Announcement

Collapse
No announcement yet.

NVIDIA OpenCL Linux Benchmarks

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    Can anyone tell me how to get the OpenCL stuff to run? I've installed the latest Nvidia drivers, 260.19.29, but when I install the phoronix-test-suite packages it says I need to install the Nvidia OpenCL package. I've googled and looked on the forums, but I can't find anything about how to get OpenCL to work.

    I have an 8800GT, gtx 460, and a gtx 470 that I'd like to benchmark.

    Comment


    • #17
      The benches require that the openCL headers be placed in /usr/include/CL/. The headers you can get when you install the cuda sdk available here:

      http://developer.nvidia.com/object/c...downloads.html

      Comment


      • #18
        Thanks! I'll give that a shot.

        Comment


        • #19
          Originally posted by deanjo View Post
          Not even close, many changes occurred in the computing capability in the GTX 2xx series (64-bit precision being one big one). In fact many generations have carried changes.

          You might want to look at this for a good comparison:

          http://www.geeks3d.com/20100606/gpu-...arative-table/
          I don't see anything about 64bit precision over there. I see more registers, more threads and warps per multiprocessor and couple of atomic operations support. I'm not really familiar with gpu computing, but these all look like minor improvements to me.

          Comment


          • #20
            Originally posted by bug77 View Post
            I don't see anything about 64bit precision over there. I see more registers, more threads and warps per multiprocessor and couple of atomic operations support. I'm not really familiar with gpu computing, but these all look like minor improvements to me.
            Let me highlight for you:



            http://en.wikipedia.org/wiki/Double_...g-point_format

            Comment


            • #21
              Why does PTS not install the headers locally?

              Comment


              • #22
                Originally posted by deanjo View Post
                Hehe, that's a CC 1.3 feature and GT220 only supports CC 1.2.

                In short, I was wrong, there were some changes with regard to CC since the 8000 thousand series. But my point was there's a mistake (two actually) in the article. It claims that the GT220 is faster than 9800GTX (48 vs 128 CUDA cores, who are we kidding here?) and also that we should expect it to be faster.

                Comment


                • #23
                  Originally posted by bug77 View Post
                  Hehe, that's a CC 1.3 feature and GT220 only supports CC 1.2.

                  In short, I was wrong, there were some changes with regard to CC since the 8000 thousand series. But my point was there's a mistake (two actually) in the article. It claims that the GT220 is faster than 9800GTX (48 vs 128 CUDA cores, who are we kidding here?) and also that we should expect it to be faster.
                  Understandable, when I was thinking of "200 series" I was thinking of the GTX line. Anyways I noticed that my previous bench of the GTX-580 was running the wrong mandel test. Here are the complete results again with the correct test included.

                  Comment


                  • #24
                    Originally posted by dyna View Post
                    You mean something like this? As described on the last page...
                    I did not. Michael failed to describe an architectural change that made that possible. For all I know, the performance discrepancy is a bug.

                    Comment

                    Working...
                    X