Announcement

Collapse
No announcement yet.

SUSE Loses Another Former RadeonHD Developer

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

  • SUSE Loses Another Former RadeonHD Developer

    Phoronix: SUSE Loses Another Former RadeonHD Developer

    Matthias Hopf, a SUSE developer working on the X11 stack for the past seven years and one of the original xf86-video-radeonhd driver developers, has left the company...

    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
    Who cares?

    RadeonHD has been dead for a while.

    Comment


    • #3
      I had no idea SUSE had PhDs working on open source projects. That in itself is news:

      Comment


      • #4
        Originally posted by LLStarks View Post
        Who cares?
        RadeonHD has been dead for a while.
        Having used radeonhd in its heyday, I care (and Mathias contributed to other projects as well). It is nice to hear that he will have a job in the academic world where he will have a chance to spark the next generation of Linux video hackers. I wish him well.

        Comment


        • #5
          Originally posted by Qaridarium
          i care to!... the radeonHD was really cool,,, and they only lag on support! Bridgman was one of the man how kill the radeonHD because he prefer ..
          I stopped reading right about there

          Comment


          • #6
            Matthias Hopf will be at XDC2011 Chicago in just over one week. Along with Hopf there will be his former colleagues Egbert Eich and Luc Verhaegen of the defunct RadeonHD project for the X.Org Developers' Conference.
            So after RadeonHD went defunct, did they ever use their expertise and contribute to the radeon driver? If not, why not?

            Comment


            • #7
              Originally posted by Qaridarium
              i care to!... the radeonHD was really cool,,, and they only lag on support!

              Bridgman was one of the man how kill the radeonHD

              because he prefer more in-house developing means today more amd dudes do the ground work and the documentation and spec follow.

              back in the radeonHD time the Spec was first and after that they write the driver clean room.

              clean room mean the "Bridgman" version isn't clean room the Bridgman way is to have full access to the catalyst source code and internal secret documentation.

              one of bridgman's job is to make sure secret stuff be secret in the future.

              mean bridgmans job is to make sure the firmware stay closed and the UVD crypto internals stay closed and the ATOM-Bios stay closed!

              because an FIREGL card compared to an normal cheap card is only 2 thinks: different driver OpenGL-optimations AND AtomBIOS settings.

              the RadeonHD hurt this unwriten LAW because they don't use the ATOM-BIOS means in the RADEON-HD world every cheap radeon card is a FIREGL card!

              in an real free world there are no "fireGL" or "Quadro" and no AtomBIOS and no FIRMWARE and no secret UVD crypto unit to force the DRM.

              Bridgman is the Economic Hitman manager to make sure no unwriten rule "FireGL"-"Atom-Bios"-"UVD-DRM"-"Firmware" are ever be hurt.

              the open-source driver is fine if they don't touch this rules.
              Yeah, I became same opinion after he started taking arguments to prevent acceleration the driver development instead of being initiative.
              The positive thing is that he makes sure the driver is legally clean, the negative is that the driver is doomed to be second class for ever. It won?t match fglrx, not in a year, not in 10 years, never.
              Once any serious improvements in opensource driver happen, they will ported to fglrx. But fgrlx critical optimisations will never be present in opensource.
              Do you think opensource driver lack of speed is due to resorting to atombios? I hate the whole thing about firegl, the approach is completey rubbish. Why not associate the price exlusively with priority support? Why not be able to register the card and recieve priority support from AMD instead of selling hardware? Why not charge for something REAL instead behaving like microsoft?

              Comment


              • #8
                Originally posted by Qaridarium
                back in the radeonHD time the Spec was first and after that they write the driver clean room.

                clean room mean the "Bridgman" version isn't clean room the Bridgman way is to have full access to the catalyst source code and internal secret documentation.
                Neither is "clean room" implementation because either those writing the spec or those writing the code are under contract with AMD and bound by all their agreements. A real clean room implementation would take two groups, BOTH outside AMD. One group would have to reverse engineer the hardware and write a thorough specification free of any copyrighted material, the other group must be "untainted" and implement that specification. If you depend on Bridgman in any way, then it's not clean room.

                Comment

                Working...
                X