Graphics Are a Gimmie. What's NEXT?

At last month's Game Developer's Conference, Naughty Dog founder Jason Rubin (Jak and Daxter, Crash Bandicoot) gave a lecture entitled "Great Game Graphics: Who Cares?" Coming from a company known for pushing systems to the limit and selling tens of millions of games due (in part) to having some of the most incredible graphics available for their respective system, the tone of the presentation was a bit of a shocker.

Rubin wasn't trying to say that graphics in games were no longer important. He was saying that the breakneck pace of graphical development that defined the first 30-odd years of the industry is no longer going to be the driving force behind the types of games people make and the types of games that sell. By way of example he traced the history of several game genres. For example: Tennis. You start with Pong, which had the essential elements: two players and a ball. By the time you get to the Intellivision or Nintendo 8-bit era, you can switch perspectives, allowing you to show the height of the ball and adding depth (literally) to gameplay. Later hardware was actually able to put 3D-modeled players on a 3D court, strengthening gameplay by giving gamers realistic player positions and timing. With the current generation of hardware, the player models are so good that game developers feel obliged to zoom in on them after each volley to show them off. Video game tennis is, from a distance, indistinguishable from a real live tennis match on TV.

Graphically, it's still possible to do more: In subsequent generations of hardware, we could (for instance) render each hair on the tennis player's head, or add in beads of sweat. Their eyes could follow the ball. Hell, we could render each individual racket string. But here's Rubin's point: "How's it gonna change the game?" he asks. Current graphics technology is able to provide all the fundamentals needed for a realistic tennis game.

You can trace a similar evolution with fighting games from 2D sprites to digitized people to 3D flat-shaded models to what we have today -- fully rendered 3D environments and realistic 3D characters moving within them. Or first-person shooters. Each generation of graphical technology added a whole new depth to the gameplay ... until today. There's no doubt that the lighting effects of DOOM 3 or Deus Ex 2 are visually stunning, but the technological leap won't affect gameplay as much as the leap from, say, the "flat-3D" fakery of DOOM to the full 3D environment of Quake.

Graphics No Longer Need to be "Cutting-Edge" to Impress

The success of games like Grand Theft Auto 3 and the Tony Hawk Pro Skater series underscore the point. Both games use the Renderware game engine, which (owing in part to its flexibility) isn't a state-of-the-art system. Your typical screenshot from Tony Hawk isn't pushing nearly as many polygons or showing off nearly as many effects as a cutting-edge PS2 game. But most gamers are hard-pressed to notice the difference; the game still works and it's damn fun. With previous generations of hardware (such as the PSOne) the difference between a cutting-edge game and typical games was significant; gamers noticed the difference between a few hundred polygons or a few thousand, and many games (ones with intricate levels, for instance) weren't even possible without pushing the envelope. The situation is different today: The difference between a 50,000 polygon model and a 150,000 polygon model is mostly cosmetic. Everything necessary to create living breathing characters capable of depicting action and emotion onscreen is available. Future enhancements are just gravy, not gameplay.

This is a pretty big shift, and it's going to shift the way people make games and the way people buy them. "What are we going to do to expand what we do?" Rubin asked his audience. "How are we going to draw people in?" His talk left the question unanswered. In fact, it was a kind of challenge directed at the audience. What's next?

What Will Sell in 2004 and Beyond?

I'm going to take a stab at picking up where Rubin left off. Obviously one way to bring people to a new game is to attach it to a license, be it a sequel or a movie license. But if you're like me, you're a lover of original games, fresh ideas, and new characters created for gaming alone. So I'm going to tackle it from that angle: What will make original games sell?

The short (but by no means definitive) answer I suggest consists of two things: Artistry and Interactivity. Let's assume that great graphics are a given. Within the next generation of tech (or perhaps you can argue the time is already here), even run-of-the-mill developers will have the ability to render fully functional 3D worlds with as much detail as needed to impact gameplay. What will the crucial differences be?