Announcement

Collapse
No announcement yet.

OpenGL 3.1 Not Likely In Mesa Until 2013

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

  • OpenGL 3.1 Not Likely In Mesa Until 2013

    Phoronix: OpenGL 3.1 Not Likely In Mesa Until 2013

    The release plans for Mesa 8.1 and Mesa 8.2 have been proposed. Unfortunately if you were hoping to see OpenGL 3.1 compliance in this open-source graphics driver library this summer, it looks like that won't come until 2013 and support for newer OpenGL specifications are even further out...

    http://www.phoronix.com/vr.php?view=MTA5Njc

  • #2
    What a pity!

    Comment


    • #3
      Not as bad as it looks

      It's true Mesa is not at the same level as the OpenGL standard. However, if you look at the state, many features for OpenGL 3.X are already done. Since they are incremental updates, they should be easier to achieve than going to OpenGL 4.X. Even some 4.X features are already done.

      The real question is : is there so many applications using OpenGL 4.X out there not able to use 3.X?

      It is important to keep the pace, but it's also important to improve performances. That's what I'm the most worried about.

      Comment


      • #4
        Yawn.

        That's rather pessimistic, I'd expected GL3.1,2,3 to come rather quickly now, and my geometry and tessellation shader support in gallium/nvc0's been sitting there bored for half a year already ... guess everyone's busy fixing bugs.
        I should be, too; plenty of bugs to squash. And optimizations to be done. No time to work on core mesa.

        Comment


        • #5
          I can't understand it. Mesa has so great support from Intel, Red Hat and others big giants of IT, and they couldn't implement 3 years old specs, where nVidia and AMD could do it in so short time. What is the main reason? No so much ppl or money for work?

          Comment


          • #6
            Originally posted by olbi View Post
            I can't understand it. Mesa has so great support from Intel, Red Hat and others big giants of IT, and they couldn't implement 3 years old specs, where nVidia and AMD could do it in so short time. What is the main reason? No so much ppl or money for work?
            Good Question. Some of the big vendors (AMD or Intel, are you listening?) should help out in this area. It would buy a lot of community goodwill.

            Comment


            • #7
              Originally posted by olbi View Post
              I can't understand it. Mesa has so great support from Intel, Red Hat and others big giants of IT, and they couldn't implement 3 years old specs, where nVidia and AMD could do it in so short time. What is the main reason? No so much ppl or money for work?
              FOSS projects do not have a lot of developers. The idea that there are millions of people willing to contribute to FOSS may be true in the broadest sense, but it's certainly not true when you narrow things down to the actually important, useful, and/or relevant FOSS projects.

              If you add up all of the people actively working on Mesa, GNOME, the desktop-related bits of the kernel, X11/Wayland, glibc and the rest of the core GNU system, and so on, you will still not have as many people as Microsoft has test engineers. The last goofy little 6-month 2D game project I worked on had a larger team than all of Mesa has.

              Companies that actually make large profits can hire tons of $120k+/year developers. FOSS gets the scraps; a handful of paid engineers that the low-profit FOSS companies can afford, a number of hobbyists working part-time after work, and a shitload of whiny self-entitled users who bitch and moan constantly about a free OS while contributing absolutely nothing.

              Developers with the necessary many years of experience and who can work 40+ hours/week do not often pop out of nowhere and start writing code for FOSS projects instead of working full-time at a proprietary software house. Somebody somewhere needs to be tossing them six-figures plus benefits every year. The most well-off FOSS company may be Red Hat -- they hit $1B yearly revenue (note: that's not profit) last year, or an average of $250M per quarter. AMD has a quarterly revenue of around $1.5B while Intel gets around $2.7B per quarter. Google managed $10B last quarter. Microsoft, on the other hand, pulls in almost $20B in quarterly revenue while Apple pulled in just shy of $40B last quarter. Needless to say, there's a shitload more highly experienced and talented developers at the big proprietary companies than there are at the little FOSS shops.

              Comment


              • #8
                All that said, Now would be a good time for the core Mesa team to begin approaching companies like Valve, AMD, and EA.

                F

                Comment


                • #9
                  Originally posted by russofris View Post
                  All that said, Now would be a good time for the core Mesa team to begin approaching companies like Valve, AMD, and EA.

                  F
                  Valve will hopefully be doing some things... Well, they have already been doing some things to the point that the Intel guys already promised me at least a beer at XDS in September for getting some things arranged with them
                  Michael Larabel
                  http://www.michaellarabel.com/

                  Comment


                  • #10
                    mesa desperately needs someone with deep pockets to back it up

                    Comment


                    • #11
                      Originally posted by 89c51 View Post
                      mesa desperately needs someone with deep pockets to back it up
                      Why? Do you really need OGL 4.2 now?

                      Comment


                      • #12
                        Originally posted by Nedanfor View Post
                        Why? Do you really need OGL 4.2 now?
                        probably not (though it'd be good to have) but i thing the FOSS stack needs Video Acceleration, OpenCL, bug fixing, and all the other things that are left behind due to lack of manpower.

                        Comment


                        • #13
                          I did a piglit run on fglrx the other day,

                          http://people.freedesktop.org/~airli...it/fglrx-r600/

                          so we can say we are a lot better in terms of actually passing tests :-)

                          That is a piglit run from two AMD drivers a year apart.

                          but really its only Intel working a lot on core mesa at the moment, with others expending time as jobs allow. Its not as is Red Hat can ship GL3.0 drivers anyways so working on them isn't a great spend of our time. and working on GL doesn't get you CL or video decode or anything. Also it not as if Red Hat can ship video decoders, so again no reason for us to invest heavily in them.

                          Perhaps some of the distros that do ignore patents could invest more in these.

                          Comment


                          • #14
                            Originally posted by Nedanfor View Post
                            Why? Do you really need OGL 4.2 now?
                            Why not? It would nice being able to play OilRush at the highest quality... but again performances are the main issue. Also we need at least 3.2 to get the best from wine.
                            ## VGA ##
                            AMD: X1950XTX, HD3870, HD5870
                            Intel: GMA45, HD3000 (Core i5 2500K)

                            Comment


                            • #15
                              Originally posted by darkbasic View Post
                              Why not? It would nice being able to play OilRush at the highest quality... but again performances are the main issue. Also we need at least 3.2 to get the best from wine.
                              Support in Mesa != Support in drivers. At the same time we need power management, performance, OCL, hw decoding, etc. etc. so I guess that only a little minority of us would really need OGL 4.x over all these feature.

                              PS. Sandy Bridge and Ivy Bridge only support OGL 3.x so I think that Intel isn't directly interested in supporting OGL 4.x.

                              Comment

                              Working...
                              X