Announcement

Collapse
No announcement yet.

OpenGL vs. Vulkan On The AMD Ryzen 3

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

  • #31
    Originally posted by bridgman View Post
    Because it's a different driver, different people, different design, different ideas ?
    This test shows nothing. Can be, yet..
    Originally posted by bridgman View Post
    We do look at using the Mesa driver for workstation from time to time and are looking at it again - but adding compatibility mode to Mesa would be a big pile of work and is not accepted as a good idea by much of the Mesa community. The closest thing to a workable solution I have seen so far is keeping the current -PRO GL driver for workstation and using Mesa for consumer Windows drivers, although that would require some changes in DOOM to work with core profile rather than compatibility profile. There are probably some other Windows OGL-only games which require compatibility profiles though, and most of them are not likely to change.

    So a big honkin' pile of work for relatively little benefit. Hasn't made sense yet, as I have said multiple times already.
    Interestingly, we are now talking only about OpenGL and the Vulkan or even about the Radeon Settings, OpenCL (which generally consist of the Windows driver) ?
    I would like to have "more open and free the driver" on all platforms.

    P.S. Thoughts on some cards doesn't even work with the Vulkan.

    Comment


    • #32
      bug77, thx

      It will be nice to see how polaris utilize cpu

      Comment


      • #33
        I looks like a RX 480 regression in the latest mesa. I had much better results with R9 380 2GB + Ryzen 5 1600X in Mad Max. I know that my CPU is better than Ryzen 3 from the test, but Mad Max on Linux doesn't need 12 threads, so difference between those 2 CPUs in this game should be much smaller. Those are my results:
        OpenBenchmarking.org, Phoronix Test Suite, Linux benchmarking, automated benchmarking, benchmarking results, benchmarking repository, open source benchmarking, benchmarking test profiles

        Vulkan test didn't run properly and benchmark was executed just for OpenGL (Vulkan test skipped very fast... and results was copied from OpenGL test). I verified results via Steam FPS counter.

        Comment


        • #34
          Originally posted by mphuZ View Post
          This test shows nothing. Can be, yet.
          If you look at other recent tests it seems to show a regression in Mesa, although that has not been confirmed. It is possible that performance is actually wildly lower on a 4-core without HT than a 4-core with HT but that seems less likely.

          Originally posted by mphuZ View Post
          Interestingly, we are now talking only about OpenGL and the Vulkan or even about the Radeon Settings, OpenCL (which generally consist of the Windows driver) ?
          I would like to have "more open and free the driver" on all platforms.
          I don't understand. You were talking about Mesa earlier and using it on Windows, and that was the topic of all the questions you claim I did not answer (despite having answered a bunch of times).

          The OpenCL driver has already been open sourced and work is proceeding on open sourcing the Vulkan driver.

          Originally posted by mphuZ View Post
          P.S. Thoughts on some cards doesn't even work with the Vulkan.
          Which cards ? AFAIK all the cards supported by AMDGPU-PRO are working with Vulkan.
          Last edited by bridgman; 18 August 2017, 08:27 PM.
          Test signature

          Comment


          • #35
            nadro, I don't think so, this is not look like polaris bug
            this my result for dota2. i5 haswell (stock 3.8GHz and trimmed to 2GHz) + rx470 OC (pretty the same as 480):
            OpenBenchmarking.org, Phoronix Test Suite, Linux benchmarking, automated benchmarking, benchmarking results, benchmarking repository, open source benchmarking, benchmarking test profiles

            i5@2GHz is still too times faster... this is never gonna happend

            Comment


            • #36
              There is definitely something odd going on here. Looking at Michael's previous CPU scaling article...

              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


              ... there is definitely some CPU scaling going on with OpenGL but DOTA 2 at 1920x1080 is getting almost 100 fps with R580 and i7600 vs 36 fps with a 480 and Ryzen 3 in this test. Seems way beyond what CPU scaling can explain.

              Vulcan numbers are even stranger - the previous article shows essentially no CPU scaling between 7600 4 thread, 7700 8 thread and R1800 16 thread (7600 was actually fastest), with over 100 fps in all cases on R580 vs 25 fps on R480 with Ryzen 3 in this article.

              First thought was maybe the performance governor not used with Ryzen ? Michael explicitly stated use of performance governor in previous article but I didn't see a mention either way in this one.
              Test signature

              Comment


              • #37
                Originally posted by bridgman View Post
                First thought was maybe the performance governor not used with Ryzen ? Michael explicitly stated use of performance governor in previous article but I didn't see a mention either way in this one.
                The answer is:
                Code:
                Processor Details-
                Scaling Governor: acpi-cpufreq ondemand

                Comment


                • #38
                  I have some other OpenGL game benchmarks coming out this weekend on Ryzen 3 and so far it's looking like there may be a regression mostly affecting the Dota2/SS3/MadMax/Dow3 tests as at least for my other (games without Vulkan renders), the performance appears more normal in relation to GTX 1050/1060.
                  Michael Larabel
                  https://www.michaellarabel.com/

                  Comment


                  • #39
                    Originally posted by eydee View Post

                    Because Mesa is not too efficient on the CPU side. When I switched from Phenom II to Ryzen 5, performance on my old 6850 increased by up to 400%. In old games, not the latest titles it couldn't run anyway.
                    probably the CPU was the bottleneck.

                    Comment


                    • #40
                      Originally posted by bridgman View Post
                      Which cards ? AFAIK all the cards supported by AMDGPU-PRO are working with Vulkan.
                      This is on Windows. On HD7850 and RX480 (the other person) Vulkan just crash. With other games from ID, too many problems.

                      Comment

                      Working...
                      X