First, watch this video. Make sure you watch the whole thing. (this video is not realtime – it took far longer than this to render)

First of all this video is really cool. But it also provides an glimpse of what might be possible when the extreme computational requirements of universally-physically-simulated games become manageable.A lot of people think physics is just an advertising buzzword. Unfortuantely, as of 2007, they are mostly right.I think that physics has a lot more potential than we’re currently appreciating. We’re going to need to look at how a physics system capable of simulating a lot of bodies at high accuracy could actually affect game mechanics. Oddly enough, I think it will allow simpler game mechanics, while simultaneously increasing the complexity of the game.All real-world phenomena are emergent from a small set of physical rules. With extreme processing power, we may be able to simulate a lot of game systems via physics instead of directly hardcoding them. We could actually hardcode less and simultaneously get more behaviours emerging from the game system.Once that starts happening, I hope we will see a lot of interesting and subtle emergent differences in gameplay. Little things which physics simulation adds in could magnify into gameplay-relevant elements.

Even better, we all intuitively understand physics, so this would allow us to add another layer of complexity to the game system, without adding any learning curve. That’s great for a game design.

At the same time, it would help reduce the ‘game-y’ weirdness in all games that detracts from immersion and emotional impact. Except for game mechanics reasons, there would be no reason you couldn’t do anything you can do in the real world. We will be limited only by the game we want to create, and the interface between machine and player.

I look forward to the day when we can wrestle with this.