March 23, 2017

Pixel Pretty

There are many ways to think about what makes a game beautiful. A good number of theories and ideas regarding beauty have been developed over the years to address this. Some are very old, going back hundreds of years, having been refined by application and re-application to every relevant artform to have ever existed. Some of them are recent, brought into existence by the emergence of videogames as a new artform. I’d like to posit that applying the idea of possibility space to a game’s audiovisual design can also be a valid and helpful way of understanding them as aesthetic objects.

Videogames are often touted as being “unique” in their interactivity. A proponent of this viewpoint might, for example bring up in comparison a traditional object of visual beauty, a “static” image such as the following:


Galatea of the Spheres, 1952. Beautiful, isn’t it? It’s definitely one of my favorites. And the reason it looks so great is because the artist, Salvador Dali, chose for it to look that way, had total control over what the final image would look like, and exercised that control to ensure the image you saw matched the vision in his mind.

On the other hand, one might compare that to a more “dynamic” visual experience, such as these scenes from Super Mario Bros.:

It’s nice to look at, but is it in quite the same way? None of the flying fish are permanent fixtures. Furthermore, the visual effect of the level changes depending on how you play it; If you stop running, the arcs of the fish become much more prominent, which is visually interesting in its own right. Vary your speed throughout the level, and the fish seem to follow a wider array of trajectories. And by the way, the number of fireworks that appear at the end of the level is dependent on the digit at the end of the timer when the level is completed. Do you want your final note to be short point of punctuation to the action or something long and held out, drawing attention to itself?

Of course, the so-called “uniqueness” of videogames is often overstated. As Games and Arts critic Zolani Stewart points out in an episode of the audio essay podcast Critical Switch, all art forms are “interactive” to some degree, as engaging with art is always an active process:

“The experience of engaging with an art object is always reciprocal. It is always reciprocal! Whether it is a videogame, or a poem, engaging with art is usually a back and forth thing, where I’m not just sitting there doing nothing while a movie transfers senses into my brain, or where I can just completely full-heartedly impose all of my will onto a videogame without it ever acknowledging the nature of my actions, although there are games and movies that do both these things.”

Furthermore, the issue of a visual space changing based on the angle from which you view it is not unique to videogames. Theater, for instance, gives viewers a wide range of angles from which to see the set, which can change the impression that set makes upon the viewer. (Does a set seen from high and far away, or at an angle, seem as imposing as one seen from below and close up?) On top of that, performance art allows considerable opportunity for play and interaction (and without need for computers!), as do the various strands of interactive and internet art that have risen to prominence over the last few decades.

At some point during this back and forth over where videogames “fit” in with other media, we have to stop and ask ourselves what a “medium” even is. The assumption that interactivity is the hat that makes videogames special implies the assumption that a medium can even have a “hat” at all. In reality, artforms are not defined by a single gimmick. We wouldn’t ignore the obvious role that sound plays in cinema, and we wouldn’t call them “photographed theater” just because they involve photographed images and scripts. We can recognize these various forms of information as being distinct, but we can recognize the format in which they are intermingled as also being something distinct. Theatre itself is never mistaken for performed literature, even though their stage plays are text, often bound in book form, and even though Shakespeare’s playwriting makes for a great library checkout.

It is my belief that the identity of a medium has less to with the kind of information being conveyed (visual? Aural? Tactile?) and more to do with the way that information is “packaged”, for lack of a better word. Static images with text captions? Static images with text inside the images themselves, in a bubble above the speaker? Text spoken aloud w/o visual information? Just text, maybe with the occasional illustration? Even mediums made entirely up of elements shared by other media (text, moving images, music, the input-output systems we commonly refer to as “gameplay”, see where I’m going with this) can add to form something larger than the sum of its parts, a form with a distinct mode of engagement, rather than a mere copy, a “knock-off” version of anything.

Which is to say that, though they may not be totally unique in some overblown exceptionalist way, videogames can’t be considered entirely derivative, either. They are at least distinct enough that we can recognize them as their own form, after all, and as such there is a point at which talking about in terms exclusively taken from the study of other media becomes reductive in its own right. Videogames are not more special than other artforms, yes, but they are also not less special, and like other artforms, they deserve to be approached on their own terms from time to time.

This becomes extra important when we have to deal with those problems that are the fallout of videogames’ particular amalgamation of content types. And here’s a videogame problem: There is a much smaller gap in experience between two theatergoers on opposite ends of the room than there is between a player who dashed through World 2-3 without stopping and a player simply stood still and let the clock count down to zero. Or worse, between the player who dashed through World 2-3 and the player who skipped over World 2-3 altogether. You might never see the fireworks at all.

Everyone sees the game from a different angle. No two players will play the same way as each other, so one player might see or miss things that a different player did not, or see them under different circumstances, or see them in a different form. The less linear a game is, the more extreme this becomes.

The problem is that a lot of the stuff we talk about being beautiful when we play a videogame is stuff that other players never end up seeing. The “experience” is often only ours, not just in its subjective reception, but in the very concrete details of “What happened when you played this game?” But if that is the case, how do we begin to talk to each other about the beauty of a game in its totality?

Worst of all, it begs the question: If an event is so rare and seemingly coincidental, is it really reasonable to count as a valid part of the intended play experience?

If we’re going to thoroughly address this question, (and to those who don’t have time to finish this, the non-thorough answer is: Yes, yes it is) a good starting point might be another quote from that Critical Switch episode, titled “Expanding Interactivity”:

“Videogames are often very static objects. They are collections of script files, and model objects, and image files, and sound files. They’re not actually responding to you because they are pre-built to make consistent responses. Every CD of a videogame you buy, at a store, is the same CD. The dynamicism that we see in videogames is an interpretation. It’s a perspective that we project onto them because we learn to process phenomena in a certain way, the end result of what’s happening in front of us.”

I like this way of thinking about games, as mediated directories, not in the least because it places significant emphasis on context instead of deifying “mechanics”. It also ties into an existing idea called the possibility space. The possibility space is usually used not to describe music or visual elements but gameplay, the idea being that the game is made up of a series of descriptive statements, alternately called rules, laws, or mechanics depending on who you ask. These laws form the boundaries of what is possible within the gameworld, and it is within the space where these laws intersect that you find the game’s “possibility space”, containing all of the actions and events possible within the game. (I can’t remember where I first heard the term, although Will Wright has been using it at least as far back as 2006.)

One might argue that a videogame is its possibility space, the loose directory of images, sounds, and other files and the systems which organize them into a meaningful structure, one greater than the sum of its parts. Not the program, but the whole folder. In that case, videogames allow arguably less real freedom than performance art. But it’s not about how many possibilities there are, it’s about how they’re arranged, how they’re connected and structured. These laws all intersect eventually, and the space taken up by the implications of those intersections are incredibly important.

This brings me to the main point of this post: that the “possibility space” concept, while originally intended as a way to describe gameplay, can also be applied to the way a game looks and sounds.

But before we continue, a brief disclaimer: Although I am focusing primarily on one aspect of videogames (visual information and feedback), it should be clear that the various elements of a videogames are not in reality isolated features that in exist in a vacuum, or on an IGN review rubric.


The idea that the visual identity of a videogame exists in a possibility space immediately gives two scales on which to think about that identity. On the micro scale, which comprises the individual elements that make up the image, characters, objects, the background and foreground, effects, enemies, even the HUD can factor in. On the macro scale, we can also take into account how all of these disparate elements come together to form an image that composes and recomposes itself.

To take an example from earlier in this post: In World 2-3, Mario, the Cheep-cheeps, and the tiles that make up the bridges and their support columns are all elements of a larger picture, the exact makeup of which can vary considerably from one playthrough to the next. The image is not beautiful because of how well any of the individual elements were drawn, or because of their precise placement in the level, which the designers could not entirely control. It was because of the peculiar way the elements of the game screen were programmed to interact with each other, creating the potential for any number of beautiful images. The early Super Mario Bros. games were especially good at this, rearranging their limited assortment of square tiles into all kinds of new and interesting pictures. The blocks that make up a staircase are later repurposed as support pillars for the bridges, and the sprite used for Mario’s fireballs are recombined into the hypnotic spin of the deadly firebars. At its best, the visual style of the early Mario games can be compared to a collage of carefully matched squares.

In the case of Super Mario Bros. 3, its pastel colors, cute faces, and simple linework suggest a collage cut by Sanrio. (Source)

For early 8-bit games, which had limited memory and a Picture Processing Unit that only worked with tiles, this approach to composing the screen was a practical necessity, although the benefits of such an approach are being rediscovered by the curious tinkerers of the Wii U’s Super Mario Maker. I could go on to talk about how the tile-based architecture of the early Mario games were at the core of their entire design philosophy, even today, but I risk digressing from the point of this post. For now, I’ll bring up another example of a game from that time period that uses these restraints to its advantage. In Clu Clu Land (Unknown, 1984),tiles representing golden gemstones are arranged into a hidden picture that you uncover, piece by piece, by swimming over them. Over the course of the game, the same tiles, along with other elements of the playing field such as trampolines and vortexes, are re-arranged to form such images as person, a house, a bear, or even a submarine. Part of the fun of the game is how the picture unfolds around you as you play, as though you are creating it yourself. If the appreciation of beauty is “the re-creation of the artist’s moment of inspiration” (H.E. Huntley, The Divine Proportion, 1970, p. 22), Clu Clu Land literalizes this idea by having you literally re-create the image in question.

The reason I like pixel art so much is that it encapsulates this holistic micro-macro idea at the atomic level. Every pixel, every atom of color has to be chosen not just based on how pleasing it is by itself, but also where it fits in a complete image with an overarching color palette. A single dot in the right context can represent motion, texture, lighting. The best pixel art tends to have an impressionistic feel to it, and these ties to classical Impressionism are especially important in an artform where effective impressions of a world have to be made to make up for a lack of pure detail. I mean, we’re good, but we’re not Pixar. We don’t have the option of being able to spend hours rendering a single frame. Hence, Pixel Pretty.

Even then, it’s not as simple as getting the individual blocks to fit together into a single image. That’s because there is no “single image.” When we say “in macro”, do we mean the image framed by the screen, or the gameworld as a whole, or even the game as a whole? Even if you will never be able to see all of the gameworld on screen at once? Even if a beautifully planned city will only ever look from the player’s point of view like an indecipherable series of straight paths?

I believe the most complete way of understanding the way that a game is or is not beautiful is to look at the visual potential of the game in its totality. This isn’t that different from the already well-established concept of “possibility space”, which describes the full range of events possible within the boundaries of a game’s design. The game is a system of systems. It is a system capable of assembling beauty, but it can’t be treated like a singular object with a definite visual form. It can, however, be treated like a system governed by preset principles, a game with an art style, an aesthetic which, far from being a vague sense of the game’s “look”, can in fact be a strong harmonizing force coursing through the very heart of the experience.

This is what I mean when I say that a game’s elements “fit” together, like puzzle pieces that can be rearranged a million different ways. The puzzle pieces are all cut from the same block of wood. The colors and forms can complement each other or play off each other, but they still fit into a unified aesthetic.

This concept of unified aesthetics is interesting because when we talk about a game looking good, we often talk about the game’s aesthetic. We talk about “Wind Waker‘s art style” or “Okami‘s art style” or “Shadow of the Colossus‘ art style” as much as we talk about any of the singular visual moments we experience in the games themselves. And there are also games like Geometry Wars 2 where the art style is one of the game’s main draws. This is probably as close as we can get to looking at a videogame as a composed image, by seeing it not as a constructed visual space framed by a screen, but as a constructed space of visual possibility, whose only “frame” is the limitations of its own code.

Looked at this way, even games like Space Invaders and Pac-Man can be considered to have an aesthetic. Pac-Man‘s cute, brightly-colored neon aesthetic was a huge part of its success. And even though I feel weird that Space Invaders has become kind of a cultural shorthand for videogames, I’ll admit that it does look nice in its own way. I don’t hear this talked about much, but I’ll bet part of the reason it was so popular was because its “Invaders” were extremely cuteInvader

The end result of this process is a game with a strange inner glow that shines through in every possible interaction. A game whose aesthetic has been thoroughly articulated through its every piece, such that it can look beautiful even when its pieces are scrambled because the game is totally glitched out, as in the work of Tumblr glitch artist postmodernCorruption. Games that you can pause anywhere and still be wowed by. It is the kind of game I like of call “screenshot bait”; It manages to look good no matter when you pause it, and it’s overflowing with screenshot-worthy moments.

It’s the kind of game, to put it succinctly, where every frame is worth hanging up on a wall.

Now, that previous statement about the best tiles being beautiful even when arranged haphazardly isn’t meant to imply that the arrangement of these tiles is unimportant. On the contrary, it’s what allows that inner glow to come through at its strongest. It’s what allows a game like Geometry Wars to look the way it does, despite the fact that its action is totally random and completely chaotic. There is control, but it isn’t the direct control of a painter. It is, rather, a sense of indirect but no less meaningful control: Instead of arranging the pieces, one designs a system that allows the pieces to arrange themselves.


This in turn implies a kind of intermediary between the will of the designer and the game as it finally appears to the player, almost a sort of robot director. This is not new. Consider Left 4 Dead, in which a program known as the “AI Director” changes the game on the fly as you play, placing items in convenient places and adding more enemies if it thinks you’re having too easy a time.

But there’s no reason to assume that the rest of the play experience isn’t mediated in a similar matter. The term “AI Director” implies that no other part of the game has been “tampered with” by the game’s systems, that the rest of the game is a sort of pure, raw, unadulterated experience. Of course, we should know better than to believe that. Given that the game is essentially a series of static files, the game you see doesn’t just spring up onto your screen fully-formed. The game comes disassembled, and clearly something is doing the assembling.

I wouldn’t go so far as to say that this something is literally alive (I just happen to love metaphors), but I want to stress that a videogame is a self-organizing system by nature, and that there is no part of it that is not “directed” during play, even if all that needs to be done is to set the hallway up for you to walk down. The unique set of methods by which a game’s resources are organized lends that game a unique style and personality. Perhaps this organizing system should be named. Conductor? Coordinator? Composer? Medium? Intermediate? Assembler?

This is most obvious when it comes to music, specifically adaptive soundtracks. The conductor in Hyrule Field and the AI composer known as Emily Howell are cut from the same cloth, on opposite ends of the same thread with regards to complexity, bound to each other by their modes of operation.

However, they are also bound by their limitations. Even the most adaptive of adaptive soundtracks cannot make up music using sounds that lie outside its files and code. In most cases, this means that the songs they play will tend to be made up of snippets pre-written by a composer, preexisting raw material to be remixed by a virtual DJ. Even Emily Howell’s amazing taste is just a series of logical rules hard-coded into her by her creator, which, as you can image, places a hard limit on her “creative” capacity (although within the bounds of that capacity are some truly beautiful songs), reducing her to an extension of the composer David Cope, not much different from the algorithm written by Derek Yu to devise Spelunky’s caves.

The point I’m trying to make is that, insomuch as the developers’ rules make up its style and personality, the limits of what it is able to accomplish, that personality could be considered an extension of the person or people who put it together, an echo, the developer’s ghost in the machine, here to do the performer’s job in the absence of a performer. Such a personality can be said to sleep in any game (and you could have yourself some fun trying to anthropomorphize it!) Every game has its Howell.


Early last March, I decided to take a photo of a tree that was blossoming outside my home. When I realized that the moon and the tree were just so positioned that I could take a photo of the moon as seen through the tree’s branches, I stayed up until the Moon was in just the right part of the sky before looking up and snapping the picture. I liked how it came out.

The reason I bring up photography as an example of this process is because it’s the closest analogue I can think of to beauty in games, in that photography is exposing the beauty in a self-composing system. The world as it is has always been capable of moments of beautiful coincidence, despite the day-to-day mundanity that hides it from any casual observer. Early 20th Century photography made a name for itself by beautifying what is so often filtered out and mystifying the seemingly innocuous, from the still photography of Henri Carter Bresson, to the early Soviet silent documentary Man with a Movie Camera (Dziga Vertov, 1929), truly a photographer’s film, which strung and spliced its individual moments together into a series of points, counterpoints and motifs strengthened through repetition and timing, bringing potential patterns out into the open to wrench meaning from chaos, cutting a gem out of the raw ore of everyday life. (Although if you intend to click the link, it’s only fair to warn you that the film does include very quick footage of a live birth. That said, I’m already it making sound worse than it really is.)

Could we say that the blossoming tree and the area surrounding it were always this beautiful? Probably not. We could, however, say that the tree always had the potential to be beautiful, to be a part of that beautiful image. The tree could be said to exist in a perpetual state of potential beauty, and the same can be said of the most visually arresting videogames. A beautiful videogame is a synchronicity engine.

Just as important as the potential for beauty in a videogame, however, is the frequency with which it delivers on that potential. You could even go so far as to place a game on a plane, with one axis relating to the magnitude of the game’s most beautiful moments, and the frequency with which it becomes beautiful, beauty per second. Beauty in videogames is a distinctly two-dimensional concept.

And when I talk about synchronicity, I mean it in the Jungian sense, in the sense of events that are connected not by causality but by meaning. Beautiful coincidences or moments of poetic justice. Coincidences like these are rare enough in reality, and when they do take place they are just that, coincidences. But of course, this is not reality. This is a videogame. You can design the game to spit out as many coincidences as you want, and they can mean whatever you want to make them mean and be as beautiful as you want them to be, so that all of the pieces fall perfectly into place over and over and over again, no matter where you turn. Sometimes the effect will be little more than a superficially pretty sight, as cute as it is fleeting.

But sometimes it will be an experience that rocks a player to her very core, engrossed and enraptured as the world around her melts away.


For the first hour or so of Metroid Prime, I was in a completely reactive position, stranded on a lonely, abandoned planet, being jerked this way and that by distress signals, runaway space pirates, and other low-level concerns. When I stumbled into an area called the Artifact Temple, my Prime Directive was simply to move forward by any means necessary.

And then I saw the statues.



Huge metal structures, belonging to an old dead race of birdlike aliens known as the Chozo, and often fashioned in their image, seemed to pop up everywhere I went, in every Metroid game I had ever played up to then. This was unremarkable in itself; For them to appear on planet Zebes, which had been acknowledged in the series lore for years as their old home (and the childhood home of series heroine Samus Aran), seems reasonable enough. But to find them here, on a planet in the middle of nowhere, seemingly overlooked by the rest of galactic society, was an epiphany.

It was one of those oft-cliched moments when you realize just how great the scale of the universe is, how deep the reach of life, how old its civilizations, and how small you are, the sheer size of everything. It was the realization that, for all the rightful exultation we give to the power of love and the struggle of good versus evil and the arc of justice and all those other millennia-long conflicts by which we define the history of mankind, the history of mankind was just the tip of the iceberg, and I had just been plunged face-first underwater. Forget “Human Heritage.” Embrace the heritage of sentient life.

But at the same time, I realized that I wasn’t alone, and that I wouldn’t be, no matter how far I went. This universe, far from being empty and cold, was in fact full of vitality. It was the stage of a vast cosmic drama. A story I had inherited for sharing the same condition that they did. Samus Aran enters, stage left.

The rain poured down around me. I felt connected with a force and a history greater than myself and the banality of my immediate survival. I felt a profound sense of togetherness with a group of people who had long since disappeared, as if they were reaching through time to console me, as if I were surrounded by loving ghosts.

I stared into the sky as the music swelled to a crescendo. I knew they were out there.

It was a contemplative moment. It was beautiful. It was all my synapses firing at once. It was heaven.

It will never happen again.

I’d say you have to see it for yourself, but I doubt it would matter anyway. It would never have the same effect, or at least an effect as strong.

But that doesn’t mean that it happened for no reason. The set up was certainly there. The series revels in making you feel lonely; The worlds you visit are carefully designed to be discomforting and alien, and records of previous life tend to devolve into a series of bleak autopsies, with signs of deeper conflicts drifting in and out of the background, cosmic catastrophes the details of which you can just barely understand, leaving only the barest evidence of their existence. The games reward investigation with such lovely sights as beautifully rendered ruins and ancient artwork. Much of the game’s canon is shrouded in mystery (although that doesn’t stop fans from speculating like wild), and the culmination of years spent layering backstory on backstory is a universe that feels profoundly, terrifyingly old. And so Metroid Prime exists in an aesthetic space that allows moments like these to happen.

If there are aspects of this image that could “obviously” be compared to the monoliths in 2001: A Space Odyssey, the only reason they stood out to me is because I was looking for them, and the only reason I was looking for them was because the specific details of my particular playthrough had primed me for them.


So with all this said, where does this leave us? With a long list of unqualified statements, statements that leave a lot of unanswered questions and plenty of room for expansion and criticism. To be honest, it seems like little more than a sloppy starting point.

I can accept that. I should stop now. If I didn’t, this would go on for even longer. If any of what I’ve written helps anyone in any way, whether it be a jumping-off point for further inquiry, a topic of discussion, or even just a mistake to be learned from and avoided by others in the future, I will be satisfied.

In the meantime, I look forward to finding more echoes, more ghosts. The cartridges are all haunted.

Note: If you liked this post, consider supporting my writing through Patreon by becoming a patron.

EDIT 11/7/2016: Fixed a major typo (incomplete sentence.)

Radical Helmet
About Radical Helmet 8 Articles
A purely digital, low-poly entity. I love my helmet. You can reach me at
%d bloggers like this: