Video in 2010 will support hardware akseleratsiyu calculations related to artificial intelligence in games, writes TechRadar. Over the relevant procedures are already working Nvidia and AMD, the major manufacturers of graphics cards.
The developers claim that 90 percent of the procedures of artificial intelligence are a few basic algorithms. In particular, the video will help identify the characters, whether they see the player and each other, as well as to calculate the optimal way in the game world.
The main advantage of this approach is the ability to cheat a large number of parallel in the world of computer characters, and there is no need to develop an artificial intelligence system from scratch.
Video is much better than conventional processors, adapted to the parallel processing of large volumes of identical data. To implement this capability exists standard OpenCL, describing the modes of transmission of computer graphics processing power. This technology will greatly accelerate the normal application.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment