Announcement

Collapse
No announcement yet.

AMD Kaveri: Gallium3D vs. Catalyst Drivers

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

  • AMD Kaveri: Gallium3D vs. Catalyst Drivers

    Phoronix: AMD Kaveri: Gallium3D vs. Catalyst Drivers

    Now having looked at the Windows 8.1 vs. Ubuntu Linux performance, carried out an initial GPU comparison, and compared several Intel/AMD processors, our latest performance investigation with AMD's A10-7850K "Kaveri" APU is looking at the Radeon R7 Graphics performance on Ubuntu Linux when using the open-source "RadeonSI" Gallium3D driver and then using the closed-source AMD Catalyst driver.

    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
    I wonder how Trinity and Richland fare. Are they in the same boat as Kaveri?

    Comment


    • #3
      Originally posted by molecule-eye View Post
      I wonder how Trinity and Richland fare. Are they in the same boat as Kaveri?
      Much, much better.

      I wonder, is DPM supported/enabled on the A10-7850k? I know that 3.13 is supposed to enable it on RadeonSI, but it would explain the huge difference in performance.

      Comment


      • #4
        Originally posted by molecule-eye View Post
        I wonder how Trinity and Richland fare. Are they in the same boat as Kaveri?
        Not at all. I have a 5800k htpc that gets within 10% the performance of Catalyst in Minecraft and HoN, and it has EXA 2d so that also runs fast.

        If you want an AMD APU based machine right now under Mesa, a 5800 or 6800k is definitely a better buy. Just like how a 6970 still beats any SI card under Mesa.

        Though I do know it took Mesa 10 and kernel 3.12 for the hdmi audio to work, and it still doesn't resume the hdmi output if I suspend it.
        Last edited by zanny; 19 January 2014, 10:37 AM.

        Comment


        • #5
          Originally posted by molecule-eye View Post
          I wonder how Trinity and Richland fare. Are they in the same boat as Kaveri?
          Interesting... the performance looks more like the "before" numbers in Michael's 3.12 vs 3.13 radeonsi article.

          Trinity and Richland use r600g instead of radeonsi so there are more performance optimizations in place for those parts. That said, radeonsi was putting out some much improved numbers in yesterday's article as well (albeit on Pitcairn & Tahiti dGPUs with ~3x and ~4x the shader power respectively) so there might be something else going on here.

          Based on limited experience with my half-a-Kaveri dev box I was expecting about 2x this performance, ie roughly 1/2 of Catalyst performance rather than 1/4. Anyways, thanks Michael for running the tests.
          Last edited by bridgman; 19 January 2014, 10:43 AM.
          Test signature

          Comment


          • #6
            Question: What were the clocks and timings of the memory? Memory bandwidth makes a very big difference in the performance of an AMD APU's graphics component, and latency makes a small-but-measurable difference in CPU-bound tests.

            Comment


            • #7
              Originally posted by A Laggy Grunt View Post
              Question: What were the clocks and timings of the memory? Memory bandwidth makes a very big difference in the performance of an AMD APU's graphics component, and latency makes a small-but-measurable difference in CPU-bound tests.
              DDR3-2133MHz is what the memory is run at for all Kaveri tests; the timings were whatever the profile was specified at; I don't recall off-hand and sadly not exposed to user-space so not recorded. PTS can only chart the memory frequencies when run as root.
              Michael Larabel
              https://www.michaellarabel.com/

              Comment


              • #8
                Were the Benchmarks with or without DPM?

                Comment


                • #9
                  Originally posted by ObiWan View Post
                  Were the Benchmarks with or without DPM?
                  DPM is defaulted on in 3.13, so the huge performance gap is with all the gpu cores in use and at their rated frequencies. I think. They may not have DPM code for Kaveri in yet.

                  Comment


                  • #10
                    Top end Kaveris like/want/need 2400MHz DDR3 RAM.

                    Comment

                    Working...
                    X