Announcement

Collapse
No announcement yet.

AMD's opensource lies exposed

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

  • #31
    I wish that kind of thread would get deleted on sight.

    Comment


    • #32
      Originally posted by bridgman View Post
      Maybe I'm missing something, but AFAICS the "3dfx legacy IP lesson" is that it was a lot easier to support open source development before DRM was a big issue.
      really no one care about the DRM(the bad one) stuff and no i don't care about UVD video acceleration because my cpu is fast to decode any video.

      really drm/UVD don't care.

      but the Patent stuff really hurt.

      to not have openGL3/4 support because of Patent stuff (s3tc,floadingpoint) really shots down any hope for an better means opensource world.

      be sure many more people think like this DRM(the bad) and UVD don't care but s3tc and floadingpoint graphics really care

      Comment


      • #33
        Patent-Haters Gotta Hate

        Originally posted by Qaridarium View Post
        really no one care about the DRM(the bad one) stuff and no i don't care about UVD video acceleration because my cpu is fast to decode any video.

        really drm/UVD don't care.

        but the Patent stuff really hurt.

        to not have openGL3/4 support because of Patent stuff (s3tc,floadingpoint) really shots down any hope for an better means opensource world.

        be sure many more people think like this DRM(the bad) and UVD don't care but s3tc and floadingpoint graphics really care
        And *none* of that is either ATI's fault, or AMD's fault.

        If I recall correctly, that method of texture compression was never owned by ATI Technologies, or AMD (or even Microsoft, which also licensed it). It was, in fact, owned by S3 Technologies (which was in turn acquired by VIA Technologies). It was popular with *all* the major graphics players because it was easy to implement in both software *and* hardware (S3, even though it owned the patent, wasn't even the first to implement it in hardware; that honor went to ATI Technologies with the original 3D Rage graphics accelerator - on the software side, Microsoft licensed it in their DirectX runtimes, where it still appears today). What drives the patent-haters round the twist is that it has held on for so long.

        Comment


        • #34
          Originally posted by Qaridarium View Post
          really no one care about the DRM(the bad one) stuff and no i don't care about UVD video acceleration because my cpu is fast to decode any video.

          really drm/UVD don't care.

          but the Patent stuff really hurt.

          to not have openGL3/4 support because of Patent stuff (s3tc,floadingpoint) really shots down any hope for an better means opensource world.

          be sure many more people think like this DRM(the bad) and UVD don't care but s3tc and floadingpoint graphics really care
          I know your just stating that you don't care about UVD because your cpu might be fast enough, but just putting it out there, there might be other people who care about UVD if their cpu isn't as capable as yours.....

          Comment


          • #35
            Originally posted by Quaridarium
            really no one care about the DRM(the bad one) stuff and no i don't care about UVD video acceleration because my cpu is fast to decode any video.
            Just what kind of viewpoint is this? What about people with low-end or small scale machines? Then they are forced to use windose to view html5 VP8 HD or HD flash content and its OK with you? So here is your answer to them when they ask for this support:

            "No way pal, you can't accelerate HD video on this OS and AMD/ATI hardware because your system is a low-end piece of shit and my CPU can kick any ass you find on the market, so either upgrade it or get the hell out of my sight and use your windose to play your videos."

            Comment


            • #36
              "No way pal, you can't accelerate HD video on this OS and AMD/ATI hardware because your system is a low-end piece of shit and my CPU can kick any ass you find on the market, so either upgrade it or get the hell out of my sight and use your windose to play your videos."
              And this is where your opensource driver love FAILS to embrace the linux community.. This is the point where opensource drivers are more proprietary than nvidia's WORKING binary blob...

              Comment


              • #37
                Originally posted by Danny View Post
                I know your just stating that you don't care about UVD because your cpu might be fast enough, but just putting it out there, there might be other people who care about UVD if their cpu isn't as capable as yours.....
                on an long therm the other just buy an faster cpu in the future.

                means no one care about UVD on an long therm of thinking.

                Comment


                • #38
                  Originally posted by glxextxexlg View Post
                  Just what kind of viewpoint is this? What about people with low-end or small scale machines? Then they are forced to use windose to view html5 VP8 HD or HD flash content and its OK with you? So here is your answer to them when they ask for this support:

                  "No way pal, you can't accelerate HD video on this OS and AMD/ATI hardware because your system is a low-end piece of shit and my CPU can kick any ass you find on the market, so either upgrade it or get the hell out of my sight and use your windose to play your videos."
                  this is only right if you have an very low distance view in the future.

                  the speed doubles every 2 years and yes also on the lowend systems.

                  means your point is NOW right but in 2-4 years you are wrong.

                  and hey the UVD unit can not handle VP8/webM means your statement is pointless for HTML5+webM

                  Comment


                  • #39
                    I think Q what more pointing out what was important for him, and just him, not for everybody in general.
                    I personally would rather have the ability for open source driver OpenGL 3/4 before video decode (ok, ideally, both at the same time, but priorities are priorities), but if (read: when) I have an interest in using AMD's APUs in an htpc, then I'd probably want video decode first.

                    Comment


                    • #40
                      Originally posted by glxextxexlg View Post
                      And this is where your opensource driver love FAILS to embrace the linux community.. This is the point where opensource drivers are more proprietary than nvidia's WORKING binary blob...
                      The linux community? not the Xorg community? What about the wayland community?

                      Comment


                      • #41
                        Originally posted by Qaridarium
                        this is only right if you have an very low distance view in the future. the speed doubles every 2 years and yes also on the lowend systems.
                        yes they double and so the resolution and codec complexity of the videos!
                        and hey the UVD unit can not handle VP8/webM means your statement is pointless for HTML5+webM
                        hey, who said new chips won't incorporate this support?

                        Originally posted by mirv
                        but if (read: when) I have an interest in using AMD's APUs in an htpc, then I'd probably want video decode first.
                        All the web is moving towards HD content, even the cheapest and smallest linux devices can do it if they have nvidia hardware onboard and if you think video acceleration is only for htpc then discussion is pointless.
                        Just tell me the name of your planet and... naah not again :P

                        Comment


                        • #42
                          Originally posted by glxextxexlg View Post
                          yes they double and so the resolution and codec complexity of the videos!


                          hey, who said new chips won't incorporate this support?



                          All the web is moving towards HD content, even the cheapest and smallest linux devices can do it if they have nvidia hardware onboard and if you think video acceleration is only for htpc then discussion is pointless.
                          Just tell me the name of your planet and... naah not again :P
                          Where, and really please point this out, did I even remotely mention that video accel was only for htpc's? And please, refrain from the insults, or you might just bite off more than you can chew.

                          Comment


                          • #43
                            Originally posted by glxextxexlg View Post
                            yes they double and so the resolution and codec complexity of the videos!
                            i don't think so. on 1900X1200 pixels with steroscoopic view on 60hz peer eye on an long therm of period you will not get any better than that in the mass market.

                            be sure. today you only get 1650x1050 and only 24hz peer eye and mostly no stereoscoopic view.

                            double the resolution is a no way on the mass market.

                            and codec complexity no i don't think so mpeg2 is used today LOL and h264 is years old on an long time of period you will net get any better than h264-stereoscoopic view.

                            and i think VP8 is less complex than h264-





                            Originally posted by glxextxexlg View Post
                            hey, who said new chips won't incorporate this support?
                            yes sure they are but its pointless thats because VP8 is fast on normal cpus.




                            Originally posted by glxextxexlg View Post
                            All the web is moving towards HD content, even the cheapest and smallest linux devices can do it if they have nvidia hardware onboard and if you think video acceleration is only for htpc then discussion is pointless.
                            Just tell me the name of your planet and... naah not again :P
                            right but HD content means low quality 1650x1050 pixels at 24fps with no stereoscoopic view.

                            on an long period you will not get any better than 1900x1200 pixels at 60fps with stereoscoopic view.

                            means no you don't get 4K resolutions with 100hz and multidimmensionalstereoscoopic views

                            means no.. every bullshit cpu in 2 years can handle this shit .

                            Comment


                            • #44
                              I really don't see the purpose of this thread other than flame baiting

                              Comment


                              • #45
                                Originally posted by Qaridarium
                                right but HD content means low quality 1650x1050 pixels at 24fps with no stereoscoopic view.
                                on an long period you will not get any better than 1900x1200 pixels at 60fps with stereoscoopic view.
                                means no you don't get 4K resolutions with 100hz and multidimmensionalstereoscoopic views
                                means no.. every bullshit cpu in 2 years can handle this shit .
                                What happened to superscalar thinking? Are we still in the times of this chip?



                                What if the user not only watches a video but does something like this at the same time?

                                http://craftymind.com/factory/html5v...nvasVideo.html

                                Destroy the video for several times and see the CPU cycles cry. It gets near the computation limit of my Athlon 64 X2 5600+ which is clearly a high-end class hardware. And this is only a low-res video
                                This is why video acceleration is so important. People won't be just watching the video but interacting with it and offloading video decode to the gpu is a good idea.
                                Your "low-end system does everything" theory can break easily with emerging technology so be careful with it...

                                Comment

                                Working...
                                X