Game Design


A work of art is both a single object and a collection of individual choices – sentences in a novel, assets in a game, instrumental parts in a piece of music, each of these is added and shaped with intent to achieve the overall goal of the piece. This is pretty self-evident, but often is not explicitly thought about by the creator during the creation. In some ways, it’s better not to think about it – for reasons I’ll get into in a moment.

You have the theme or message or whatever of the work: Sometimes you know exactly what this is, sometimes you have to hone in on it carefully in the process of working on the piece. With each new stroke you add to the composition, you can choose to support this theme, to add to its message by echoing it; you can contrast with this theme, to push against it and by so doing ground and emphasize it; or to layer new elements onto the theme, add details that seem completely disconnected but add complexity.

Say you have a picture of a gigantic statue: To support how gigantic this statue is, you could add comparatively tiny human figure to show how it dwarfs the scale of humanity; to contrast, you could add a field of stars behind it or pull the viewpoint back, to show how even the greatest creations of humanity are minuscule in the greater scale of the universe; or you could add something different, a mural or a small scene between characters or some strange creature, to show that the story of this statue and the world it lives in is more complex than we might at first imagine.

Naive artists will, given the choice, always pick the first of these. I have been this kind of naive, and still often discover this kind of naivete in myself. It makes sense: I’m an artist, I know the impression I want to create, I should use everything I have in my toolbox to create the feeling I’m going for. And yet, most of the time, this kind of approach leads to something which feels flat, manipulative, and obvious. All bombast, all sorrow, all silliness, with no leavening by contrasting or diverging emotion, will inevitably feel flat and numbing.

This is why I said it’s probably better that most artists don’t think explicitly about their high-level intent and how to achieve it most of the time: The mindset of trying to achieve a specific emotional impact is difficult to separate from the mindset of how to most effectively bolster that tone in each particular instance. Much better to take freely from the chaos of the mind, to harness opportunities to create threads that flow alongside, flow against, or flow perpendicular to the main thread of the narrative as they occur to us.

However, for those of us who have a hard time not thinking about intent, have a hard time getting out of our heads and have a hard time not hammering the same points home with each individual component of a work, it might be worth it to keep these three thoughts in mind: Support, Contrast, Layer.

A tapestry is not woven out of only threads in parallel.


I’ve been watching old videos of the original The Binding of Isaac, and it’s strange looking back. As many huge improvements as Rebirth, the remake that came out a couple of years after, made to the base game, still it feels like something was lost in translation. Several things, actually…

Maybe it would be best to start with talking about all the reasons people generally regard Rebirth as categorically superior to the original. The first game had notorious framerate issues, many items didn’t work properly with each other, and it was built using technology that made it impossible to expand – many people say it reached the limits of Flash, Adobe’s multimedia tool, but Isaac was actually not just built in Flash, but built using Actionscript 2, the version of Flash’s scripting language that was deprecated in 2007. Since I’m building my own game in Flash (technically AIR, the standalone equivalent), this is a narrative that I feel compelled to correct whenever it comes up. Rebirth could have easily been built in Flash. But I digress: The point being, Rebirth fixes all these issues, so when viewed entirely within the scope of the shortcomings of the original it definitely seems like a superior game.

Looking back though, something seems off with what we have now – and it’s interesting to examine why that is. There are aspects of the design, art, and music that just fail to click in quite the same way.

The least contentious of these is the music. Nearly everyone preferred the music from the original game, composed by Danny Baranowsky, to that in Rebirth, composed by Ridiculon (Matthias Bossi and Jon Evans). The new soundtrack actually does some cool stuff, with music layers that fade in and out based on what’s currently happening in the gameplay – but this actually undermines part of what made Danny B’s score so amazing. With parts fading in and out, it becomes necessary to create a consistent base track for these to play on top of, which makes it impossible to construct an overall narrative flow to the music. Consequentially, Ridiculon’s music is background music in the truest sense, just providing accompaniment to the experience of the game, whereas Danny B’s score actually defines the tone of the game and creates its own narrative high and low points which interplay with the gameplay highs and lows to create a more complex experience. Combined with a generally more melancholy and creepy tone, it makes the overall musical experience of playing Rebirth rather lacking comparatively.

Aesthetically, I have a bone to pick with the game similar to that regarding the defamation of Flash. When they announced that Rebirth was going to have a “16-bit” art style, I thought that was a peculiar choice, but was willing to see what they came up with. What they came up with was, unfortunately, kind of a pathetic excuse – which seems harsh, but I promise I have a reason for saying that.

First, let’s talk about the art in the original. Isaac used vector art, a specialty of Flash: Vector art is a style of rendering that stores images and a set of drawing instructions, a list of lines and colors. This is a powerful tool because these instructions can be easily rotated, scaled, color-shifted, and so forth with no loss of quality, but it pays for this in making detailed art very processor intensive. Rebirth, conversely, uses raster images for its assets: Raster images are what we’re generally used to working with in photoshop and other editors, just a grid of colors which can look realistic at its native resolution but looks notably blocky at lower resolutions. 16-bit games used raster images at a set low resolution to create a crunchy but vibrant look that is still beloved today. However, the entire design of Isaac was based around arbitrarily scaling and coloring assets which, as mentioned, works a lot better with vector images than raster images. However, for whatever reason the Rebirth team didn’t want to work with vector images, so to conceal the shortcomings of scaled, rotated, or otherwise processed raster images they used super low-resolution raster images and called the resulting look “16-bit”.

This is kind of insulting. There’s no coherence to the resolution – even when the pixels align along the grid the objects that own the pixels move with subpixel accuracy, creating a smoothness that’s impossible in a true 16-bit environment, and as game objects scale up or down in accordance with the mechanics they turn into grotesque pixellated bullshit. Also, because they use such low-res assets, there’s no room for detail in any of the enemies: The original enemy designs, though crude, have an expressiveness to their lines that makes them creepier and more compelling. While pixel art has a great deal of expressiveness in its own right, within the context of Rebirth that expressiveness is curtailed by being constantly squashed and stretched, one of the ugliest things you can do to pixel art.


The design issues with the game snuck up on me. In general, the gameplay choices made in Rebirth are very smart, limiting boring and overpowered tactics in favor of more interesting and aggressive ones, expanding the possibility space for encounters by adding lots of new items and enemies and rooms, and generally spicing thing up by adding new interactions. However, something weird started to happen as more and more items were added. I first noticed it with the item “Gimpy”, which is… exactly what it sounds like.


…And this comes to a fairly fine point about what Isaac is and is not. The Binding of Isaac has a lot of kind of gross and shocking content, but all of it is contextualized by the understanding that this is a child’s conception of the world, and all the weird gross things in it are exactly the sorts of weird gross things that kids tend to develop obsessions with – bodily functions, deformities, and so forth. Up until Rebirth, Isaac items tended to fit one of three themes: Everyday objects granted extraordinary significance, religious symbols, or video game references. These make total sense from the perspective of a weird shut-in kid who only knows his toys, the random things he finds, and the creepy religious stories his family tells him. But once you add S&M gear to the mix, it no longer becomes about expressing Isaac’s character, about life in the mind of an isolated and possibly abused child, but just about being weird and gross for the sake of weird and gross. By itself Gimpy is just one item, but it indicates an overall trend away from being expressive and meaningful and towards adding stuff to the game just for the sake of having it there.

In the end, Rebirth’s flaws are covered up by the simple expedient of repetition. Once you play the game for a thousand hours, you don’t care that the music lacks narrative flare, you don’t even hear it any more. Once you play the game for a thousand hours, you don’t notice that everything is in a different resolution – the game just looks the way the game looks, why would it look any different? Once you play the game for a thousand hours, you don’t see a gimp mask, you see a way to restore health in difficult situations. You see the frame rate stable at 60 frames per second, you see the hundreds of weird and interesting item interactions. It may have made total sense to prioritize the things they did in developing the game: In so doing they’ve made a game that people who love Isaac can play for thousands of hours and still enjoy.

However, they’ve also made it so the chill I felt when I first played the game, the genuine sense of visceral discomfort and confusion and striving understanding, are now obscured behind a layer of generic video game.


Pong and other early arcade games feel so primitive to us now in 2017. It seems intuitive to credit technological advances for the massive difference in complexity between the light arcade games of 35 years ago and the many more complex forms of video game we play today, but how much does technology really have to do with that shift? Complexity of input and gameplay systems increased over time, but not nearly in proportion with technological advances. No, the biggest difference between games then and games today is that today’s games are made to be played by ‘gamers’, people who have played other games before and been shaped by the experience.

When Pong succeeded, it did so because it was a game that could be played by anyone. Though many think of Pong as being the first arcade game, many players today would likely prefer to play the actual first arcade game, Computer Space, a game of careful movement and twitch reflex which was released more than five years before Pong – however, it was not especially commercially successful – not the way Pong was – likely because it was too complicated and unintuitive for the mass audience, who were by-and-large not technically inclined and were also slightly inebriated.

With no prior exposure to video games, members of the general population had not built up the dexterity and coordination required to successfully play a multi-button interactive game. “People learned how to play video games as a group over time,” says Bushnell. “I think they could have handled it much better two or three years later.

45 years after Pong, as much as game technology has progressed, game literacy has progressed much further. Most people with any interest in the medium have built up huge specialized skillsets that allow us to appreciate nuanced designs that would be universally mystifying 45 years ago. We have progressed beyond the two-directional wheel, to the four-directional joystick with one button, to controllers with a directional pad and 4 buttons, to controllers with a directional pad, two joysticks, and 10-buttons-plus-you-can-click-the-joysticks-down-like-buttons-too. In the meanwhile, games were developed for PC which had the keyboard as standard equipment, a controller with 50 buttons and nothing else. For a while games tried to use as many keys as possible, but developers eventually realized this was ridiculous and tried to mostly restrict themselves to the left side of the keyboard – particularly as the mouse gained popularity as a standard input mechanism.

Video game complexity increased hand-in-hand with video game literacy until the early 2000s, at which point popular games were intricate or finicky enough that only people who had been playing games for much of their lives understood them. A kind of equilibrium was reached, for a while, where the really complex games mostly went to PC for a smaller but more dedicated audience and gaming consoles got most of the lighter more arcadey fare. However, game companies tried to unify these audiences, and managed in so doing to create games that were still intimidatingly complex to people who had never played games, but at the same time insultingly condescending and easy to people who had been playing games all their lives, thus satisfying exactly no one with the precision only the way the invisible hand of the market can achieve.

Now we’re at an interesting point: Because games can be readily developed at many different team sizes and price points, we have something of a reemergence of the console/PC paradigm but with a softer barrier: Very simple games like Super Hexagon mostly find their way to mobile platforms as primary targets, and very complex games like XCOM mostly find their way to PC and, to a lesser degree, consoles as primary targets, but both also find themselves frequently ported to each other’s dominant platform. We’re finally beginning to establish, however haphazardly, a curriculum of game design, by which new players can discover how to engage with a vast and expanding medium

It’s strange now to look back and realize that no game experience is really self-contained. Each game teaches us to play itself and, as game developers learn by playing games, the many clashing ideas of what a game is or can be propagates through them memetically. Every game you have played before contributes to the experience of the game you are playing now, for good or ill, by shaping the vocabulary you use to explore it. And like that we build a bridge into the future, trying to construct a path to the experience we imagine a game can one day make manifest.


One of the hardest lessons for me when it comes to art is that it’s not always possible to take a piece that is unsatisfactory and work on it for long enough to make it good. This is counter-intuitive, especially when it comes to digital work: After all, it’s just a collection of pixels, infinitely malleable, so surely if I just work long enough the pixels will come aligned in just the right way, come together to form the incredible work of art that I know could be there. So they may, given enough time and effort, but it’s just as likely, or more so, that all that work merely serves to blur the outline that first gave it energy and potential.

At first I was hesitant to work in physical media for this reason. It terrified me to think that each stroke, each movement of the brush and each fleck of pigment, was an irrevocable decision that shaped the piece, with no undo button available. Well, I’m still not entirely comfortable with that, but it’s become more apparent to me over time that pixels aren’t so different. In theory, they are completely controllable: In practice, that is not how artwork is made, and each stroke, each randomly generated fleck, becomes part of the history of a piece, shapes its final form in subtle ways.

Though I’m speaking here in terms of visual art, the principle holds true in all forms. The words we choose matter, even if they are edited out later; the code we write matters, even when it gets refactored; the melodies that go unused, the frames that are redundant, the level that gets cut for time, these all still make their mark on the finished work. Every step we take in creating changes the creation, even those we take pains to erase – perhaps especially those we take pains to erase.

However, no stroke matters more than the first. The first motion of brush or pen, the first sentence, the first object you orient in your program, these form the shape of all that is to come. This is why there’s such a focus on doing quick works, in gesture drawing and game jams; it’s all about practicing that initial motion, so that when the time comes for polish we find ourselves polishing gems rather than turds.

Thus artists have a complex and tense relationship with ideas. The idea is the spirit of that first stroke, and placed properly it can form the basis of a masterpiece. Or, perhaps, rather than being the first stroke it might be the accent, the perfect twist that adds the character that distinguishes a work. However, ideas are in all cases worthless without the effort and mastery to put them into practice.

Additionally, the difference between an idea and the idea is the difference between the lightning-bug and the lightning*. Bereft of context, there’s no such thing as a good idea: It is the context within which an idea is made manifest that makes it a fumble or a master stroke. An idea is simultaneously invaluable and valueless, priceless and worthless; we write them down and hoard them because we never know which idea we will need, but also know that no idea is valuable unless it can actually be realized.


The scariest thing about art to me now is not the tyranny of the blank page, but the certainty when I begin that I have no real idea where I’m going.

I have meticulously planned out every moment of my game, and right now the version of it in my head is good but I don’t think it’s great. This is scary to me, because this is a huge chunk of my life to spend on something if I’m not going to be satisfied with the result – I suppose that’s true for a lot of game development, but since I’m not sharing the development with a team all that weight falls squarely on my head. However, what I know about art and what I know about game development is that the magic isn’t in the plan, it’s in the moment of creation, in the poignant details and pivotal moments.

Nevertheless, as Eisenhower said, plans are worthless, but planning is indispensable. Every decision I make now is still going to be important, just not important in the ways I expect. There’s a butterfly’s lifetime worth of hurricanes between here and there, and my control over how this plays out is simultaneously absolute and negligible.

Right now, I have the blueprints for a bridge, but I don’t know where I’m going to be building it or what I’m going to be building it out of. With what I have now, I know that – depending on what comes next – it could either collapse or it could create a pathway to someplace no one has ever been before, and which of those happens depends partially on me, my expertise and artistic instinct, but just as much on chance and happenstance.

And I have to build it. This is how the job is done. This is how art is made.

We walk by falling forwards into each new step, over and over again*. To expect to be ready for the fall, to be certain of the recovery, is too much to ask, so each step we take forward is a tiny leap of faith. We keep doing it because to do otherwise is to stand in place, and any room can be purgatory if you stand there long enough, and with each moment you wait to take a step your legs will just get heavier.

So what I’m saying is that this is scary, and I think my game could be bad, but it’s also necessary, and I think my game could be great – and that I expect to always feel this way, forever, even after EverEnding, and that if I ever lose this feeling then something has gone wrong, and I will find that all of a sudden I am standing in place – and, even if the scenery looks like it’s moving, it’s just the flickering of a screen that someone forgot to turn off. Then it’s time to walk, or fall, again.

*Thanks Laurie Anderson


Thanksgiving was a couple of days ago, but I’m still feeling the holiday spirit, such as it is.

Thanksgiving is a good holiday wrapped in a bad holiday. It’s a time for coming together and taking stock of the many good things that we are lucky enough to have, wrapped in a horrific lie about a dinner party with the people we murdered and enslaved. America is weird. Maybe it’s that every country likes to lie about itself but our lies are still fresh enough to be clearly disproven and false, I don’t know. We couldn’t bury the lie, it’s still cracking out of the ground like a telltale heart, and that’s where we are now.


Things have been strange, and bad, and promising to get stranger and worse. It’s a good time for gratitude. It’s a good time to be happy to be alive and able to live a fairly fulfilling life, and I wish more people could do so, and I get furious thinking about all the ways in which they can’t, and it just keeps on coming up once you see it, the magic eye trick doesn’t go away, the 3d picture comes right out of the page and grabs you by the neck. Every nice thing you see, you can’t stop wondering how much blood went into it, and whose, and when. Everything has become so diffuse – we act not just as people but as part of an economy, and when that economy plunders we are culpable.

The problem with seeing more threads is its so easy to imagine your hands tied. All I want is to make interesting and beautiful, albeit perhaps sometimes disturbing, things. To paint a path to another world, to show the silhouettes of our flaws and aspirations.

I’ve never been one of those people who takes objects apart to see how they work – I’ve always been the sort who wants to make interesting things happen, and only care about the inner workings of objects insofar as that leads into the interesting things those objects do. However, when your artistic medium is one as technical as video games, that means looking at how things work, or fail to work, a lot. And it’s a hard habit to turn off. To design a game is to tie two systems together – one, a mechanical system that is predictable and quantifiable but quite complex, and the other is a system of incentives and desires, a system of which the player is the center piece, which is unquantifiable and much much more complex. If you play enough with a critical eye, you see where systems break down and stop working, where challenges stop being fair, where lazy strategies dominate.

I can’t look anywhere now without seeing broken systems. A world of paintings hung slightly askew. The problem, the big meta-problem, is that there’s no clear way to fix these systems. You could perfectly identify what was going wrong and an optimal solution, create a plan that would definitely and demonstrably work to fix the issue, and end up no closer to actually solving the problem. That’s politics, baby.

There’s a good reason it works that way. Imagine you have absolute power. Whoops, you just divided by 0, the world crashed. Fuckin owned, scrub. We make sure lots of other people have a chance to look at the systems that are created, get their own say, and that the whole thing has to go through a process designed to weed out more bad ideas than produce good ones, just because we don’t want to see another crash.

That’s what it’s designed to do, anyway. Design, like intent, is not magic, and the road to hell is paved with good intentions, which turn out to be exceptionally easy on the soles of bad people. We’ve got a memory leak – the memories of past crashes leak right out of our heads, and things fall apart, the center puts us on hold, and we all join the Cassandra Club.

Still, I’m grateful, really. I’m still around to look at the mess, and I can be one of the billions of hands on the steering wheel, even if it feels too late to swerve. I can still make art. I can still survive. It’s something. For a while it may be everything.


As games have become larger and more complicated, development teams have grown bigger – and, as development teams have grown bigger, we break down the work that needs to be performed on a game into smaller and more specific categories so that each person on this now-gigantic team knows precisely what they’re supposed to be doing. Pragmatically, this is a very effective way to make sure a game actually gets made. Unfortunately, it also creates a flawed mindset about what a game is and is made of. When we experience a game, we don’t experience it as its game design plus its art plus its music, we experience it as the totality of these things filtered through the particular lens of our input. That is to say, regarding the game’s design, its story, its art, as separate aspects is extremely limiting. All of these, even if they are manifested very differently, affect each other.

Yeah okay maybe this sounds like 101 baby bullshit, but it can be surprising how often this discrete-field attitude manifests. It used to be that even review scores were broken down in this nonsense way, scoring audio and graphics and ‘fun factor’ separately. It’s a prevalent attitude because it makes it easier to think about the game when you can think about little bits of it at a time. You can’t really separate the design from the art from the story from the audio, though. They all affect each other – often in surprising ways.

For instance, I have spent way too much time playing Team Fortress 2, and much of that playing Spy. As the spy, you need to be able to infiltrate the enemy team and get close to them without being noticed in order to do most of your work, whether that’s destroying equipment or assassinating vital team members. In order to infiltrate, you must rely on distractions and otherwise concealing your motion. Under these circumstances, a lot of the game’s art and audio design become incredibly important to you: How noisy a weapon is determines how easily you can sneak up on an enemy using that weapon, how long a gun is determines whether you can hide behind a corner while holding it, and if you happen to disguise yourself as the man wearing an elaborate hat that shoots sparks and cost him a hundred bucks he’ll probably notice you running by wearing it.

In other games, a particularly common example of aesthetics affecting gameplay arises as visual obfuscation. Tall grass and foliage often act as soft cover in action games, allowing players to conceal certain moves from each other, and the particular lighting of a room has separate concurrent effects on the gameplay, the narrative, and the aesthetic of the room. Trying to adapt the soundtrack to the current action has the interesting side-effect of informing the player of what the game believes the current action to be – that is, if there’s a combat music that plays when one or more living enemies are around, the player might know that an area that looks cleared out still harbors enemies, since the combat music hasn’t stopped yet.

The interplay of aesthetic and design becomes particularly relevant when regarding issues of accessibility. If a particular aesthetic aspect of the game becomes core to the design, and certain players of the game are unable to partake of that aspect due to disability, the design may just quietly break, leaving the game subtly unplayable. The most common example of this is probably the puzzle game with vital audio cues being played to an unhearing player – these cues, unaccompanied by any visual change, are not only possible to miss but are possible to miss without having any clue you missed them. This affects players not only on the basis of physical capability, but on the particular hardware they’re playing on as well – as a child, I got stuck on a particular dungeon in Final Fantasy 6 for months because, as it turned out, the television I was playing on was so dark that the switches I needed to press to progress were invisible to me.

The core design of the game is something which can only be expressed through the aesthetic choices that form the representative layer of the game. These cannot truly be separated, even if approaching them as separate disciplines makes it easier to develop the project.

Perhaps this is yet another reason that Dark Souls has resonated so much with people. In a world full of games where code and design drive the combat with animations and effects merely added afterwards to describe the action, the Souls approach of having each weapon’s animation directly influence its attack pattern, of having each swing be, not just a nice animation applied to a hit volume, but having the most important property of every attack be the motion which it attacks with, is refreshingly consistent. And it’s worth wondering, now, whether that kind of cohesion is possible when you have a designer and an artist and a musician each doing their own work, in separate rooms, in separate cities, hoping that in the end one and one and one will, after all, turn out to be four.