Announcement

Collapse
No announcement yet.

X.Org Server 1.20 RC4 Released, EGLStreams For XWayland Might Still Land

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

  • #31
    Originally posted by lucasbekker View Post
    Not merging EGLStreams for XWayland will only hurt Wayland adoption, not NVidia...
    Exactly. My users use scientific software that uses CUDA. Not having Nvidia cards is not an option. Unfortunately all this will do is force us to stay on X.

    Comment


    • #32
      Originally posted by DerCaveman View Post
      So that 0.5% should buy AMD or use Windows - period
      I've *never* had a good experience with AMD cards under linux. There's always some gotcha bug. Last time it was couldn't get dual monitors to work.

      Comment


      • #33
        Originally posted by pgoetz View Post

        Exactly. My users use scientific software that uses CUDA. Not having Nvidia cards is not an option. Unfortunately all this will do is force us to stay on X.
        Stay on X then. Who are your users and why should I care whether they're using Wayland?

        Not having Nvidia cards is an option. I'm not suggesting that Nvidia should be excluded from Wayland, I'm suggesting Nvidia should be forced to support GBM like everyone else who cares about a healthy FOSS ecosystem. And that force should come in the form of a certain quote from Linus whenever Nvidia push EGL streams. And if they don't want to support GBM then that's up to them. In that case, if you're an Nvidia user who wants to use Wayland then sucks to be you.

        Comment


        • #34
          Originally posted by pgoetz View Post
          Exactly. My users use scientific software that uses CUDA. Not having Nvidia cards is not an option. Unfortunately all this will do is force us to stay on X.
          So you will always have an option because X isn't going away anytime soon. No one forces you into Wayland, and I bet your scientific software won't support or require Wayland anyway anytime soon too.

          Comment


          • #35
            Originally posted by pgoetz View Post

            Exactly. My users use scientific software that uses CUDA.
            So nvidia has got you/your users in a tight vendor-lock-in grip and you prefer to keep it that way ?

            I always thought best business practice is to avoid becoming dependant on one vendor.


            Comment


            • #36
              Originally posted by DanL View Post
              Whether it gets merged or not has nothing to do with ideological reasons.
              And that's exactly the problem. It should have a lot to do.

              Originally posted by DanL View Post
              Are you the same Lucio guy that subscribed to the mailing list to tell the devs how evil EGLStreams is? If so, please don't waste their time with that crap. They don't care. If you have something technical to say about the code, that's great, but don't try to start a holy war on the Xorg mailing list FFS.
              Here I am. I'm free to subscribe to the mailing list and they are free to ignore me if they want. You are free to waste your time reading my crap and so they are, it's not my choice, it's yours and theirs. If it's really only crap they will ignore me for sure and I won't start any war. I just wanted to make sure that my opinion reached who's in charge. But you are not ignoring me, so I assume it's not only crap to you.

              TBH, you seem scared by my crap.
              Last edited by lucrus; 11 April 2018, 12:43 PM.

              Comment


              • #37
                Originally posted by LoneVVolf View Post

                So nvidia has got you/your users in a tight vendor-lock-in grip and you prefer to keep it that way ?

                I always thought best business practice is to avoid becoming dependant on one vendor.
                Currently there are only two desktop GPU vendors which offer some degree of performance. So, making some software GBM-only or CUDA-only, both mean vendor locking.

                Comment


                • #38
                  Originally posted by Khrundel View Post
                  Currently there are only two desktop GPU vendors which offer some degree of performance. So, making some software GBM-only or CUDA-only, both mean vendor locking.
                  One tries hard to lock you in on purpose. The other doesn't. Also, in a few years we are probably going to see a third vendor enter the high end GPU market and they sure won't support Nvidia proprietary technology.

                  Comment


                  • #39
                    Originally posted by Khrundel View Post
                    Currently there are only two desktop GPU vendors which offer some degree of performance. So, making some software GBM-only or CUDA-only, both mean vendor locking.
                    FYI:
                    GBM is for display, CUDA is for computing, great apples and oranges comparison.

                    Were you afraid of talking about our lord and saviour OpenCL?

                    And even then, GBM is not proprietary and patented, any vendor can use it. CUDA is, and whoever tries to offer CUDA on their hardware will get sued to hell by NVIDIA.

                    Comment


                    • #40
                      Originally posted by kaprikawn View Post
                      I'm suggesting Nvidia should be forced to support GBM like everyone else who cares about a healthy FOSS ecosystem.
                      Spoken like a true commie thug.

                      Comment

                      Working...
                      X