Loading Profile...
Speed up synth with nVidia or ATI GPU!
You need to make this work with nVidia's latest GPU's and drivers, so the processing can be done by a graphics card. My new 8800 GS card kicks butt on folding@home data (team 84451) so why couldn't be utilized to speed up the synth process?
10
people like this idea
I like this idea!
Tell me when this idea gets some attention.
The more people who like this idea, the more it gets noticed.
The more people who like this idea, the more it gets noticed.
The company has this under consideration.
-
Inappropriate?There are some steps in the synthing process that could be GPU accelerated. I agree, this would be really fun to do.
-
Inappropriate?Although I agree that in some cases you can speed stuff up a bit using the GPU for general computation, in the case of folding@home, it's because the normal CPU code for the GROMACS core is only 32-bit, single-threaded, and overall very inefficient. It's such that it's actually slower on my 64-bit Core 2 Duo than on my single-core, 32-bit Pentium 4. I've tossed around the idea of helping them out by making a 64-bit, multi-threaded, vectorized version of it, but it'd take a lot of time that I don't have right now. When Larrabee gets out, then using GPUs for general computation should really take off.
I’m unsure
-
Inappropriate?NVidia CUDA development lanquage is the most advanced of the GPU computing programs. More info on CUDA and the latest GPU's:
http://www.realworldtech.com/page.cfm...
1 person thinks
this is one of the best points
EMPLOYEE
