jmke

Advanced Micro Devices said that Nvidia Corp. had specifically altered its PhysX application programming interface (API) so that it could not take advantage of multi-core central processing units (CPUs) while making physics effects computations. According to AMD, the reason for such modifications was to increase importance of graphics processing units (GPUs) that are used to process physics effects in select games that are powered by PhysX.

“The other thing is that all these CPU cores we have are under-utilised and I'm going to take another pop at Nvidia here. When they bought Ageia, they had a fairly respectable multi-core implementation of PhysX. If you look at it now it basically runs predominantly on one, or at most, two cores. […] I wonder why Nvidia has done that? I wonder why Nvidia has failed to do all their QA on stuff they don't care about – making it run efficiently on CPU cores – because the company doesn't care about the consumer experience it just cares about selling you more graphics cards by coding it so the GPU appears faster than the CPU. It's the same thing as Intel's old compiler tricks that it used to do; Nvidia simply takes out all the multi-core optimisations in PhysX,” said Richard Huddy, AMD’s Worldwide Developer Relations manager, in an interview with Bit-tech.net web-site.

Nvidia has been proclaiming importance of physics processing on graphics processing units with the help of PhysX API it acquired with the takeover of Ageia back in 2008. The company has been also showing benchmarks that showed substantial performance advantage of GPUs over CPUs when it comes to processing of PhysX effects. However, AMD believes that such performance advantages are artificial.

http://www.xbitlabs.com/news/multime...PhysX_API.html

__________________