Announcement

Collapse
No announcement yet.

NVIDIA vs. AMD OpenCL Linux Benchmarks With Darktable 2.2

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

  • NVIDIA vs. AMD OpenCL Linux Benchmarks With Darktable 2.2

    Phoronix: NVIDIA vs. AMD OpenCL Linux Benchmarks With Darktable 2.2

    Given this weekend's release of Darktable 2.2 as a big upgrade to this open-source RAW photo workflow software, here are some fresh benchmarks of NVIDIA GeForce and AMD Radeon graphics cards under Linux when making use of the program's OpenCL support, which did see some improvements during this v2.2 cycle.

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    Wake me when AMD and company fixes the broken libclc that currently sits at September 26 release and also completely broken against Mesa 13.x and it's Clover OpenCL stack. It's getting really boring seeing the same pointless AMDGPU-Pro tests.

    Comment


    • #3
      The RX 480 seems the go-to GPU for OpenCL, just slightly slower than the 1070 & 1080, but much cheaper.

      Comment


      • #4
        Originally posted by Marc Guillot View Post
        The RX 480 seems the go-to GPU for OpenCL, just slightly slower than the 1070 & 1080, but much cheaper.
        No, it's the RX 470. It performs as well and it costs less.
        ## VGA ##
        AMD: X1950XTX, HD3870, HD5870
        Intel: GMA45, HD3000 (Core i5 2500K)

        Comment


        • #5
          He, he, if not that $99 RX 460, probably per watt too... GTX 1050 does not come anywhere near

          Comment


          • #6
            Originally posted by Marc Driftmeyer View Post
            Wake me when AMD and company fixes the broken libclc that currently sits at September 26 release and also completely broken against Mesa 13.x and it's Clover OpenCL stack. It's getting really boring seeing the same pointless AMDGPU-Pro tests.
            In case you missed my response to your previous post, we switched focus from working on the Mesa/Clover OpenCL to opening up our closed source OpenCL stack some months back, so testing AMDGPU-PRO is the right thing to do even for the open stack.
            Test signature

            Comment


            • #7
              Originally posted by bridgman View Post

              In case you missed my response to your previous post, we switched focus from working on the Mesa/Clover OpenCL to opening up our closed source OpenCL stack some months back, so testing AMDGPU-PRO is the right thing to do even for the open stack.
              He is using Debian, where amdgpu-pro isn't avalable and oss driver CL unusable... since some focuses are shifted around zero point

              And benchmarking does not tells that so he get nothing in practice, from that POV he misses which isn't an answer... so misses nothing, double or even triple nothing

              Comment


              • #8
                Personally, this is the only benchmark I would want to look at when I need to buy a new computer (darktable performance being the only datapoint I'm really interested in), so i hope you continue to run this benchmark regularly.

                Comment


                • #9
                  The GTX 680, 760, 780 TI, 950, 960, 970 and 980 all finishing within margin of error in the Masskrug test is pretty intriguing. What's going on there? It looks like they're all limited by some hardware resource other than the traditional available compute units or memory bandwidth.

                  My guess would be that threads are stalling due to the special function units (which in CUDA-based GPUs are separate from the CUDA cores and much fewer in number) are being used to their capacity and threads are having to wait to use them. If that's the case, then there's probably some optimization work that could be done for much improved performance as a GPU with 2048 cores and a 256 lane wide memory interface should not perform within margin of error of a card with 768 of the same cores and a 128 lane wide memory interface both working at roughly the same clock rate.

                  Comment


                  • #10
                    r7 260x its much faster than r7 370 in this test. curious

                    Comment

                    Working...
                    X