Announcement

Collapse
No announcement yet.

Radeon Evergreen HDMI Audio Code Is Still M.I.A.

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

  • #21
    Worth noting. When you use the words "everyone" and "industry" to talk about a grand total of 5 or 6 companies, it means that there would be antitrust issues investigated if our government wasn't bought by these companies to begin with.

    Comment


    • #22
      Originally posted by Qaridarium
      I'm sure amd can make profit in making hardware for 1-2% of the most computer addicted people on earth.

      thats the REAL point!
      1-2% of a market where many are just as much "free as in beer" as "free as in freedom", on a platform that has no high-end games to speak of and that will lack many features simply because it's not within AMDs power to deliver them legally so people will continue to whine and not buy it. Your point has no basis in reality....

      Comment


      • #23
        Originally posted by DaemonFC View Post
        *rant*
        The easiest AMD could do is to be like nVidia. Source? No. Specs? No. Buzz off and go away. All you get a blob and the license says "No Reverse Engineering. Customer may not reverse engineer, decompile, or disassemble the SOFTWARE, nor attempt in any other manner to obtain the source code."

        Is AMD completely and utterly committed to making open source drivers? Hell no, for one thing they'd like to stay in business. Sure, they could just post the full source of their proprietary driver here, they'd only get sued to oblivion and have all their graphics cards blacklisted. So because they're not doing that, they're not trying. *rollseyes*

        You are starting your raving at the completely wrong end. AMD doesn't have any benefit from DRM, if it all magically went away they'd sell graphics cards like before. Maybe even more. It starts with Hollywood, they want DRM. They pressure Microsoft and say either you accept DRM, or Windows won't play BluRays - only standalone players. So Microsoft turns around and says okay, to get the Windows logo the graphics card must implement DRM. And if you want to deliver to 95% of the market, actually 99% because it's same with Apple then you say yes. What do you want from AMD, corporate suicide? Make Hollywood stop demanding DRM, then AMD will stop delivering it. If not, Hollywood will go to someone who will.

        Comment


        • #24
          Originally posted by Kjella View Post
          The easiest AMD could do is to be like nVidia. Source? No. Specs? No. Buzz off and go away. All you get a blob and the license says "No Reverse Engineering. Customer may not reverse engineer, decompile, or disassemble the SOFTWARE, nor attempt in any other manner to obtain the source code."

          Is AMD completely and utterly committed to making open source drivers? Hell no, for one thing they'd like to stay in business. Sure, they could just post the full source of their proprietary driver here, they'd only get sued to oblivion and have all their graphics cards blacklisted. So because they're not doing that, they're not trying. *rollseyes*

          You are starting your raving at the completely wrong end. AMD doesn't have any benefit from DRM, if it all magically went away they'd sell graphics cards like before. Maybe even more. It starts with Hollywood, they want DRM. They pressure Microsoft and say either you accept DRM, or Windows won't play BluRays - only standalone players. So Microsoft turns around and says okay, to get the Windows logo the graphics card must implement DRM. And if you want to deliver to 95% of the market, actually 99% because it's same with Apple then you say yes. What do you want from AMD, corporate suicide? Make Hollywood stop demanding DRM, then AMD will stop delivering it. If not, Hollywood will go to someone who will.
          Item 1: It's obvious that Red Hat's lawyers don't think those EULA sections are enforceable. If they are, then WINE is also illegal. Though courts in the US have traditionally sided with the victim of the lawsuit when this kind of nonsense ends up in the court system, provided that they do their reverse engineering like Nouveau and WINE do.

          It's surprising that Nouveau in many cases is on par with AMD with no help from Nvidia. Obviously the documentation and occasional skeleton driver code contribution by AMD is helpful, but there would probably be a sister project to Nouveau if AMD decided to close up the Radeon specs and go full on evil again.

          Item 2: Of course AMD isn't doing this because it's right. They're doing it because FGLRX is an unholy mess and they want free labor to make a driver they can port to Windows. Phoronix has had several articles about the same driver code being made proprietary to go into Windows CE. Free labor is a nice thing to have. The Catalyst mess is probably far from optimal for real Windows, much less the embedded thing calling itself that. (It will also be nice to have if they want their hardware to go into tablets.) Other than FGLRX being utter crap, it heavily depends on an X server for the 2d rendering which embedded Linux distributions like Android don't have. It also has no OpenGL ES support at all, which would mean they'd need a shim like Google's ANGLE riding on top of a bloated OpenGL driver that they already aren't fully licensed to use themselves that ships in their proprietary Linux and Windows drivers already.

          Item 3: They've already said they do benefit from DRM indirectly in that it's easier to roll over and ask the MAFIAA to screw AMD's customers with no lube than it is to draw a line and tell them to fuck off and come back when things are less contaminated. They're all too happy to collaborate with criminals to build this stuff into their hardware, so it entirely AMD's fault that they can't document parts of their own hardware. You can lob any stupid argument against this you want but I won't budge.

          If you read Theo de Raadt's "Intel is only OPEN for business" article from 2006 where they were being boner biting bastards about their hardware documentation and firmware redistribution, I think you'll find a lot of it is unfortunately relevant to the situation with AMD. Unfortunately it is still more of a fit for Intel and Nvidia's continued behavior which is why I'm not using them. Nvidia hands down makes better hardware, but when it comes to openness, they want to kill with fire any technical details about their cards that can get out there. If Nvidia had open drivers, we wouldn't be discussing this and I'd just be using Nvidia like I was 2-3 years ago before AMD had any open specifications and were every bit as nasty.

          BTW, here's Theo de Raadt's article on Intel I promised with some choice quotes I feel apply a lot to AMD right now.



          Before we ask a vendor, we have already lost (ie. the device does not work). When a vendor says no, we have lost nothing further -- there is no way we can lose further than having the device not work. We can only win, and then the device works. So there is no point in giving up until we win back the rights to write software for the hardware that we have purchased.
          This fits the HDMI audio situation like a glove.

          These vendors often want a quiet private discussion, because in a quiet private discussion they can continue to dismiss the requests and in the end do absolutely nothing. They do not want a noisy public discussion, because then they look bad. But they DESERVE TO LOOK BAD, because they are being bad to those who bought their hardware!
          Yes, AMD has a few people here to do PR and damage control. After the article about HDMI audio support missing, they came to spin the situation back into AMD's favor and to derail the conversation with half-truths, impotent gestures, and outright obfuscations and lies. It is technically better than not responding at all I guess, a lot of companies that have been approached by open source developers just asking for rights to redistribute their non-free firmware have had their requests go totally unanswered for a long time. (See OpenBSD v. Texas Instruments). The fact that they have yet to reply in a fully honest and open manner does bother me. Perhaps they can rectify this in the future.


          In this particular case, we would like more documentation for the Intel wireless chips. Damien [Bergamini] has already written drivers that make the devices work quite well... but there are still bugs, since all of this is based on reverse engineering efforts. The drivers could be better. Intel stands in the way of your devices working as well as they should.
          (emphasis mine)

          Again, replace Intel with AMD and you have the same situation with HDMI audio on Radeons.

          Quite frankly, Intel has been a royal pain in the ass. Not to us, but to people who bought their devices.
          Replace Intel with AMD.

          Intel also must grant these rights freely
          Replace Intel with AMD again here. Actually I'd rather AMD be total dickheads here because then I end up using free software that was the result of reverse engineering rather than a really long strand of obfuscated code that has no commenting and would be even harder to figure out. (Nvidia did this with nv where they had an insane formula just to represent a constant).

          By withholding, Intel is being an Open Source fraud.
          Replace Intel with AMD.

          -----

          Note that this is not a rant. A rant doesn't provide facts and counter-arguments that disassemble the propaganda spread by the original comment. A rant is merely an attack without any constructive criticism at all. I am mad that AMD is voluntarily withholding programming specs, that they lie about why they do it, and use the words "Industry" and "everyone" to talk about a few companies and lobbying groups (The unholy triad at Apple, Microsoft, and the MPAA) that should be investigated under acts including Sherman, Hobbs, RICO, and at least in Microsoft's case, Sarbanes-Oxley.

          Thank you.
          Last edited by DaemonFC; 06 December 2011, 10:22 AM.

          Comment


          • #25
            Originally posted by DaemonFC View Post
            Item 2: Of course AMD isn't doing this because it's right. They're doing it because FGLRX is an unholy mess and they want free labor to make a driver they can port to Windows.
            Windows Embedded Compact, nobody is going to replace FGLRX on any normal consumer version of Windows. And for what it's worth FGLRX works just fine *on Windows*, complain all you want about how it works under Linux but almost all those issues are platform specific from trying to hack it to run on top of or instead of X11.

            Comment


            • #26
              Hm, would you like a list of issues I've seen on Catalyst for Windows?

              Comment


              • #27
                Originally posted by Kjella View Post
                Windows Embedded Compact, nobody is going to replace FGLRX on any normal consumer version of Windows. And for what it's worth FGLRX works just fine *on Windows*, complain all you want about how it works under Linux but almost all those issues are platform specific from trying to hack it to run on top of or instead of X11.
                Everyone else seems to make drivers that work fine under X11 but AMD with FGLRX.

                FGLRX is the name of the Linux display driver, the Windows one has been called Catalyst for as long as I can recall. Catalyst Linux is just a rebranding of their same old shit.

                Comment


                • #28
                  If/when this new code sees the light of day, will it also fix the blank screen/no signal issue that is seen on some monitors with hdmi audio on <= hd 4xxx? We just got a led-lcd and whenever radeon.audio=1, no signal seems to be recognized by the monitor, but on our regular lcd, everything works swimmingly.

                  Comment


                  • #29
                    Dang it, have there been any developments since october [link]? How far along is the fixed code compared to the new code for evergreen+?

                    Comment


                    • #30
                      Originally posted by Qaridarium
                      i think the official amd opensource HDMi support will be released in February 2012.
                      Wow. Maybe another "Q already knows what bridgman doesn't" situation?
                      We will see...

                      Comment

                      Working...
                      X