Announcement

Collapse
No announcement yet.

For Now Intel's Beignet Seems Better Off Than Radeon Gallium3D Clover

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

  • #41
    Originally posted by blackiwid View Post
    why is it then unofficial?
    Same reason the devs keep hanging out here rather than forcing everyone to the AMD forums -- it was there first and there was no compelling reason to replace it.

    Originally posted by blackiwid View Post
    Just hurry to make this same kernel parts for radeon and the blob, to get rid of 50% of the code of the blob, and then work fast to replace the other 50% I dont get this double strategie, every year it gets less understandable, in the beginning radeon could not do much and was slow but every 5% it gets faster and has more feature, it gets more insane to keep that garbage.
    If we hurried any more I don't think you would like the result.

    Originally posted by blackiwid View Post
    Just "get it out of here", seriously who buys amd hardware to use under linux this blob? If I want to use a blob I know nvidias driver under linux is better. It makes not even sense economical.
    Pretty much the entire workstation market. Also some gamers, but the open source stack is progressing fast enough to make that almost a non-issue.

    Originally posted by blackiwid View Post
    I get that u want to keep the old drivers for a few cracy people that use this blob, but just move the 100 developers to make radeon better and keep 10 to maintain the blob not the other way around.
    One more time (I may have said this once or twice in the past ) if the Linux Catalyst driver was a completely separate code base with a separate development team it would have been open source a long time ago. That's just not the case though -- the reason proprietary drivers exist is because most of their code can be shared across multiple OSes and only a small part of the stack (and development team) needs to be OS-specific.
    Test signature

    Comment


    • #42
      Originally posted by bridgman View Post
      One more time (I may have said this once or twice in the past ) if the Linux Catalyst driver was a completely separate code base with a separate development team it would have been open source a long time ago. That's just not the case though -- the reason proprietary drivers exist is because most of their code can be shared across multiple OSes and only a small part of the stack (and development team) needs to be OS-specific.
      So there are less people that port the catalyst driver to linux than there are people developing the free driver?

      And another question, why not port it the other way around, why not port the free driver to windows from linux Ok maybe that would not be good if windows would get a nearly bugfree driver linux would loose one selling point

      Comment


      • #43
        Originally posted by blackiwid View Post
        So there are less people that port the catalyst driver to linux than there are people developing the free driver?
        That actually wouldn't surprise me at this point.

        And another question, why not port it the other way around, why not port the free driver to windows from linux
        Well, there are a ton of reasons, but you can probably start off with the fact that Direct3D is the only API that really matters on Windows and Mesa doesn't even attempt to support it. So you'd basically be starting almost from scratch, minus the gallium framework.

        Comment


        • #44
          Originally posted by bridgman View Post
          If we hurried any more I don't think you would like the result.
          Does this mean that AMD didn't dropped effort to switch to Radeon as kernel driver for Catalyst?
          No details needed, I only care if this going to happen or not...

          Comment


          • #45
            Originally posted by bridgman View Post
            Same reason the devs keep hanging out here rather than forcing everyone to the AMD forums -- it was there first and there was no compelling reason to replace it.



            If we hurried any more I don't think you would like the result.



            Pretty much the entire workstation market. Also some gamers, but the open source stack is progressing fast enough to make that almost a non-issue.



            One more time (I may have said this once or twice in the past ) if the Linux Catalyst driver was a completely separate code base with a separate development team it would have been open source a long time ago. That's just not the case though -- the reason proprietary drivers exist is because most of their code can be shared across multiple OSes and only a small part of the stack (and development team) needs to be OS-specific.


            Mr. Bridgman i wanted to ask something serious for a long time. I don't really get the workstation market. Why a gamer gpu and a workstation gpu have the same performance in games but 5x difference on static graphics? Is this needed? I mean if there wasn't for workstation gpus, those people will mostly bay the strongest and expensive with more ram and slots and software gaming gpus anyway, and most probably more of them, so the lock is it needed? Don't you also thing that companies that they don't use those locks (like Intel) will eventually beat you? And then why those locks are driver related in the first place, can you do it with your bios instead? Don't you thing that if are driver related, the open drivers will unlock them? Is there any other reason except this wrong control, that you use Catalyst on workstations? Don't you thing a really open source driver policy for all OpenGL on all OSs (like: if you need OpenGL install Mesa) will give you an advantage like that of consoles (many studios will prefer you for OGL games, users to). Don't many people use Mesa and its state trackers on Windowz with Catalyst so many years now? Don't you think that merging Nine will give you an advantage now?
            Last edited by artivision; 17 September 2014, 04:54 PM.

            Comment


            • #46
              There is no chance that the open source driver will beat Catalyst in the workstation market: AMD assigned too few manpower to the FOSS team to make it competitive and they will never pay for the certifications which are needed for such market.
              ## VGA ##
              AMD: X1950XTX, HD3870, HD5870
              Intel: GMA45, HD3000 (Core i5 2500K)

              Comment


              • #47
                Originally posted by artivision View Post
                Why a gamer gpu and a workstation gpu have the same performance in games but 5x difference on static graphics?
                Just wonder do you have any proof-link that this was ever true for ATI / AMD GPUs?

                Originally posted by artivision View Post
                Is this needed? I mean if there wasn't for workstation gpus, those people will mostly bay the strongest and expensive with more ram and slots and software gaming gpus anyway, and most probably more of them, so the lock is it needed?
                Also you really missing point of workstations GPUs. It's not about performance, it's about guarantees of rock solid stability on certified hardware with specific OS and software version.

                Companies like Autodesk have zero interest in official support for hundreds of gamer-grade GPUs, tons of OSes and so on. This why they and their partners sell hardware with workstations GPUs and pre-installed OS where drivers likely would be never updated. And there is companies who ready to pay for such hardware and services.

                It's same reason why companies pay for RHEL or any other enterprise services: ETAs, SLA, official support, etc.

                Comment

                Working...
                X