Announcement

Collapse
No announcement yet.

ZINK: OpenGL Over Vulkan Comes As New Mesa Gallium3D Driver

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

  • ZINK: OpenGL Over Vulkan Comes As New Mesa Gallium3D Driver

    Phoronix: ZINK: OpenGL Over Vulkan Comes As New Mesa Gallium3D Driver

    Collabora has been developing a new "Zink" Gallium3D driver for Mesa that gets OpenGL running over Vulkan...

    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'd hate to sound like I'm being overly negative here, but I fail to see the point in this. With DXVK and MoltenVK the point is kind of obvious, both being well supported but completely proprietary APIs, but any platform that supports Vulkan is virtually guaranteed to support OpenGL up to a pretty recent version.

    Sure, there may be some performance benefits similar to DXVK, but with all the AZDO work I'm having a hard time imagining they're going to be all that substantial.

    Don't get me wrong, I'm not saying that developers shouldn't spend their free time on this and I imagine it's quite an interesting set of challenges, but I still fail to see the point in this.

    Comment


    • #3
      opengl over vulkan? ive been ZINKing about this since the day vulkan launched

      Comment


      • #4
        Originally posted by L_A_G View Post
        I'd hate to sound like I'm being overly negative here, but I fail to see the point in this.
        Should this be(come) performant and complete, graphics driver makers will only need to write a vulkan driver to support everything including OpenGL instead of having their own or even another mesa openGL driver implementation.

        On Desktop with the AMD and nVidia this may not be an issue because drivers already exist for them, but for the embedded platforms, Adreno, Mali, Power3d etc it could simplify driver development a lot and give users a better experience because openGL support is almost automatic.

        Comment


        • #5
          I am hoping this or DXVK will improve my hardware utilization on Arma 3 or other modern games using legacy rendering APIs. And I think to myself, What a wonderful world...

          I tested it last night on Windows (DX11) and Linux (OpenGL). I have not played it about a year and since there are so many DLCs maybe they optimized the game. In some instances I got 9 FPS while running in a town with lots of weather effects, my CPU was at 9% and GPU at 12% utilization. In other areas with less buildings and props I got > 100 FPS with 100% GPU utilization (test where done in single player mode). I have yet to disable battle-eye and play with DXVK.

          I'm sure I would have profiling in place once ZINK reaches OpenGL implementation required by Arma 3.

          Comment


          • #6
            Performance benefits compared to what? Neither GL nor DX map directly to Vulkan. So there will always be overhead.

            Comment


            • #7
              Currently AMD (who currently support 5 versions of GCN) have to write 5 OpenGL drivers, 5 Vulkan drivers, 5 DirectX 9 drivers, 5 DirectX 10 drivers and 5 DirectX 11 Drivers*. And then the hundreds of game specific patches need to be implemented 5 times each (assuming each game only supports one API).

              If these conversions layers work out being as good as native then AMD will only have to write 5 Vulkan drivers with every thing else converted, and game specific patches would only need to be applied to the relevant conversion layer once instead of multiple times for each driver.

              *The different versions of GCN have a lot of overlap so some of the driver can be shared between each version.

              Another advantage is that currently the drivers have a lot of hacks in order to incorporate every game specific patch, AMD tried to fix a bug in one game a few months ago and broke the game specific patches for a 15 year old game engine in the process. With these conversions layers you can store multiple versions, so that if a game has a really broken DirectX 9 implementation you can create a specific version for that game and always use that version for that game without the need to add the patches to the mainline.

              Comment


              • #8
                Originally posted by log0 View Post
                Performance benefits compared to what? Neither GL nor DX map directly to Vulkan. So there will always be overhead.
                Overhead over Vulkan performance, but they can run faster than the current OpenGL implementation.

                Comment


                • #9
                  Ok, Guys... When systemd over Vulkan?





                  sorry I couldn't resist...

                  Comment


                  • #10
                    Originally posted by Danielsan View Post
                    Ok, Guys... When systemd over Vulkan?





                    sorry I couldn't resist...
                    But will it be written in Rust or not?

                    Comment

                    Working...
                    X