Announcement

Collapse
No announcement yet.

XvMC support

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

  • bridgman
    started a topic XvMC support

    XvMC support

    Moving this discussion out of the open source 6xx/7xx thread...

    Originally posted by Dieter View Post
    Look at

    http://www.x.org/wiki/RadeonFeature

    XvMC is listed as TODO across the board. When can we expect to
    see FLOSS AMD/ATI drivers supporting XvMC?
    There hasn't been much interest in XvMC so far -- general feeling seems to be that even a laptop CPU can handle MPEG2 decoding well enough. There seem to be an increasing number of HD MPEG2 use cases, and we have already released enough information to implement MC on 5xx (and, as of today, 6xx/7xx I guess) but nobody has even asked how to implement it, which surprises me. We have IDCT on the list of hardware to try and open up, but given the lack of interest in MC it doesn't seem like a real priority (MC eats more CPU time than IDCT and is probably easier to implement).

    If the issue is simply that not enough developers know how to implement XvMC then we could probably put together a sample implementation to get things started, but nobody seems to even ask about XvMC let alone show any interest in implementing it. I guess the issue is that the only place XvMC really buys you much these days is playing HD resolution MPEG2 streams, typically from off-the-air HDTV (ATSC, DVB), and not many people seem to do that.

    EDIT -- I might have found the answer. The "classic use case" for XvMC was European digital TV, which was heavily standardized on MPEG2 at HD resolutions. Looks like many countries have already jumped ship to MPEG4 for most of their HD broadcasts, so the demand for HD MPEG2 acceleration seems to have evaporated. Given that, I think interest in XvMC is going to continue to be lukewarm until there is some agreement on an API which cleanly handles H.264 and VC-1 as well, whether it be an XvMC extension or something new like VAAPI, XVBA or VDPAU.

    Originally posted by Dieter View Post
    > There hasn't been much interest in XvMC so far

    Perhaps you missed Phoronix's "2008 Linux Graphics Survey"?

    The #1 activity was video playback, listed by 37.90%

    "the second area with the most interest is seeing video improvements to X.Org"
    Chart says 21.26%

    Video playback/acceleration was listed by 15.09%

    "When it comes to video playback improvements, the leading open-source solution
    right now is through XvMC"

    A related feature, TV-out was listed by 19.6%

    http://www.phoronix.com/scan.php?pag..._results&num=1

    > I guess the issue is that the only place XvMC really buys you much these
    > days is playing HD resolution MPEG2 streams, typically from off-the-air
    > HDTV (ATSC, DVB), and not many people seem to do that.

    Does the date 2009-02-17 mean anything to you? Do you actually think that
    not many people watch TV?

    So there is need/interest. When will FLOSS drivers accelerate mpeg decoding
    on AMD/ATI GPUs?

  • duby229
    replied
    I understand completely. I can only wish that the right thing be done, but just like with everything else there is always some excuse as to why it cant be done. Its OK though, maybe someday when the content industry fails they'll realize they've been doing it all wrong, and give you guys the opportunity to finally document the technology they've been trying to push all along.

    Leave a comment:


  • bridgman
    replied
    Sure, but if we could do that then we would have already released the information to the open source development community.

    Leave a comment:


  • duby229
    replied
    Originally posted by pingufunkybeat View Post
    The fglrx drivers are not open-source, so putting the code in a public git is not really an option.
    And that is exactly what the problem is. Putting that code in the public under an open source agreement would solve this issue once and for all.

    Leave a comment:


  • Ex-Cyber
    replied
    Originally posted by Kano View Post
    The lib is there, just not the headers. What would be hidden in the headers?!
    An implicit commitment to continue supporting the interface?

    Leave a comment:


  • Kano
    replied
    The lib is there, just not the headers. What would be hidden in the headers?!

    Leave a comment:


  • pingufunkybeat
    replied
    The fglrx drivers are not open-source, so putting the code in a public git is not really an option.

    Leave a comment:


  • duby229
    replied
    Originally posted by bridgman View Post
    If you're saying that we should have hidden the work-in-process XvBA code better so that people would not speculate maybe that's fair. If you're saying that we should have taken legal action to stop people from publicly speculating, that's not the way we like to do things.
    He might be saying that, but I'm definitely not. I think the work in progress code should be made available in a public git so that we can --see-- what work is being done. This is the only way to make it fair. If we could see the code then there wouldnt be any speculation.

    Leave a comment:


  • bridgman
    replied
    Originally posted by Zajec View Post
    That explains your disappointment, but as you noticed by yourself, that has nothing to Bridgman's information. Where did you read this UVD announcement actually?
    I would like a copy as well, please

    The only UVD-related announcement we made was the statement that we did *not* (and I stress *not*) plan to open up UVD, however I did commit to looking into whether it was possible after we had the other core functionality in place.

    On the fglrx side, Michael drew some perfectly reasonable conclusions from what he saw in the driver binaries and heard from his contacts around the industry; even if things didn't work out that way this time in general his guesses are quite good.

    If you're saying that we should have hidden the work-in-process XvBA code better so that people would not speculate maybe that's fair. If you're saying that we should have taken legal action to stop people from publicly speculating, that's not the way we like to do things.
    Last edited by bridgman; 07-18-2009, 11:04 AM.

    Leave a comment:


  • Zajec
    replied
    Originally posted by mirak63 View Post
    at the time I bought a hd4850 ATI was claiming linux support, and also UVD support was announced.

    now we know what it was worth

    It's not about bridgman here, however I think the less ati talks about things that ati can't support right now or officialy at some point, the less arm it will do for ati image.

    And probably phoronix didn't do anygood by anouncing stuffs based on the fact some lib was seen in some binary
    That explains your disappointment, but as you noticed by yourself, that has nothing to Bridgman's information. Where did you read this UVD announcement actually?

    Leave a comment:

Working...
X