Kodi Community Forum

Full Version: VDPAU API for Linux released by NVIDIA today - GPU hardware accelerated video decoder
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Oooh, some benchmarks posted! http://www.phoronix.com/scan.php?page=ar...dpau&num=1

Looks like VC1 and WMV3 acceleration aren't on all cards. Do we care about those accelerations so much? If so might need to get Davilla a better card? Any other devs looking to work on this as well?

Note that they clocked the Intel CPU down to just 1.8ghz and got rock bottom CPU usage. Apparently only a few profiles are supported so much work to be done but WOW I'd love to have a cooler quieter machine for my next build! I wonder if they had the CABAC stuff working too....

Edit: Hrm, wonder if an ATOM could now be a viable HD capable CPU?Shocked
BLKMGK Wrote:Looks like VC1 and WMV3 acceleration aren't on all cards. Do we care about those accelerations so much? If so might need to get Davilla a better card? Any other devs looking to work on this as well?

Could you post a link to that information?
Supported list is here:
http://www.nvnews.net/vbulletin/showthread.php?t=123091

I can say that here (c2q q6600, 8800gt) for the samples that work, almost 100% of the decoding is offloaded.

When playing a 1080P sample of Gladiator, cpu usage sits at < 1%. Compare that to ~80% without the vdpau.

VERY few movies actually work, though I'm sure this will be remedied soon.

I think the possibility of smooth 1080p playback with a VERY lightweight CPU and a decent nvidia chipset >= 8xxx is a reasonable one.

TheUni
I wonder if the 9300 and 9400 motherboards support video vc1.
malloc Wrote:Could you post a link to that information?

Inferred from this statement -> http://www.phoronix.com/forums/showpost....stcount=12 and I thought I saw it elsewhere in some of the NVIDIA posts - something to the effect that some cards would have more functions or somesuch but looking back I'm having trouble finding that statement. This morning I pretty much dug all over the place for info on this so no telling where it was or if I understood it correctly but that was what I inferred anyway.

Edit: found more info -> http://www.nvnews.net/vbulletin/showpost...ostcount=8 Further on I found that the G92 based 8800GT cards should support MPEG but not the older G82 based cards. VC-1 will not work on G92 -> http://www.nvnews.net/vbulletin/showpost...stcount=19

H.264 is my primary focus anyway to be honest, I recode everything into it and it looks goood! To answer your second question, yeah I think a 9300 would do itNod

If a 9x base is decided to be a better choice for coding I might be willing to pop for one of these -> http://www.newegg.com/Product/Productcom...6814261031 as really they aren't much different in price. Figures there's no low profile that's fanless though - grr!

Mmmm, an Atom or Celeron based low powered XBMC box... /me likes! Nod

P.S. I see someone updated ffmpeg in the SVN, has someone begun adding these patches in? sadly I cannot access the Sourceforge log - 502 Bad Gateway error... If so do we need to update to the beta video driver (shiver) to testHuh?
theuni Wrote:Supported list is here:
http://www.nvnews.net/vbulletin/showthread.php?t=123091

I can say that here (c2q q6600, 8800gt) for the samples that work, almost 100% of the decoding is offloaded.

When playing a 1080P sample of Gladiator, cpu usage sits at < 1%. Compare that to ~80% without the vdpau.

VERY few movies actually work, though I'm sure this will be remedied soon.

I think the possibility of smooth 1080p playback with a VERY lightweight CPU and a decent nvidia chipset >= 8xxx is a reasonable one.

TheUni

The reasoning for this is that apparently only one H.264 "profile" has been accelerated to date. I believe I picked that tidbit up from the Pharonix testing or from one of the posts in their forums that naturally I cannot find at the moment. From the sounds of it they accelerated one profile as an example and are maybe hoping others will work out patches for others? I believe the max profile was 4.1? I'm not "up" on the nomenclature but am pretty sure that was what I'd read and that hopefully more will be coming as people tune up on this? The ffmpeg guys sound a bit critical of the patches <shrug>

Some additional discussions: http://www.nvnews.net/vbulletin/showthread.php?t=123095 and http://lists.mplayerhq.hu/pipermail/ffmp...56054.html
theuni Wrote:Supported list is here:
http://www.nvnews.net/vbulletin/showthread.php?t=123091

I can say that here (c2q q6600, 8800gt) for the samples that work, almost 100% of the decoding is offloaded.

When playing a 1080P sample of Gladiator, cpu usage sits at < 1%. Compare that to ~80% without the vdpau.

VERY few movies actually work, though I'm sure this will be remedied soon.

I think the possibility of smooth 1080p playback with a VERY lightweight CPU and a decent nvidia chipset >= 8xxx is a reasonable one.

TheUni

Are the Linux patches for this going to be usable with the Windows platform?
Livin Wrote:Are the Linux patches for this going to be usable with the Windows platform?

Absolutely, without a doubt, no. This is a UNIX only API. Windows has a separate API, which is why I said before, OpenGL is still the best cross platform, cross GPU way to do this. The last thing we want to do is lock ourselves into one graphics vendors proprietary API which they could abandon at any point.
Here's a potentially useful PDF -> http://www.nvidia.com/object/IO_43029.html

Good question on the patches also supporting Winders. Hrm...
BLKMGK Wrote:Here's a potentially useful PDF -> http://www.nvidia.com/object/IO_43029.html

Useless. Doesn't talk about 9xxx series and this is for PureVideo (Windows), not VDPAU (UNIX).
No, it doesn't talk about the 9x series but it may give some indication as to what the 8x series does. I realize that it's Windows geared but if the hardware doesn't exist for Windows then it certainly isn't going to be there for Linux either. It does suck that it doesn't cover the 9x series however so I will keep reading. So far that mplayer discussion seems to have some decent information - I'm surprised the ffmpeg list is as quiet about this as it is. <sigh>

Oh and the NVIDIA guys seem to be active on the mplayer discussion too which is nice!
theuni Wrote:Supported list is here:
http://www.nvnews.net/vbulletin/showthread.php?t=123091

What does that say about your reading abilities?
You expect me to remember 6hrs ago?
BLKMGK Wrote:The reasoning for this is that apparently only one H.264 "profile" has been accelerated to date. I believe I picked that tidbit up from the Pharonix testing or from one of the posts in their forums that naturally I cannot find at the moment. From the sounds of it they accelerated one profile as an example and are maybe hoping others will work out patches for others? I believe the max profile was 4.1? I'm not "up" on the nomenclature but am pretty sure that was what I'd read and that hopefully more will be coming as people tune up on this? The ffmpeg guys sound a bit critical of the patches <shrug>

Just FYI -- I dual-boot my XBMC box (Nvidia 8400) to Vista (very rarely) and have the GPU decoding working fine for x264 mkv files. While 720p has worked just fine, it doesn't support any "scene" 1080p releases due to some of the encoding options used. So --- if during any testing of GPU decoding under linux some stuff doesn't play, it may be a limitation of the GPU and not the linux implementation.
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28