AMD’s developer relations manager, Richard Huddy, explained to Custom PC that the most common AI tasks involve visibility queries and path finding queries. ‘Our recent research into AI suggests that it’s not uncommon for gaming AI to spend more than 90 per cent of its time resolving these two simple questions,’ says Huddy. He also added that these two queries are ‘almost perfect for GPU implementation,’ because they ‘make excellent use of the GPU’s inherently parallel architecture and are typically not memory-bound.’Huddy says some middleware providers are looking to pack up a GPU AI library for games, while some developers are looking to transfer their existing AI code from CPU to GPU. NVIDIA expects the first games with GPGPU-accelerated AI may arrive within 12-18 months.
Nvidia’s director of product management for PhysX, Nadeem Mohammad, agrees,telling Custom PC that ‘all the simple, complex operations’ involved with path finding and collision detection ‘are all very repetitive, so path finding is one of the algorithms which does work very well on CUDA.’
ATI and NVIDIA: AI will move to the GPU
Posted on Thursday, January 22 2009 @ 18:45 CET by Thomas De Maesschalck
AMD Developer Relations Manager Richard Huddy told CustomPC that future games will use the GPU to calculate artificial intelligence, in the same fashion that NVIDIA has enabled PhysX on its GPUs. Huddy claims the first games with GPGPU-accelerated AI may arrive as soon as early 2010.