Skip to main content
Mathematica

You are not logged in. Your edit will be placed in a queue until it is peer reviewed.

We welcome edits that make the post easier to understand and more valuable for readers. Because community members review edits, please try to make the post substantially better than how you found it, for example, by fixing grammar or adding additional resources and hyperlinks.

Required fields*

4
  • 1
    $\begingroup$ Before spending a lot on this, be aware that only a limited set of things can be accelerated using the GPU. Take a good look at what you are running, and experiment a bit on a colleague's machine that does have an nvidia card. Otherwise you might very easily get disappointed. Parallelization is not easy to get right. GPU computing is even more difficult. (BTW I heavily rely on Mma's parallel tools for what I do, but it would not be possible to use GPUs for the same thing) $\endgroup$ Commented Feb 24, 2012 at 0:44
  • 1
    $\begingroup$ As Szabolcs already pointed out it very much depeneds on your problem domain whether you can expect substantial gains from GPU computing. Why do you need CUDA, would OpenCL work for you? Your HD4670 is supporting OpenCL. You could do a few tests and see what acceleration you can expect. $\endgroup$ Commented Feb 24, 2012 at 1:51
  • $\begingroup$ @Matariki I didn't realize OpenCL existed and I had no idea Mathematica has some support for it. It may require too much development for what we need to do in the time we need to do it, but it certainly looks like a possibility. Thanks. $\endgroup$ Commented Feb 24, 2012 at 6:11
  • $\begingroup$ @Jagra Mathematica has builtin OpenCL support ( reference.wolfram.com/mathematica/OpenCLLink/tutorial/… ) $\endgroup$ Commented Feb 24, 2012 at 8:33