1,000 Players Can't Be Wrong
The primary responsibility of a game AI programmer is to anticipate player behavior, and provide satisfying responses for every anticipated scenario. With each generation of games this gets more difficult, as virtual worlds become more realistic with additions like real-time physics simulation. I’m working on a research project that flips the traditional approach to game AI on its head. My goal is to capture gameplay from 1,000 multiplayer game sessions, analyze this data, and use it to generate AI for a new single player game this summer.
Read more and download game at:
http://theRestaurantGame.net
The simple game I’ve created does not have real-time physics, but it does require conversation between players via typed text, which proves equally challenging -- in both cases no matter how many games you observe, you will never see the exact same scenario play out twice. It is unrealistic to believe human AI programmers will ever account for the nuances of every possible scenario. Think about an analogy to animation – when animators want convincing human movement, they go straight to the source and use motion capture. Animators blend, layer, sequence captured animations to generate new motions beyond those recorded. Why should we believe that authoring AI behaviors is any different – perhaps we can get the most believable results by starting with the source material, and procedurally blending, layering, and sequencing behaviors as we see fit.
I’m not expecting people to play my game and get nothing in return. Players will be rewarded with Game Designer credit on the new game for contributing to this effort. It has never been easier to earn Game Designer credentials! I see this collaborative community-oriented approach as falling in line with other successful examples of leveraging the power of the network to overcome insatiable demands for content such as Spore, the ESP game, flickr, and myspace.
Of course, collecting gameplay data is only half the battle, but it is a start. The game produced this summer will employ the first of many iterations of the learning algorithm, and the corpus of data collected will be available for others to use as well (with player identifiers removed).
We are running our second set of pilot tests Thursday Feb 22 at 9pm EST AND 9pm PST. You are welcome to download and play any other time too, but wanted to set a time when people would be sure to find other players (because the game requires two players per server).
Read more and download game at:
http://theRestaurantGame.net