Announcement

Collapse
No announcement yet.

May 2011: Gallium3D vs. Classic Mesa vs. Catalyst

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

  • #21
    Originally posted by elanthis View Post
    That would be just about everything. Literally, I don't think Heaven can run on the Mesa drivers at all, period.
    http://phoronix.com/forums/showthrea...eaven-on-r300g

    I'm not sure how it is on r600g, but I remember reading that it was working too. Not all GL3 stuff works, obviously.

    Comment


    • #22
      I'll be glad when

      I'll be glad when Texture Compression, etc are enabled by default.

      But until then these little posts about performance:
      a. Are always subject to debate by the people who have enabled the hidden tweaks the rest of the userbase has not a single clue about. Which leads me to suspect one of us needs to go to Ubuntu Brainstorm and add a proposal to have the tweaks enabled by default. Texture compression, odds and ends, filesystem modifiers

      b. Should actually be run from fluxbox so that limited resources are used and we see the raw results.

      c. I'm actually very happy running LTS copies of the operating system with one exception. 8.04 / 8.10 were the prime for my laptop in terms of raw performance. Although I'm neglecting the I915 driver. Ubuntu 11.04 is the prime for that chipset. See you can't have your cake and eat it too.

      d. Distributions need a build system that incorporates these performance exceptions by compiling mismatches in terms of New Xorg with Old userbase. New userbase with Old Xorg. Puppy is about the only distribution I've read about who does this.

      But who am I, what do I know of such things. I'm but a sentinel watching the progress of slime sliding down a dirt hill.

      Fin

      Comment


      • #23
        Fluxbox is getting compositing this gsoc, you can't then say it brings the best performance

        Comment


        • #24
          Originally posted by bongmaster2
          heaven runs fine withoutany corruptions or errors. engine even reports optional rxtensions available.looks exact like on windows. its just a bit slow.
          Yeah just tried it, except tessellation everything seems to render fine. But I only get about ~2fps on my HD 5750.

          Comment


          • #25
            Originally posted by mangobrain View Post
            I have a 6950, one of the last remaining unsupported cards, it would seem. Well, unsupported in terms of 3D acceleration, anyway - KMS works, but no direct rendering. I really want to like GNOME Shell, but can't give it a fair try, because if I use the open-source drivers or llvmpipe it goes into fallback mode, and if I use the proprietary drivers I get creeping texture corruption until the session eventually becomes unusable.

            Wake me up when Gallium3D support arrives for Cayman....
            I'm in the same boat. I have a 6970 and usually just switch to my integrated 4200 chipset for Linux to avoid using the VESA driver.

            Comment


            • #26
              Originally posted by pingufunkybeat View Post
              http://phoronix.com/forums/showthrea...eaven-on-r300g

              I'm not sure how it is on r600g, but I remember reading that it was working too. Not all GL3 stuff works, obviously.
              Yes, Heaven works on r600g too. Unigine Tropics works as well, which is a pretty cool benchmark too.

              In fact, any Mesa driver exposing ARB_texture_float is Unigine-ready.

              Comment


              • #27
                Does anyone play Quake Live with the radeon driver? It's about the only game I play that is a "must work". I don't own an ATI card so I can't test it myself.

                Comment


                • #28
                  just another stupid question/idea of mine:

                  how complicated/useful/performant/whatever would it be to have some layer for computation: What i mean is something that hides all computational cores and then distinguishes which/how many cores to use for a certain program call.

                  So every graphics would be run in software renderer using opencl on the gpu for instance (or on multiple GPUs) dividing work according to the cores (high performant GPUs for 3D works and integrated ones for 2D and say CPU for C-Ray calculations)

                  or is the idea just stupid as programs themselves have to decide where they want to be run on and into how many threads they can divide?

                  hasnt there been some article recently on CUDA for kernel uses?

                  Comment


                  • #29
                    Originally posted by AnonymousCoward View Post
                    Does anyone play Quake Live with the radeon driver? It's about the only game I play that is a "must work". I don't own an ATI card so I can't test it myself.
                    I just tried nouveau with my nvidia card. All of the graphical features I use work with Quake Live, and FPS stays around 125fps. However, there are pauses for half a second here and there. Doesn't happen with the binary drivers. Also, games run under wine are far too slow, even old games like Half-life 1.

                    Pretty impressive nonetheless, but not quite there yet. KMS is nice, though.

                    Comment

                    Working...
                    X