metal shield”, and I want equipping the shield to feel good, but I also want to avoid all the equipment (and buffs) adding up to close to 100.

Percentage reduction works well for categorical resistances - a fire elemental takes only 25% damage from fire attacks, or plate mail reduces slashing weapon damage by 50%. But that’s more a system on top of the base damage formula. Using percentage reduction as the base - in particular summing up the individual armor pieces to get a total percentage reduction - is hard to work with.

Instead of additive you can do something multiplicative - if you have 50 defense boots and a 50 defense hat instead of 100% reduction (zero damage taken!) you reduce 50% then 50% again, so 75% reduction total. This multiplicative approach is less numerically problematic but it’s much less understandable. In that case “total armor” isn’t a meaningful stat - instead we probably need to label the boots and hat with “50% damage reduction” and then have a total readout that says “75% damage reduction.” This isn’t that hard to understand with simple numbers. But if you have hat, gloves, helmet, body and shield with numbers like 14%, 5%, 20%, 40%, 30%, who knows what the total damage reduction should be? You take 70% of 60% of 80% of 95% of 86% of the damage - to many players this won’t make sense beyond “bigger numbers better.” Is the final number the average, the sum? Something else? Is it better to upgrade the 5% gloves to 10%, or the 20% hat to 25%?

This multiplicative approach makes it easier to assign numbers to items in a spreadsheet without worrying about total reduction approaching 100, but it doesn’t solve the problem that damage reduction matters more the more of it you already have, or the problem that small numbers feel inconsequential.

Those problems are intrinsic to the formula. League of Legends switched from “cooldown reduction”, which used this sort of formula, to a different formula termed “ability haste”, because cooldown reduction had the same problem - the more you had the better it got.

Anther problem with this formula is that it’s boring. However the “fixed” version of this formula shares that problem, so I’ll move on to that one.

Formula 3: Effective Health

Damage = Attack * (100/(Defense+100))

• When Defense = 100, Damage =Attack /2

• When Defense = 200, Damage = Attack /3

• When Defense = 300, Damage = Attack /4

With this formula each point of defense increases effective health by 1%. When you have 100 physical defense you effectively have double the HP, since you’re taking half damage. When you have 200 physical defense you effectively have 200% additional HP, since you’re taking 1/3 damage.

Like the previous formula this one reduces all damage by a percentage, but it handles high defense numbers more gracefully. It’s very forgiving. As a designer you can’t choose dramatically wrong values for individual items, or values that combine poorly. It passes the “what number do I type into the spreadsheet” test - it doesn’t break if you enter 5 or if you enter 500. It’s not as understandable as “40 defense equals 40% damage reduction” but it’s not too bad, especially if players think in terms of effective HP. (“40 defense = 40% more HP”)

The problem with this formula is that it has very dull characteristics.

Consider the following scenario: you’re playing a tactics game where one of your squad members is a thief who attacks quickly with a low damage dagger, and another character is a knight with a huge slow hammer.

You’d naturally expect the thief to excel against unarmored units, and struggle against heavily armored ones. Conversely, you’d expect the knight with the giant hammer to excel against heavily-armored units. It makes sense intuitively and in video-game logic. A thief is an assassin type unit meant to get behind enemy lines and take out soft targets. And in real life stabbing a vulnerable person does a lot of damage, while stabbing someone in heavy armor with a thin blade doesn’t do much. Using a hammer to crack open armor makes sense - like using a crab-hammer to eat crab. I don’t know that using a giant hammer against armored knights is a valid military tactic - the point here isn’t absolute realism, just plausibility and intuitiveness.

A subtractive system supports this naturally. If three stabs have 15 attack each and a giant hammer has 30 attack these do different amounts of damage against different defense values. At no defense the stabs clearly prevail, doing 45 damage total. But versus 15 defense the three stabs do no damage while the hammer still does 15.

This is a nicely organic way of making different units, loadouts, and attacks (like weak vs strong in a God of War style game) more interesting and meaningful. It’s not a lock-and-key system like “water is good vs fire” or “gauntlet weapons are effective against shielded enemies.” It’s not hard-coded, it’s a natural consequence of the system. With a subtractive damage reduction system, weapons with different attack speeds and attack powers naturally have different uses.

I worked on an action-adventure game that used a subtractive damage formula. Against weak enemies I found my most effective combos used weak attacks, since they were quick, safe, and effective. But when I added high-defense enemies those combos did very little damage, and I had to switch to using the harder-hitting slower moves in my arsenal. It just felt right. I was using armor-cracking attacks against tough enemies, and it gave different moves and weapons more clarity of purpose.

In a percentage reduction system you get none of that - the higher DPS weapon is always better. If you want to differentiate hammers and daggers you need to do that in other ways.

Personally I think percentage reduction is also a bit dumb. The goal of a damage formula probably isn’t realism, but armor reducing the damage of a nuke and a spork proportionally is a little silly for my tastes.

League of Legends and Dota 2 both use an effective health setup. For those games it makes a lot of sense. You only control one character and that character has one basic attack. While you can buy items you can’t swap out weapons and armor the same way you can in an RPG or action-adventure game.

In League of Legends if you want to counter an enemy with high armor you don’t pick a character with a giant hammer. Instead of you pick a character that does magic damage, or that naturally builds armor penetration, or that lines up well against them in terms of abilities or macro strategy. Games like League of Legends already have plenty of counter strategies and character counter-picks, so adding an additional layer of “this character with a dagger can’t damage Malphite because he’s too rocky” would be overkill.

These games have enough other things happening in each 30 to 60 minute match. And while MOBAs are full of combat the combat systems themselves aren’t the focus. So this formula works well for that genre - it’s numerically stable, hard to screw up as a designer, and easy enough to understand as a player.

For a game like Fire Emblem or Dark Souls the formula is still attractive in how forgiving it is, but I’d be worried about the blandness.

The first purely subtractive formula is “broken” in that characters can do zero damage. But in Fire Emblem it’s cool that when a thief hits a giant armored General you hear a ping sound and do no damage. Formula two, the percent reduction, encourages you to invest in defense either a lot or a little, since defense is more valuable when you already have a lot. That may also be broken in some sense, but at least it presents two distinct paths - you can invest in defense and become a super-tank or blow it off and focus on something else.

This effective HP formula is almost too fair. Investing in defense a little, somewhat or a lot all differ by the same degree. It’s like a system where each point you invest into “vitality” gives you 10 extra HP - fine but totally unexciting. And, as covered earlier, it reduces weapon effectiveness to DPS, since incoming damage of all amounts is scaled proportionally.

Formula 4 (The Last): Subtraction Revisited

This one is hard to type out so here’s a picture. The two lines are the same formula written two different ways, where s and s2 are damage, a is attack and d is defense

The second is easier to read and parse:

1. When attack is higher than defense, damage = attack - defense/2

2. When defense is higher than attack, damage = attack * attack / (2 * defense)

The first version of the formula is the one I came up with originally - it looks more complicated but I think the logic is easier to grasp:

1. Start with half the attack as a baseline damage

2. Then, depending on the ratio of attack to defense, either add or subtract damage, such that the total range is between attack and zero

I’ve created a Desmos page with an interactive formula - from this page you can change attack and defense values and see what the formula produces, and also see a graph of attack vs a particular defense value. I’ve also included a graph of the Fire Emblem formula, for comparison.

Green / top line = our formula, purple/ bottom line = Fire Emblem

When attack is higher than defense this formula is the Fire Emblem formula in shape. But as defense outpaces attack the damage dealt approaches zero rather than quickly crossing that threshold.

I don’t know that this is “fixed” exactly, but it improves on the “what number do I enter for boss defense?” problem of the original. It’s more forgiving in that if you enter 50 for defense and the player only has 40 attack the player doesn’t do zero damage, they just do greatly reduced damage. (In a way that’s more sophisticated than a minimum cap) At the same time, it retains the strengths of a subtractive system - high defense penalizes low attack more than it does high attack.

I’ll revisit the case of our triple-hitting dagger vs hammer, with the Fire Emblem formula and our new one

Fire Emblem formula:

• 30 Attack (hammer) vs 0 Defense: 30 damage

• 15 Attack(dagger) x3 vs 0 Defense: 45 damage

• 30 Attack vs 15 Defense: 15 damage

• 15 Attack x3 vs 15 Defense: 0 damage

New formula:

• 30 Attack vs 0 Defense: 30 damage

• 15 Attack x3 vs 0 Defense: 45 damage

• 30 Attack vs 15 Defense: 23 damage

• 15 Attack x3 vs 15 Defense: 24 damage

• 30 Attack vs 100 Defense: 5 damage

• 15 Attack x3 vs 100 Defense: 3 damage

Against a low defense target 3 quick stabs does more damage than one hammer blow, at medium defense they even out, and against very high defense the hammer blow is superior.

Tweaks and Additions

Our graph is asymptotic on the left side and linear on the right side - it could be more than linear on the right. I believe that some of the From Software games do this - if your attack is much larger than enemy defense you do bonus damage, which I imagine helps players quickly clear through “trash mobs” once they’re powerful.

Some games also take player and enemy level into account as a safety valve. With a naïve damage formula a level 1 character with the stats of a level 40 character effectively is level 40, which may not be what you want. By factoring in levels you can cheat a little: if a player is way over-leveled an enemy becomes easier, regardless of raw stats, and if a player is way under-leveled an enemy is tougher, regardless of the other math. This allows grinding to be a difficulty release valve, and prevents exploits like level 10 characters beating a level 90 boss by equipping end-game items. (Though you might consider this a feature)

Where Next

When considering a damage formula, “what number should this formula produce?” is the question you ultimately need to answer since you have to generate a number and substract it from HP.

But the best way to think about the problem, in my opinion, is to consider the two user groups.

When the player goes from a wooden sword to an iron sword what should happen? How significant should that feel? What about when the player encounters a way-too-tough enemy, or revisits the starting area and runs into way-too-easy ones? If the player equips a dagger vs a hammer how should that feel? Do we want the hammer to be an armor-cracking weapon, or should hammer and dagger have the same effective DPS and vary in other ways, like reach and play style?

And consider the other users - the coworkers (including sometimes yourself) who have to type numbers into spreadsheets. Do designers understand the range of those numbers? Can they reliably determine correct-ish values that make sense and play well together? How likely are designers to enter values that break the game, effectively do nothing, or have little underlying logic?

The math can get quite complicated, with floors and ceilings, exponents, natural logs, etc.

For example here’s a reverse-engineered Dark Souls 3 formula

This is easier to understand in graph form - damage ranges as an s-curve from 10% to 90% of attack power, where higher defense stretches the curve horizontally. (More or less)

As another example here’s how various Fallout games handle damage. Reading between the lines you can see the designers grappling with some of these issues - preferring subtraction, preferring percentage reduction, doing both, doing both but swapping the application order, and finally settling in Fallout 4 on a formula designed to handle a high range of values gracefully. (I think)

That formula, similar to the one I ended up on, uses the ratio of attack to defense, and when they’re equal attacks do half damage, same as mine. But the shape of the curve is different. Again: it can get quite complicated.

I’ve looked into various From Software formulas and, based on their use of cubes, logs, etc, my guess is they have a math whiz on the team. A small team might lack that math whiz, but for a small team complex formulas are probably overkill. A large team probably has a math whiz on it, even if it’s not you.

But the math is a means to an end - a way to accomplish the particular goals of the game.

This was a lot - and I never provided a real answer to “what damage formula should I use?” But hopefully I’ve provided at least a starting point.

(If you’re dying for a concrete answer: start with formula 4 and adjust to taste)

The game awards don't represent gamedevs. They represent the people abusing gamedevs. Where are the gamedev awards?

Anyone have a Game Awards ticket for sale? It's for a friend!

Any friends going to the Game Awards? I will be!!

This is not an evaluation of The Game Awards. I’m not looking to convince you that the show was disrespectful, that producer Geoff Keighley was cowardly for not mentioning layoffs or Palestine, that the show was sexist for not featuring enough women, or of any other of the common complaints. But if you believe these criticisms are valid - that the show is sexist and cowardly and celebrates abusers, that the industry being honored fuels anti-Arab resentment and indirectly kills them - why would you attend it?

That’s not a rhetorical question. It’s not a trick question, and I don’t think it’s an unfair one.

Goofus and Gallant: Influencer Edition

Goofus has heard that Chick-fil-A is anti-gay but he hasn’t looked into it and he’s not really “into politics”. When a friend says “let’s go to Chick-Fil-A” he responds with “naw, I’ll just stay in,” mostly because he’s lazy.

Gallant on the other hand, being well-informed and politically active, knows that Chick-fil-A is anti-gay. One day Gallant, while a bit hungry, walks by one. Next door is an In-And-Out but here’s the thing: Gallant likes chicken sandwiches more than hamburgers. So Gallant does what any reasonable modern influencer type would do: he takes a picture of the Chick-fil-A sign, tweets it out with “yikes homophobic much?” - then gets in line.

Can you see where I’m going with this?

Gallant regularly excoriates The Game Awards. They’re juvenile. They focus too much on Hollywood celebs. They’re all ads and product placement. They celebrate abusive higher-ups who lay off workers and gin up hatred against minorities. When Gallant says these things people like and retweet with “ur so brave Gallant, thank u for speaking up.”

But here’s the thing: speaking up against The Game Awards makes Gallant feel good and appear influential, but so does attending them.

So Gallant does what any reasonable influencer type would do: he attends the awards but live tweets his disgust.

Activism is Sacrifice

Sacrifice gives activism power, because real sacrifice indicates real belief. Someone standing in front of a moving tank probably isn’t “virtue signaling.” Even if you think their principles are way off you have to admit they at least have some.

Mike Rose, from game publisher No More Robots, advertises himself as an “ally.” (Before you get mad I’m not burying Mike Rose, keep reading) A few years ago there was an event called “Pro Indie Dev” that was a series of talks aimed at aspiring game developers. Here’s what the website first looked like:

I’m not the most perceptive of or invested in diversity topics. I don’t have pronouns or acronyms in my bio or a profile pic of me wearing a shirt with “this is what a male feminist looks like.” I’m not conservative, I’m not “anti-woke” and I’m not a “centrist / traditional liberal” (which is usually just a different way of saying “anti-woke”) - but I would never stick an “in this household we believe” sign in my yard.

Even I immediately noticed that this picture indicates a very limited character creator. (I saved this image as “allloooksame.jpg”)

Mike Rose quickly dropped out of the event so that someone more diverse1 could take his place.

This was a sacrifice. Not a huge sacrifice - he didn’t stand in front of a tank. But dropping out gives up the exposure these events confer.

Another presenter at this event, who also curated good guy ally vibes, said he didn’t notice the lack of diversity on the panel (even after the website was published!), then claimed that expecting him to research the exact makeup of the panel was an unfair burden. He was an ally up until it involved the backbreaking labor of quickly glancing at the event website. (I’m sure he would describe it very differently)

Science fiction author John Scalzi is well-known for being “progressive” - maximum air quotes. He once launched a series of tweets angry that a science fiction convention had too many panels full of old white guys and not enough minorities. At the same time he complained that he was only on a couple panels. Maybe you think diversity at conventions is a big issue, or maybe you think it’s “woke” nonsense, but either way I think you have to roll your eyes at “there are too many white guys here - one of you other guys should step aside.” (Nudge nudge)

A game developer once announced that he was boycotting Nintendo - he had no games on Nintendo systems and was at best starting work on porting to Wii U, which was probably a money-losing endeavor. When the Switch took off his boycott ended. People pledged to boycott PAX because the creators of Penny Arcade “mocked rape victims”, but when they had game collections to promote or panels perform at - when PAX could help them make money - those boycotts similarly ended.

It’s easy to take a zero-consequence moral stand. And in the age of social media, especially in politically charged circles where moral grandstanding is a currency, it can be lucrative. But if you’re unwilling to make even the smallest sacrifice for the principles you claim to hold you’re just posturing.

Skipping an Awards Show is Very Very Easy

I’ve read so many pieces to the tune of “there’s no such thing as being apolitical - if you don’t take a side you’re supporting the status quo - an active and conservative political choice.”

Attending an event is an active show of support2 - when you attend the event you are the event. The organizers get your money and attention, they record your excited yells and enthused faces. You make up the attendance numbers they show to advertisers.

“If you don’t approve of the show you should skip it” makes skipping sound like the effortful choice. But the natural state of the universe - the path of least resistance and most inertia - is you not attending the show. You don’t have to choose to skip it; if you do nothing at all the show will pass on by.

To attend a show you have to get dressed up, get a ticket or finagle an invite. You have to get to the event location, which for many includes flying in. That’s a lot of work! Simply doing something else is no work at all.

Eat the Rich but Also Eat Finger Sandwiches with Them

Carolyn Maloney (pictured above) and Alexandria Ocasio-Cortez both attended the 2021 Met Gala.

It did not go over well.

The Met Gala is an event you attend to hobnob with rich people. It’s fundamentally rooted in inequality - a celebration of the haves.

The picture above is particularly striking. There’s not even the pretense of equality. The stars - the rich people - wear whatever they want and the help wear uniforms. The servants wear masks but the celebrities don’t. That’s particularly galling in that Democrats are supposedly the “follow the science” party - “the science” doesn’t say that only waitstaff can communicate COVID.

AOC wore a dress that said “tax the rich.” But this party is full of people who pay their accountants good money to avoid taxes.

Prior to the Game Awards, Kotaku ran a number of serious critiques. They also had someone attend and complain in real time. After the event their front-and-center story is….a celebration of the show fashions.

The tonal whiplash from this-cowardly-show-won’t-acknowledge-Palestine to check-out-these-people-glitzing-it-up is really something.

It’s not that some people are wagging their fingers at The Game Awards and other people are treating them like a glamorous party, and when you combine those groups you get an aggregate hypocrisy. It’s (often) the same people.

I wasn’t impressed by AOC’s “tax the rich” dress, but hey, at least she wore that dress. I’m looking at these pictures of fashionable Game Awards attendees and I don’t see many “free Palestine” dresses or even discrete pins. Would that have ruined the drip?

The Judgement-Free Conclusion

I don’t think attending the Game Awards makes you a bad person, on the whole or in that moment. Personally I think the show is….fine. Better than the Spike TV incarnation, at least.

But you should abide by your own loudly stated morality.

If you think the Game Awards celebrate abusers don’t attend them! It’s not complicated and it’s not “nuanced.” Yes there’s some nuance but the word “nuance” isn’t a magic shield.

It’s fine to say “you know what? I don’t think a video game awards show it the place for geopolitics and labor disputes - let’s just have fun.”

It’s also fine to be mad for real - to skip the event and do literally anything else.

What’s not fine is being pretend-mad.

You can’t be very angry about the event while eagerly attending it, especially if you know exactly what to expect. (And you do!) In that case you’re Gallant bashing Chick-fil-A while standing in line at one. AOC wearing a “tax the rich” dress to an event celebrating tax-avoiding billionaires.

You can’t effectively protest the thing you actively participate in, when not participating is the tiniest sacrifice imaginable.

Before you come back with “but there’s no ethical consumption under capitalism” read the second half of that sentence again please.

Attending The Game Awards doesn’t make you a bad person. But if according to the morality you espouse the event is unworthy of your support, just don’t support it! Instead of attending the event do absolutely anything else. That is, quite literally, the very least you can do.

If I believed an awards show was racist, sexist and morally repugnant I would simply not attend it.

That’s just me.

Subscribe now

I occasionally get asked why my criticism isn’t “constructive” - why I criticize games but rarely broach how to improve them. My answer: that question reflects confusion about what “constructive” criticism is and what function critics serve.

The Meaning of “Constructive”, Real and Imagined

If you replace “writer” above with “game developer” you get a real sentiment rather than a joke one: that “constructive criticism” is when someone says “I’m not a game developer but….do that.”

One example of this mindset is this misguided ResetEra forum post.

Many people make the mistake of pointing out a flaw in a game, and explaining in detail why it is a problem, and leaving it at that. That is not constructive. Constructive criticism must explain how something can improve, not just where it falls short.

For some reason “constructive” has come to mean “possessing suggestions.” But suggestions are usually less useful than a well-articulated explanation of a problem. Even at game development studios a spot-on description of a problem isn’t common. A programmer is just as likely to say “this animation looks wonky” as “this would look better if the initial frames used hand-tweaked tangents.” An artist might complain that the camera is “sluggish” without pegging the culprit as frame rate, inertia, or input lag. A detailed explanation of a problem is extremely useful. It’s often more useful when it’s not mixed in with a specific suggestion for a fix, since that shifts the focus to debating a likely ill-considered solution.

If players tell developers exactly what they dislike that serves a useful purpose - it’s “constructive.” The developer can come up with the solution - that’s what they get paid for.

Critics Aren’t Game Developers

Critics don’t make games1. It’s not their job to improve them, nor is that a job they’re good at. As a purely technical point, most criticism is post-release. If your review of Mario Wonder includes “constructive” comments like “I think the game would be better if Mario could jump twice as high” you aren’t constructing anything beyond an idle musing. There’s almost no chance the dev team is going to patch the jumping to please you, and your off-the-cuff suggestion is probably bad anyway. Criticism in aggregate can improve a live service game or inform a sequel, and criticism can in theory improve the medium as a whole over time, but criticism of a particular game is unlikely to “construct” anything.

It’s commonly understood that players and playtesters are good at identifying problems but bad at identifying solutions - or really are good at identifying the symptoms of problems. But some people expect critics to excel at identifying solutions. That’s not their job and not something they have practice in or training for.

On GameDeveloper.com I have an older blog titled Ostensible Improvements: When Better Isn't. It includes this anecdote:

I once worked on a game that was sent out for a mock review / analysis. It came back with a list of suggested improvements, and when we made all the improvements and resubmitted it the response was “you did everything we wanted but ultimately the game isn’t any better.” Which was disappointing but unsurprising — if mock reviewers knew which specific changes to make to improve a game they’d probably be game designers.

In that case it is the job of ex-media consultants to offer suggestions, but I think that’s a job with limited usefulness. Doing mock reviews to predict critical reception makes sense, it’s a logical extension of doing game reviews. But an ex-reviewer offering design consultation is like a movie critic offering script consultation - it could work I guess, but I wouldn’t pay for it. At least the reviewer-turned-consultant is learning game design or has some affinity for it - there’s no reason to expect a currently-working reviewer to have special design insights. (Though game reviewers might on average make better game designers than a person plucked at random)

None of this is a knock on games media. Media members can successfully transition to game development - Insomniac’s creative director used to write for EGM, for example. But reviewing games and making them are separate skills. There’s no reason someone who writes about Apple products couldn’t become an electrical engineer, but there’s also no reason to expect them to have meaningful electrical engineering insights.

Of course some critics are game developers - myself, for example. But that I could play someone else’s game, think about it for a few hours, then articulate exactly how to fix it is far-fetched - I can’t reliably do that for the games I’m paid to work on! Fixing up up games is a job, not a one-off essay.

I’ll close out this section with an exchange between film-maker Sidney Lumet and film critic Pauline Kael.

The real rift between Lumet and Kael came on "a very difficult evening" when the two of them got involved in one of those boring conversations about the function of a critic. "There were two other people present," Lumet recalls, "and she said to them, 'My job is to show him' -- pointing to me -- 'which direction to go in.' I looked at her and said, 'You've got to be kidding.' She said, 'No, I'm not.' I said, 'In other words, you want the creative experience without the creative risk.'

“Here’s How I’d Fix that Starfield Rain”

Above is a photo-mode Starfield screenshot - as you can see it’s a bit borked. The rain only falls in a visibly small area around the player, and most noticeably the squiggly distortion effect is also constrained to that area.

There was a discussion of this on Reddit with many people saying “they could easily fix this by parenting the rain to the camera instead of the player” as an “industry-standard” solution. In particular, one person who claimed to be a developer argued it was an easily-fixed unforgiveable error that exposed the developers as clueless.

It takes a certain arrogance to play a game for 10 hours, think about it for 1, then write “here’s how I’d fix it.” This guy looked at a single screenshot and immediately knew how to fix it and that the people responsible for the error were incompetent.

The Redditor who claimed the Starfield rain was an unforgiveable sin worked on Battlefield.

According to the Washington Post

“Battlefield 2042,” its developers at Dice and publisher, EA, have been harshly criticized by players for its buggy state and lack of industry standard features since its November 2021 release.

Oh.

Fundamental attribution error is, more or less, the belief that when other people screw up it’s because they suck, but when you screw up there’s a reasonable explanation that lets you off the hook. If you asked this Battlefield dev why their games are buggy their answer wouldn’t be “my own fundamental incompetence.” It would be lack of time and budget, unrealistic expectations, maybe bad management or bosses, a flawed process (at a different org level than this guy occupies, of course), etc etc.

When a Battlefield dev on Twitter (side note: it’s so often Battlefield devs - what’s going on guys?) complained about the Elden Ring UI people pointed out that a) Battlefield 2042 didn’t ship with a functional scoreboard, something that’s been standard for decades and b) the design of checkboxes made it impossible to tell if an option was on or off.

If you asked this dev why their game shipped in that state their answer wouldn’t be “I’m just bad at my job.” So it doesn’t seem fair to assume that the Starfield or Elden Ring devs are bad at theirs.

Here are a few points I’d make about the Starfield rain:

First, virtually every AAA game ships with more severe bugs - crash bugs, progress blockers, falling through the floor, quests refusing to unlock, framerate issues, etc. To pretend that this is an extraordinary failing is absurd.

Second, there are a dozen plausible explanations for why this problem exists and why an “industry standard” solution wasn’t used. The photo mode could have been a late addition. The rain may correctly be parented to the camera, but the photo mode uses a second camera. Parenting the rain to the camera may have been tricky due to engine and workflow-specific reasons. The photo mode may have been given to a junior dev without much oversight. It could have been a quick implementation that was always going to be improved later, but other issues became more pressing.

Or it could just have been an oopsie.

That’s not to make excuses. The rain is a problem and ideally would be fixed. But the game devs saying “here’s how I’d easily fix it” have, without question, shipped games with bugs as bad or worse, including bugs they’re personally responsible for. So a natural rejoinder to “here’s how I’d fix your game” is “why not fix your own?”

Improving Games is Hard

The gist of Ostensible Improvements: When Better Isn't is that sometimes “objective” improvements don’t make games appreciably better. Adding more mechanical depth in Tacoma didn’t make it better than Gone Home. Improving the combat of Silent Hill with Homecoming resulted in one of the worst games in the series.

In Why Artifact Failed I wrote what I consider the definitive2 examination of the problems with Valve’s Artifact. Now here’s the shocking follow-up: Valve released Artifact 2.0, which fixed many of the problems I identified - and was a worse game.

“Fixing” the game stripped away some of the character and replaced it with effectively crowdsourced content. Some of the solutions were worse than the problems - a big issue in 1.0 was that the board was spread out over three screens. The solution - cramming it all into one screen - made the game harder to parse visually and significantly uglier. A standout feature of Artifact 1.0 was strong production value; it wasn’t that fun but it looked well-crafted. Artifact 2.0 looks pedestrian by comparison. In some cases Artifact 2.0 fixed problems only to repeat variations on them. Confusing card text was removed then replaced with different confusing card text. Overly-complicated systems were simplified only for an overly-complicated shop to be added.

The Artifact 2.0 changes were close to what I would have suggested, but that didn’t result in a better game. Which is why I don’t often make suggestions about games I’m not working on. When I am working on a game a suggestion is the first step of a process.

How I’d Fix: The Callisto Protocol

As a refresher, here’s what I wrote about The Callisto Protocol:

Here’s how I’d fix the game: I wouldn’t.

There’s low-hanging fruit to pluck: improve the misleading tutorial flow and text. Allow for players to dodge just-in-time instead of requiring well-in-advance inputs. But I don’t think fixing the tutorials would do much, because as I often argue, tutorialization can’t fix flawed systems, and often can’t even make them more understandable. The game is a conceptual miss with a low ceiling - hard to greatly improve without redoing at least the combat.

How I’d fix: Lies of P

Here’s what I wrote about Lies of P:

One of the issues with dodging in the game is that when you dodge into an enemy you don’t collide-and-slide off of them, instead you get awkwardly stuck on them. So my solution would be…fix that.

Now how to fix that: no idea. Maybe the character controller has no ability to collide-and-slide off of other character controllers, so that would have to be added. Maybe collide-and-slide is turned off because it made the character feel too slippery, and should be turned back on in certain cases. Maybe it’s an issue with the character controller interacting with limbs, and the solution depends on the exact physics implementation, depenetration logic, etc.

“It’s bad that you get stuck on guys when you roll so make that not happen” is the best I can do - as an outsider I could phrase that to sound more sophisticated I can’t come up with a more sophisticated solution. Nor do I need to - the team that made Lies of P was competent enough to make the game, so they’re probably competent enough to figure out a solution to this particular problem.3

A Series of Disclaimers and a Conclusion

Advice and suggestions aren’t always bad. Sometimes teams can become blind to serious issues. When developing a game you become saddled with a complicated history - maybe you tried something two years ago, it didn’t work, and you mentally filed it away as a bad idea, but circumstances have changed. In that case an outsider suggesting even an “obvious” old idea is helpful.

While game development is hard and often takes specialized knowledge or training, I don’t want to give the impression that everyone who isn’t a certified professional is an unsophisticated rube who should keep their mouth shut. It’s not impossible for players or critics to have genuinely useful suggestions, they just don’t have a high hit rate. And personally I find critique that focuses on “here’s what I’d have done instead” is often more ego-driven than interesting.

There’s nothing wrong with offering suggestions, if it’s not done in a sneering style. But there is something wrong with expecting suggestions, or dismissing criticism that doesn’t come with them as “not constructive.” The majority of players and critics are better served focusing on clearly identifying issues rather than brainstorming solutions and new features. Some critics write “poor level design” or “unengaging combat” and leave it at that.4 If you can’t say why the combat is unengaging your proposed changes are almost certainly underbaked.

As a critic it’s fine to say “it’s flawed and here’s why” and leave it at that. And as a developer you have to swallow your ego and consider that feedback, rather than reject it as impolite.

Subscribe now

This first ran on GameDeveloper in 2018. While I consider it evergreen it feels especially relevant now with the release of Baldur’s Gate 3. (And to a lesser extent, Armored Core 6)

I often write pieces that push back against conventional wisdom but I try not cross over into contrarianism. When pressed about his piece “Don't Play Untitled Goose Game” Ian Bogost admitted people should play it - that’s just trolling.

This is not trolling. Topics like difficulty and accessibility are often used as culture war footballs, but I’m treating accessibility as a game design (and sales and marketing) issue. It would be easy to write an “accessibility is good and if you’re against it you’re doing a gatekeeping” take, or an “accessibility is code for ‘wokified’ ” one, but hopefully I haven’t written either of those.

I’ve edited this reprint a bit more than usual for readability and to sharpen some points.

The Many Meanings of “Accessibility”

I often hear that lack of proper terminology is holding back game design and criticism. This makes less sense with each passing year and it was flimsy to start - it's hard to believe that new jargony terms are needed when we're so bad at using existing ones.

I can't count the number of times I've seen a walking sim described as "unique" despite using not only familiar mechanics (walk around, examine objects, read graffiti, listen to audio logs) but a familiar central story as well: explore the abandoned place and figure out where the people went. Or how often "core gameplay loop" is used to describe something that is neither gameplay nor a loop. "Everything is political" is a common phrase these days, but if everything is political then the word necessarily has no meaning. (One could just as easily state that "everything is ham")

Thanks to the work of advocates like AbleGamers there's been a strong uptick in the game industry's understanding of accessibility issues. But the game industry regularly uses "accessibility" to refer many distinct concepts, most of which are better described by other more specific words. Those concepts include:

1. Actual accessibility

2. Immediate Understandability

3. Approachability

4. Difficulty

5. Conformance to Preference

In this blog I'll argue for a narrow conception of "accessibility" while examining (and largely rejecting) other uses of the term.

Accessibility Meaning 1: Actual Accessibility

What I'm calling "actual accessibility" is the one related to disability and access. Making games more accommodating to people with disabilities is just a nice thing to do, and those changes often help everyone. Subtitles are useful when playing at low volume, for non-native speakers, or can rescue a game with sound mixing issues. (Such as in Xenoblade Chronicles X, where the music in cutscenes is regularly mixed too loud to hear the dialogue) Icons that look distinct not only in color but some other aspect are easier to pick out even for those without vision issues. It's up to each developer to determine the effort they put into accessibility but broadly speaking that effort’s a good thing.

This use of "accessibility" — the plain English meaning — is fairly new in games1. For a long time "accessibility" has primarily meant a mixture of lowering difficulty and "streamlining." A developer interview from a decade ago about making their sequel more “accessible” is almost certainly not about actual accessibility but instead about something like moving a game from PC to console and making it palatable to that audience.

Accessibility Meaning 2: Immediate Understandability

Disclaimer: when it comes to understandability, approachability, complexity, etc, there are accessibility issues related to cognitive functions. The following sections should be read as addressing "casualization," streamlining etc for the sake of mass appeal, while granting that those overlap with "actual accessibility." It's left to the reader to determine which sorts of streamlining changes are in service of genuine accessibility concerns and which are not.

We frequently use "accessible" to mean "streamlined" or "simplified." Why don't we just say that then? I'll suggest one reason here that I'll revisit later: "accessible" conveys a certain moral weight and importance; it's a word with a very positive connotation, when a term like "simplified" reads as neutral or negative.

If you're a Deus Ex 2 developer you probably don't want to say that your game has smaller maps and universal ammo because you think console gamers have smaller brains than PC gamers — that sounds bad! That the game is being "made more accessible" sounds good — who could possibly be against accessibility?

One downside of using "accessibility" as a euphemism is that gamers are now trained to roll their eyes and think "here we go again.” Not because those gamers are against actual accessibility changes (though…some are) but because they're against sacrificing core elements of a franchise in an attempt to broaden market penetration.

Is Immediate Understandability Good?

I’m opposed to using "accessibility" to mean "understandability" and I'm also not convinced that immediate understandability is even a good thing. At least, not across the board.

It’s wise to be wary of conventional wisdom, especially when delivered in slogan form separate from any original justifying arguments. There are two related conventional wisdoms I'd like to examine: "easy to learn, hard to master" and the idea that while depth is good complexity is bad.

Or take League of Legends. This game breaks so many rules of “good design”. It is a clone. It is over-complicated to the point of utter indecipherability. It is fussy, baroque, full of arbitrary, non-intuitive details (Last hitting? Inhibitors??). It makes no attempt to teach the player or draw them into its labyrinthian systems.

I certainly don't agree with everything in Against Design by Frank Lantz but the above point seems unassailable. (Well...minus the "clone" claim) And what's written above could apply to the entire MOBA genre or any number of Steam survival games. 

Just as there are people drawn to tall dark mysterious strangers there are people who like coy games that don't readily reveal their mysteries of content and construction.

I often hear people use Street Fighter 2, one of the most popular games of all time, as an example of "easy to learn, hard to master." But Street Fighter 2 is not particularly easy to learn. Compared to contemporary arcade games it was much more complicated with virtually zero instruction. The game introduced concepts that, while well-understood today, were novel then: blocking by holding back, circular and charge motions, combinations enabled by hit stun, cross-up attacks, cancelable attacks. The game had six buttons at a time when most games had two.

Street Fighter 2 wasn’t intuitive or streamlined. Is was the polar opposite of streamlined, not "accessible" at all in the marketing-speak sense. What Street Fighter 2 was, however, was fun for people of all levels. It had great graphics and animations, detailed backgrounds and cool line-scrolling parallax foregrounds. It had catchy music. Even if you didn't know what you were doing you could pick Chun-Li and mash out some "wind kicks", or mash fierce punch as Guile to alternate (for a beginner seemingly at random) between a sweet backfist, a devastating uppercut and a suplex. It's not that Street Fighter 2 is easy to learn, it's that it's fun even when you haven't learned it. And many people never learn it, content to wiggle the joystick and mash buttons forever.

Similar to "easy to learn, hard to master" is the notion that depth is good but complexity bad. This is a meme-level proposition typically supported with meme-level arguments, usually analogies involving Chess. (The world needs another appeal to Chess like I need a hole in my head) There's a simple "proof of the pudding is in the tasting" counter-argument to the idea that complexity is bad: plenty of people like it.

League of Legends is extremely complex. Not only are there many rules but as in a collectible card game the distinction between rules and content is thin. To play the game at a reasonable level you have to understand dozens of characters and hundreds of abilities, and to play at a high level you have to understand over a hundred characters. As well as items, jungle monsters, XP formulas, non-intuitive ability interactions, and an increasingly complex rule set.

There's a notion that if a game is popular an "accessible" version of that game will be even more popular. But the streamlined version of League of Legends, Heroes of the Storm, is less popular. Maybe the right level of complexity for MOBAs is very high because breadth is a main reason people like that style of game. When Street Fighter 2 came out 8 selectable characters seemed like a lot. Now players complain if a fighting game launches with less than 20. They demand breadth.

Every so often someone attempts to make a more "accessible" fighting game. Fantasy Strike didn't come close to reaching its goal on Fig and appears to be one of the worst-performing games on the platform. Rising Thunder was canceled while still in alpha. There doesn't appear to be high demand for streamlined fighting games, at least the non-transformative kind that subtracts complexity from the Capcom formula. (As opposed to a more transformative, more apples-to-oranges game like Nidhogg)

The belief that a streamlined game will have more mass appeal is for many a non-falsifiable religious one. When a streamlined game does well it's evidence that the theory is sound, but when one does poorly it's dismissed as an irrelevant data point. For every Monster Hunter: World - a simplified and successful entry - there's a Marvel vs Capcom: Infinite; the track record of games that have been made more "accessible" for broader audiences is very mixed. Often these games lose their built-in audience while failing to capture a new one. I installed the demo for Dawn of War 3, played the first tutorial and never touched it again, even though Dawn of War 1 is my favorite RTS. When I read that Marvel vs Capcom: Infinite was reducing team size and removing assists I was worried, and that worry proved justified — personally I'm not interested in a Simple 2000 Marvel fighting game and the new audience the streamlining was supposed to unlock never materialized. Few people seem happy with the simplification of multiplayer maps in Titanfall 2 or the weapon system of Destiny 2, or with the 2-gun limit of Resistance 2. For these cases and hundreds more you can claim that removing complexity is a solid strategy that was just implemented poorly. But then can't I claim that improved graphics and a seamless map are the reasons Monster Hunter: World did well, and it would have done even better had it retained more complexity from previous versions? Can’t I claim that increasing complexity is a great strategy and any counter-examples were simply poor execution?

Of course none of this means that all games should be needlessly complex or that obtuse UIs and unclear inputs are good. But I do think that "easy to learn, hard to master" is ultimately more a bumper sticker than a well-realized thought. Looking at Twitch popularity or Steam sales numbers I don't see any indication that "easy to learn" is a slam-dunk value proposition. If anything the PC games market often leans towards games that have more "needless" complexity and that are somewhat fiddly and baroque — towards "PC-ass PC games." (For example: Kingdom Come: Deliverance) Even the "hard to master" half doesn't look so solid in an age of "cinematic experience" content-tourism games that don’t demand much of players.

Not only is using "accessibility" to mean "immediate understandability" or "streamlining" not good word choice it also encapsulates an overvalued concept.

Accessibility Meaning 3: Approachability

If meaning 2 of accessibility is "easy to grasp" then meaning 3 is "apparently easy to grasp"; meaning 2 but from the outside looking in.

This meaning is based on apparent nature, so it can be a function of marketing rather than development. We see titles like Madden 2018 and Assassins's Creed: Origins instead of Madden 27 and Assassin's Creed 14 in part because those titles would indicate a buildup of mechanical plaque or essential narrative.

“Approachable" or "welcoming" captures the intended meaning here much better than "accessible." And as with understandability I'm not sure that apparent understandability is as attractive as we think it is, for the same reasons I'm not sure about understandability itself. So once again I'm opposed to both the word choice and the concept it captures.

Looking at the Steam Top 100 for 2016, by my estimation about half of the non-bronze titles have high apparent complexity. (I’m ignoring bronze purely for simplicity) The Stellaris description on the store reads “from the makers of Crusader Kings and Europa Universalis.” One of the first reviews reads "Stellaris is one of those games where you have to invest a lot of time and effort into learning how to play.” These are presented as selling points, not negatives.

In some sense by definition players want approachable games. But the games people want to approach may not be the games we typically consider “approachable.”

Especially for games with the potential for community, a baroque appearance can promise some fun communal problem-solving. Decades ago you might have heard the trick to beating King Hippo on the playground. In elementary school a girl lent me her copy of Shadow of the Ninja (a hidden NES gem) in exchange for an explanation of how to get past a gap in the sewer level of Ninja Turtles. (Just tap the jump button so that you don't bonk your head on the ledge above you. Get it together Kristina!) The modern equivalent is playing ARK with your friends online and sharing theories on how to best tame creatures or which items to craft first.

Some games foster community with multiplayer mechanics, but even a mostly single-player game like Dark Souls can foster community through lore and systems that encourage players to band together intellectually. I got over the hump in Dark Souls when my friend from out of state visited and helped me get a Black Knight Sword and defeat the Gargoyles. And for multiplayer games players who fumble through an experience together may end up appreciating both the experience and each other more, similar to boot camp or fraternity hazing. I've seen people remark that the Sea of Thieves beta had very little instruction but that’s often framed positively, as it encouraged players to work together and led to wacky high-seas hijinks.

So again: we shouldn’t use “accessibility” to mean “approachability” and approachability is probably overrated anyway.

Accessibility Meaning 4: Difficulty

There's a relationship between difficulty and accessibility though it's often oversimplified. Take a bullet hell game where the player can't see the bullets due to some visual impairment. In that case dramatically lowering the difficulty may make the game beatable but it doesn't address the real issue and delivers a very different, lesser, experience. In many cases difficulty has no bearing on accessibility at all: lowering the difficulty of a game doesn't help someone who has problems making out the dialog in cutscenes or who can't play the game due to epilepsy concerns. Accessibility guidelines encourage granular options and describe a range of disabilities; being bad at videogames is not a disability, though it can be the result of one.

It’s hard to object to optional difficulty tweaks, even as someone with contrarian leanings. While these topics may be contentious among fans or games media they aren’t particularly contentious among working developers. When Nintendo began introducing explicit helper methods in their games, starting with the Super Guide in New Super Mario Brothers Wii, there was an outcry from some corners but I sense only a small percentage of developers had any issue with it. The Mario games are made for all ages and the Super Guide is an opt-in way to help younger or less skilled players. Similar to the debate over ludology vs narrative, which is performed almost entirely by academics rather than working game developers, the debate for or against systems like the Super Guide is performed almost entirely by outsiders. 

Unlike the overrated concepts of real and apparent understandability, adjustable difficulty has been historically underrated. (Difficulty options in older games are rare) And while there is a distinction between accessibility and difficulty they are at least related, so it's hard to object too strongly to the terms being intermingled, even if they aren’t synonyms.

Meaning 5: Accessibility as Catering to Preference

This section is going to based heavily on Now Ubi’s opened the door, can we have our “Skip Boss Fight” button?, so you should probably skim that before continuing.

There are two arguments made in favor of the titular button: one is that boss fights present an accessibility issue, and the other is that the author just doesn't like boss fights. These are unrelated and the conflation of personal taste with accessibility concerns does the piece no favors.

Accessibility guidelines are typically highly specific. I can't remember seeing "bosses" as an accessibility category. If I had to sum up accessibility guidelines in one sentence it would be "give people fine-grained control", and a button to skip bosses is the opposite of that.

It's hard to imagine that someone who likes boss fights but has a disability would be happy to skip boss fights except as a failsafe. If I lost my hand in a farm accident (don't laugh, it happened to my college roommate) I'd still want to do boss fights and have the same experience as other players, which might mean adjusting the difficulty downwards. I certainly wouldn't want to skip them entirely — losing a hand wouldn't make me also lose my love of boss battles.

The idea that skipping inaccessible content is an accessibility solution is strange. It's like telling your wheelchair-bound friend that the solution to a concert venue that's not wheelchair-accessible is to skip the concert because the band playing it is awful. That only makes sense if your friend didn't want to attend in the first place! Otherwise the solution is just the problem restated.

The piece conflates inclusiveness, difficulty and accessibility. "Gaming has always been inclusive"? Accessibility concerns in games are relatively recent — show me an NES game with a colorblind mode or an SNES game that allows you to tone down bright flashes. Games are historically non-inclusive of people with disabilities, which is why organizations like AbleGamers exist. Yes, some older games included cheat codes, but the Konami code is a hidden difficulty setting, not an accessibility setting. Using different words for different concepts is good.

The biggest weakness of the article is that accessibility and inclusivity concerns appear in the piece to prop up and add import to the author's personal tastes. The author wants a “skip boss fights” button because they don’t like boss fights, not out of deep empathy for disabled players,

Providing accessibility options feels like a kind thing to do - there's some moral weight to accessibility concerns. There is, in my mind, no moral weight to preference concerns. Game development is a not a service hospitality industry in which the customer is always right. I've seen all sorts of wonky debates comparing mediums: well on a DVD you can skip the parts of a movie you don't like. Well in a movie theater you can't. All that nonsense aside, if you can't skip the parts of a game you dislike: you'll live. In an age of carefully crafted "we're listening to each and every fan" PR responses I suppose that's not something you're supposed to say but there it is.

To ignore accessibility concerns is to perpetuate de facto, if unintentional, discrimination. (That is why the Americans with Disabilities Act exists) "People who dislike boss fights" are not a historically oppressed people in need of redress and creating a product that people dislike is not discrimination. An accessibility issue is when someone wants to experience something but can't, not when they can experience something but don't want to. The way these two are intermingled in the piece, as if they are on even footing, only makes accessibility concerns appear trivial. To be as blunt as possible: if some disability prevents you from playing a game that's bad. If some preference prevents you from enjoying a game that's life. Accessibility is important in a way that preference simply isn't. They don’t belong in the same conversation.

Words: They Mean Stuff

There's a strong temptation to overuse weighty words and erode their meaning. "Problematic" is used to describe things that pose absolutely no problem to anyone. "Mansplaining" and "gaslighting" once both had real meanings; now they both simply mean "someone disagreed with me." The right uses "treason" to mean "unclapfulness"2 and "cultural marxism" to mean "anything I dislike."

Historically “accessibility“ was often trotted out as a euphemism for questionable streamlining efforts in the way that "VR content experience" is used in place of "short tech demo." But “accessibility” is increasingly used in a specific literal sense - that's a good thing. We already have perfectly good words to describe understandability or approachability or “things that conform to my preference.”

It's easy to consider only the most obvious accessibility cases: someone is colorblind or missing a hand. Expanding that to someone with a repetitive stress injury or chronic fatigue or whatever else is appropriate. I've seen suggestions that games that run poorly on old computers is an accessibility issue. Being unable to afford a new computer isn't a disability but it is literally access-limiting. So a wide definition of accessibility is good. What isn’t good is blurring the lines between accessibility, inclusivity, preference and marketing strategies, such that decisions to increase mass appeal are touted as accessibility, or that adding Burger King styled "have it your way" options are "inclusive" of people with strong preferences.

Addressing access and addressing preference can look similar but these are distinct concepts. “If you don't like it buy something else" doesn't work for people with disabilities as their issue is access, not preference. That line is fine, however, if someone just doesn't like the graphics on level three.

One strong similarity between the food and game industries is how fad-driven they are. Molecular gastronomy was all the rage for a while - molecular gastronomy is when you order steak and potatoes and get frozen steak chalk with potato foam. “Deconstructed” food was a trend that posed the question “what if you only did the prep?”

Plating is mostly aesthetic so it’s extremely faddish. You can serve a scoop of ice cream but if you’re fancy you can quenelle or rocher instead. Serving food over a smear was popular for a bit, as were spots and drizzles. There’s not much functional difference between these so they change periodically for the sake of change.

Which brings us to We Want Plates. We Want Plates is a twitter account / reddit sub / vital social movement that says “plating does have some functional requirement so please give me a vessel that at least properly holds the food.”

This sort of plating is increasingly common even though it fails at the most basic level. It’s not even novel or interesting anymore - it’s just annoying.

Today I’m looking at video game features whose popularity outstrips their worth - overused features, or once-ubiquitous features now sheepishly binned. Games don’t have many functional requirements so I can’t say many of these are objectively bad - but often their inclusion is a net negative.

It’s not too interesting to just list bad fads; I’ll also discuss the difference between justification and rationalize, and the kind of thinking that leads to faddish design.

Post-Processing in the 360 Era

Low quality bloom was an epidemic in 360 era. (This page has many good examples) Bloom can be hard to get right technically, but it seemed like some games weren’t even trying, instead leaning into glowing bald heads and white shirts.

There’s not much to say about the “piss filter” beyond “yuck.” In theory it makes for “gritty realism” but real life isn’t uniformly brown so I’m not buying it. Sometimes new effects are overdone to show off - better too noticeable than not noticeable enough. The first normal maps were high amplitude on shiny surfaces - normal maps looked cool when they were new so that’s understandable. But turning the screen brown was never impressive.

The red jam effect may be the worst of this era’s full screen effects. We determined that players need more visual cues when they’re taking damage or low on health - fair enough. But somehow that birthed the strawberry jelly effect.

Red jam on your screen is supposedly diegetic and realistic: if you get shot in real life the consequence is a glob of red jam gets smeared on the giant windshield in front of your face….? Much of the time this effect didn’t attempt to be a “blood in your eyes” effect or even a “blood on the camera lens” effect. Instead it was jam on a windshield.

Something I’ll keep coming back to is at the time developers rationalized these effects. But that we don’t see them much today reveals that these were rationalizations, not justifications.

In hindsight these effects look ugly and have aged poorly. The Last Remnant on PC has nice models and textures but the post effects look awful. The same is true of the remastered Kingdoms of Amalur: Reckoning - it looks nice until the presence of bad bloom, most noticeably on VFX.

Turret Sequences

I suppose the appeal of turret sequences is they provide a change of pace. But these sequences are most common in first and third-person shooters, and are often just worse versions of the game you’re already playing.

In Resident Evil 5 or Dead Space you can switch weapons, use melee attacks…move. Then you get to a turret sequence and it’s the same game but with one stationary weapon.

These sequences can work. The turrets in Monster Hunter are fun in small doses, since the game is primarily melee and the ballista and machine guns usually appear in set-piece battles. But it’s rare for them to work - more often than not they’re low points.

Not many game designers will admit “truth be told we put turret sequences in our game because it was the fashion at the time.” They may not consciously recognize that as their prime motivation. Instead they’ll say “turret sequences provide a palate cleanser.”

I consider that a rationalization rather than an explanation because much of the time it’s simply not true.

Slow Walking

Forced slow-walk sequences are something that’s just become A Thing Games Do.

"If every game is doing this there must be a reason” is seductive thinking. And sometimes it’s correct - if every game is doing something you don’t understand you may be missing something. But sometimes there’s nothing to miss.

NFL teams used to always punt on 4th down. Regardless of the score, the time remaining, or the yardage required for a first down - they would simply always punt.

A strictly rational actor would go for it on 4th down when appropriate. But while a coach could be fired for failing a 4th down attempt they would never be fired for punting, especially when every other coach in the league would also punt in that situation. So always punting was, for a long time, the accepted always-correct thing to do.

If every coach always punts on 4th down how could it be wrong? If every high-budget game has a slow-walking sequence clearly that’s correct…right?

It’s easy to say “question everything” but if you’re stopping to question every aspect of making a video game it’s going to take an extra 10 years. But we could probably afford to question more than we do, especially when it comes to design trends.

Questioning the win conditions of fighting games is how you get Smash Brothers. When Dark Souls released many complained that the attack buttons were on the shoulders - now I often hear people complain when other games use face buttons for attacks. In their minds what was considered “good design” is now “bad design” and vice-versa - the dogma has swapped. And those were well-liked features, not features like forced walk-and-talk or trite hacking minigames.

The slow-walking in No More Heroes, during which you listen to increasingly elaborate and unhinged phone calls coming from the Wii Remote speaker, is fun! But the majority of time I play a game with a slow-walking section my thought is simply “please just let me play the video game.” Walk and talk segments often seem to only exist because they’re standard practice - like punting on 4th down.

Smart people are good at convincing themselves of silly things. I think some game developers are good at disguising “other games are doing it” as more sophisticated reasoning.

Features of successful games have strong gravitational pull. If you’re making an open world game you probably want to add climbing and a glider, and while you can rationalize that the root motivation is likely “BOTW did it.”

Cover-based Shooting

There was a time when any shooter without a formal cover system got dinged in reviews. It wasn’t enough to have a natural cover system - if you hide behind a metal wall you’re safe, if you hide behind a wooden door you take some damage. No - a game had to have a “stick to cover” button with discrete protection rules, some mix of lean and blind fire mechanics, etc.

We don’t see a lot of that these days, for what I think is a simple reason: most of those games weren’t very fun.

Not only was cover shooting often not very fun, but in cases like Quantum Break it derailed the game. As I wrote once before:

One of the main complaints about Quantum Break was that it was a standard third person cover game with weak TPS mechanics that overshadowed the more unique gameplay elements . The rebuttal to that was “you’re playing it wrong” — you aren’t supposed to hide behind cover, you’re supposed to run and gun while using your powers. But then why does the game have familiar chest-high walls, a formal cover system and regenerating health? The level and system designers sure put a lot of work into things you’re supposed to ignore, things that indicate "stop and pop" rather than "run and gun" gameplay.

They made a game premised on using cool run-and-gun powers then saddled the game with “stop and pop” mechanics and level design.

Regenerating Health and 30 Seconds of Fun

Something that happens too often in game design is a game does something neat, it becomes a GDC presentation, that morphs into a memefied context-free distillation, then other developers repeat a warped version without understanding the original intent.

Halo’s “30 Seconds of Fun” is a prime example.

“30 Seconds of Fun” is, I believe, largely responsible for popularizing the pervasive “core gameplay loop1” talk we get today, and the idea that if your game is fun for 30 seconds you can loop those 30 seconds ad infinitum to create a hit game.

That’s not what “30 Seconds of Fun” was supposed to mean.

"All of which would be fine," he continues, "but the worst part is, everyone uses it to mean exactly the opposite of what I meant when I said it! They use it to say you only need 30 seconds of fun, and if you repeat something that is fun for 30 seconds over and over you have a game like Halo. There was a whole second half of the quote that got cut out of the Vidoc [video documentary] where I talked about taking that 30 seconds of fun and playing it in different environments, with different weapons, different vehicles, against different enemies, against different combinations of enemies, sometimes against enemies that are fighting each other. No 30 second stretch of Halo is ever repeated; the missions are constantly changing the context on you."

Constantly changing context is a hallmark of some of the best games. Resident Evil 4 and God of War (PS2), for example, have a huge variety of combat setups while rarely straying into one-off gimmicks or alternate mechanics. But “30 Seconds of Fun” has come to mean “endlessly repeat the same 30 seconds”, not “use the same great mechanics in widely-varying scenarios.”

Regenerating health is a key component of “30 seconds of fun” because it breaks gameplay into discontinuous 30-second blocks. You do 30 seconds of shooting, you hide behind a wall as your health regenerates, then you do another 30 seconds of shooting.

I have no complaints with the regenerating shields in Halo - it’s a scifi game so regenerating shields fit. In theory it opens up gameplay complexity - maybe you have a weapon that’s strong against shields but weak against health. But non-Halo games adopting this approach have been a very mixed bag.

An advertised strength of regenerating health is that it serves as a sort of save scumming, atomizing difficult segments of a game into discrete parts. In a game where you can’t heal between encounters the difficulty is additive- the more enemies you encounter the more you’re subject to attrition. If your health regenerates between encounters there’s no attrition, and the difficulty is more of a max function: if you can beat the hardest encounter you can beat them all. Is that what we want? The downsides of this approach are obvious.

The Getaway is often considered the game that popularized regenerating health. The Getaway was a game with two main gameplay modes - very fun open-world GTA-style driving sections and awful on-foot combat sections.

The on-foot sections in The Getaway have exactly the problem you’d expect: you can recover health between encounters, which means the game is trivially easy unless the encounters are very difficult - which they are. So the game is both too easy and too hard. In old beat-em-ups like Teenage Mutant Ninja Turtles you could cheese the game by barely inching forward, making enemies spawn in one at a time rather than in groups. That’s how you play The Getaway - you inch forward, fight an enemy, rest against a wall to heal up, repeat. It’s 30 seconds of tedium on loop.

When Doom 2016 released many were impressed that the game used glory kills to refill health rather than regenerating it. Glory kills aren’t health packs exactly, but they’re conditional health packs with different theming. Game industry institutional memory is so short, and we’re so fad-driven, that sometimes we forget about mechanics that work, or confuse “old” with “dated.”

That’s not to say the regenerating health is a fad that should die, but I suspect many games would be made stronger by ditching it.

The Jell-O Salads of Gaming

It’s not hard to look back and identify bad fads. But as game developers we should strive to identify valueless fads in the moment.

I don’t think Chicken Kiev or cheese fondue are popular these days but they’re fine dishes. Then there’s the Jello-O salad - a food that was never ever good. I wouldn’t say that regenerating health is a Jell-O Salad - regenerating health can be fine. But the “piss filter” is pretty close to Jell-O Salad territory. Most turret sequences are a serving of Jell-O Salad.

“RPG-style” colored loot in non-RPGs, where “Rare Ruby Batgloves” give you “a 25% chance to do 10% extra damage every third hit against Bane-type enemies” are Jell-O Salad. (I’m playing Stranger of Paradise: Final Fantasy Origin and on one level I got 30 of what are basically the same gloves) So are many skill tree implementations that space out meaningful rewards with imperceptible stat adjustments. I don’t like a lot of MTX implementations but I understand why they exist- they may not make for better games but they can make more money. But complex currency systems in games without MTX are usually Jell-O Salad.

Which brings us the modern Jell-O Salad that irks me the most.

We Want to Not Squeeze Through Cracks

That’s right - I wrote this piece to complain about “squeeze holes.”

I have two main objections to “squeeze holes.”

• They rarely serve a useful purpose (that isn’t equally served by something better)

• They lend the appearance of low quality

Second one first -Square-Enix tweeted out a video of Final Fantasy 162 that features a squeeze hole and this is the first reply I see:

The picture topping this section is from the first Unreal Engine 5 demo. It too features a squeeze hole.

Players associate squeeze holes with tech limitations - our understanding is that they’re fancy loading screens. When you add a squeeze hole to your game you’re adding a loading screen, and the Unreal Engine 5 demo is supposed to demonstrate the elimination of loading via the SSD and Nanite.

This a functional failure of the demo, not a subjective or aesthetic failure. The inclusion of the squeeze hole makes the demo less successful an advertisement for Unreal Engine 5, because it highlights an engine limitation. (That doesn’t actually exist)

There are rationalizations for squeeze holes but they read like rationalizations.

“The are less gamey and more realistic.” You could live for 1000 years and if you’re not into caving never come across a real life squeeze hole. Audio logs and blood-drawn graffiti are gamey, despite those too being “realistic.” Anything becomes “gamey” when it becomes a common gaming affectation.

“They imply one-way traversal, like a door that closes behind you” is another I’ve seen a lot. This one just doesn’t make sense: nothing about a narrow passageway implies it’s one-way.

“They can be used as gates.” So can gates and doors, magical barriers and a million other things. “They can be used to break up areas visually, and make for dramatic reveals” - sure but so can passageways, fog doors, etc. All of these are “to a man with a hammer everything looks like a nail” - to a team used to squeeze holes everything looks like a job for a squeeze hole.

These rationalization are flimsy, and they don’t apply to the Unreal Engine 5 demo since that’s not a game. There are no combat arenas to gate. There’s no need for one-way passages in a video.

You can convince yourself that squeeze holes make sense, but if they weren’t already in our lexicon as a way to mask loading I don’t think many would consider them the best solution to non-loading scenarios.

Why did the demo include a squeeze hole?

Epic wanted to demonstrate that Unreal Engine 5 could be used for next-gen AAA games, and those games have cracks you shimmy through, so their demo has them too. Fifteen years ago the demo might have had a turret sequence for the same reason.

Epic had to clarify that the crack traversal in their demo wasn’t a technical deficiency. God of War Ragnarok developers had to similarly clarify.

Butter served on a rock is goofy but functional; syrup spilling off of a board and onto your lap isn’t functional. Squeeze holes are like the worst “We Want Plates” offenders. I may not like turret sequences but maybe you do - taste is subjective. But adding the appearance of a loading screen and technical deficiency to a game is self-sabotage.

It turns out “Dodging in the Callisto Protocol Doesn’t Work” was the first entry in my own dodge-themed blogging universe - my Tom Cruise’s The Mummy.

Welcome to Part 2.

Player suggestions are often bad and their observations often wrong, but player impressions are never wrong. When players feel a certain way it’s important to understand why, and you can’t just ask them because they usually don’t know.

Many thought the dodging in The Callisto Protocol felt off. When I looked into it I found that I didn’t like the design, but the design also wasn’t well-implemented.

I see a lot of people panning the dodging in Lies of P and it’s similar story. I don’t think the dodging in Lies of P is broken, but there’s ample room for improvement.

One of the most important skills of a game developer is the ability to go beyond “this feels bad” and into “for these specific reasons”, as that naturally leads to “and here’s how we fix it.”

I agree that the dodge feels bad, for specific reasons I’ll lay out in the rest of this post.

Dodge is on Button Press, Not Release

A move like Ryu’s charged Denjin Hadoken using button release makes sense - you hold the button to charge up and release the button to unleash the move. Testament in Guilty Gear and Carl Clover in BlazBlue both use button releases to control a second character as their central gimmick. But if you’re making a game with a punch - just a plain-old nothing-fancy punch - it should probably activate on button press. This is even more true of dodging in an action game as dodging is extremely time-sensitive.

In Lies of P if you hold dodge your run, and if you press and quickly release you get a dodge. I’m not a fan of dodge and run on the same button but maybe they just ran out of buttons. In some games holding the button would transition the dodge animation into a run, but maybe that didn’t work due to the animations and properties of the dodge. So I get it, even if I don’t love it.

At first I thought dodge being on release was the main issue, but someone pointed out that in Dark Souls dodge is also on button release. I don’t remember ever having a problem dodging in Dark Souls, and honestly I didn’t even remember that dodge was on button release. So while dodge on release isn’t ideal it’s clearly not a deal-breaker.

With that in mind I decided to compare dodging in Dark Souls 2 (because that’s the one I have installed) to the Lies of P demo.

Timing Dodges vs Attacks

I ran some standardized tests on both games, which I’ll get to shortly. In doing them I noticed some odds and ends I’ll call out here.

In Lies of P if you aren’t locked onto an enemy you get a dodge roll, but if you are locked on you get a dodge hop. The dodge roll goes farther and has a more recognizable animation so it’s not just a cosmetic difference. When you’re just running around the game world doing nothing in particular you have a dodge that that travels far with a very recognizable animation, but when you’re in combat (presuming the game wants you to lock on in combat) you have a dodge that’s less distinct and covers less distance.

The two dodges also control differently, which I’ll get to in-depth later and is very strange.

Comparing Dark Souls to Lies of P it’s a lot harder to time dodges in the latter; I’m not sure why that is exactly, though I suspect enemy animation style is one culprit.

Enemies in Lies of P do have wind-up tells, but the time between the wind up and the attack is highly variable. This is a good thing to include in moderation - otherwise the player can just dodge with the same timing whenever they see the enemy wind up. But the timing is probably too variable, especially for early-game enemies.

In Dark Souls 2 the starting enemies have pretty slow attack animations, and in particular they accelerate into the swing pretty slowly. In Lies of P the once the enemy starts to swing they accelerate into a full swing very quickly.

The two of these together means that you see a wind-up pose you know you should be ready to dodge, but that’s not actually when you should dodge. Once their attack starts in earnest it’s fast enough that it’s hard to dodge on reaction.

The wind-up poses aren’t very distinctive. The starting enemies in Lies of P have two attacks: a big vertical overhead swing and a small vertical overhead swing, that both begin with the same arm on the same side. You can tell them apart - for the big swing the enemy turns their back to you to wind up - but they look more similar than different.

There may be other factors as well. I’m not saying these are factors - just that they could be.

• There could be a technical implementation issue where the opening frames of the dodge aren’t invincible - an order of operations issue perhaps

• The hitboxes in Lies of P could have more aggressive leading edges. (I’ve handily illustrated this below)

  

• The hitboxes might switch on earlier in the attack animation

• The window where you can cancel an attack into a dodge may open later (for this post I’m mostly considering dodges from neutral)

Whatever the reasons I often react what feels like a few frames too late to an enemy attack, in a way that feels “unresponsive”. That’s in quotes because I don’t think it’s actually unresponsive - in my unscientific testing the input lag didn’t seem bad.

Now on to the tests.

Test #1: Buffered Dodges

This is the first test:

1. Dodge to the left, then while you’re still dodging to the left:

   1. Press the left stick to the right

   2. Press and release dodge

   3. Release the left stick

This is what happens in Dark Souls 2. I would consider this correct.

Maybe another time I’ll go in depth on input buffering, but in short you want to buffer the player intent.

As the player my thought process is “I’ll dodge to the left, then dodge to the right, then start walking again.” So I input a dodge left, then I input a dodge right, then I start moving the left stick. Because of input buffering I can input the right dodge while I’m still in the middle of the left dodge - without buffering I’d need frame-perfect timing to dodge twice in a row without a gap in between. The key point here is that my intent for the second dodge is to input a dodge right.

This is what happens in Lies of P

I dodge to the left twice! This type of thing is subjective but that said: this is wrong. I input a right dodge but the game gave me a left dodge.

When the buffered dodge fires in the video I’m not holding left on the stick, I’m in neutral. So even if the game didn’t remember my intended direction I would expect the second dodge to be neutral - in Lies of P neutral dodge is a backstep, so I would expect left dodge → backstep. But instead I get two left dodges.

At the end of the demo you can unlock the ability to turn your dodge into a longer distance dodge by pressing the button again, the way you can in God of War and Stranger of Paradise: Final Fantasy Origins. Maybe it makes sense that extending a dodge would make the extension continue in the same direction . But I don’t have that ability at this point - I don’t even know it exists. And I’m explicitly holding right on the stick when I input the buffered dodge, not left or neutral. And I’m not double tapping the input, I’m inputting the second dodge slowly. Maybe the code has to deal with a lot of different cases including optional unlocked abilities and got messy, but what it’s doing doesn’t make sense based on my input and my current abilities.

But that’s not all.

Here’s a second test. It’s very simple:

1. Dodge to the left

2. While in the middle of the dodge, input dodge again while the left stick is in neutral

Before watching the video what do we think will happen? In the first test the game didn’t really care what the left stick was doing when I input the second dodge - it seemed to act like the stick was in neutral. So now that the stick is in neutral I’m expecting the same behavior: I’ll dodge left twice.

Huh.

What actually happened is I dodged left then dodged right. Or really, I dodged left and then did a backstep (which is the neutral dodge) while facing left. At the tail end of the video you can see that if I input left dodge, neutral dodge, then hold the stick as the buffered backstep fires I do the backstep opposite the current direction of the left stick.

Maybe in theory it’s cool that you can roll left, buffer a backstep, and then choose the backstep direction later, but the way this works is getting awfully complicated. Here’s the rules as I understand them:

1. If you buffer a neutral dodge it comes out based on the direction you’re holding when the dodge fires

2. If you’re hold the left stick when you buffer a dodge the game buffers a directional dodge, but it doesn’t care what direction the left stick was in, merely that it was being pressed

   1. If you’re holding the left stick when the directional dodge fires you dodge in the current left stick direction

   2. If you aren’t holding the left stick when the dodge fires you continue in the direction you were moving

While I can explain this logic my fingers don’t grasp it in the moment. When I enter an input my brain moves on to the next input. If I enter left dodge, right dodge, special move, I don’t expect that I can revise my right dodge into a left dodge by holding left when the right dodge fires - I’m done with the right dodge I’ve moved on with my life. This breaks my mental model of how an input buffer works: I input moves then they fire off when available. Instead this system remembers only some details of the moves you input, and draws other details from when the move fires.

It’s ok for a system that “just works” to be opaque. You don’t have to understand the exact details of the system if it makes intuitive sense. But this is hard to explain and understand and it doesn’t “just work” - it just feels weird. There’s a muddy relationship between what I enter and how my character responds.

But wait that’s not all.

What if you’re locked onto an enemy? When you’re locked on you do hops instead of rolls. Take a guess what happens with various inputs then watch this video.

So…

Here if you input left dodge, then buffer a right dodge, you get a backstep. I wrote above that if you input a directional dodge while already dodging it’s buffered but with the direction TBD. It turns out that’s not right, or at least incomplete. If you do that while locked on it buffers a neutral dodge (a backstep), not a directional dodge. The same input that gave you two left dodges when not locked on gives you a left dodge then a backstep when you are locked on.

You shouldn’t need a PhD in dodge mechanics to dodge in a video game.

Something has gone wrong here. The implementation has technical bugs, or logical bugs, or the design is fundamentally misguided, or it’s accommodating future abilities I don’t have yet. Maybe the dev team got too used to the system and has grown blind to the quirks. But it just feels off.

To understand this section you don’t have to perfectly understand how the system works. The point is it’s hard to articulate and hard to make use of - it’s confusing to both brain and fingers.

In Dark Souls 2 I tried a bunch of different dodge timings and inputs and it “just worked.” If you pay close attention there are some oddities - it seemed easier to buffer a directional dodge than a neutral dodge. That’s interesting but not bad - it never felt wrong, my guy did what I wanted. That’s what an input buffering system should do: capture player intent. In Dark Souls 2 my guy does the right thing and in Lies of P my guy frequently does his own thing.

Test #2: IFrames

I promise these other tests are a lot simpler!

Test #2: roll into a guy as he’s attacking to see how much invincibility dodges have.

Dark Souls 2

Lies of P

Since Lies of P has different dodges while locked on I tested that was well:

My conclusion: whatever they all feel fine.

I was honestly surprised by Lies of P. The dodging “feels bad” and I’ve seen people complain about IFrames but they seem reasonable. Generous even. It’s visually well-communicated - when you dodge forward the part of the animation where you begin to stand back up is about when you become vulnerable, which is what you’d expect.

That said, something that came up in this test (and became a frequent issue) is the physics of dodging into enemies. In Dark Souls 2 when you dodge into an enemy the second half of your dodge slides off and to the side and you end up going around them. In Lies of P you get stuck on the enemy.

This isn’t necessarily a problem, and there might be reasons for it: later you unlock a post-dodge attack (or so I’ve read), so sliding past an enemy might make that attack awkward to use. But I frequently found myself hung up on enemies while trying to dodge past them rather than into them, which was a problem.

Test #3: Enemy Tracking

Tracking is how much the enemy rotates to face you during an attack.

Tracking is composed of three main elements:

1. How aggressively the enemy tracks, IE how many degrees per second they can rotate to face you

2. How much of the animation they track for - they might track only during the windup, or during both the windup and the attack, or somewhere in between

3. Any special caveats - for example not tracking at all if the target is doing something special like dodging or teleporting

Here’s a test of Dark Souls 2 tracking:

This enemy has very slow tracking, and maybe also the tracking is only during the windup. It’s hard to tell if they track you while you dodge but I’m leaning towards no.

Either way your dodge is very effective - it goes far, it slides past the enemy instead of getting stuck on them, and the enemy doesn’t track well at all.

Here’s Lies of P

There are a bunch of not-great things happening here.

1. One of my dodges gets stuck on the enemy’s head

   1. I said before that not sliding past enemies when you dodge into them was fine in theory, but in practice this happened way too often by accident. In the video I was trying to dodge to the side and got stuck on the enemy’s big dumb hat.

2. The enemy tracking is pretty aggressive, lasts for a long duration, and it tracks well during a dodge. There are a couple points in the video, particularly at the end, where I roll and the enemy tracks me close to perfectly during the entire roll, then hits me. This is the first enemy in the game!

Here’s Lies of P while locked on:

The enemy tracking seems the same, with the hop subbed in for the roll.

The hop covering a short distance combined with the good tracking on the enemy attack means it’s easier to dodge if you dodge late, to take advantage of Iframes. But timing dodges at the last second is tough for reasons discussed earlier.

Enemies track well rotationally but they don’t seem to adjust for distance - the early enemies have short-ranged attacks and they don’t step into them much or appear to dynamically adjust if you’re farther away. So effectively they track side hops very well but backsteps not at all.

It’s perfectly valid for dodging to be more about creating space than abusing Iframes, but it feels unbalanced here in that only the backstep creates appreciable space, while also rendering the aggressive tracking moot and avoiding getting stuck on enemies.

A Last Weird Behavior of Note

There’s one other behavior I want to call out.

In this video I’m holding up on the controller and trying to dodge into the enemy. Instead of repeatedly dodging into the enemy I get trapped in a sort of negative feedback loop where I dodge more and more to the side - by the end it almost seems like I’m dodging away from the enemy.

I’m not sure what’s happening here - it may be a consequence of the camera movement. I hold towards the enemy and press dodge, and I slide off the enemy a bit to the side. As I do that the camera moves left and rotates. Now when I hold up on the stick I’m not holding exactly towards the enemy since they’ve drifted off to the right. So I dodge to their side even more, and as the process repeats and I become increasingly misaligned?

This may be exacerbated by how dodge direction is determined. (Oh no not this again!) I expect the game to remember my desired world-space dodge direction at the time I input it. (World space: translate the controller input into an absolute direction in the world, not relative to the player or camera) But the game determines the dodge direction at the time the dodge fires, not when input. Between the time the dodge was buffered and the time it fires the camera can move - in this case it moves a lot. So my original intent was to dodge towards the enemy (I was holding up AKA towards them), but by the time the dodge fires the camera has moved and “up” on the controller means something different?

Similar problems could arise for other in-combat camera movements - mini-cinematic events, manual camera adjustments, potentially even camera shake. Whenever the camera moves my directional dodge intent is distorted.

Pressing up on the controller and pressing towards an enemy aren’t precisely the same thing, since the enemy isn’t exactly centered on the screen. But with this lock on camera system, which keeps the enemy in front of you and largely centered, I expect “up” and “towards” to be roughly equivalent. In this weird case my character is moving what looks like 90 degrees or more off of what I intend. I’ve effectively lost the ability to dodge forward. Even if I pause a beat before dodging toward I still dodge to the side, as if the game itself got dizzy.

And it looks weird animation wise - it looks like my character begins to dodge forward then turns and goes to the left. I think this is just a quirk of the camera angle and animation.

I tried repeatedly dodging into an enemy in Dark Souls 2 and it seemed fine - you go off angle a bit but don’t end up in a negative feedback loop - you still dodge roughly into the enemy. It “just works.” (Though in fairness due to stamina limits I could only dodge 3 or 4 times in a row)

Nitpicking is the Job

All of this may seem rather nitpicky. But nitpicking is the job. Nobody is paying me to write this so it’s not my job exactly. But it’s the job of game developers to nitpick their own game.

When players say “this feels bad” you can respond with “they’re playing it wrong” or “what do they know?” but that’s not going to make the game better. (Or sell better) We should take common complaints seriously. If you want to increase game performance the first step is figuring out why it’s slow. If you want to fix “this feels bad” the first step is figuring out what that means exactly. That involves a lot of nitpicky detailed analysis but sweating the details is the job. Almost every game you love had team members who sweat the details.

What’s a Hook?

According to this oft-referenced video a hook is “something that gets stuck in people’s heads” and is both surprising and desirable. A hook can be both pre and post play.

Other frequent definitions include a “unique selling point”, a splashy concept, a novel mechanic, or a memetic / viral element.

Note on Ryan Clark

Clark is a good commentator. Video game advice is often poor, and sales / marketing advice especially so. It consists largely of hindsight, just so stories, and analysis that’s merely plausible rather than correct. It’s often characterized by low economic literacy; I can’t count the number of times I’ve seen people lament median sales data without grasping the implications of “median”, or independently discover the concept of supply and demand.

Clark’s central idea is that predicting the sales of games is a skill you can practice and develop. Anyone can write “here’s 10 reasons Flappy Bird sold well”- that requires no knowledge or insight; it’s a creative writing exercise. “Here’s 10 reasons Flappy Bird is poised to sell well” is much harder analysis to reliably produce. Nothing I’m writing is meant as a criticism of Clark - I like his approach.

Hooks and Video Game Jargon

A hook is sharp bent thing you use to catch a fish.

A music hook is well-defined: it’s the catchy part of a song. Hooks are an important part of how music has traditionally been sold. Bands create albums, and from those choose the most immediately-appealing (“hooky”) songs to package as singles for sale or radio play. (Or often, they set out to create radio-friendly singles)

The rough game equivalent of a single is a demo, and the hook would be the most immediately appealing part of that demo. Given how few games have demos, and how inaccessible demos are compared to radio songs, this should already be setting off small alarm bells.

A hook in music is not a concept. The opening notes of Under Pressure in Ice Ice Baby are a hook; Vanilla Ice saying “we’ll play a catchy riff and then I’ll start rapping” is not a hook. “It’s a concept album about the year 2112” is not a hook. Cool album art is not a hook, though it may prompt sales. Word of mouth is not a hook. “Hook” has a specific meaning.

The purpose of jargon is to simply encapsulate well-defined concepts - something game development jargon often struggles with. In geometry a convex hull is, intuitively, a shrink-wrapped tight shell. The exact definition is complicated. The Unreal Engine button labelled “generate convex hull” is a lot simpler than “generate the convex intersection of a collection of hyperplanes that minimize the sum of the positive signed distances of each point in a set to that set of hyperplanes.”1 It’s convenient to just say “convex hull.”

Useful jargon has a precise, if complex, meaning. “This song needs a stronger hook” is a useful critique of a single because it’s specific and actionable.

The video game “hook” refers to a grab bag of unrelated elements. It can be a concept, screenshots, an art style, a unique mechanic, a sales angle, something viral. It can even be a “post-play” hook. As such “this game needs a stronger hook” isn’t useful - it’s not actionable. All it really means is “the game isn’t compelling”, or simply “I don’t like it.”

The Definitional Problems with “Hooks”

I don’t believe a “post-play hook” makes sense. Why hook a fish after you’ve already caught it and put in your cooler?

A hook, according to Clark, can be a reason someone buys the game, but also a reason someone enjoys the game after they’ve bought it - these are two wildly different things! Using the same term for both makes little sense. I’m going to reject “post-play hooks” out of hand because we can call that “word of mouth” or “customer satisfaction.” Strong word of mouth can drive sales but that’s not a hook.

Clark’s video posits that a hook is surprising yet desirable, similar to the keys to a successful restaurant. Are those the keys to a successful restaurant? I’m dubious. What about menu, head chef, location, rent, cost of service workers and food, regional preferences, competition, front of house issues, etc? Poking holes in the analogy has limited use, but I don’t think the analogy works. Certainly the key to a successful movie is not surprise. If you think otherwise check out the 2022 box office and the success of films like Top Gun Maverick.

Many best-selling games aren’t surprising at all - if anything their conformance to a narrow formula is the selling point. Some modern hook analysis attempts to clumsily address this with the idea of “anchors” - positing that games need hooks but also the opposite of hooks…

The “desirable” half of “surprising and desirable” is tautological - for a game to sell well people have to desire it, yes. I don’t think “surprising” is necessarily true, and is subsumed by “desirable” anyway - it’s one reason gamers might desire something. All this definition says is that for a game to sell well people have to want to buy it.

I suppose one could argue that a hook has to be surprising simply by definition, and that any other reason someone might desire a game doesn’t technically qualify as a hook. But why would we elevate surprise above good graphics or high production value or underserved genre or any other sales driver?

In music a hook is just one type of sales driver, but it’s a prime one - it is (or was, in the radio era) a centerpiece of sales and marketing machinery . I don’t see any reason to believe that surprise in games is more relevant than any other sales driver, and there’s no sales and marketing machinery that centers surprise.

Do Steam Top Sellers Have Hooks?

Let’s look at the Steam 2021 best sellers. (I’m going to mostly ignore free to play titles, because almost none of the conventional wisdom around sales / marketing / hooks addresses them)

Does Battlefield 2042, a best selling game, have a hook?

Does it have a splashy interesting concept? Not really. Does it have a unique selling point? Is it surprising? No and no. Of course it has some differences from previous versions but that’s true of nearly every sequel. “Specialists” are new to the series but the idea of “heroes” is well-worn. It has 128-player matches which is bigger than 64 but in a very predictable way. 128 players is a new maximum for Battlefield (I think…) but MAG had 256-player matches.

I don’t believe that Battlefield 2042 sold well because it has great hooks - I think it sold well because it’s a new Battlefield, with enough changes to make it plausibly not the same game as last time.

That formula holds true for a huge number of games on the list. Forza Horizon 4, FIFA 22, NBA2K. Lots and lots of sequels that sell well because they are known quantities. CIV6, Sims 4, Age of Empires 4. Mass Effect: Legendary Edition is repackaged old games.

Back for Blood was sold on it’s similarity to Left for Dead - the “hook” there is “it’s like that other game you like.” What is the hook of DOTA2? What’s the hook of GTA5? It’s the next game in the series, it has high production value, it has multiplayer - are those hooks?

Let’s look at the games that arguably have hooks. I don’t really know what NARAKA:BLADEPOINT is but sure, let’s say it has strong hooks. Valheim is interesting - you could argue the graphics are a hook, but if the game had flopped I suspect people would point to graphics (“it looks like a PS2 game!”) as one reason why. Arguably the concept is “hooky” - but is it hooky because it’s surprising and new or because it’s solidly in a genre Steam users like?

Dead by Daylight - no arguments here. Same with It Takes Two. These are games that have a splashy unique-ish concepts. There are a few more games like those on the list. But then there’s also a Monster Hunter sequel, a Crusader Kings sequel, an Elder Scrolls game, Total War, Guilty Gear, Flight Simulator.

I’m not going to do this for the Steam 2022 games but the same analysis applies. Many of the top sellers in 2022 are sequels to top sellers in 2021, or are the same game. CounterStrike, Rainbow 6, Monster Hunter, Dead by Daylight. I have the two pages open next to each other in my browser and at one point I got confused and thought I had 2021 open twice.

The idea that people want new and exciting games sounds great - it’s optimistic and compelling. But I don’t see any reason to believe that it’s true. The Wandering Village is a game I think embodies the hook (and “anchor”) ideal. It’s “anchored2” in a genre PC gamers like, and it has a splashy original concept. But I don’t see it high on the list. Instead I see the highly original and surprising….Call of Duty. As well as Battlefield 5, two different Forzas, two different FIFAs. It’s rough out there for original concepts. There’s Stray. Ok.

It may seem unfair to compare The Wandering Village to Call of Duty. If you compare The Wandering Village to games in the same genre with approximately the same budget it still doesn’t come out looking great. (This is left as an exercise for the reader) If you compare Call of Duty to “hooky” games in the same weight class - well which games are those exactly? AAA publishers don’t appear to value hooks very highly. Presumably they believe the return on investment is higher for safe games.

Here’s the Steam most-wishlisted games3. #1 is Hogwart’s Legacy. #2 is The Day Before.

The Day Before is an open-world MMO survival set in a deadly, post-pandemic America overrun by flesh-hungry infected and survivors killing each other for food, weapons, and cars.

This is a lot less hooky and a lot more “give the people something predictable.” I wrote this section before the controversy around The Day Before arose. One aspect of that controversy is that it uses shots and staging cribbed from other games and trailers. That it looks like a mishmash of existing games is why it’s anticipated.

What’s the hook of any of the games on the top 10 November 2022 NPDs?

Hindsight Analysis (Is Easy)

A major issue with hook analysis is that it’s mostly post-facto. (Which is why I appreciate Ryan Clark’s predictive approach, even if I don’t agree with his hook take) If you believe that games need hooks to sell and a game sells well you might skip over “does this game have a hook?” and right to “this game must have a hook, so what is it?” Conversely if a game does poorly it must be due to a weak hook, and your job is to identify that weakness.

What’s the hook of Overwatch? It’s a hero shooter - hardly unique. It’s a hero shooter mixed with TF2 or Counterstrike or something - isn’t that also Paladins and Valorant? Also Gigantic and Battleborn to a lesser degree? And a bunch of other games I’m forgetting.

People claim that the heroes are the hook. Every hero-shooter has heroes. Maybe Overwatch has the coolest-looking heroes? But here’s the rub: if Overwatch had done poorly people would say “the heroes weren’t compelling.” It’s a lot of broad archetypes / stereotypes. Cowboy-man, emo guy with a skull mask, Japanese Ninja / Samurai guys who say things like “swift as the wind.” There’s Sexy Girl, Sexy Girl with ‘Tude, Sexy Girl with a Mech Suit, Sexy Girl with a Suit that Looks like a Mech Suit but Technically isn’t a Mech, Sexy Girl with a High Tech Suit Who is Different From Previous Girl Somehow, Sexy Sniper Girl. And Neinhalt Sieger.

On paper Lawbreakers has stronger hooks: gravity shenanigans and “the fastest game around.”

What’s the hook of Hollow Knight? The trailers for the game do what marketing experts tell you not to do - it’s mostly generic. “Master new powers.” “Tight, fluid gameplay.” You could claim the art style is unique but an indie platformer with a unique art style might be the single least unique indie game pitch. “You can use mana for spells but also to restore health” is a mechanical detail, not a hook. If you’re a person who believes games need hooks you can probably find them in Hollow Knight but that’s motivated reasoning - AKA cope.

If Gotham Knights sells poorly you can claim it lacked hooks - it’s a Batman game without Batman and it ditches the well-liked Batman combat system. If it sells well you can claim it had strong hooks - it’s a Batman game with co-op and 4 selectable characters with unique play styles.

That’s the vast majority of video game marketing and sales analysis - explainers for past events. When sales and marketing gurus try to predict rather than explain they often admittedly aren’t particularly good at it.

William Goldman famously wrote about the movie business “nobody knows anything.”

Nobody knows anything...... Not one person in the entire motion picture field knows for a certainty what's going to work. Every time out it's a guess and, if you're lucky, an educated one

In video games everybody knows everything. There’s so much advice on every topic, including sales and marketing, from people who have no apparent expertise and are never deterred by being wrong.

Why People Buy Games

Chris Zukowski wrote two related blogs about how players navigate steam and how they shop during a sale. These were well-received, I think because they match the experience most readers have, but contradict some established video game wisdom. In particular shoppers look for genre fit rather than novelty.

When trying to pitch your game to publishers or press, the general advice has always been “tell them why you are unique. What distinguishes your game from all the rest?” That is still true for gate keepers and curators. But in the hours of user testing I did, none of the participants said “I wonder what makes this game unique?” or “What is the unique selling proposition for this game?” or “hmm why is this game different from all the others” Instead I heard a lot of people saying “Oh so this game is just like game XYZ” or “this is right up my alley because I like these games” or “Ah this is my type of game.” 

What gets someone to pay for a video game? I don’t have any hard data so this is just my read, based on which games sell and how I and my friends shop.

1. The game is a sequel to a game they like

2. The game is from a publisher / developer they trust, so they at least consider the game in question

3. The game is a style they like - if not a sequel it’s a spiritual successor or in a preferred genre

4. The game looks expensive and well-made - it has high production value and seems “AAA” (this is more applicable to console games, particularly Sony and MS)

5. The game seems buzzy due to a combination of word of mouth and marketing

These aren’t mutually exclusive with having a hook. Stray has a hook and also looks expensive. But if you look at sales figures and probably your own buying habits you might be surprised at how rarely the “hook” matters.

What is the hook of God of War: Ragnarok? I don’t know that Ragnarok even has a twist. The marketing message is “it’s another God of War” - it’s “bigger, better and more badass” in non-specific ways. The same is true of Horizon: Forbidden West - it has underwater sections is that a hook? (Gamers famously love underwater sections!)

Some sequels are at least plausibly different from the preceding game. Monster Hunter: World is different from Monster Hunter: Generations, and Monster Hunter Rise is genuinely different from World. (Largely by circling back to the older game style…) I don’t think there’s anything markedly different about the latest Madden or FIFA or NBA2K or Call of Duty. I’m sure superfans of those games would tell you about small mechanical changes that make them different, in the same way I would assure people that Street Fighter 2: Hyper Fighting is totally different from Street Fighter 2: Champion Edition - basically a whole new game!

A hook doesn’t seem important if you have an established franchise or an established IP. It’s probably not very important past a certain budget level - the “hook” of the The Last of Us is that it’s another Sony-exclusive expensive cinematic AAA game from the makers of Uncharted.

But what if you’re making a low to mid-budget game with no established franchise or IP? Maybe then the hook matters a lot? I’m still not convinced.

Yes there needs to be a reason someone would buy your game. But that’s not necessarily a hook, unless “hook” is incredibly broad. What about all those Steam survival games? Or light-RTS / Sim-City medieval village builders? A lot of those do well and from a high level look similar.

The Misguided Quest for Hooks

From Reddit’s game design sub:

I'm just now getting an idea for a video game I want to design with Phaser.js, and although I like the concept, I'm sure we can all agree that it's quite uninspired sounding: the concept is to choose one from a bunch of Public Domain characters to fight creatures from mythology, gaining levels, recruiting other public domain characters, and using currency to buy more monsters along the way.

I don’t think this sounds particularly uninspired.

“Public domain characters fight creatures from mythology” sounds a lot like God of War or Age of Mythology or Fate/Grand Order. If you think your own idea is uninspired players are likely to as well, but there’s nothing particularly uninspired about this idea.

When described in simplest terms most games sound uninspired. In Diablo you go into a big hole and click on some skeletons. I’m phrasing this for dramatic effect but I think you have to admit that doesn’t elide much. Starcraft is Warcraft with different races. What is inspired about Elden Ring? An open-world fantasy game where you run around with a sword and bop creatures on the head oh wow totally original!

When talking to a friend recently I described Souls games as games with a specific coherent vision, but the individual elements of that vision are mostly generic. You fight against a guy with a sword. You fight against a different guy with a different sword. Maybe a guy has a hammer instead of a sword. You can fight a dragon, and a different kind of dragon, and a third kind of dragon. I’m not being dismissive - my point is that high-level novelty is hard to find.

Souls doesn’t have splashy unique mechanics either. The Team Ninja souls-like games (Nioh, Nioh 2, Wo Long: Fallen Dynasty) have way more mechanical jeuje: ki, morale, stances. Elden Ring is more Street Fighter than Guilty Gear.

People who offer game sales advice often push for twee game ideas. If you present a Steam capsule description4 like "an RTS town-builder where your band of Vikings establishes a foothold in a new and foreboding land" they'll tell you that's boring and samey, but "you play as a telepathic dog who teams up with an alien baby to solve The Mystery of the Missing Snausages" is dynamite. Which of these games will sell better? An unfair question with too many variables but probably the first one.

The quest for hooks often leads developers to mechanical “unique selling points” that are detrimental to game quality. Devil May Cry (DmC) has color-coded enemies immune to some weapons. Only half your arsenal works against some enemies - is that fun? Bayonetta 2 has flying sections with unintuitive controls that similarly reduce your moveset. Xenoblade Chronicles 2 has many novel mechanics that add nothing or detract from the game - field skills that compel you to tediously navigate menus, a combat system designed for boss battles that leads to inflated HP on normal enemies. Each new Mario RPG has a gimmick like using stickers or splashing colors when what people want is traditional mechanics like levelling up.

Arguably systems like these help sell the game even if they don’t make the game better, by serving as marketing bullet points. But is anyone buying DmC because it has red and blue enemies?

Effective marketing often doesn’t require much in the way of underlying mechanics. Telltale got great mileage out of “Suzy will remember this” but that’s the illusion of mechanic more than actual mechanic - the branching in those games is based on discrete major crossroads not cumulative micro-decisions. Almost every The Last of Us 2 preview and review mentions that enemies have names, which is supposed to make killing them more emotional. Final Fight had named enemies - that’s not a novel feature. Another major talking point was that Ellie can jump.

The new gameplay features being implemented in The Last of Us Part 2 are rather mundane from a broader context. For example, The Last of Us Part 2 will allow Ellie to jump. However, for The Last of Us Part 2, it's a major addition. Ellie being able to leap over obstacles grants her a level of dexterity that Joel could never have managed. It opens up a ton of new gameplay options that Naughty Dog has to build The Last of Us Part 2 around, as well.

This is a preview that’s very excited about jumping - a mechanic that’s been in a zillion games since Donkey Kong and Pitfall. This is smart marketing not unique mechanics.

Having a mechanical unique selling point is different from simply saying you have one - sometimes the latter is enough. And I’ve played so many games where the unique mechanical gimmick is one of the worst aspects. Tactics Ogre for PSP added multiple novel mechanics that uniformly made the game worse, then Tactics Ogre: Reborn removed those and added some new novel mechanics that, while better than the PSP version, also made the game worse.

In some games the unique mechanic is the game - you can’t have Octodad with standard controls. But in many, many games with “unique” mechanics those mechanics are non-essential and annoying.

In Conclusion

Maybe it’s unfair to ask “do games need hooks?” Not many people passionately argue that games absolutely need hooks. But we do talk about hooks an awful lot.

“It’s good if your game has a hook” isn’t controversial, but it’s also not useful. Your game needs a good trailer or good screenshots or a good concept or a good elevator pitch or a unique mechanic or a unique art style or it needs to go viral somehow or maybe it needs a “post-play hook” AKA make it high quality and enjoyable - well sure.

A player might say an action game “has loose controls” or “feels unresponsive.” An articulate game developer would hopefully say something more meaningful: “the camera has high inertia and low acceleration, so it feels unresponsive” or “the engine itself has significant input lag” or “the game has apparent input lag because animations have long, subtle blend-ins.” Saying that a game feels unresponsive isn’t wrong and it’s minorly helpful but it’s not real analysis - at best it kicks off real analysis.

Hook analysis is similarly not real analysis - it’s someone shrugging and saying “I dunno it’s just not exciting - it’s not grabbing me.” That’s fine from a layman but not from an “expert.”

This was a long winding road so summarize:

1. “Hook” is too broad to be useful, and narrower definitions are often arbitrary or simply don’t make sense

2. There’s not much reason to believe that hooks, whatever their definition, impact sales all that much

3. The search for hooks often leads developers to add annoying, superfluous elements to games

4. “This game needs a hook”, “what’s the hook?” and similar statements are observations you’d expect from people with limited knowledge of games, not domain experts

(* Honestly who knows)

How Callisto Protocol Dodging Works (In Theory)

This explanation is easy to understand but reading it makes me a bit nervous. This is a game with action combat but this sounds like a very “cinematic” approach. The timing doesn’t matter. The direction you dodge in doesn’t matter except in a mini-game style way. Good action combat includes timing, reaction, anticipation, observation and mastery of systems, and this mechanic includes none of those. You don’t have to time, react to or anticipate enemy attacks, nor do you have to observe their incoming direction or range. You just hold left. There’s no system to master here because there’s no system - this is effectively a QTE.

Dodging In Games (And In Real Life)

Arguably jumping over a Goomba is dodging, but let’s consider video game dodges as a separate verb. Mega Man 3 had the slide. King of Fighters 94 has the sway dodge. Mike Tyson’s Punch-Out!! had a dodge in 1987. Dodging has been common in 3D action and adventure games since Ocarina of Time.

These dodges consist of two parts: moving out of the way of attacks (sliding under a projectile as Mega Man, slipping an uppercut as Little Mac) and using invincible frames (i-frames) to pass through an attack. (The KOF sway)

Real life has no i-frames but you can still dodge by moving out the way. In boxing this is “slipping a punch.”

Typically you slip a punch to the outside, as in this gif. (Towards the incoming direction)

This is also how you use i-frames to dodge attacks in games: you dodge into the attack, not with it, to minimize the time your hurtbox overlaps the enemy hitbox.

Video game dodges are nearly always systemic - you enter the command, your character performs the dodge, and the outcome is the result of interacting systems. If you dodge late and have many i-frames you’ll be safe. If you dodge late and don’t have i-frames you’ll be hit. If you dodge too early and in the wrong direction your i-frames may run out while the attack is still intersecting you. If you act early and your dodge has a lot of movement you may be safe simply from the space created, regardless of i-frames.

There’s no confusion about what happens when fighting multiple enemies. If those attacks don’t reach you or they only reach while you’re invulnerable you’re safe. Otherwise you get hit. If two enemies swing at you maybe you can use i-frames to avoid one and use the space created to avoid the other. Maybe you can pass through both attacks at once, or if the attacks are spaced out dodge one then dodge again to avoid the other. If you understand the basics of the system you understand how it applies to multiple opponents, and typically to projectiles and other damage sources as well.

Dodging In Callisto Protocol

Dodging in Callisto Protocol doesn’t work like the dodging in anything I can think of. I’ve seen people compare it to Punch-Out!! but that game is all about timing. (And in Super Punch-Out!! the direction you dodge sometimes matters) When making a game you don’t have to stick to what other developers have done; I dislike the game design school of thought of “we’re going to copy/paste from existing games and add our small twist.” But if you’re going to deviate from a well-established element - to break the rules of video game grammar - you need a good reason.

It’s very easy to technically describe what happens when you dodge in Dark Souls or Bayonetta or hundreds of other action-adventure games. Your character moves and/or is invincible for some amount of time. It’s hard to describe what actually happens when you dodge in Callisto Prototocol. Do dodges have invincibility or is it more that enemies are coded to pass through you if dodged in the correct direction? Put another way, if you successfully dodged an enemy attack and halfway through that dodge a fireball from someplace else passed through your body would you get hit?

Does the movement of your character matter? Can you dodge the wrong direction but still avoid getting hit because the hitbox flat-out missed your hurtbox? Does the game have hit and hurtboxes? Maybe not every gamer knows that terminology but they know that if an enemy sword intersects your head that’s a hit. If you dodge the correct way in Callisto but the enemy limb still intersects your body is that a hit? If you dodge the wrong way but the enemy limb totally misses your body is that a miss?

In Super Mario Brothers you press A and Mario jumps and where he lands is up to you. In games like Assassin’s Creed you hold the stick and your character sorta jumps around on his own. The Callisto Protocol dodge is a lot like the latter. You hold left and your character chooses when to dodge on his own, in a way that feels suggested rather than direct.

Is that fun?

In Dark Souls you have to choose precisely when to dodge and what direction to dodge in - that’s a major element of mastery in the action portion of the game. In Callisto Protocol you hold left.

If you want to grow as a game designer you should try to understand game design decisions even if you don’t like them. I don’t enjoy the basically automated platforming in games like Uncharted but I understand why it exists and I’ll concede that it may be the best choice for that game given the target audience. But it’s hard for me to understand why this is how dodging in Callisto Protocol works. I’ve heard the game described as a brawler or even a one-on-one fighting game but as someone who enjoys those genres I don’t get any enjoyment out of this system. I don’t understand what’s supposed to be enjoyable, other than that you can get into a certain rhythm and that feels good I guess?

It seems very animation-driven rather than action-game-mechanic driven. For the first 20 minutes of the game you slow walk through corridors, squeeze through tight passages, do cinematic-QTE-style vaults over objects, and lumber like a well-animated Frankenstein1. The game is full of mini-cutscenes that take away player control. Towards the beginning of the game there's an exciting sequence where you have to crawl towards the cockpit and avoid getting sucked out into space and I'm genuinely unsure if I was controlling my character or if that was just a cinematic. I held up on the stick and occasionally pressed left and right and beat that section - did I have to use the stick? If so I was clearly giving vague suggestions to an almost entirely authored set-piece.

So my suspicion is that the dodge mechanic is also very animation-driven. The animators carefully crafted enemy attack animations and successful and failed dodge attempt animations. In order to look good those animations have to line up in space and time, which is why your character feels stuck facing enemies and why you have no control over the dodge timing - they are synchronized animations, more or less. That’s my best guess because it’s hard to understand what’s appealing about “hold left” as a combat mechanic.

Why Is Dodging In Callisto Protocol Hard?

If you hold left, regardless of timing, you automatically dodge any attack. Then you hold right and automatically dodge the next attack. This sounds trivially easy - really you should never get hit, at least when fighting one opponent. You walk up holding left, then after dodging switch directions and you’re invincible. Before playing the game myself I was ready to complain that the dodging is too easy.

But dodging in Callisto Protocol is weirdly hard!

I think this is because the implementation is just broken. But we’ll get there.

One of my video game design mantras is that a tutorial can’t make something that’s unintuitive seem intuitive. The Callisto Protocol tutorial is well-written and understandable2 - that’s not an issue. But the system it’s describing is just strange. It’s just not how dodging works, in video games or real life, ever.

There are also a number of unfortunate decisions that make it hard to understand. When enemies attack their animations often include a very deliberate anticipation pose where they pull their arm way back to one side. If you fail to dodge an attack that clearly swung from left to right you might think “I dodged the wrong way”, because the animation makes the direction of the attack seem important. There’s a ton of animation weight given to a thing that doesn’t actually matter.

In theory the only reason to fail a dodge is you dodged in the same direction twice in a row. So if you fail a dodge should you dodge in the same direction the next time or the opposite one?

Say I dodge left successfully, then dodge left again and fail because you have to alternate sides. Now which direction do I dodge in? In one view the correct pattern was left, right, left - so I should dodge left again to get back on pattern. On the other hand if I dodge left again then I’ve entered left twice in a row (actually 3 times in a row) and the system is supposed to make me alternate. Maybe if I pressed left twice I need to get back on rhythm by pressing right next time - to re-establish the left-right pattern.

Left or right? Neither answer is more logical because the mechanic is so arbitrarily mini-gamey - either answer gets you back in rhythm, depending on your perspective. In most games I would adjust my timing later to take advantage of i-frames, or dodge earlier to create distance, or try to dodge into the attack rather than away. Figure out what I did wrong and correct it. With this mechanic there’s no logical wrong here, I just didn’t do what the game designers wanted me to do.

I have one other question that leads into the final “it just doesn’t work” section: how long does it take for the game to forget your last dodge direction?

What happens if I dodge an enemy attack and then trigger a 30-minute cutscene. When that cutscene ends do I have to dodge in the opposite direction? Ok 30 minutes seems awfully long but what about 5 seconds? 2 seconds? What if I dodge, attack once, then need to dodge again. What if I dodge, take one tiny step, then need to dodge again? In that case do I need to dodge in the opposite direction or have I returned to a neutral state?

Again it’s hard to know because the system has no underlying logic. There’s no reason why I should have to alternate dodges at all - that’s just how someone decided it should work. That same person also decided how long the system remembers your last dodge direction and holds it against you. There’s no way to hazard a guess other than they probably didn’t choose an extreme value like 1 millisecond or 10 minutes.

The Dodging is Fundamentally Broken

The dodging is hard because it’s not governed by any real logic, runs counter to the long-established history of dodging both in video games and real life, and is murky in the details despite the clear tutorial text. But also, even if you accept that the dodging is weird and just roll with it, it simply doesn’t work correctly.

At 4 seconds into this video why do I get hit? At 20 second in why do I die?

At 4 seconds in I’m clearly pressing left well before the attack hits me. According to the tutorial: “there isn’t a timing window. Just be holding left or right before the enemy hits you.”

I have just aggroed this enemy - before that he was doing a little mini-cutscene. Did I get hit because I dodged left before the mini-cutscene started and my new left broke the alternating rule? If so that’s wildly unintuitive. This is, in my mind, a fresh encounter. Maybe the game has horrible input lag and my left press didn’t register? I did press it pretty late but I definitely pressed it well before I got hit - I pressed it with the timing that would give you a perfect dodge / block in many action games.

At 16 seconds in I successfully dodge left. At 19 seconds in I dodge left again and die. Does that count as dodging left twice in a row? That’s 3 seconds later, which is an eternity in an action game. It’s not like the enemy did two strikes back-to-back, or even strike, wait a beat, strike. He swung once, walked backwards, waited, walked forward again, and then swung. This is very clearly (in my mind) a reset to neutral.

Here’s a second clip of the full tutorial encounter:

This is a mess. At 4 second in I dodge an attack and then the enemy grabs me. There’s a ‘Y’ QTE prompt and I hit him a couple times. Then at 11 seconds the next enemy attacks and I dodge the “wrong” way. You can see the game wants me to press right but I press left at first.

It’s been 7 seconds! Did the game really remember that I pressed left 7 seconds ago and hold that against me?

That’s what I thought at first. But it’s even worse than that. When I fight the first enemy I dodge left then right. When the second enemy attacks me it forces me to dodge right again. It doesn’t even register when I press left. So the game is breaking it’s own rules in the very first combat encounter. The game says that which direction you dodge in doesn’t matter, but it will only accept right and not left here. And right is the same direction I dodged in last time, when the system is supposed to make you alternate.

This is the tutorial encounter!

Why do I have to press right to dodge the second enemy? Is the answer that they only animated one dodge direction and prompt, and the one they chose to was incorrect by their own rules?

Halfway through the video I meet the enemy in the first clip. On his first swing I hold left and dodge successfully. In the first video I also held left and it didn’t work. Then I hold right and…get hit. Maybe there’s input lag but my character clearly dodges to the right before taking damage, which is the correct alternating direction. So I hold left and successfully dodge, then hold right, play a dodge right animation…and get hit?

How does this game work?

I get hit by the next attack as well. At this point I have no idea what’s happening. From my perspective the dodging fails at random, and once it fails for reasons I don’t understand I don’t know which direction to press to get back on pattern.

How Does This Actually Work?

The tutorial text says “there isn’t a timing window” and simply hold “before the enemy hits you.” But it also says “before the enemy attacks.” These are two very different things!

In every game I can think of the timing of the enemy attack - as in when the attack animation starts - is irrelevant. What matters is when the hitbox intersects your body. If you want to block or dodge you don’t enter the command as soon as you see the attack animation start up - a common trick in action games is giving enemies very long wind-up animations that fool the player into committing too early.

But I think in this system you have to input the command well before the attack hits you. At a certain point the game decides to play either a “you dodged successfully” animation or a “you dodged the wrong direction” animation. If you aren’t holding the correct direction well before the attack lands the game can also play a “you dodged in the correct direction but too late” animation.

Which would be fine except according to the tutorial that’s not how the game works. That’s also counter to how most games reward you for dodging / blocking as late as possible - with Perfect Blocks, Perfect Dodges, Dark Step3, White Blocking4, Just Defense5, Parry6, Guard Impact7, Universal Low Parry8, Witch Time9 - I can do this all day.

Point being I (and many other players) have years of muscle memory from games actively rewarding late dodging, and i-frame mechanics organically encourage late dodging as well. And here’s a game that specifically states that timing doesn’t matter, then punishes you for using what would be optimal timing in 98% of games.

I think. To be honest I’m guessing. I defy anyone to watch the videos above and explain how it works in a way that meshes with the tutorial text.

If it does work the way I now suspect - that you have to input very early - well that’s weird. And what does “before the enemy attacks” mean exactly? Does the wind-up part of the animation count as the attack or is that technically pre-attack?

Maybe the word “hold” here is key and you have to hold (well in advance?) then keep holding?

Let’s not even get into how this works with multiple enemies.

I said before that the tutorial text is clear - I have to amend that. If the mechanic just worked that would be true enough. Maybe “hold” vs “press” is a bit nebulous, sure. “Enemy attack” and “enemy hit” shouldn’t be used synonymously, if you want to be picky. But this is tutorial text not a detailed legal document - I’m only carefully studying the wording because the mechanic doesn’t “just work.”

And again: I don’t understand why you’d use this system in an action game anyway. (And not say in Asura's Wrath) Did they create elaborate dodge animations then realize they looked bad when played late, so the system is a workaround for that? Was the system designed by people who prioritized animation and this was their way to force synchronization? Or did a combat designer genuinely think this was a good system from a mechanics perspective?

I don’t get it. I don’t love how the mechanic reads on paper - even if it worked perfectly it sounds more Dragon’s Lair than anything else, which is not what I’m looking for in an action game. In theory it’s way too easy. I don’t like how opaque the technical details are and how it feels less like a system than a one-off. I don’t like that you can’t ever choose to dodge - instead you tell your character to walk left and he decides to dodge when he thinks the time is right.

But also it seems fundamentally broken. Or maybe the tutorial text (and the tutorial itself) is just wrong.

The section above might strike some as nitpicky, but as a reminder I’m approaching this as a game developer. You have to consider these things if you’re a gameplay programmer or a combat designer, even if they’re invisible to the average player.

In Conclusion: Git Gud?

I’ve seen a lot of complaints about dodging in Callisto and a lot of defenses - but most of those defenses are “it works fine lol game journalists don’t get it. Git gud.”

A developer (not one who worked on Callisto, to be clear) I won’t name shared a video of impressive-looking (in their mind, at least) combat along with this:

This is how The Callisto Protocol's combat is actually supposed to work, btw. For those of you who are passing judgement from footage. Not in any way implying you have to LIKE this, but this is much more accurate than some of the launch day gameplay I've seen.

What does “accurate” mean here?

To claim that your experience is more accurate than someone else’s is like looking at someone else’s photo and saying “what a bad selfie this looks nothing like me.”

Maybe by “accurate” they mean “representative” but that doesn’t seem true - based on the complaints and reviews (including Steam reviews, not just media reviews) the “bad” gameplay seems more representative of the average experience.

As a developer “everyone is playing it wrong” is rarely a wise comment. It’s part of your job to understand what the average experience will be, and to make sure players grasp enough of the game to have fun. Players don’t have to understand everything - the vast majority of players won’t. But they need to understand enough.

Plenty of people will play Bayonetta and never use dodge-offset, the ability to continue combo strings after dodging. When Street Fighter 2 first released players would pick Chun-Li and mash buttons to get lightning legs. You can have fun in Madden without knowing what a zone blitz is. These games are designed to be enjoyable at a variety of skill and understanding levels.

When you say “everyone is playing it wrong” the intent is “everyone but me is an idiot”, but the implication is “the designers failed at their job.” “Failed” is maybe a little harsh but video games are consumer products designed to used, not conversation pieces.

If you’re a professional door designer and everyone is using your door wrong you should ask why, instead of saying “it says push anyone pulling is an idiot.”

To their credit I haven’t seen this argument from the developers of Callisto Protocol, but it’s a common defense of the game. I’m far more receptive to “git gud” arguments than most critics but it’s just not applicable here. In particular I don’t know what I’m supposed to get good at - I seem to be operating the game correctly but it just doesn’t work.

Maybe after publishing this someone will point out that I didn’t understand the tutorial and was simply doing it wrong. I don’t think that will happen but it could. (Boy that would be embarrassing - move over Cuphead guy!) But even if that’s the case you have to ask why so many people get hung up on the dodging - call it “clunky”, declare it “broken”, claim it fails at random.

You don’t have to, I suppose - but as a developer you should.