Announcement

Collapse
No announcement yet.

18-Way NVIDIA/AMD Linux Performance For Dawn Of War III

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

  • 18-Way NVIDIA/AMD Linux Performance For Dawn Of War III

    Phoronix: 18-Way NVIDIA/AMD Linux Performance For Dawn Of War III

    Today marks the highly anticipated debut of Dawn of War III for Linux (and macOS) ported by Feral Interactive. Here are a number of OpenGL and Vulkan benchmarks of NVIDIA GeForce and AMD Radeon graphics cards running Ubuntu Linux with this game.

    http://www.phoronix.com/vr.php?view=24767

  • #2
    it seems that nvidia cards provide better support both to vulkan and opengl on video games.

    A good comparison would be using mesa drivers on all the cards although I ignore if mesa drivers implement vulkan api.
    Last edited by Azrael5; 06-08-2017, 07:15 AM.

    Comment


    • #3
      Originally posted by Azrael5 View Post
      it seems that nvidia cards provide better support both to vulkan and opengl on video games.
      Vulkan performance on AMD would likly have been better then OpenGL if the proprietary driver was used.

      Comment


      • #4
        Originally posted by AJenbo View Post

        Vulkan performance on AMD would likly have been better then OpenGL if the proprietary driver was used.
        I'll be trying -PRO today to see.... Feral hadn't mentioned -PRO support at all and I didn't have the time yet as only got the final build for benchmarking ~23 hours ago.
        Michael Larabel
        http://www.michaellarabel.com/

        Comment


        • #5
          i am confused you start with vulkan having better results and then the tests show opengl is better?

          Comment


          • #6
            Originally posted by cj.wijtmans View Post
            i am confused you start with vulkan having better results and then the tests show opengl is better?
            The results are confusing indeed.

            The GoL benchmark shows Vulkan practically always doing better at 1080p, literally the direct opposite of what Phoronix is showing: https://www.gamingonlinux.com/articl...rt-report.9802

            They don't have as much of course, but it shows Vulkan doing much better than in Phoronix testing.

            Comment


            • #7
              Originally posted by gutigone View Post

              The results are confusing indeed.

              The GoL benchmark shows Vulkan practically always doing better at 1080p, literally the direct opposite of what Phoronix is showing: https://www.gamingonlinux.com/articl...rt-report.9802

              They don't have as much of course, but it shows Vulkan doing much better than in Phoronix testing.
              thats because he uses a beefier cpu.

              Comment


              • #8
                Originally posted by gutigone View Post

                The results are confusing indeed.

                The GoL benchmark shows Vulkan practically always doing better at 1080p, literally the direct opposite of what Phoronix is showing: https://www.gamingonlinux.com/articl...rt-report.9802

                They don't have as much of course, but it shows Vulkan doing much better than in Phoronix testing.
                In the CPU-bound cases, Vulkan does better. Perhaps due to them using a slower CPU.
                Michael Larabel
                http://www.michaellarabel.com/

                Comment


                • #9
                  Originally posted by AJenbo View Post

                  Vulkan performance on AMD would likly have been better then OpenGL if the proprietary driver was used.
                  Yeah, likely. It's not 100% certain though. We might still be surprised.

                  Comment


                  • #10
                    Originally posted by Michael View Post

                    In the CPU-bound cases, Vulkan does better. Perhaps due to them using a slower CPU.
                    I thought it might be, but in that case, does it not make using such a high-end CPU a little misleading for people? The vast majority of people likely have mid-range hardware, so people looking at all your benchmarks are likely seeing results they will never see themselves. We need more variation in the processors used, otherwise these are essentially pie in the sky stuff for most people.

                    Comment

                    Working...
                    X