Announcement

Collapse
No announcement yet.

The OpenGL ES 3.1 Foundation Is Being Laid In Mesa

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

  • The OpenGL ES 3.1 Foundation Is Being Laid In Mesa

    Phoronix: The OpenGL ES 3.1 Foundation Is Being Laid In Mesa

    Intel developers in particular have been trying to wrap-up OpenGL ES 3.1 support within Mesa. That work is getting closer to finally being realized...

    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 if there will be any drastic optimization work done with performance patches after mesa has finally caught up with extensions or if the mesa devs are already going full throttle on that front alongside the extension work. Would be pretty neat to see mesa benchmarks beating out catalyst results as the new norm and then start to close the gap between AMD GPU's and proprietary Nvidia driver performance.

    Comment


    • #3
      Thanks all Mesa3D/driver hackers for hard work. Everytime I try to understand and grasp work and knowledge required, I am bit taken back. Big thanks (and I start to think I will donate some money to Mesa this year).

      Comment


      • #4
        Originally posted by kenjitamura View Post
        I wonder if there will be any drastic optimization work done with performance patches after mesa has finally caught up with extensions or if the mesa devs are already going full throttle on that front alongside the extension work. Would be pretty neat to see mesa benchmarks beating out catalyst results as the new norm and then start to close the gap between AMD GPU's and proprietary Nvidia driver performance.
        It is quite likely to happen like that. First, there hasn't been any news of a newer version of OpenGL coming soon, all the focus has been on Vulkan and reducing complexity. So, they will have time to work on optimizations versus feature parity. It should be noted that they are probably trying to optimize as they can, but it isn't their highest priority.

        I think they will look into using the Vulcan SPIR-V as an opportunity to standardize the shader pipeline, allowing them to reuse the same code as Vulkan. Someone will probably look into implementing OpenGL on top of Vulcan, and that may have some benefits, keeping everyone working on the same stack. But we don't know yet how any of that will pan out.

        With the focus off of featuritus, open source OpenGL has a chance to shine.

        Comment


        • #5
          Originally posted by dragorth View Post
          I think they will look into using the Vulcan SPIR-V as an opportunity to standardize the shader pipeline, allowing them to reuse the same code as Vulkan. Someone will probably look into implementing OpenGL on top of Vulcan, and that may have some benefits, keeping everyone working on the same stack. But we don't know yet how any of that will pan out.
          What do you mean by "standardizing the shader pipeline"?

          Comment


          • #6
            He probably means theres a common standard shader input we can all target.

            Comment

            Working...
            X