Announcement

Collapse
No announcement yet.

The Fallacy Behind Open-Source GPU Drivers, Documentation

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

  • #31
    Originally posted by mirv View Post
    I suggest you go back and re-read that thread - Bridgman never said anything of the sort.
    Bridgman stated in my ATI bashing thread the following, on page 1, and I quote:

    "We work with the community to make initial support available (working code and/or docs) and the open source community does most of the enhancement work. We were told multiple times "just give us programming information and the community will do the rest" - we are doing that and more."

    Anyone else want to take a poke at me?

    Comment


    • #32
      Originally posted by SheeEttin View Post
      Hey vendors. Here's an idea: don't bother with closed-source drivers. Just make one open-source one.

      I mean, you might have a few more headaches in the beginning (with legal or whatever), but we'll love you all the more for it.
      Only open-source so a person can only run @ 1/4 the performance of a 5 generation old card running a blob on todays top of the line card? Ummm, ya, no thanks.

      Comment


      • #33
        Originally posted by Mr James View Post
        Bridgman stated in my ATI bashing thread the following, on page 1, and I quote:

        "We work with the community to make initial support available (working code and/or docs) and the open source community does most of the enhancement work. We were told multiple times "just give us programming information and the community will do the rest" - we are doing that and more."

        Anyone else want to take a poke at me?
        That response does not mean that the community is refusing assistance from AMD. Please, read it again to see what is actually being said.
        The community did, multiple times, talk about that all they needed were the programming docs - that's a request for something from AMD (or ATI, whichever time frame you're looking at) - and that AMD wouldn't need to help beyond that. AMD released the docs, are helping beyond that, and the community is embracing that assistance.

        Comment


        • #34
          Originally posted by Wyatt View Post
          You've been trolling awfully hard lately becuase you couldn't be arsed to verify that the hardware you purchased was supported in a way that was satisfying to you.
          Actually, we are talking about the FOSS drivers and all FOSS graphics drivers regardless of manufacturer suck. ATI sucks way less because the at least partitipate in providing info and trying to help. FOSS support for your card is just as lousy as for my card. So I don't see what your point is.

          Originally posted by Wyatt View Post
          Tearing into the very people that you rely on to make that hardware work and making libellous statements about them is sure to help the situation, though!
          I'm tearing into them because they, as bridgman pointed out are not letting the AMD devs participate enough.

          Originally posted by Wyatt View Post
          Please go buy nvidia, install their blob, and be done with it already.
          Why should I buy nVidia when my AMD card is recent and performs well? All I need is a proper driver, FOSS or not - and not just me but every other ATI Linux consumer as well.

          Comment


          • #35
            Originally posted by energyman View Post
            Mr James is cute, isn't he? Or pathetic. Depending on the point of view.

            Btw, my latest mobo has amd onboard graphics. I am contemplating to turn it on and connect an usused display I have standing around.
            And your problem with me is?

            Comment


            • #36
              Originally posted by mirv View Post
              That response does not mean that the community is refusing assistance from AMD. Please, read it again to see what is actually being said.
              The community did, multiple times, talk about that all they needed were the programming docs - that's a request for something from AMD (or ATI, whichever time frame you're looking at) - and that AMD wouldn't need to help beyond that. AMD released the docs, are helping beyond that, and the community is embracing that assistance.
              Now this is a decent reply. Thank you.

              I posted based on what Bridgman stated. There is no way of deducing from Bridgman's post that that which you explained. At least not by me. Again, thanks.

              Comment


              • #37
                Originally posted by deanjo View Post
                Only open-source so a person can only run @ 1/4 the performance of a 5 generation old card running a blob on todays top of the line card? Ummm, ya, no thanks.
                Well stated. +1

                Comment


                • #38
                  Originally posted by Mr James View Post
                  Now this is a decent reply. Thank you.

                  I posted based on what Bridgman stated. There is no way of deducing from Bridgman's post that that which you explained. At least not by me. Again, thanks.
                  I realised that perhaps you didn't know some of the history - if you dig hard enough, you might find "ancient" posts on Phoronix, but the community wanting the programming docs goes back to before then. I remember the r200 driver days with the radeon8500 - those drivers were fast (really fast!) - the r200 mostly worked, and ATI wasn't supporting consumer cards (blob or otherwise), hence the community wanting docs so they could take of everything.
                  Others, feel free to fill in any gaps in my knowledge here, or correct where I'm mistaken!

                  Comment


                  • #39
                    Originally posted by mirv View Post
                    I realised that perhaps you didn't know some of the history - if you dig hard enough, you might find "ancient" posts on Phoronix, but the community wanting the programming docs goes back to before then. I remember the r200 driver days with the radeon8500 - those drivers were fast (really fast!) - the r200 mostly worked, and ATI wasn't supporting consumer cards (blob or otherwise), hence the community wanting docs so they could take of everything.
                    Others, feel free to fill in any gaps in my knowledge here, or correct where I'm mistaken!
                    Nah, they will just call you a liar or a troll or some other shit.

                    Comment


                    • #40
                      Originally posted by deanjo View Post
                      Only open-source so a person can only run @ 1/4 the performance of a 5 generation old card running a blob on todays top of the line card? Ummm, ya, no thanks.
                      Um? While unrealistic, the idea is that if every attempt is concentrated on one driver only (opensource in our case) the results would be much better than the current situation...

                      Comment


                      • #41
                        Originally posted by Mr James View Post
                        Nah, they will just call you a liar or a troll or some other shit.
                        If you're not intentionally lying and/or trolling, then you sorely need to improve your reading comprehension (or stop posting).

                        Comment


                        • #42
                          Originally posted by DanL View Post
                          If you're not intentionally lying and/or trolling, then you sorely need to improve your reading comprehension (or stop posting).
                          To be fair in this particular case, if he didn't realise the history and didn't know that Bridgman was referring to history, not current events, I can see how it might have been ambiguous.
                          So misunderstanding fixed hopefully.

                          Comment


                          • #43
                            Originally posted by Mr James View Post
                            All I need is a proper driver, FOSS or not
                            And instead you chose to buy ATI and then complain about it. Nice.

                            Comment


                            • #44
                              Originally posted by mirv View Post
                              I realised that perhaps you didn't know some of the history - if you dig hard enough, you might find "ancient" posts on Phoronix, but the community wanting the programming docs goes back to before then. I remember the r200 driver days with the radeon8500 - those drivers were fast (really fast!) - the r200 mostly worked, and ATI wasn't supporting consumer cards (blob or otherwise), hence the community wanting docs so they could take of everything.
                              Others, feel free to fill in any gaps in my knowledge here, or correct where I'm mistaken!
                              I remember the discussions on slashdot back when the first GeForce cards were appearing. Back in those days, 3d chips were not that complex and this made sense.

                              The reality is that the 3d hardware has advanced a lot since then, and is the second most complex piece of hardware in your computer -- after the CPU. Perhaps even more complex than that.

                              There are hundreds of companies funding linux kernel development, and only a handful of people working on 3d drivers. That is the true problem. I think that AMD's approach is correct here -- release the documentation, release the initial code as a proof of concept, and let the community deal with it.

                              The problem is that there are not enough people in the community to take care of it. RedHat is doing a great thing by funding devs to work on 3d drivers, but there need to be more. Another 10 paid developers who know what they are doing would go a long way. Ideally the people who are hacking on the drivers in their spare time right now. Drop 60k on one of them for a year, and let him bite into it, and you should see some improvement.

                              Comment


                              • #45
                                Originally posted by Mr James View Post
                                Now this is a decent reply. Thank you.

                                I posted based on what Bridgman stated. There is no way of deducing from Bridgman's post that that which you explained. At least not by me. Again, thanks.
                                FWIW I read it the same way Mirv did. The word "refuse" does not appear in Bridgman's post.

                                Comment

                                Working...
                                X