Announcement

Collapse
No announcement yet.

Rich Geldreich: A Bad Catalyst GL Driver Is Bad For Everyone

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

  • Rich Geldreich: A Bad Catalyst GL Driver Is Bad For Everyone

    Phoronix: Rich Geldreich: A Bad Catalyst GL Driver Is Bad For Everyone

    Rich Geldreich, the former Valve OpenGL developer that left the company and has been publicly expressing the poor OpenGL driver landscape, has another new post out today...

    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 thought people in the forums were more quick to judge the validity of the test programme itself as being very nVidia biased (especially since it's theres), therefore making a nVidia v Anybody else comparison moot.

    And then every article thereafter pointing this comparison out is also, therefore, moot.

    Or am I missing something?
    Hi

    Comment


    • #3
      Originally posted by stiiixy View Post
      I thought people in the forums were more quick to judge the validity of the test programme itself as being very nVidia biased (especially since it's theres), therefore making a nVidia v Anybody else comparison moot.

      And then every article thereafter pointing this comparison out is also, therefore, moot.

      Or am I missing something?
      yes, you are. NV extensions were only used in tests where only nvidia was tested. other tests only use ARB

      Comment


      • #4
        another clickbait ?

        Anyway Rich Geldreich is a nice name if you translate English/German the proper words ... It would make a good avatar for a paid poster :-)))))

        So basicaly the test tells us that NV has the best GL4.2+ implementation (or is 4.4 needed for all the ARB extensions?). Now that Intel is well behind in GL4+ and makes up the most of the GPUs used, the test is not relevant to anything byt high-end gamers.

        Also plese can this be run on the last and current-gen consoles just for comparison ? Oh can't, last-gent does not support GL4+ ....

        Basically we are talking about evaluating the best implementation of the latest standard in an industry area that does not care about the latest standard (the largest player does not support it).

        I do however agree with the conclusion that the AMD results are a disaster for all of us.

        Comment


        • #5
          What nVidia says with this test is: "Look brothers, i am fast here" . That is OK, but that means nothing for the other implementation . Rich is right, no OpenGL driver is as fast as other driver in the same place and that is true. Devs just needs reliable API which fast & reliable on most vendor implementation, common thing as they can target majority of people with no problem, etc... so they want just that to be easy and easy supportible for them, with not so much vendor diversity included .

          With this test and nVidia's ""Look brothers, i am fast here" and "Do you see that, you can do it as well" , it is more and more understandible to me why AMD want to do Mantle on their own .

          Graphics Mafia, OS Mafia... yeah Richie boy you are right . How many drivers Linux has..."5000 drivers maybe" why that many for one normal computing needs , maybe we just need dictator like Linus for the graphics - diversity is normal thing .

          Comment


          • #6
            Wow, the comments here are so silly, it is almost sad.

            Some Nvidia dude writes some demo code for GDC14 Approaching Zero Driver Overhead presentation, to show what is actually possible with modern OpenGL, talking about x15 speedups in some cases.

            Some other dude from Aspyr Media makes this demo run on Linux, and tests how well it performs with other GPUs. He finds that AZOD only really works with Nvidia hardware, because other vendors don't give a damn about performance (or anything else) it seems.

            So it must be Nvidias fault. What the fsck is wrong with you?

            Comment


            • #7
              Originally posted by log0 View Post
              Wow, the comments here are so silly, it is almost sad.

              Some Nvidia dude wdrites some demo code for GDC14 Approaching Zero Driver Overhead presentation, to show what is actually possible with modern OpenGL using AMD spearheaded extensions, talking about x15 speedups in some cases.

              Some other dude from Aspyr Media makes this demo run on Linux, and tests how well it performs with other GPUs. He finds that AZOD only really works with Nvidia hardware, because other vendors don't give a damn about performance (or anything else) it seems.

              So it must be Nvidias fault. What the fsck is wrong with you?
              Here, fixed that for You.

              Comment


              • #8
                Originally posted by przemoli View Post
                Here, fixed that for You.
                Who cares if AMD spearheaded them? Their spear is blunt, and nvidia's is sharp. Everyone is criticizing OpenGL, and the conclusion is: modern OpenGL works *BECAUSE* nvidia exists. Others may be "working towards" or may be "spearheading stuff" but if you want decent OpenGL in 2014, nvidia is your only option.

                I'd like it very much if it wasn't like this. It's hard to find nice small laptops with nvidia chips. I'd like to be able to game a bit on my laptop. But unfortunately, AMD is not the answer.

                Comment


                • #9
                  NVIDIA only providing binary drivers is bad for everyone too.

                  Comment


                  • #10
                    Please drop Catalyst driver for Linux!!! And send all workers to r300, r600, radeonsi driver! Need petitions?

                    Comment

                    Working...
                    X