History Of Game AI
Part of the stated mission the AI Game Programmers Guild is to document the history of game AI. This is, by its very nature, not only an ongoing task, but a large and daunting one as well. Soon, the members of the AIGPG will be adding information about past and current game titles including:
- The types of AI used
- Tidbits of information that made them special
- Possible commentary from the designers and programmers
- Source code (when available)
We hope that this project will be informative and educational not only to game AI programmers, but to the game community at-large.
In order to help describe and categorize the game AI discussed here, we use some general terms. None of these is meant to be definitive but rather used as a guide for understanding our terminology and methodology.
We define the eras of game AI in order to provide a very coarse categorization to the games listed in this site. These are largely chronological periods where a particular approach to game AI was used regardless of the underlying architecture used to achieve the approach.
We define the AI architectures that are generally used in games. These aren't necessarily as specific as algorithms. Instead, they tend to be a higher-level framework.
General algorithms such as finite state machines, A*, behavior trees, etc. are beyond the scope of this project. They are also usually well-documented in other areas of the internet including Wikipedia. Algorithms will only be discussed in this project as they relate to specific titles -- particularly if one title had a unique or groundbreaking manner in which an algorithm was used that set it apart from the norm.