Announcement

Collapse
No announcement yet.

Benchmarks Of AMD's Newest Gallium3D Driver

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

  • Once the Gallium3d drivers are near par with the proprietary drivers the way bridgman mentioned, maybe the next step would be to move the open source drivers from the Gallium3d code back to more hardware specific code. Maybe that might increase odds of a "convergence" of the open source drivers and Catalyst.

    Comment


    • The problem is that there will always be new hardware and new graphics standards to support, and having a small hardware-specific layer reduces the effort required to do that (ie gives the development community a chance of keeping up). If the driver requirements ever really became static (or even static-ish) then optimizing for performance at the expense of extensibility might make sense.

      The current Gallium3D API level seems like a pretty reasonable compromise between potential performance and required development effort.
      Test signature

      Comment


      • Originally posted by chrisr
        Can you remember what was done to fix this "glitching" issue, please? E.g. was it in the kernel, the driver, or Mesa? Because I was noticing some glitchy behaviour with the 2.6.36.1 kernel and xorg-drv-ati from git recently (HD4890).
        Originally posted by bridgman View Post
        Pretty sure it was in the kernel. My guess is that you would need 2.6.37 to get it.
        I still get flickering with dynpm on R300 hardware with 2.6.37 and everything else from git. Maybe less, yes, but still annoyingly enough as to not activate it.

        Comment


        • Actually, after reading bridgman's post, I enabled dynamic pm again (with 2.6.36), and I don't have flickers or strange slowdowns anymore. It used to be quite bad. This is on rv710.

          There's almost certainly room for improvement still, but it seems like dynpm becoming a viable option. Perhaps the hardware + CPU combination plays a large role?

          Comment


          • Well, something odd I noticed is that now I don't get much flickering when doing regular GUI stuff, like browsing the web or so. I could live with that. But when stuff is displayed on a terminal it's a lot worse. So, for instance, doing "top -d 0.1" (just to get something going on) the screen flickers every 2-3 seconds or so. That means that sometimes I'll open a terminal and will get annoyed with those glitches when I start typing something.

            Comment


            • This is a funny one. Actually, the screen flickering I mentioned when doing stuff on the terminal only happens when using Yakuake, which is the terminal I use for mundane things. When using urxvt or xterm everything is normal.

              ???

              Comment


              • Originally posted by yotambien View Post
                I still get flickering with dynpm on R300 hardware with 2.6.37 and everything else from git. Maybe less, yes, but still annoyingly enough as to not activate it.
                Same here with my hd2600, and the power management doesn't even seem to bring temps down much yet, so right now I have to stick with fglrx so I don't burn my house down.

                Comment


                • bwat47, have you tried the profiles ? The "mid" profile seems to give a pretty good compromise between power and performance.
                  Test signature

                  Comment


                  • I did try the profiles but they made stuff like scrolling in firefox laggy for me, but granted this was a while ago before a lot of the recent optimizations and scrolling has gotten a lot better so I should probably try it again. Dynamic would definitely be ideal though.

                    Comment


                    • Profiles is the only viable option for me, since dynpm still causes my laptop (RS480) to freeze.

                      Comment

                      Working...
                      X