Announcement

Collapse
No announcement yet.

AMD Catalyst 14.12 Omega Driver Brings Mixed Results For Linux Users

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

  • #21
    I saw the 290 improved a lot, how does it compare to Windows catalyst now?

    Comment


    • #22
      Testing hardware fillrate NOT driver performance

      By testing at ridiculous resolution where enthusiast class cards struggle to get 40FPS, all you're measuring is the cards fill rate capacity. If you want to see the effect of driver optimization, you need to run at resolutions that aren't fill-rate limited. Instead of 2560x1600 for Ungine Valley, why not something more reasonable like 1080p so mid-range cards can get more over 10FPS?

      Comment


      • #23
        Originally posted by Adarion View Post
        2nd vote for APU tests here. Kabini, Kaveri, whatever but APUs.
        APUs! APUs! And lots of them. Yeah, lots.

        And I will be waiting for new Carrizo APUs next year with Excavator cores.

        Comment


        • #24
          again

          new driver fail again, year after year and they can provide a good driver

          Comment


          • #25
            I tested a radeon hd 6670 and it is working much better with serious sam 3 (44-55fps), previously I was getting lock downs, but now is working properly at medium settings.

            Comment


            • #26
              Hopefully the R9 285 at least got an upgrade. Since that one has the future tech. Did we at least get some new cool stuff? Like FreeSync?

              Comment


              • #27
                APU

                Originally posted by GraysonPeddie View Post
                APUs! APUs! And lots of them. Yeah, lots. ...
                Additional vote for APU tests.

                Comment


                • #28
                  Originally posted by alexThunder View Post
                  Would you mind telling me/us about your experience with the open source driver on that card? I have an R9 290x and didn't try that driver out so much. Only yesterday i quickly checked out Dota 2, but was pretty much garbage. I still wonder how other games perform. I'd like to switch to the open source drivers soon, but for now performance for (more) modern games doesn't seem to be sufficient.
                  I'd say (compared to my old HD5750) most games I had could play at a higher detail and full frame rate, but there are a few here and there that still don't play at full frame rate, or, have played significantly worse due to specific detail settings. The R9 290(X) performance has increased dramatically after every driver release, so it wouldn't surprise me if these problems go away soon.

                  Ever since switching back to Catalyst, I'm getting that deaded "GLX Context error" when running Steam, and my usual fix of removing stuff like libstdc++ or libgcc don't work. Valve seriously needs to fix that problem - it cripples games that otherwise wouldn't have an issue.

                  Comment


                  • #29
                    Originally posted by slacka View Post
                    By testing at ridiculous resolution where enthusiast class cards struggle to get 40FPS, all you're measuring is the cards fill rate capacity. If you want to see the effect of driver optimization, you need to run at resolutions that aren't fill-rate limited. Instead of 2560x1600 for Ungine Valley, why not something more reasonable like 1080p so mid-range cards can get more over 10FPS?
                    Presuming the drivers were perfect, sure, the card speed would be the only limiting factor. But as the tests show difference at this resolution they are very useful to compare driver performance. Testing at 1080p would have its merits as well, but that's a different thing.

                    Comment

                    Working...
                    X