Announcement

Collapse
No announcement yet.

OpenSUSE Says Farewell To RadeonHD Driver

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

  • OpenSUSE Says Farewell To RadeonHD Driver

    Phoronix: OpenSUSE Says Farewell To RadeonHD Driver

    The RadeonHD Linux driver that came about in 2007 following the announcement of AMD's open-source driver strategy has had an interesting history. This driver was developed by Novell's developers, but now they are even dropping it from their openSUSE distribution...

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

  • #2
    Hard to say goodbye but i hope the Novel Devs work now on "radeon".

    Comment


    • #3
      this is bond to happen, two drivers for the same thing is bad, especially when developer man power is lacking

      Comment


      • #4
        Well, this is not really surprising, is it? Everybody knew that would happen sooner or later.

        Comment


        • #5
          Finally something that should have never existed disappeared. What a damn waste of time and man power.

          Comment


          • #6
            Originally posted by bulletxt View Post
            Finally something that should have never existed disappeared. What a damn waste of time and man power.
            i wouldn't call it a waste, even though i never used it.

            it was an attempt to rewrite ati driver from scratch for new generation of cards. it got proven wrong by classic ati driver team. nothing wrong with that. besides radeonhd team was not that big to start with, and shared some developers with the other team.

            Comment


            • #7
              IIRC the ati KMS driver picked up code from the radeonHD codebase, so it's not a complete loss.

              Maybe we'd be further if they had concentrated on radeon from the beginning, but it's too late for that anyway.

              Comment


              • #8
                Why not merge?

                How hard would it be to merge any interesting bits from radeonhd and go from there?

                Folks, Open Source software evolves like this. Many related ideas/projects pop up, one will be strongest and survive, but you couldn't have told a priori. The model has proven incredibly successful. A large chunk of the computing hardware (from embedded to big iron) runs open source software, mostly GPL type. That ranges from cellphones to supercomputers. So, again, it is what it is and it works. It just gets a little bit to get used to it.

                Comment


                • #9
                  Originally posted by mendieta View Post
                  How hard would it be to merge any interesting bits from radeonhd and go from there?
                  I believe radeon has atleast the same functionality, so there aren't any interesting bits...

                  Comment


                  • #10
                    Wondering how much money Novell has invested in this project.

                    Comment


                    • #11
                      I guess SuSE wanted to be the first to have a good support for the new chips (R500 and up), then the quarrel about the use of AtomBIOS or not.
                      But sooner or later it had to happen and it was no surprise to me. I unmerged that driver 3 days ago, since I hadn't used it for a few months now (using xf86-video-ati). I just wish they had joined forces. Reinventinv the wheel, useless pride are too common in FOSS development. Yes, it's good to have alternatives but sometimes...

                      Well, it gave me a few months good 2d usability of my cards which wasn't supported by xf86-video-ati then. Later they were on about same level and now... thanks for the good time and goodbye.

                      Comment


                      • #12
                        Good riddance. Performance was crap and it was not stable (on ATI X1400 Radeon mobility).

                        Comment


                        • #13
                          Curous about your performance comments. For the last 18 months or so the acceleration code has been largely common between radeon and radeonhd - it was really modesetting where the two drivers were substantially different.

                          Comment


                          • #14
                            Thank you Egbert, Mathias, Luc, and the rest. This driver was more stable and had better EXA for a while.

                            Comment


                            • #15
                              Originally posted by bridgman View Post
                              Curous about your performance comments. For the last 18 months or so the acceleration code has been largely common between radeon and radeonhd - it was really modesetting where the two drivers were substantially different.
                              Yes, I've heard that said many times, and I'm sure you are right. However, real life empirical results for me always showed that radeonhd was massively slower at DRI, and slower in 2D-acceleration when compared to radeon. I've probably tested radeonhd about 10-15 times over the course of time, hoping it would work OK and using it until it crashed (which it almost always did eventually). Then went straight back to radeon. Of course, I've no idea what actually causes the different performance profiles of the two drivers, but they sure are different on my hardware.

                              Comment

                              Working...
                              X