Announcement

Collapse
No announcement yet.

Core Mesa Is Now Just One Step Away From OpenGL 4.3 Compliance

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

  • Core Mesa Is Now Just One Step Away From OpenGL 4.3 Compliance

    Phoronix: Core Mesa Is Now Just One Step Away From OpenGL 4.3 Compliance

    Mesa is now very close to OpenGL 4.3 compliance thanks to a massive patch series that was posted today for review...

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    For Intel, I wonder why its not possible to refactor their windows OpenGL driver code to linux????

    Comment


    • #3
      Originally posted by swoorup View Post
      For Intel, I wonder why its not possible to refactor their windows OpenGL driver code to linux????
      Lawyers

      Comment


      • #4
        FWIW, Igalia has also been working on FP64 support for Intel, based on the work I did over the summer. Based on the bugzilla entry it seems like they've made a lot of progress on it, although it probably won't be ready in time for 11.2.

        Comment


        • #5
          We put so much emphasis on catching up with the OpenGL spec, it makes one wonder just how much we care about the performance sometimes. We see that RadeonSI is slower than Catalyst in many cases, and end up chocking that up to the driver as a whole. But who's to say it isn't because of any hackish implementations of OpenGL extensions that somebody threw in with a "FIXME" comment and never looked back at?

          I don't watch the MESA git, admittedly, but most of the releases seem to be bugfixes or new features, with 1/8 or so mentioning performance.

          Comment


          • #6
            Originally posted by Daktyl198 View Post
            We put so much emphasis on catching up with the OpenGL spec, it makes one wonder just how much we care about the performance sometimes. We see that RadeonSI is slower than Catalyst in many cases, and end up chocking that up to the driver as a whole. But who's to say it isn't because of any hackish implementations of OpenGL extensions that somebody threw in with a "FIXME" comment and never looked back at?

            I don't watch the MESA git, admittedly, but most of the releases seem to be bugfixes or new features, with 1/8 or so mentioning performance.
            They might want a complete working implementation that behaves according to the specification as much as possible before they start looking into serious optimization.

            Comment


            • #7
              Originally posted by Daktyl198 View Post
              We put so much emphasis on catching up with the OpenGL spec, it makes one wonder just how much we care about the performance sometimes. We see that RadeonSI is slower than Catalyst in many cases, and end up chocking that up to the driver as a whole. But who's to say it isn't because of any hackish implementations of OpenGL extensions that somebody threw in with a "FIXME" comment and never looked back at?

              I don't watch the MESA git, admittedly, but most of the releases seem to be bugfixes or new features, with 1/8 or so mentioning performance.
              First you make the software work and be "complete", then you worry about making it fast. Very useful to have a 'rollback point' where you know you had working code, just incase your optimizations screw it all to hell.
              All opinions are my own not those of my employer if you know who they are.

              Comment


              • #8
                Originally posted by Daktyl198 View Post
                We put so much emphasis on catching up with the OpenGL spec, it makes one wonder just how much we care about the performance sometimes. We see that RadeonSI is slower than Catalyst in many cases, and end up chocking that up to the driver as a whole. But who's to say it isn't because of any hackish implementations of OpenGL extensions that somebody threw in with a "FIXME" comment and never looked back at?

                I don't watch the MESA git, admittedly, but most of the releases seem to be bugfixes or new features, with 1/8 or so mentioning performance.
                Software Development works as follows: Make it Work -> Make it Right -> Make it Fast. And quite a lot of analysis is required for step 3 so you can't just magically jump there.

                Comment


                • #9
                  Originally posted by cwabbott View Post
                  FWIW, Igalia has also been working on FP64 support for Intel, based on the work I did over the summer. Based on the bugzilla entry it seems like they've made a lot of progress on it, although it probably won't be ready in time for 11.2.
                  Sweet! I was wondering if anyone was working on finishing that code. I do follow the mailing list, but I didn't think to check bugzilla. Opengl 4+ on Intel, here we come!

                  Comment


                  • #10
                    if there is much effort we dont know, maybe from the 5 people that are on it, its really much effort, compared to proprietary drivers its maybe 1/100 effort. Maybe the factor is only 1/20 or so dont know the exact numbers but "much effort" is something relativ...

                    I dont want to bitch but valve could have interest in good free intel and amd drivers because they dont want to make their customers have to use fglrx, so they could pay 20 fulltime people, I know they do SOME stuff, thats always better than nothing, but maybe I am wrong what I read so far sounded more like 1-5 full-time developers on that front if not less.

                    We are lucky that gaming matters less and less because games get more and more uninteresting, and all games that make fun dont need highend hardware or high opengl levels, because basicly this games are pure reality-demos... but k thats maybe me talking subvective others have other priorities, I just see in forums many frustrated gamers.

                    Heck I even am not sooo excited about wayland anymore, because my "tiling" windowmanager dont work with em, and even if somebody would start at some day developing it it would take at least from then on a year to get it so far. (talking about exwm).

                    But ok thats a bit off topic. And no not the past was all better, but the todays xserver is good enough for most tasks, I am just sick on waiting many years for this marketing stuff to become true and really usable.

                    You have to take your tools into your hand, and there is no software that is good for everybody, whats good for most people is most likely garbage if you want productivity.

                    so sorry I start mumbling its late, and stop here for now

                    Comment

                    Working...
                    X