I originally posted in the NVIDIA forums, but I received a recommendation for the Radeon HD 6990 and would like to consider AMD GPUs as well. I need to buy a GPU that will be used for scientific computing with OpenCL.
Fast double precision performance
Considering: Radeon HD 6990/Radeon HD 6970/GeForce GTX 590/Quadro 5000/Quadro 6000
Stream Processors/CUDA Cores: 3072/1536/1024/352/448
Gigaflops (double precision): 1270/683/unlisted/359.04/515.2
Memory: 4GB(2GB per GPU)/2GB/3GB(1.5GB per GPU)/2.5GB/6GB
Memory Bandwidth (GB/s): 320 (160 per GPU)/176/327.7(~160 per GPU)/120/144
It looks like the memory on any of these cards would be sufficient.
One concern I have about the AMD's Radeon HD 6990 and NVIDIA's GTX 590 is that they each house two GPUs, which I'd assume slows down performance some due to memory transfer between the cards (and also complicates porting the program to OpenCL). Furthermore, I read in AMD's June 2011 OpenCL Programming Guide that multiple GPU devices are currently not supported. If I understand correctly, this means that OpenCL could only use half of the Radeon HD 6990.
At this point I'm leaning towards the Radeon HD 6970 since, putting the Radeon HD 6990 aside, it has the highest gigaflops (double precision) and memory bandwidth (per GPU).
Any advice or other suggestions would be appreciated.