Chasing realism is a trap.

This might be true of art in general, but it’s video game graphics that keep reminding me of it. Chasing realism is an expensive luxury. Unless realism is a core part of whatever you’re trying to make, it’s probably not worth fixating on it. That’s because…

…making art is hard.

Every piece of art is composed of many constituent parts and decisions, each of which requires effort, each of which contributes to the whole. Some parts contribute more, some less, and the overall combination of those separate choices creates the final product—like a collage. Creative choices that don’t pan out—or don’t offer something to the greater whole—hurt. They’re wasted effort, or even actively counterproductive.

Games, and video games, are art. Making a video game is making art. Making a video game is a huge undertaking, a large project that is difficult to coordinate, with many constituent parts involved. It really hurts when some of those constituent parts don’t work well with the greater whole to produce a cohesive final work of art.

And yet when I look at video game graphics, I can’t help but think that there are plenty of games which sleep-walked through their visual design choices and fell for the realism trap.

Realism qua realism doesn’t do much for most games.

There are obvious potential exceptions. Maybe you want to make the most realistic simulator that you can. Maybe you are portraying events in our world and want to communicate that this game isn’t taking any creative license with its depictions of those. But if you aren’t hell-bent on depicting our own reality, there are compelling reasons to take liberties with the art direction of your game. And if you’re aiming for realism because that’s the default, it’s worth asking yourself why.

Ultimately, I think this comes down to questions of longevity, relevance, and aesthetics.

Video games, and their graphics, run on hardware. That hardware has constantly changed, and become more capable over the decades.

Chasing realistic graphics limits a game’s long-term visual relevance and almost ensures its obsolescence. As old hardware becomes obsolete, the most realistic graphics that hardware could support are also rendered obsolete by newer games’ graphics. For an example of this, take a look at games lauded for their realistic graphics over the past thirty years. By modern standards, there are a whole lot of stinkers in that collection—and plenty of those stinkers don’t have unique stylistic appeal to help them stay relevant.

Worse from an audience-size perspective, new hardware is expensive; a smaller group of people will have access to hardware that can run your game’s cutting-edge realistic graphics. By the time more people have access to that hardware, your graphics will be increasingly obsolete. Without something special to set your game’s visuals apart from other realistic depictions, you risk becoming another of those once-lauded stinkers I mentioned above.

It doesn’t help that cutting-edge realistic graphics are trying to replicate the world that everyone has access to. If you get something wrong, it’s relatively easy for people to realize it. By pursuing realism, you open your project up to the risk of being stuck in an uncanny valley of almost-good-enough realism.

As far as I can tell, the trick is to chase a distinctive and evocative art style instead.

Distinctive and evocative art and design isn’t necessarily easier or cheaper than cutting-edge realism. Nothing guarantees that a game design team will comfortably gel around a given vision. Nothing guarantees that your artists will produce evocative and consistent less-realistic art for less time or effort than it would take them to produce something realistic.

But the final result of pursuing realism is a style that can be more easily superseded by the next hot thing, doomed to look dated without independent appeal.

The final result of pursuing a distinct style besides realism, meanwhile, is your own separate piece of art—something more likely to withstand the obsolescence of whatever hardware it was built to operate on.

Pursuing a distinct style also creates opportunities. A game’s graphics can pay homage to other games, to other works or schools of art or, or can fit the feel of a specific genre. They can emphasize or erase features, and create distinction that would be difficult to achieve with a strictly realistic portrayal. All of this opens the door for creative exploration… and isn’t that one of the reasons to make art in the first place?

Consider the difference between Call of Duty 2 (released Oct 25 2005 on PC) and Okami (released April 20 2006 on PS2). These games are contemporaneous, though they were released for different systems. CoD2 may still be a fun or interesting game, but there’s little to recommend it visually over some newer iteration in its long-running series. Okami, on the other hand, is still recognizable, distinct, and often beautiful.

This isn’t a totally fair comparison. CoD2 focused on mostly-historical events. It was intended to look about as realistic as it could at the time, and people might have felt weird playing a WW2 shooter with a less-realistic art style. But I think it’s telling that shooters like Borderlands, Overwatch, or Apex Legends have leaned into more cartoony visuals to varying degrees. In doing so, they’ve distinguished themselves, made themselves easily recognizable.

So. I’m not opposed to realistic graphics per se. But I do think that anyone trying to perfectly replicate reality in their art—especially in their video game graphics—had better understand what they’re getting themselves into, and make that choice intentionally.