Announcement

Collapse
No announcement yet.

NVIDIA Releases Standalone VDPAU Library

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

  • NVIDIA Releases Standalone VDPAU Library

    Phoronix: NVIDIA Releases Standalone VDPAU Library

    While NVIDIA developed VDPAU (the Video Decode and Presentation API for Unix, one awesome way of accelerating HD video playback with great results) for use in their proprietary graphics driver, the API itself is open and has been well adopted by multimedia applications. VDPAU has worked out so well and has received critical mass that there is a VDPAU back-end for Intel's VA-API and work is underway on bringing VDPAU support directly to Intel's graphics driver. VDPAU may one day end up being used in other open-source drivers too...

    http://www.phoronix.com/vr.php?view=NzU0MA

  • #2
    So what are the odds of having VDPAU support in xorg-driver-ati one day in the future?

    Comment


    • #3
      Originally posted by phoronix View Post
      [VDPAU has worked out so well and has received critical mass that there is a VDPAU back-end for Intel's VA-API and work is underway on bringing VDPAU support directly to Intel's graphics driver.
      Actually, MPEG-2 and H.264 video decode acceleration for Intel G45 (GMA4500HD et al.) is now being developed as VA API driver.

      Comment


      • #4
        Originally posted by gbeauche View Post
        Actually, MPEG-2 and H.264 video decode acceleration for Intel G45 (GMA4500HD et al.) is now being developed as VA API driver.
        Since according to Wikipedia you can use VDPAU as an VA API backend, that should not make a significant difference, I think...

        Comment


        • #5
          Originally posted by nanonyme View Post
          Since according to Wikipedia you can use VDPAU as an VA API backend, that should not make a significant difference, I think...
          This means you need an application supporting VA API to use it.

          Comment


          • #6
            Originally posted by gbeauche View Post
            This means you need an application supporting VA API to use it.
            Well, if everything had gone according to plans, all programs would now be supporting VA API, not VDPAU. Someone messed stuff up by starting to implement other frontends in video players.

            Comment


            • #7
              Originally posted by DoDoENT View Post
              So what are the odds of having VDPAU support in xorg-driver-ati one day in the future?
              Last time I checked, the documentation released by AMD lacked any info for the video decoder. Also, as Bridgman pointed out earlier, the early HD Radeons had the acceleration implemented in a way such that documenting it would compromise HDCP (or something along those lines).

              I'd say video acceleration for R600/R700 is simply not coming to opensource drivers (not from official sources), we might see it in fglrx if we're lucky.

              Such is the sorry state of linux graphics - my choices are intel (drivers in quantum state, getting all the features to work with adequate performance is next to impossible for a casual Linux user), AMD (no opensource 3D for cards less than 4 years old, proprietary driver has problems with some basic functionality (Xv) and switching between computer power modes), nvidia (no opensource driver worth using yet, proprietary driver mostly works, but lacks some common features (XR&R1.2), and there's always the creeping shadow of nvidia's 'Bumpgate'). Other graphics vendors have no drivers/hardware worth using.

              Comment


              • #8
                AMD (no opensource 3D for cards less than 4 years old
                There is opensource 3D for all currently shipping ATi cards. Your info is too old.

                It's not at the level of their binary right now, but many advanced games work (stuff like Nexuiz).

                Comment


                • #9
                  Originally posted by pingufunkybeat View Post
                  There is opensource 3D for all currently shipping ATi cards. Your info is too old.

                  It's not at the level of their binary right now, but many advanced games work (stuff like Nexuiz).
                  That's not entirely true. Although there has been some progress on the 3D support for the latest chipsets, it's not even close to being useful/stable right now.

                  Comment


                  • #10
                    I run open drivers on my rv710 at home (chipset released this year, I believe), play OpenArena and neverball for hours on end, and have all the 3d effects in KWin enabled, and I have no issues right now. So for me, it's both useful and stable, although it won't reach feature parity with fglrx for quite a while.

                    I'm sure that experiences will differ, though, as the rate of development is really fast. But I disagree that there is no OSS 3D for any card less than 4 years old.

                    The AMD/XOrg devs have been doing great work recently.

                    Comment


                    • #11
                      Originally posted by myxal View Post
                      I'd say video acceleration for R600/R700 is simply not coming to opensource drivers (not from official sources), we might see it in fglrx if we're lucky.
                      It's a damn shame.

                      Originally posted by myxal View Post
                      Such is the sorry state of linux graphics - my choices are intel (drivers in quantum state, getting all the features to work with adequate performance is next to impossible for a casual Linux user), AMD (no opensource 3D for cards less than 4 years old, proprietary driver has problems with some basic functionality (Xv) and switching between computer power modes), nvidia (no opensource driver worth using yet, proprietary driver mostly works, but lacks some common features (XR&R1.2), and there's always the creeping shadow of nvidia's 'Bumpgate'). Other graphics vendors have no drivers/hardware worth using.
                      That sounds about right. I wish someone would write an article (HEY PHORONIX, I'M TALKING TO YOU! on the lines of "The Sorry State of Sound on Linux". At least no one has figured out a way to integrate PulseAudio into the X stack to mess things up

                      Comment


                      • #12
                        Originally posted by myxal View Post
                        I'd say video acceleration for R600/R700 is simply not coming to opensource drivers (not from official sources), we might see it in fglrx if we're lucky.
                        I don't believe that that is an entirely accurate statement. There are different levels of video acceleration... the difference is in how much of the decode process is accelerated. Right now we DO have acceleration -- though only very basic Xv. Playing a full-HD video right now *does* peg any CPU that isn't at least a fairly recent 2-core or better. Offloading a -- lets call it a -- "significant chunk" over to the GPU (even without using the video decoder junk in the GPU) will take a significant chunk of the processing off the CPU to hopefully make HD playback stable on even older 2-core processors (maybe even 1-core's).

                        Now the question you need to ask yourself is this: how much acceleration do you really need? My "tv computer" is an older X2-3800 that I recently picked up for free + an RHD3650 ($40). HD video playback goes like this;
                        720P single threaded: fairly OK with the occasional chop. Very watchable.
                        720P multi-threaded: perfect.
                        1080P single threaded: unwatchable, drops about 50%.
                        1080P multi-threaded: fairly OK with the occasional chop. About the same as 720P single threaded.

                        So how much acceleration do *I* need on this "$40" computer to make 1080P perfect? The answer is *not much*. And that's on old junk.

                        Here's what bridgman has to say about video decode acceleration:
                        http://www.phoronix.com/forums/showp...69&postcount=3

                        Comment


                        • #13
                          I think you can take it for granted that video acceleration is coming to open source drivers. While we're not sure yet about UVD, there is already work being done on a shader-based acceleration stack. Cooper will be including ymanton's XvMC-over-Gallium3D code as one of the test cases for the 300g Gallium3D driver, and zrusin is planning to integrate that XvMC code into the xorg state tracker as well. Once XvMC is working all the key bits of plumbing will be there and adding support for additional video standards (or APIs) will not require much in the way of hardware-specific knowledge.

                          Even moving MC (the largest consumer of CPU cycles) from CPU to GPU is likely to save enough CPU cycles that one CPU core will be enough for most users.

                          Hey, wasn't this thread supposed to be about the new VDPAU wrapper library ?
                          Last edited by bridgman; 09-18-2009, 12:18 PM.

                          Comment


                          • #14
                            Originally posted by bridgman View Post
                            I think you can take it for granted that video acceleration is coming to open source drivers. While we're not sure yet about UVD, there is already work being done on a shader-based acceleration stack.
                            Are you saying we might see UVD, i.e. bitstream acceleration in opensource drivers?
                            Originally posted by bridgman View Post
                            Cooper has been working on the 300g Gallium3D driver and will be including ymanton's XvMC-over-Gallium3D code as one of the test cases, and zrusin is planning to integrate that XvMC code into the xorg state tracker as well. Once XvMC is working all the key bits of plumbing will be there and adding support for additional video standards (or APIs) will not require much in the way of hardware-specific knowledge.
                            Sounds great. I recall there being some limitations on XvMC. Going straight to what I care about and need the stack to provide (note: according to reports on the web, VDPAU with nvidia does this): Postprocessing of the decoded video frames, needed to support current mplayer's implementation of subtitles, OSD, etc. Does XvMC even allow this?
                            Originally posted by bridgman View Post
                            Even moving MC (the largest consumer of CPU cycles) from CPU to GPU is likely to make the difference between one core not being sufficient to one core being enough for most users.
                            The fad now is mobility - how does the power draw compare when using UVD and when using shaders?
                            Originally posted by bridgman View Post
                            Wasn't this thread supposed to be about NVidia's new wrapper library ?
                            Well the library is a wrapper for various implementations and we already know nvidia's implementation (mostly) works. We're just THAT eager to see other implementations, working with hardware unaffected by Bumpgate

                            Comment


                            • #15
                              Originally posted by lbcoder View Post
                              I don't believe that that is an entirely accurate statement. There are different levels of video acceleration... the difference is in how much of the decode process is accelerated. Right now we DO have acceleration -- though only very basic Xv. Playing a full-HD video right now *does* peg any CPU that isn't at least a fairly recent 2-core or better. Offloading a -- lets call it a -- "significant chunk" over to the GPU (even without using the video decoder junk in the GPU) will take a significant chunk of the processing off the CPU to hopefully make HD playback stable on even older 2-core processors (maybe even 1-core's).

                              Now the question you need to ask yourself is this: how much acceleration do you really need? My "tv computer" is an older X2-3800 that I recently picked up for free + an RHD3650 ($40). HD video playback goes like this;
                              720P single threaded: fairly OK with the occasional chop. Very watchable.
                              720P multi-threaded: perfect.
                              1080P single threaded: unwatchable, drops about 50%.
                              1080P multi-threaded: fairly OK with the occasional chop. About the same as 720P single threaded.

                              So how much acceleration do *I* need on this "$40" computer to make 1080P perfect? The answer is *not much*. And that's on old junk.

                              Here's what bridgman has to say about video decode acceleration:
                              http://www.phoronix.com/forums/showp...69&postcount=3
                              You make a good point here. We shouldn't spend more than 50 bucks if all you want is to watch HD content.

                              I think the problem is with people that spent 150 or more and want to get the most out of their hardware.

                              Comment

                              Working...
                              X