Announcement

Collapse
No announcement yet.

Khronos Releases OpenGL 3.3 & OpenGL 4.0

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

  • #21
    Originally posted by cl333r View Post
    OpenGL was slowly developed and overconfident back then and Microsoft had already expertise killing such big and slow beasts (like IBM for example).
    So, I give M$ credit for being smart and fast back then, but they mostly won because their counterpart was too lazy to compete and to ignorant, not because M$ is a house of geniuses (it's not).
    Now, if OpenGL goes on this way people will see (once again) that M$ is in fact very vulnerable and the world will quickly drop DirectX in favor of OpenGL once Apple's market share goes so high that game developers can't afford to ignore it. M$ is just a beast with a history of killing lazy and weak competitors and when they have to face real competition they show they can't win (for example, Google search is strong as never before in the world).
    Historically, it is safe to say that OpenGL failed because of inter-IHV tensions. 3dLabs proposed a number of (then) awesome updates for OpenGL 2.0. Unfortunately, IHVs didn't want to drop backwards compatibility (ring a bell?) and the 2.0 spec was cut down and delayed until only a shadow remained.

    Microsoft pursued a much more aggressive approach, rewriting their API until they got it right. By the time D3D7 came out, the majority of the game developers had already switched. By the time of D3D9, OpenGL was already decomposing. It wasn't until 5 years later (and 2 years behind D3D10) that the new ARB managed to resurrect the spec with v3.0 and it wasn't until 3.2 that developers started taking OpenGL seriously again. 3.3 and 4.0 continue that legacy (only 6 months behind D3D11 this time) and things are starting to look up.

    Right now, we have two major players adopting the new specs (Nvidia and Ati), one major player dragging their ass (Apple) and the final player wishing OpenGL didn't exist at all (Intel). Hopefully, Apple will realize the synergy between OpenCL and OpenGL and start supporting the latest specs sooner (it's been two years and they still don't support OpenGL 3.0). On the other hand, Intel is hopeless and only (barely) supports D3D. Their OpenGL drivers are beyond hopeless.

    One can only wonder how the 3d landscape would look now, had the ARB followed 3dLab's original suggestions back in 2001. AAA gaming on Linux might not have been such a distant dream then.

    Comment


    • #22
      Slow down this OpenGL releases, please!
      New OpenGL/DirectX specifications -> new requirements in games -> new graphic hardware -> new PCs... -> more time to implement specs in drivers... & more possible bugs with implementation of many old and new specs... :/

      Comment


      • #23
        (a) The ARB is composed by members of the various IHVs.
        (b) IHVs make money from selling new hardware.
        (c) Developers need new hardware to develop better software.
        (d) OSS driver developers cannot keep up with the breakneck speed of new hardware releases.

        (a) + (b) + (c) > (d), I'm afraid.

        Comment


        • #24
          Originally posted by BlackStar View Post
          (a) The ARB is composed by members of the various IHVs.
          (b) IHVs make money from selling new hardware.
          (c) Developers need new hardware to develop better software.
          (d) OSS driver developers cannot keep up with the breakneck speed of new hardware releases.

          (a) + (b) + (c) > (d), I'm afraid.
          Unless (d) = (d) + 1

          Comment


          • #25
            The OpenGL standards have lurched forward and caught up with DX functionality, which is great to see, but presumably from this point things will settle down and the two graphics standards will tend to move forward together.

            That said, the open source driver code base is growing as the level of functionality increases, and that growth *is* going to require a corresponding increase in the size of the development community..

            The good news is that most of the increase is required in the 3D userspace side (Mesa), and that part of the stack is "further from the metal" and has a lower learning curve than any other part of the driver, so in a sense it's easier to get involved and make a real contribution than ever before.

            A lot of the work required for GL 3 and higher is in the hardware-independent portions of Mesa, which is also good news because that work only needs to be done once rather than repeatedly for each hardware vendor/generation.
            Test signature

            Comment


            • #26
              Originally posted by BlackStar View Post
              Right now, we have two major players adopting the new specs (Nvidia and Ati), one major player dragging their ass (Apple) and the final player wishing OpenGL didn't exist at all (Intel). Hopefully, Apple will realize the synergy between OpenCL and OpenGL and start supporting the latest specs sooner (it's been two years and they still don't support OpenGL 3.0). On the other hand, Intel is hopeless and only (barely) supports D3D. Their OpenGL drivers are beyond hopeless.
              This is only true if you look at closed-source drivers. As far as the open-source drivers are concerned Intel is the player that supports most of the OpenGl-Specs, with 2.1 support + some backported extensions from 3.0 and GSGL 1.4 I think.(Although they do it via classic Mesa instead of Gallium, so as someone concerned with X.Org/Mesa already being a mess I had hoped they'd go for the way that's easier in the long run) With open development Intel is leading by a mile, being the first to get stable Memory Management and stable KMS/DRI2, probably being the first to implement most of OpenGL 3.0 in future mesa versions and so on.

              I mean it's easy to say bad things about Intel, but let's remind ourselves again which company isn't even supporting recent X.org-Apis with their closed-source driver AND recently dropped half their user base with the reasoning of having less to maintain? I'm not a fanboy of Intel and as a X3100 user I really hope they would invest in getting a little more power into their Laptop-based solutions and hardware-OpenGL-support, BUT they are the one-eyed among the blind, no doubt about that.

              And let's not talk about the Nvidia-Blob doing anybody any good. The people that use it don't care and that's okay for them. I don't mean to put harm nor any fault on anyone living inside a vault, but let's stop trying to please them, shall we.

              Comment


              • #27
                Originally posted by BlackStar View Post
                Right now, we have two major players adopting the new specs (Nvidia and Ati), one major player dragging their ass (Apple) and the final player wishing OpenGL didn't exist at all (Intel). Hopefully, Apple will realize the synergy between OpenCL and OpenGL and start supporting the latest specs sooner (it's been two years and they still don't support OpenGL 3.0). On the other hand, Intel is hopeless and only (barely) supports D3D. Their OpenGL drivers are beyond hopeless.
                Sorry but you have that completely ass backwards. openCL does not require a newer version of openGL to do it job. If anything openGL has been holding back openCL. .Sure openGL can be made better to accommodate openCL and has done so with 4.0. Apple took openCL to a industry standard in a record breakneck pace vs pretty much every other Kronos spec.

                So who's fucking up with openCL support? Ironically it is the open community.

                Comment


                • #28
                  Originally posted by deanjo View Post
                  Sorry but you have that completely ass backwards. openCL does not require a newer version of openGL to do it job. If anything openGL has been holding back openCL. .Sure openGL can be made better to accommodate openCL and has done so with 4.0. Apple took openCL to a industry standard in a record breakneck pace vs pretty much every other Kronos spec.

                  So who's fucking up with openCL support? Ironically it is the open community.
                  My point was that OpenGL 4.0 contains updates to improve interop with OpenCL. Since Apple is supporting OpenCL so much, it follows that they would wish to improve the GL/CL synergy by implementing the latest GL specs. (Although with Apple you never know).

                  Morevoer, it is incorrect to say that OpenGL is not holding OpenCL back. The specs can be implemented independently, there is nothing that says you need to have an OpenGL driver for OpenCL to function.

                  Finally, if there's anyone who is fucking up with OpenCL support, it is (surprise) Intel, who is refusing to implement the spec and distribute a runtime to its users. This means that at least 60% of all OpenCL-capable PCs will never see OpenCL support.

                  Compare that with the impact of a potential OSS driver: maybe 2% market share, out of which at least 50% does not have OpenCL-capable hardware (being generous here).

                  This is only true if you look at closed-source drivers. As far as the open-source drivers are concerned Intel is the player that supports most of the OpenGl-Specs, with 2.1 support + some backported extensions from 3.0 and GSGL 1.4 I think.
                  Intel's OSS drivers are OpenGL 2.1 with GLSL 1.2 - when they work. That's, ugh let's see, around 7 years behind the curve? Remember, OpenGL 4.0 with GLSL 4.0 were just released.

                  Even if Intel's OSS are great, this doesn't get them of the hook for their atrocious closed-source driver. Remember that Intel's OSS marketshare is tiny compared to their closed-source install base (which is larger than either Nvidia or Ati). It is this huge marketshare that makes it impossible to deliver a modern 3d application (game, UI, whatever) without a D3D code-path.

                  Think about this for a moment.

                  Comment


                  • #29
                    Originally posted by suokko View Post
                    Unless (d) = (d) + 1
                    (d) = (d) + 1 <=> (d) - (d) = 1 <=> 0 = 1

                    Which means you can prove anything you like, as per the principle of explosion.

                    Comment


                    • #30
                      Originally posted by BlackStar View Post
                      Rofl Xkcd is awesome!

                      Comment

                      Working...
                      X