Announcement

Collapse
No announcement yet.

An Optimized Open-Source Driver Tries To Compete With AMD Catalyst

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

  • #16
    Taxi but the decoding runs on the cpu only mpeg2/1 is currently hardware accelerated.

    Comment


    • #17
      Yes, it clearly says it's using ffmpeg's h.264 codec implementation, only the display part is done by the vdpau implementation, and that part works just as well with xvideo.

      Comment


      • #18
        Originally posted by TAXI View Post
        Well,VDPAU via 3d shaders seems to work
        The last time I checked, it only supported mpeg 1 and 2.

        Comment


        • #19
          it seems the biggest bottleneck roght now it is on the lack of shader packetizer and threading but the fixed functions seem pretty competitive right now wtih catalyst

          Comment


          • #20
            Originally posted by ChrisXY View Post
            Yes, but if it has 60+ FPS for the applications you need competing (optimizing remaining "bottlenecks") is kind of pointless. That's competing enaugh for me.
            I wouldn't call it "remaining bottlenecks" as long as the binary blob is still 2-40 times faster.

            Originally posted by ChrisXY View Post
            It would be definitely very good if it was supported for more video codecs, but what CPU does your HTPC have and what quality of videos do you want to play?
            My HTPC has an AMD X2 4050e with an onboard AMD HD 3???. It is capable of playing Full-HD videos with the binary blob. However I'm currently using a NVidia GTX430, as XBMC or the Catalyst driver has/had problems with vba (got display errors with some hd-videos) and suspend-to-ram. I'm pretty much done with AMD, my next system will be an Intel Atom with a Nvidia chipset. Nvidia is just the only problem-free solution on Linux.

            Comment


            • #21
              Two comments about the graphs:

              1. At some point, the order changes, Catalyst moves position to between the two mesa results, which is confusing.

              2. Why not have vertical indicators at 30, 60, 120 ... fps instead of automatically determined by the results? From my point of view 30 is ok, 60 is the maximum for most displays, and more than 120 is uninteresting in practice.

              Also, why not put the Xonotic results directly after Nexuiz if you refer to them anyway?

              Originally posted by enrico.tagliavini View Post
              I wonder if enabling S3TC might improve the performance of the radeon driver. Well this can be an idea for the next article
              It does, in applications that enable texture compression the fps are higher when libtxc-dxtn is installed.

              Comment


              • #22
                Great article Michael!

                Its quite sad to see how little these tweaks boosted performance on the r600 (46xx and 48xx) series. Maybe, just like AMD, all the resources are being put into the newer generation cards.

                Comment


                • #23
                  Originally posted by gururise View Post
                  Great article Michael!

                  Its quite sad to see how little these tweaks boosted performance on the r600 (46xx and 48xx) series. Maybe, just like AMD, all the resources are being put into the newer generation cards.
                  hey i do have the same problem here: http://phoronix.com/forums/showthrea...n-driver/page2

                  hd4770 13fps hd6770=4xfps

                  but its 6xx shaders vs 800 shaders and 750mhz vs 800mhz

                  the hd6770 should not be 4 times faster maybe 1/3 faster.

                  Comment


                  • #24
                    Thanks for this benchmark. The problem is that I already knew the results. The OSS driver performs bad and misses plenty of features. Catalyst driver performs good for 3D but is a piece of crap under all the other aspects.

                    Under Linux you can never have 1 thing that works good for all scenarios. You have to choose between 2 or more things depending on your needs and will never be 100% satisfied because one day, soon or later, you will surely need that missing feature that only the other choice has.

                    Comment


                    • #25
                      So it's pretty much the same old story: despite all the effort spent tweaking the OSS drivers, they're still running at a tiny fraction of the speed of the closed-source binary blobs. Other than supporting older chipsets that Catalyst has put out to pasture, it's difficult to put a positive spin on this.

                      Comment


                      • #26
                        I did hear that the nouveau driver performs much better in relation to the blob.

                        I don't really understand why the open ati driver is still that slow after several years and with ATI itself as the major contributor.

                        Comment


                        • #27
                          They needed to catch up on multiple generations - there are now three open AMD drivers - r300g, r600g and radeonsi.

                          Then there's all the work they have done that wasn't been able to be released.

                          Comment


                          • #28
                            Originally posted by archibald View Post
                            They needed to catch up on multiple generations - there are now three open AMD drivers - r300g, r600g and radeonsi.

                            Then there's all the work they have done that wasn't been able to be released.
                            So? The main problem seems to me to be that there's about 3 developers, 5 managers and 25 lawyers working on this.

                            Comment


                            • #29
                              Originally posted by barkas View Post
                              So? The main problem seems to me to be that there's about 3 developers, 5 managers and 25 lawyers working on this.
                              Actually 4 developers, a fraction of a manager and no lawyers but don't let me slow you down

                              IIRC the main differences between radeon and nouveau in terms of performance progress are :

                              1. the NVidia hardware equivalent of two performance improving features (what we call tiling and HyperZ) are implemented in such a way that "turning them on" requires less driver work than doing the same on ATI/AMD hardware. Jerome and others been making great progress here recently.

                              2. the VLIW shader core probably requires relatively more shader compiler effort to get to comparable levels of HW utilization. I'm not sure this is an issue but we should know more once the SI open source driver starts to run beneficially.
                              Last edited by bridgman; 06-13-2012, 01:55 PM.

                              Comment


                              • #30
                                Mr. Bridgman, please stop letting facts and reason get in the way of complaints and conspiracies ;-)

                                Comment

                                Working...
                                X