Announcement

Collapse
No announcement yet.

ATI dropping support for <R600 - wtf!?

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

  • Originally posted by Stormking View Post
    For five years, I got promises. About how the driver will be better, soon. Month after month after month.
    This is the part I just don`t get. I have gone through every single press release, interview and on or off-record employee comment I could find on the internet and haven`t found *anything* like that. There is the occasional announcement talking about open source drivers or workstation drivers but that`s all I have been able to find.

    Originally posted by Stormking View Post
    And now? More promises. About how great the open source driver will be some time in the future. I don't care about a year from now. I want to use my laptop, *NOW*. I haven't been able to do that (to the full extent) for the greater part of its lifetime. In a year or two (when I assume the OSS driver will be feature complete, optimized and stable) I will very likely not use this hardware, anymore.
    Yeah, this is the heart of it. Your laptop. Designed for Windows, tested by the OEM on Windows, bugs fixed for Windows before shipment. Desktops are fairly generic and (with the exception of AGP and high memory) different users tend to see similar results. Laptops are much more heavily customized, and the reason they work so well with Windows is that all of the qualification and platform-specific driver code is worked out with the OEM before the laptop ships.

    That doesn`t happen with Linux, and (with the exception of a few OEM preloads) may never happen. The reason so many people are beating on you about getting more involved with the development effort is that so far this has been the only way to make up for the fact that platform integration is still done pretty much completely on Windows and MacOS, not Linux.

    If you want the same kind of platform-specific diagnosis and fixing that goes along with a Windows or MacOS system (and the much larger market share), then either OEMs are going to have to be convinced that shipping with pre-loaded Linux (along with all the associated qualification testing) is a good investment, or *someone* with your specific system and priorities is going to have to get involved personally, using the open source drivers.

    Don`t wait a couple of years. Make a list of what doesn`t work on your laptop today with the open source drivers and I think you`ll find that the issues are either already fixed, are in the pipe for fairly short term resolution, or need fixes specific to your platform which are only going to happen with your involvement.

    Comment


    • Originally posted by bridgman View Post
      Yeah, this is the heart of it. Your laptop. Designed for Windows, tested by the OEM on Windows, bugs fixed for Windows before shipment. Desktops are fairly generic and (with the exception of AGP and high memory) different users tend to see similar results. Laptops are much more heavily customized, and the reason they work so well with Windows is that all of the qualification and platform-specific driver code is worked out with the OEM before the laptop ships.
      First of all, before I bought that damn thing, I checked every feature I wanted to use for the availability of linux drivers. I had positive experiences with NVIDIAs closed source drivers for linux. Not in a dream I would have thought that those of its direct competitioner ATI could be so crappy.

      Then: I know that it is possible for my card to work under Linux. All the features I want to use worked at some point. Just not all of them with the same driver version. And there were driver version that even did it all (the one before 8.22.x). But one can't stay with an old kernel, forever.

      Comment


      • Originally posted by duby229 View Post
        I think you need to read up on what open source actually means.
        Well, my arrogant friend, I presume I pretty well know what opensource means; however, we seem to disagree on the question whose responsibility the development of driver code for hardware advertised as "supported on linux" is. If you find that this question is covered by the definition of opensource or the legal content of different opensource licenses, feel free to drop a link.

        Oh, by the way: in which mode are you going to contribute to the development of the opensource drivers?

        Comment


        • Originally posted by bridgman View Post
          Don`t wait a couple of years. Make a list of what doesn`t work on your laptop today with the open source drivers and I think you`ll find that the issues are either already fixed, are in the pipe for fairly short term resolution, or need fixes specific to your platform which are only going to happen with your involvement.
          I think we already met in the X-randomly-freezes-thread in the FOSS driver subsection of this forum.

          Also, one of the most important drawbacks of the FOSS driver is the OpenGL performance. 60 - 70% of the closed source driver's performance is not an acceptable replacement for the "loss" of it.
          Last edited by Stormking; 03-08-2009, 01:34 PM.

          Comment


          • OK, thanks; I'll go back and check that thread.

            re: performance, the 60-70% number is an estimate averaged across applications and across GPU generations.

            The simpler apps which get run the most under Linux will tend to run faster than the most demanding ones, since the hardest optimization work is usually related to bandwidth-constrained situations on specific applications.

            Older GPUs will typically have less of a performance delta than newer GPUs, since the biggest benefit from a more advanced shader compiler comes with the superscalar shader core in 6xx and above. For 5xx and below, with vector pipes rather than superscalar pipes, the shader compiler in fglrx has much less opportunity to optimize. A lot of the really slow performance in 3D on the open source drivers today is from software fallbacks related to missing GL 2.0 features, not the inherent performance of the driver stack.
            Last edited by bridgman; 03-08-2009, 03:45 PM.

            Comment


            • @bridgman

              The Win drivers will be updated still every 3rd month (that's what was written in a news article in german), will the Linux drivers get then updates again too?

              Comment


              • No, we're going to be doing ongoing support for Linux via the open source drivers rather than quarterly updates of fglrx.

                The rationale is that most of the people using fglrx on older GPUs in a consumer environment are looking for additional features and functionality, not just minor bug fixes and keeping it working on new X, kernel and distro releases. Given that, going with the open source drivers seems like the right approach.

                Comment


                • Originally posted by bridgman View Post
                  Just to be clear, we're changing support level for all OSes, not just Linux. There's no new OS support for Windows either, although in fairness the impact on Windows users is less because Windows tends to keep ABI compliance and also releases new OS upgrades at a much slower rate.
                  So, no Windows 7 for my laptop?

                  Comment


                  • Originally posted by DirtyHairy View Post
                    Well, my arrogant friend, I presume I pretty well know what opensource means; however, we seem to disagree on the question whose responsibility the development of driver code for hardware advertised as "supported on linux" is. If you find that this question is covered by the definition of opensource or the legal content of different opensource licenses, feel free to drop a link.

                    Oh, by the way: in which mode are you going to contribute to the development of the opensource drivers?
                    The hubris of the misinformed....

                    You can call me arrogant all you want, but it still doesnt change the fact that it doesnt matter what AMD promised. AMD could promise the earth and the moon and it doesnt matter. The simple matter of fact is that as of this announcement the open source driver --WILL-- improve far beyond what it otherwise would have. I've already explained this, but I'll explain it again if I must. The power behind an open source development model is simply put: The size of it's user base. The bigger the user base is the more potential developers that project will have. And that really is the bottom line. What AMD has done here is they have dramatically increased the size of the open source drivers user base. And what they'll need to do in the future is better organize the manpower in the developer base. And that is nothing but good.

                    You attack AMD for not supporting there drivers, but the reality is that what they have done is actually much better then just that. Simply supporting the hardware isnt enough.

                    Comment


                    • Originally posted by DoDoENT View Post
                      So, no Windows 7 for my laptop?
                      The Windows developers have a different view of ABI compatibility from what I can see. Vista could load XP drivers (albeit with limited functionality AFAIK), and my understanding is that Win7 will work with existing Vista drivers.

                      What you won't get with older GPUs is the newer WDDM 1.1 driver model, which support some additional features such as using DX10 for DWM. You will get Vista-level functionality and morewith Vista drivers, including Aero Glass.

                      If a new Windows release causes an existing binary driver to stop working, the Windows devs tend to treat that as *their* problem and their responsibility to fix. If a new Linux kernel or X server causes an existing binary driver to stop working, it's regarded as a driver problem (or "a good start", depending on the developer ).

                      Perversely, that makes graphics driver maintenance for Linux even *more* expensive than the same level of support for Windows.

                      While building Windows 7, Microsoft is attempting to resolve scenarios that managed to successfully handicap Windows Vista in terms of compatibility. In this context, in order not to break devices that currently work with its precursor, Windows 7 will come to the table, from the get-go, with support for all Vista-certified drivers. Compatibility with devices designed for Vista ensures that users will have a seamless upgrade/migration experience. Grant George, the VP of Test for the Windows Experience, revealed that Microsoft had full compatibility with Vista-certified drivers for Windows 7 as a primary goal.
                      http://news.softpedia.com/news/The-W...y-102076.shtml
                      Last edited by bridgman; 03-08-2009, 03:54 PM.

                      Comment


                      • Recently I have been forced to use my old x800gto. Open source drivers work marvelously. For stuff that the regular user does on a daily basis they're more than enough. Movies, DE effects, games (most of then, and most certainly not WINE games), all function reasonably well. The advances being worked on will only make them better (3D speed, more complete support for newer chips, etc.). I can imagine most users or pre r600 cards being very happy with open source drivers.
                        For me the only real advantage of fglrx is 3D optimization. Speed in 3D applications is always a good thing, and the more you have the better, and if this is of importance to you chances are you have a rather new card, which are in turn supported by fglrx. I really don't see what the big fuss is about, specially on the Linux side.

                        Comment


                        • Frankly, even though I don't typically use the fglrx drivers, I am disappointed in this move. For example, at the present moment doom3 is only playable with the arb render path with the open souce drivers and, even then, the performance is horrible compared to fglrx. quake4 is unplayable with the open source drivers due to a lack of support for decompressing s3tc textures on the card. libtxc_dxtn is available ( http://homepage.hispeed.ch/rscheideg...3tc_index.html ) but has issues with multitexturing on the r300 driver and makes nearly every texture in the game flicker. ut2004 also gains a huge speedboost from that library, but experiences the same problem.

                          While I have no problems with AMD creating a legacy driver package that gets updated to support new servers and kernels, I think it's quite foolish to stop supporting them in the binary drivers altogether. Any users of those games on r300-r500 hardware (which are perfectly capable of playing those games on fglrx with good quality and performance) will be left in the cold when it comes time to upgrade their X server or kernel.

                          Once those items get ironed out in the open source drivers, and the remaining issues such as power-management get resolved, I'm all for dropping the from catalyst.

                          Adam

                          Comment


                          • Originally posted by Melcar View Post
                            For me the only real advantage of fglrx is 3D optimization. Speed in 3D applications is always a good thing, and the more you have the better, and if this is of importance to you chances are you have a rather new card, which are in turn supported by fglrx. I really don't see what the big fuss is about, specially on the Linux side.
                            3D optimization AND power management! And I think the power management is the most important feature that R300-R500 laptop users will miss until FOSS drivers start supporting it. I kind a expect this by June in Fedora 11, but I'm a bit too optimistic person .

                            And about 3D optimization - I'm a bit sad I wouldn't be able to play Scorched 3D on my Radeon 9550R (R300 card) for a while - but I'm planning to solve this by not upgrading to Jackalope or Koala. Eventually I'm planning to put openSuSe 11.2 in November or December on that computer and by this time FOSS drivers will be using Gallium3D and will be more optimized for 3D than are today.

                            Comment


                            • OK, maybe I'm missing something here, but...

                              We *are* moving the older GPUs into a legacy branch, which means you go from monthly releases to quarterly at best. That part is no different from what any other vendor does. Given that, the earliest date for the first quarterly release would be some time in June, with the next one happening some time in September (for example). Our decision was not based on the state of the open source drivers today, but our best guess about which approach will work out best at the time when the legacy driver updates would happen (let's stay with June 09 and Sept 09 for argument).

                              That's the time frame I'm talking about when I say that the open source approach is likely to seem like the right decision, *not* today.

                              The relative state of the open source and fglrx drivers *today* is only really relevent to the extent that the current state and current work-in-progress allow you to predict how the two approaches will compare later in the year.
                              Last edited by bridgman; 03-08-2009, 04:10 PM.

                              Comment


                              • Originally posted by bridgman View Post
                                OK, maybe I'm missing something here, but...

                                We *are* moving the older GPUs into a legacy branch, which means you go from monthly releases to quarterly at best.
                                This is what the phoronix article says. Please correct me if it's wrong:

                                AMD will be moving the R300/400/500 support to just a single legacy driver, but this branch will not be maintained. In fact, do not really look for legacy driver updates after April, as AMD does not intend to add support for newer kernel / X.Org server releases to this driver.
                                Having a legacy branch that is never updated to support newer kernels and X servers is not much different from dropping support entirely.

                                If you are thinking that the open source approach will be best approach come June or September, does that mean there are plans to tackle the s3tc issue in that time frame?

                                Is there some sort of development roadmap that users can look at? Over the past few years, I've seen lots of "It'll be done when it's done" answers to questions about time frames for support of specific functionality. Does this mean that there are now official time frames for certain features? While the developers will sometimes give you best guess estimates, they are often off by months/years. Are you really that confident that the r300 driver will be up to par with fglrx in terms of 3D performance and power management sometime between June and September?

                                Adam

                                Comment

                                Working...
                                X