Announcement

Collapse
No announcement yet.

Mesa 7.11 Release Candidate 1 Arrives

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

  • Mesa 7.11 Release Candidate 1 Arrives

    Phoronix: Mesa 7.11 Release Candidate 1 Arrives

    While this is non-news for many Phoronix readers that are Mesa Git followers and already onto tracking Mesa 7.12-devel after using Mesa 7.11-devel for months already, for everyone else, the first release candidate of Mesa 7.11 is now available...

    http://www.phoronix.com/vr.php?view=OTY1MA

  • #2
    This should be a great release, a lot has been done since 7.10.

    Comment


    • #3
      Completely working power management and I'd switch to the open source stack in a heartbeat. Video acceleration would be the only thing I would really miss then, but my desktop's CPU is fast enough anyway (AMD Phenom II X4 940) and I don't watch HD videos on my notebook (AMD Fusion E-350).

      Comment


      • #4
        It will be great to see some optimizations for r300g, because even Open Arena is hardly playable.

        Comment


        • #5
          Originally posted by kraftman View Post
          It will be great to see some optimizations for r300g, because even Open Arena is hardly playable.
          r300g is comparable to Catalyst in performance now.

          Has been for quite a while.

          Comment


          • #6
            Originally posted by pingufunkybeat View Post
            r300g is comparable to Catalyst in performance now.

            Has been for quite a while.
            Well, in the last benchmark Michael did, r300g had ~70% the speed of the blob.

            Comment


            • #7
              Originally posted by kraftman View Post
              It will be great to see some optimizations for r300g, because even Open Arena is hardly playable.
              There are indeed new r300g optimizations in 7.11. From the top of my head:
              - faster vertex uploads (also on r600g)
              - new optimized winsys that completely bypasses (doesn't use) libdrm_radeon
              - new fragment shader register allocator

              Comment


              • #8
                What are the minimum requirements for drm,libdrm,and co ?

                Comment


                • #9
                  Originally posted by pingufunkybeat View Post
                  r300g is comparable to Catalyst in performance now.

                  Has been for quite a while.
                  Then I wonder why it sucks so much on my box?

                  Comment


                  • #10
                    Originally posted by marek View Post
                    There are indeed new r300g optimizations in 7.11. From the top of my head:
                    - faster vertex uploads (also on r600g)
                    - new optimized winsys that completely bypasses (doesn't use) libdrm_radeon
                    - new fragment shader register allocator
                    Thanks. I will check this out when 7.11 will get into Arch stable repo.

                    Comment


                    • #11
                      Originally posted by fatih View Post
                      What are the minimum requirements for drm,libdrm,and co ?
                      libdrm 2.4.24 is required.
                      I believe it is supposed to work with any drm version, but some features and optimizations will be disabled unless you are using drm from at least linux 2.6.39.

                      Comment


                      • #12
                        Originally posted by kraftman View Post
                        Then I wonder why it sucks so much on my box?
                        Is your card AGP by any chance?

                        Comment


                        • #13
                          Setting clock rates manually

                          Originally posted by kraftman View Post
                          Then I wonder why it sucks so much on my box?
                          That's what I had been wondering with my evergreen card for months until I found out the clock speeds are not set correctly on my system.

                          If you haven't already make sure the clocks have been set to something reasonable:
                          Code:
                          cat /sys/kernel/debug/dri/0/radeon_pm_info
                          I suspect power management is currently the biggest problem Radeon users have with Mesa.

                          I have slowly been doing something to improve the situation because my laptop has several useful clock modes that are not used at all even if I manually set "low", "mid", or "high"! Those almost randomly picked clock modes or the 2 that are used by dynpm are just not good enough in my case.

                          I wonder if it would be acceptable for mainline inclusion to make a kernel patch and an user mode program that could be used to manually select sane clock modes for fixed profiles and dynpm? Perhaps it could even be part of driconf?

                          Comment


                          • #14
                            Originally posted by kraftman View Post
                            Then I wonder why it sucks so much on my box?
                            Isn't your hardware slow?

                            Comment


                            • #15
                              Originally posted by curaga View Post
                              Is your card AGP by any chance?
                              No, it's a PCI E. X1600XT. I will have to check some other games, too. Maybe there's a problem with OA only.

                              Comment

                              Working...
                              X