Announcement

Collapse
No announcement yet.

AMD Has A New Linux/OpenGL Driver Relationship Representative

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

  • AMD Has A New Linux/OpenGL Driver Relationship Representative

    Phoronix: AMD Has A New Linux/OpenGL Driver Relationship Representative

    Beginning next week there's going to be a new member of AMD's developer relationship group that manages the ties between the graphics driver developers and game developers. This new employee is an experienced Linux developer and will focus on Linux/OpenGL cooperation...

    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
    Something else, did something to Tim Writer happened? I never see him Posting (Like Bridgman).

    Comment


    • #3
      With mesa-git i do have less problems than with catalyst. The performance feels about the same like catalyst 14.12.
      In theory it should be worse but honestly: i do not recognize it. As a nice + gallium nine makes wine games run and in everey case much smoother.

      Will Matth?us make me switch back? I do not think so

      Comment


      • #4
        So Chajdas, any idea of what happened to Mantle on Linux?

        Comment


        • #5
          ??

          Originally posted by Dukenukemx View Post
          So Chajdas, any idea of what happened to Mantle on Linux?
          amd don't care about linux, mantle is only for windows

          Comment


          • #6
            Originally posted by rikkinho View Post
            amd don't care about linux, mantle is only for windows
            My guess is that AMD doesn't see any future for Mantle. That's why they opened it up to Khronos Group so they can reuse as much as possible.
            This would explain why there seem to be already working OpenGL Next drivers (Demos during GDC). This is eventually how Mantle will make its way to Linux and to Windows for NVIDIA cards. With DX12 and OpenGL Next there is simply no reason for Mantle to exist. AMD is happy that mantle basically gets used. NVIDIA will be happy that they did not have to implement AMDs Mantle even though they kind of did.

            Let's hope that the OpenGL Next implementation quality won't have as much difference between vendors as with legacy OpenGL. So devs can just port to OpenGL Next and everyone on Linux can enjoy performance that hits the hardware bottleneck and not the implementation quality bottleneck.
            Last edited by blackout23; 13 February 2015, 06:07 PM.

            Comment


            • #7
              That makes sense. Let's hope the theory matches the reality check

              Comment


              • #8
                Oh, I've been following this person on twitter for a year or so, didn't even realize. Good for him!

                Comment


                • #9
                  Originally posted by Dukenukemx View Post
                  So Chajdas, any idea of what happened to Mantle on Linux?
                  Screw Mantle, Gallium OpenGL Next within 6 months of release please.

                  Comment


                  • #10
                    Originally posted by blackout23 View Post
                    My guess is that AMD doesn't see any future for Mantle. That's why they opened it up to Khronos Group so they can reuse as much as possible.
                    This would explain why there seem to be already working OpenGL Next drivers (Demos during GDC). This is eventually how Mantle will make its way to Linux and to Windows for NVIDIA cards. With DX12 and OpenGL Next there is simply no reason for Mantle to exist. AMD is happy that mantle basically gets used. NVIDIA will be happy that they did not have to implement AMDs Mantle even though they kind of did.

                    Let's hope that the OpenGL Next implementation quality won't have as much difference between vendors as with legacy OpenGL. So devs can just port to OpenGL Next and everyone on Linux can enjoy performance that hits the hardware bottleneck and not the implementation quality bottleneck.
                    that shouldn't be a problem if Khronos keeps word on 2 things. really low level and strict compatibility testing. but, the thing that would make me happiest would be some common opengl layer on top of glNext working on all vendors. that alone would solve whole lots of troubles, since old gl wouldn't be as random as it is across the vendors

                    Comment

                    Working...
                    X