Announcement

Collapse
No announcement yet.

The Most Comprehensive AMD Radeon Linux Graphics Comparison

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

  • #21
    I remember someone saying like "Michael we don't need your conclusions - we saw the graphs, we're not stupid, we can draw conclusions ourselves".
    Now it's like "Yo yo wherez duh conclusion?? I need conclusions!"
    You can't please anyone.

    Comment


    • #22
      i always read only conclusion on long texts, but when something interested is in text i read it also, this is too long.

      Comment


      • #23
        Originally posted by zeealpal View Post
        I dont see any ads with the Premium subscription... :/ Since early 2011
        Are you joking? Did you see http://phoronix.com/forums/showthrea...ng-advertising ?
        ## VGA ##
        AMD: X1950XTX, HD3870, HD5870
        Intel: GMA45, HD3000 (Core i5 2500K)

        Comment


        • #24
          Originally posted by cl333r View Post
          I remember someone saying like "Michael we don't need your conclusions - we saw the graphs, we're not stupid, we can draw conclusions ourselves".
          Now it's like "Yo yo wherez duh conclusion?? I need conclusions!"
          You can't please anyone.

          We can draw our own conclusions when it's a 2 or 3 page article with one card, e.g. a HD4850, testing Catalyst vs r600g. When it's a 38 page article (well, 1 for me, but I forgive all the non-premium users) and each page has about a megabyte worth of advertisements and several more megabytes of test result images, no one except premium users are going to take the time to look at all the graphs. So a conclusion would really help for such a long article.

          My (definitive, but biased) conclusion: if you're using r600g, anything more complex than Quake 3 or a compositing manager is going to choke at an unplayable framerate and/or crash your system. In other words: both new and old ASICs are supported as well (or as poorly) as they have ever been, for 2 - 3 years. The only recent improvements have been a few micro-optimizations, and bringing up the same (bad) level of support on new ASICs.

          It's sad that even I believe that the community effort around r600g (both AMD employees and volunteers) is the best thing going in the whole world for open source graphics right now, and that effort still is doing very little to achieve performance numbers that approach the capabilities of mid to high-end cards. The effort seems almost futile at times: even when we have significant developer manpower, and official corporate backing, and open documentation, a grab-bag of showstopping issues, from hidden/DRMed hardware (UVD), to software patents, to bottlenecks in the driver architecture that no one can quite figure out how to eliminate, continue to drag down the effort and make it all seem for naught.

          It's like watching Old Yeller die in slow motion: for the years that I've been watching, supporting, advocating and testing the open source graphics drivers, the world around the drivers seems to be accelerating at a pace that we can't keep up with, so despite all of our progress, things look more bleak every day. I want them to succeed, but the goal is a moving target, and it seems for every two steps we take forward, the goal moves three steps away.

          Comment


          • #25
            I mostly just read the graphs, very informative.

            Comment


            • #26
              With OpenGL 3 finally around the corner and increasingly timely support for new GPU generations, I don't feel that things are that bleak. We are catching up on 3d functionality, slowly but surely. For majority of native Linux 3d stuff, the open drivers are either fully sufficient right now, or at least an option with some caveats (mostly performance-related).

              What is more worrying is performance, powersaving and OpenCL. These things require time and know-how, and there are many things still left to do. These are also the areas that have changed little over the years. There were some respectable performance gains for r600g, but they are intermittent, and dynpm hasn't really moved anywhere. At least OpenCL is receiving some love, but even that is still quite a long way off.

              Comment


              • #27
                The 2900XT did surprisingly well, performing as good as a Radeon 4770...

                Why this is not the case in Windows, is a mystery. Most probably, the Linux devs don't cripple older hardware in the drivers like the Windows devs do.

                I'd love to see more 2900XT benchmarks.

                ...also, why did the 3850 perform so poorly?

                Comment


                • #28
                  I haven't seen much in new developments in the AMD graphics drivers lately. I wonder what they are working on? I know the Intel people are working on GLSL and AMD people are working on the next generation architecture, OpenCL, and video. What else is being worked on?

                  Comment


                  • #29
                    After reading most of the first page, I skipped quite a lot on my quest for knowledge. Disappointed I was, yet still satisfying (a little) was it. /yoda speak

                    I don't use adblock; I prefer to support the sites I visit when I can. However, I do disable flash on all sites without exception. It's unnecessary for secondary content to use so many resources and slow down my computer so much, and prevents me from fulfilling the purpose of my visit (or significantly slows me down).

                    Comment


                    • #30
                      Originally posted by mockingbird View Post
                      The 2900XT did surprisingly well, performing as good as a Radeon 4770...

                      Why this is not the case in Windows, is a mystery. Most probably, the Linux devs don't cripple older hardware in the drivers like the Windows devs do.

                      I'd love to see more 2900XT benchmarks.

                      ...also, why did the 3850 perform so poorly?
                      I think this is just about the 4000- and 3000-series performing below their full potential, and the 2900XT closer to its full potential. Not about the 2900XT suddenly having some secret powers unlocked

                      Comment

                      Working...
                      X