Reinventing the Art

This article appeared in the February 2008 issue of Develop magazine.

Last month, Electronic Arts took the unusual, although not unprecedented, step of permitting the release of the source code of Maxis's landmark simulation game SimCity to the public under the Gnu Public License version 3.  This is not the first time that game code has been released into the wild, but it is one of the first times that source code from a hit game that was not developed by the notorious maverick John Carmack has escaped the confines of the development house.

Now, it's far too soon to say if anyone will actually do anything interesting with the code, which is named Micropolis due to the need to protect EA's trademarks.  But even if a programmer or two manage to come up with something brilliant, it is unlikely that even a creative spin based on a genuinely great game is going to generate much attention due to that which is now the great bane of the game development industry.

To put it in a nutshell, the problem is art.  Game art, to be specific, the amount and the expense of it required in games today. 

Although my game-playing dates back to the Apple II, Akalabeth, Swashbuckler and the original Castle Wolfenstein, I didn't actually start designing my own games until the MCGA days.  There were two options then, 320x200, 256-color resolution or  640x480 with 16 colors.  Like most would-be developers, my friend and I began by copying a game that we quite liked that we thought we could do a little better, in our case, Warlords from SSG.  Our game was going to be called MythWars, wherein the player was a god from one of the various pantheons, which determined the various army types available.

With such low resolution and a 2D environment, the tiles were so small that two non-artists were perfectly capable of creating what were, at the time, very professional looking graphics.  The rolling hills gave way to majestic, snow-covered mountains and the various monstrous infantries and cavalries really looked quite good against the backgrounds.  For those who can remember those primitive days, it actually looked prettier than Warlords or QQP games like Conquered Kingdoms.  Nowadays, of course, looks crude beyond belief, something a child would be embarrassed to put on Facebook.  

We never finished MythWars, as completing college and then leaping right into the exciting new horizon of 2.5D technology turned out to be a permanent distraction.  We tried a few different approaches, especially with video capture, but we quickly learned that our art skills had reached their limits and 2.5D required hiring real artists from the local art school.  The budget for our first game was only $125,000, which paid for the two full-time artists who worked on it.  These days, that wouldn't cover the cost of the graphics used on a single game level.

The problem is that while games are visually incredible these days, they often aren't actually any more fun to play.  Consider Guitar Hero, for example.  While it's got very realistic graphics, they're really not very important to the game; the player primarily derives his enjoyment from rocking out with something that feels like a real guitar in his hands.  It's the interface that's key, not the visuals, and if you think about it, all games really are, at root, is amusement interfaces.

And art isn't only less fundamentally relevant to games than one might think, but its cost actually creates genuine design problems that are completely unrelated to the art itself.  Because budgets are so massive these days, more people have to sign off on every project and there's greater financial pressure on games to appeal to the widest possible market.  This is not the way to stimulate creativity and design brilliance, but rather imitation and design mediocrity.  This isn't anyone's fault, it's just the natural evolution of the industry.  If it weren't for the fact that a few of the industry's most innovative minds also happen to be some of its most vastly successful ones, we might all be reduced to developing clones of the latest clones of the previous clones.

Is there a way out of this artistic bottleneck?  It seems hard to imagine, since no one wants worse graphics and I don't know very many talented artists who are inclined to work for nothing.  Perhaps EA's release of “Micropolis” may hint at the way, after all, an awful lot of game art is pretty similar to the art used in other games.  What if instead of releasing game code, developers were to release their old textures and models into an online pool from which everyone could draw as needed, thus reducing the need to draw yet another space laser or oak tree?  Obviously, there would still be a need for new art, but at least everyone woudn't be constantly paying to reinvent the dying monster animation.

No doubt there are a million and one reasons why an Internet Art Pool could never come to pass, but in the unlikely event it does, I have some nice 32x32 mountains in case anyone needs them.