Login | Register For Free | Help
Search for: (Advanced)

Mailing List Archive: MythTV: Users

FYI - New HW video accellAPI from NVIDIA

 

 

First page Previous page 1 2 3 Next page Last page  View All MythTV users RSS feed   Index | Next | Previous | View Threaded


nico at youplala

Nov 14, 2008, 10:50 AM

Post #1 of 65 (8158 views)
Permalink
FYI - New HW video accellAPI from NVIDIA

On Fri, 2008-11-14 at 10:28 -0800, Andy Ritger wrote:
> I'm pleased to announce a new video API for Unix and Unix-like platforms,
> and a technology preview implementation of this API from NVIDIA.
>
> The API is called VDPAU (Video Decode and Presentation API for Unix).
>
> The current API documentation is here:
>
> ftp://download.nvidia.com/XFree86/vdpau/doxygen/html/index.html
>
> Some highlights of VDPAU:
>
> * Defines an API for GPU-accelerated decode of MPEG-1, MPEG-2, H.264, and
> VC-1 bitstreams.
> * Defines an API for post-processing of decoded video, including
> temporal and spatial deinterlacing, inverse telecine, and noise
> reduction.
> * Defines an API for timestamp-based presentation of final video
> frames.
> * Defines an API for compositing sub-picture, on-screen display,
> and other UI elements.
>
> Note that VDPAU does not address content protection.
>
> Some highlights/limitations of NVIDIA's current implementation:
>
> * Supported on NVIDIA GPUs with the NVIDIA second generation video
> processors (see the end of this announcement for a complete GPU list).
> * Currently, only one video stream can be decoded at a time; we hope
> to lift this restriction eventually.
> * Available in the 180.06 NVIDIA public beta release:
> http://www.nvidia.com/object/linux_display_ia32_180.06.html
> http://www.nvidia.com/object/linux_display_amd64_180.06.html
> http://www.nvidia.com/object/freebsd_180.06.html
> http://www.nvidia.com/object/solaris_display_180.06.html
>
> The VDPAU support in the NVIDIA 180.06 beta release is still very
> preliminary. We are aware of cases of visual corruption and in some
> cases GPU hangs. We will be working on these issues over the next
> several NVIDIA driver releases.
>
> While NVIDIA's VDPAU implementation is not ready for end user use yet,
> it should be far enough along that interested application developers
> can begin working with it.
>
> Additionally, NVIDIA has developed patches to ffmpeg and MPlayer to
> demonstrate a video player using VDPAU:
>
> ftp://download.nvidia.com/XFree86/vdpau/mplayer-vdpau-3076399.tar.bz2
>
> These patches include changes against libavcodec, libavutil, ffmpeg,
> and MPlayer itself; they may serve as an example of how to use VDPAU.
>
> Once we do some further testing, bugfixing, and cleanup, we will
> contribute the MPlayer patches to the MPlayer developers.
>
>
> If other hardware vendors are interested, they are welcome to also
> provide implementations of VDPAU. The VDPAU API was designed to allow
> a vendor backend to be selected at run time.
>
>
> Thanks,
> Andy Ritger
> Manager, NVIDIA Linux Graphics Driver
>
>
>
> VDPAU is currently supported on the following NVIDIA GPUs:
>
> Desktop GPUs:
> GeForce 200 Series
> GeForce 9 Series
> GeForce 86xx Series
> GeForce 85xx Series
> GeForce 84xx Series
> GeForce 8800 GTS 512
> GeForce 8800 GT
> GeForce 8800 GS
>
> Mobile GPUs:
> GeForce 98xxM
> GeForce 9700M
> GeForce 96xxM
> GeForce 9500M
> GeForce 9300M
> GeForce 9200M
> GeForce 8800M
> GeForce 8800M GTS
> GeForce 8800M GTX
> GeForce 8600M
>
> Motherboard GPUs:
> GeForce 9400
> GeForce 9300
> GeForce 9100
> GeForce 8300
> GeForce 8200
>
> VC-1 support in NVIDIA's VDPAU implementation currently requires GeForce
> 9300 GS, GeForce 9200M GS, GeForce 9300M GS, or GeForce 9300M GS.
>
> _______________________________________________
> xorg mailing list
> xorg [at] lists
> http://lists.freedesktop.org/mailman/listinfo/xorg

_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


chris at westnet

Nov 14, 2008, 11:32 AM

Post #2 of 65 (8079 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

Some additional information:

http://www.phoronix.com/scan.php?page=article&item=nvidia_180_vdpau&num=1


==========================================================
Chris Candreva -- chris [at] westnet -- (914) 948-3162
WestNet Internet Services of Westchester
http://www.westnet.com/
_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


travis at tabbal

Nov 14, 2008, 12:03 PM

Post #3 of 65 (8083 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

Considering the cost of the supported GPUs, it seems that the 3Ghz dual core
CPUs are the way to go for now.

Now, if it supported older mobo chipsets, like the 61xx series, it could be
quite useful for many of us. I see no need to put a powerful and expensive
graphics card in my frontend when the CPU can do it just fine for about a
$40-$50 premium over the CPU I would need anyway.


nico at youplala

Nov 14, 2008, 12:10 PM

Post #4 of 65 (8062 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

On Fri, 2008-11-14 at 13:03 -0700, Travis Tabbal wrote:
> Considering the cost of the supported GPUs, it seems that the 3Ghz
> dual core CPUs are the way to go for now.

A 8400 card is about $30 at Newegg, using passive cooling, without fan.

http://www.newegg.com/Product/Product.aspx?Item=N82E16814121235

and this one is among the expensive models, too much RAM on it.

Nico

_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


nikos.f at gmail

Nov 14, 2008, 12:15 PM

Post #5 of 65 (8061 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

On Fri, Nov 14, 2008 at 12:50 PM, Nicolas Will <nico [at] youplala> wrote:

> On Fri, 2008-11-14 at 10:28 -0800, Andy Ritger wrote:
> > I'm pleased to announce a new video API for Unix and Unix-like platforms,
> > and a technology preview implementation of this API from NVIDIA.
> >
> > VDPAU is currently supported on the following NVIDIA GPUs:
> >
> > Desktop GPUs:
> > GeForce 200 Series
> > GeForce 9 Series
> > GeForce 86xx Series
> > GeForce 85xx Series
> > GeForce 84xx Series
> > GeForce 8800 GTS 512
> > GeForce 8800 GT
> > GeForce 8800 GS
> >
> > Mobile GPUs:
> > GeForce 98xxM
> > GeForce 9700M
> > GeForce 96xxM
> > GeForce 9500M
> > GeForce 9300M
> > GeForce 9200M
> > GeForce 8800M
> > GeForce 8800M GTS
> > GeForce 8800M GTX
> > GeForce 8600M
> >
> > Motherboard GPUs:
> > GeForce 9400
> > GeForce 9300
> > GeForce 9100
> > GeForce 8300
> > GeForce 8200
> >
>
Does anyone know what the AppleTV nvidia chip (nvidia G72M 64MB) maps to -
and whether it will benefit from this API? If the AppleTV chip is supported
would increase it's utility significantly


adeffs.mythtv at gmail

Nov 14, 2008, 12:16 PM

Post #6 of 65 (8081 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

On Fri, Nov 14, 2008 at 3:03 PM, Travis Tabbal <travis [at] tabbal> wrote:
> Considering the cost of the supported GPUs, it seems that the 3Ghz dual core
> CPUs are the way to go for now.
>
> Now, if it supported older mobo chipsets, like the 61xx series, it could be
> quite useful for many of us. I see no need to put a powerful and expensive
> graphics card in my frontend when the CPU can do it just fine for about a
> $40-$50 premium over the CPU I would need anyway.

those older chipsets just don't have the design to handle processing
that data, it's not a matter of driver support.


--
Steve
http://www.jobs-khakis-chicks.com/MythTV/
Before you ask, read the FAQ!
http://www.mythtv.org/wiki/index.php/Frequently_Asked_Questions
then search the Wiki, and this list,
http://www.gossamer-threads.com/lists/mythtv/
Mailinglist etiquette -
http://www.mythtv.org/wiki/index.php/Mailing_List_etiquette
_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


adeffs.mythtv at gmail

Nov 14, 2008, 12:18 PM

Post #7 of 65 (8072 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

On Fri, Nov 14, 2008 at 3:15 PM, Nick F <nikos.f [at] gmail> wrote:
> On Fri, Nov 14, 2008 at 12:50 PM, Nicolas Will <nico [at] youplala> wrote:
>> On Fri, 2008-11-14 at 10:28 -0800, Andy Ritger wrote:
>> > I'm pleased to announce a new video API for Unix and Unix-like
>> > platforms,
>> > and a technology preview implementation of this API from NVIDIA.
>> >
>> > VDPAU is currently supported on the following NVIDIA GPUs:
>> >
>> > Desktop GPUs:
>> > GeForce 200 Series
>> > GeForce 9 Series
>> > GeForce 86xx Series
>> > GeForce 85xx Series
>> > GeForce 84xx Series
>> > GeForce 8800 GTS 512
>> > GeForce 8800 GT
>> > GeForce 8800 GS
>> >
>> > Mobile GPUs:
>> > GeForce 98xxM
>> > GeForce 9700M
>> > GeForce 96xxM
>> > GeForce 9500M
>> > GeForce 9300M
>> > GeForce 9200M
>> > GeForce 8800M
>> > GeForce 8800M GTS
>> > GeForce 8800M GTX
>> > GeForce 8600M
>> >
>> > Motherboard GPUs:
>> > GeForce 9400
>> > GeForce 9300
>> > GeForce 9100
>> > GeForce 8300
>> > GeForce 8200
>> >
>
> Does anyone know what the AppleTV nvidia chip (nvidia G72M 64MB) maps to -
> and whether it will benefit from this API? If the AppleTV chip is supported
> would increase it's utility significantly

Roughly equivalent to the GeForce Go 7300.


--
Steve
http://www.jobs-khakis-chicks.com/MythTV/
Before you ask, read the FAQ!
http://www.mythtv.org/wiki/index.php/Frequently_Asked_Questions
then search the Wiki, and this list,
http://www.gossamer-threads.com/lists/mythtv/
Mailinglist etiquette -
http://www.mythtv.org/wiki/index.php/Mailing_List_etiquette
_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


adeffs.mythtv at gmail

Nov 14, 2008, 12:18 PM

Post #8 of 65 (8063 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

On Fri, Nov 14, 2008 at 3:18 PM, Steven Adeff <adeffs.mythtv [at] gmail> wrote:
> On Fri, Nov 14, 2008 at 3:15 PM, Nick F <nikos.f [at] gmail> wrote:
>> On Fri, Nov 14, 2008 at 12:50 PM, Nicolas Will <nico [at] youplala> wrote:
>>> On Fri, 2008-11-14 at 10:28 -0800, Andy Ritger wrote:
>>> > I'm pleased to announce a new video API for Unix and Unix-like
>>> > platforms,
>>> > and a technology preview implementation of this API from NVIDIA.
>>> >
>>> > VDPAU is currently supported on the following NVIDIA GPUs:
>>> >
>>> > Desktop GPUs:
>>> > GeForce 200 Series
>>> > GeForce 9 Series
>>> > GeForce 86xx Series
>>> > GeForce 85xx Series
>>> > GeForce 84xx Series
>>> > GeForce 8800 GTS 512
>>> > GeForce 8800 GT
>>> > GeForce 8800 GS
>>> >
>>> > Mobile GPUs:
>>> > GeForce 98xxM
>>> > GeForce 9700M
>>> > GeForce 96xxM
>>> > GeForce 9500M
>>> > GeForce 9300M
>>> > GeForce 9200M
>>> > GeForce 8800M
>>> > GeForce 8800M GTS
>>> > GeForce 8800M GTX
>>> > GeForce 8600M
>>> >
>>> > Motherboard GPUs:
>>> > GeForce 9400
>>> > GeForce 9300
>>> > GeForce 9100
>>> > GeForce 8300
>>> > GeForce 8200
>>> >
>>
>> Does anyone know what the AppleTV nvidia chip (nvidia G72M 64MB) maps to -
>> and whether it will benefit from this API? If the AppleTV chip is supported
>> would increase it's utility significantly
>
> Roughly equivalent to the GeForce Go 7300.

oops, forgot to paste the linky
http://en.wikipedia.org/wiki/Nvidia_Quadro


--
Steve
http://www.jobs-khakis-chicks.com/MythTV/
Before you ask, read the FAQ!
http://www.mythtv.org/wiki/index.php/Frequently_Asked_Questions
then search the Wiki, and this list,
http://www.gossamer-threads.com/lists/mythtv/
Mailinglist etiquette -
http://www.mythtv.org/wiki/index.php/Mailing_List_etiquette
_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


travis at tabbal

Nov 14, 2008, 12:38 PM

Post #9 of 65 (8060 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

On Fri, Nov 14, 2008 at 1:10 PM, Nicolas Will <nico [at] youplala> wrote:

> On Fri, 2008-11-14 at 13:03 -0700, Travis Tabbal wrote:
> > Considering the cost of the supported GPUs, it seems that the 3Ghz
> > dual core CPUs are the way to go for now.
>
> A 8400 card is about $30 at Newegg, using passive cooling, without fan.
>
> http://www.newegg.com/Product/Product.aspx?Item=N82E16814121235
>
> and this one is among the expensive models, too much RAM on it.
>


Hmmm... I missed that, thanks for the pointer. It will be interesting to see
how it shakes out as NVidia debugs the driver. If one could use a $30 vid
card and a slowish CPU (1Ghz Single Core?) it would make a pretty cheap
frontend.

At that point, the case might be the most expensive single part. :)


nico at youplala

Nov 14, 2008, 2:01 PM

Post #10 of 65 (8082 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

On Fri, 2008-11-14 at 13:38 -0700, Travis Tabbal wrote:
> On Fri, Nov 14, 2008 at 1:10 PM, Nicolas Will <nico [at] youplala>
> wrote:
> On Fri, 2008-11-14 at 13:03 -0700, Travis Tabbal wrote:
> > Considering the cost of the supported GPUs, it seems that
> the 3Ghz
> > dual core CPUs are the way to go for now.
>
> A 8400 card is about $30 at Newegg, using passive cooling,
> without fan.
>
> http://www.newegg.com/Product/Product.aspx?Item=N82E16814121235
>
> and this one is among the expensive models, too much RAM on
> it.
>
>
> Hmmm... I missed that, thanks for the pointer. It will be interesting
> to see how it shakes out as NVidia debugs the driver. If one could use
> a $30 vid card and a slowish CPU (1Ghz Single Core?) it would make a
> pretty cheap frontend.

I've just build an FE system on Newegg for less than $200 that would be
fully HD compliant with this API:

Western Digital Caviar WD800BB 80GB 7200 RPM IDE Ultra ATA100 Hard Drive
$35.99

ASUS M3N78-VM AM2+/AM2 NVIDIA GeForce 8200 HDMI Micro ATX AMD
Motherboard (DVI+HDMI+Optical audio)
$74.99

Kingston ValueRAM 1GB (2 x 512MB) 240-Pin DDR2 SDRAM DDR2 800 (PC2 6400)
Dual Channel Kit Desktop Memory Model KVR800D2N5K2/1G
$20.99

AMD Sempron 64 3400+ Manila 1.8GHz Socket AM2 62W Single-Core Processor
Model SDA3400IAA3CW
$19.99

Linkworld Black body/ Silver strip Steel 6280-01 Micro ATX Media
Center / HTPC Case with 300W Power Supply
$38.99

Total: $190.95

Add a CPU Fan...

Ugly and probably noisy.

Now I'd like to see an Atom 330 board with PCIe 16x or an NVIDIA
chipset...

When I though that Intel would be forst with H.264 decoding on its 965
chipset, NVIDIA beats them and does more.

Nico

_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


mitchell.gore at gmail

Nov 14, 2008, 2:25 PM

Post #11 of 65 (8051 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

> When I though that Intel would be forst with H.264 decoding on its 965
> chipset, NVIDIA beats them and does more.
>
> Nico
>
> _______________________________________________
>

So, am i reading this right? nVidia should have h.264 hardware
acceleration!? That mean the proc required for the HD-PVR will go
substantial down right and play ripped blu-rays?

Also, I wonder if this has all the negatives of XvMC. Will we get full
color OSD? Will there be buffering/choppy video issues? I assume the
internal player will require patches. yet another reason i may be moving to
trunk when that is released.....

Mitchell


nico at youplala

Nov 14, 2008, 2:35 PM

Post #12 of 65 (8050 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

On Fri, 2008-11-14 at 16:25 -0600, Mitch Gore wrote:
>
> When I though that Intel would be forst with H.264 decoding on
> its 965
> chipset, NVIDIA beats them and does more.
>
>
> Nico
>
> _______________________________________________
>
>
> So, am i reading this right? nVidia should have h.264 hardware
> acceleration!?

yes, h.264, VC-1, MPEG2, and MPEG-1

Add deinterlacing, noise reduction,




> That mean the proc required for the HD-PVR will go substantial down
> right and play ripped blu-rays?

Yeah!


>
> Also, I wonder if this has all the negatives of XvMC. Will we get
> full color OSD?

There is an API for compositing sub-picture, on-screen display, and
other UI elements.

That should help.



> Will there be buffering/choppy video issues?


hmmm, I'm not sure what you mean exactly. I guess not. After all, that's
the goal of HW acceleration.


> I assume the internal player will require patches.

The internal player uses ffmpeg libraries.

NVIDIA has already patched ffmpeg (as well as mplayer) and associated
libs to work with its API.

If MythTV picks up the right code, as it wants to ship all that by
itself (I always wondred why), all should come to MythTV in good time
without much effort (I've been known to be wrong and I am no MythTV
dev).


> yet another reason i may be moving to trunk when that is released.....

That should certainly help the undecided.

Nico


_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


cynical at penguinness

Nov 16, 2008, 2:06 AM

Post #13 of 65 (7934 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

Steven Adeff wrote:
> On Fri, Nov 14, 2008 at 3:03 PM, Travis Tabbal <travis [at] tabbal> wrote:
>> Considering the cost of the supported GPUs, it seems that the 3Ghz dual core
>> CPUs are the way to go for now.
>>
>> Now, if it supported older mobo chipsets, like the 61xx series, it could be
>> quite useful for many of us. I see no need to put a powerful and expensive
>> graphics card in my frontend when the CPU can do it just fine for about a
>> $40-$50 premium over the CPU I would need anyway.
>
> those older chipsets just don't have the design to handle processing
> that data, it's not a matter of driver support.

I have a problem with this idea.

According to the PDF on the Nvidia site[1][2], every PCIe chipset (and
some of the AGP ones as well) back to the GeForce Go 6600 is supported
by PureVideo, which includes "H.264 Decode Acceleration".

Looking further, the only major difference between the 6xxx and 7xxx+
series is support for "MPEG-2 Inverse Telecine" for HD, which is only
supported for the 7600 GT and up.

If NVidia can get it[3] running under windows with older chipsets, I
don't see any reason for the Linux equivalent to be crippled in
comparison. The only reason I can think of is they went for the newer
chipsets, which may have the larger share of the NVidia/Linux market,
never mind that the removed XVMC support from the 8XXX and up chipsets
under Linux (which is the main reason I was looking at NVidia cards when
building my current FE/BE).

IMO, if NVidia pulls their collective heads out and makes this new
driver API about the same as PureVideo when discussing chipset support,
this may do a lot about repairing their bruised reputation within many
FOSS circles.

[1] Which appears to be based on version 158.18 of the windows drivers
[2] http://www.nvidia.com/page/purevideo_support.html
[3] It being H.264 Decode Acceleration and other such things
_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


nico at youplala

Nov 16, 2008, 2:25 AM

Post #14 of 65 (7952 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

On Sun, 2008-11-16 at 02:06 -0800, Justin The Cynical wrote:
> Steven Adeff wrote:
> > On Fri, Nov 14, 2008 at 3:03 PM, Travis Tabbal <travis [at] tabbal> wrote:
> >> Considering the cost of the supported GPUs, it seems that the 3Ghz dual core
> >> CPUs are the way to go for now.
> >>
> >> Now, if it supported older mobo chipsets, like the 61xx series, it could be
> >> quite useful for many of us. I see no need to put a powerful and expensive
> >> graphics card in my frontend when the CPU can do it just fine for about a
> >> $40-$50 premium over the CPU I would need anyway.
> >
> > those older chipsets just don't have the design to handle processing
> > that data, it's not a matter of driver support.
>
> I have a problem with this idea.
>
> According to the PDF on the Nvidia site[1][2], every PCIe chipset (and
> some of the AGP ones as well) back to the GeForce Go 6600 is supported
> by PureVideo, which includes "H.264 Decode Acceleration".
>
> Looking further, the only major difference between the 6xxx and 7xxx+
> series is support for "MPEG-2 Inverse Telecine" for HD, which is only
> supported for the 7600 GT and up.
>
> If NVidia can get it[3] running under windows with older chipsets, I
> don't see any reason for the Linux equivalent to be crippled in
> comparison. The only reason I can think of is they went for the newer
> chipsets, which may have the larger share of the NVidia/Linux market,
> never mind that the removed XVMC support from the 8XXX and up chipsets
> under Linux (which is the main reason I was looking at NVidia cards when
> building my current FE/BE).
>
> IMO, if NVidia pulls their collective heads out and makes this new
> driver API about the same as PureVideo when discussing chipset support,
> this may do a lot about repairing their bruised reputation within many
> FOSS circles.
>
> [1] Which appears to be based on version 158.18 of the windows drivers
> [2] http://www.nvidia.com/page/purevideo_support.html
> [3] It being H.264 Decode Acceleration and other such things

You are talking about PureVideo.

NVIDIA is really talking about PureVideo HD. It looks like the circuitry
on-chip is quite different, thus needs different code.

If the marketing names are about the same, it doesn't mean that the
engineering is.

Nico

_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


mtdean at thirdcontact

Nov 16, 2008, 6:01 AM

Post #15 of 65 (7916 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

On 11/16/2008 05:06 AM, Justin The Cynical wrote:
> If NVidia can get it[3] running under windows with older chipsets, I
> don't see any reason for the Linux equivalent to be crippled in
> comparison. The only reason I can think of is they went for the newer
> chipsets, which may have the larger share of the NVidia/Linux market,
> never mind that the removed XVMC support from the 8XXX and up chipsets
> under Linux (which is the main reason I was looking at NVidia cards when
> building my current FE/BE).

So, how does that hand taste? You know, the one that feeds you (NVIDIA
drivers)?

Just be happy with the fact that they did anything for us. Besides,
have you looked at the prices of the cards on the list of supported
GPU's? We're talking $20 to $30--not to mention the fact that /many/ of
the integrated GPU's are supported, too.

http://www.newegg.com/Product/Product.aspx?Item=N82E16814141068
http://www.newegg.com/Product/Product.aspx?Item=N82E16814121240
http://www.newegg.com/Product/Product.aspx?Item=N82E16814121235

Oh, and there's also the whole, you really don't /need/ GPU-assisted
decode, anyway. My system happily handles US MPEG-2 HDTV. If you're
using an HD-PVR to record HDTV to MPEG-4 AVC from (rented)
cable/satellite STB's (with a monthly cable bill), the extra $20 to $30
you spend to get a supported card is really a negligible part of your TV
cost, but you can always buy a system that can play that back even
without the GPU assist (and the support for HD-PVR output is getting
better every day).

Mike
_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


nico at youplala

Nov 16, 2008, 6:35 AM

Post #16 of 65 (7908 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

On Sun, 2008-11-16 at 09:01 -0500, Michael T. Dean wrote:
>
> Oh, and there's also the whole, you really don't /need/ GPU-assisted
> decode, anyway. My system happily handles US MPEG-2 HDTV. If you're
> using an HD-PVR to record HDTV to MPEG-4 AVC from (rented)
> cable/satellite STB's (with a monthly cable bill), the extra $20 to
> $30
> you spend to get a supported card is really a negligible part of your
> TV
> cost, but you can always buy a system that can play that back even
> without the GPU assist (and the support for HD-PVR output is getting
> better every day).

I'm quite with you, on most part. $30 is cheap compared to the cost of
an HD-capable MythTV system.

But, I differ on your point about not needing something like GPU decode.

Using the generic CPU works if you go the expensive way. This also means
noise on the FE, as the CPU fan will start to spin up. Agreed, if you
watch HD, you will probably have some sound covering the fans, this is
what happens at my place.

But going with a cheaper CPU, a cheap GPU that can assist HD decoding
gives less noise and less money spent.

An for most, why waste a good CPU when there is a GPU than can do the
same job with less efforts, heat and noise?

Nico

_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


mtdean at thirdcontact

Nov 16, 2008, 10:36 AM

Post #17 of 65 (7910 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

On 11/16/2008 09:35 AM, Nicolas Will wrote:
> On Sun, 2008-11-16 at 09:01 -0500, Michael T. Dean wrote:
>
>> Oh, and there's also the whole, you really don't /need/ GPU-assisted
>> decode, anyway. My system happily handles US MPEG-2 HDTV. If you're
>> using an HD-PVR to record HDTV to MPEG-4 AVC from (rented)
>> cable/satellite STB's (with a monthly cable bill), the extra $20 to
>> $30
>> you spend to get a supported card is really a negligible part of your
>> TV
>> cost, but you can always buy a system that can play that back even
>> without the GPU assist (and the support for HD-PVR output is getting
>> better every day).
>>
> I'm quite with you, on most part. $30 is cheap compared to the cost of
> an HD-capable MythTV system.
>
> But, I differ on your point about not needing something like GPU decode.
>
> Using the generic CPU works if you go the expensive way. This also means
> noise on the FE, as the CPU fan will start to spin up. Agreed, if you
> watch HD, you will probably have some sound covering the fans, this is
> what happens at my place.
>
> But going with a cheaper CPU, a cheap GPU that can assist HD decoding
> gives less noise and less money spent.
>
> An for most, why waste a good CPU when there is a GPU than can do the
> same job with less efforts, heat and noise?

I'm not saying don't use it if you've got it (when it's available and
working in the app of your choosing, of course), I'm simply saying that
it's not like NVIDIA is forcing you to go out and buy a new GPU. If you
don't have and don't buy a supported GPU, you have exactly the status
quo--you're right where you were before they announced a new feature
that many in the community will appreciate, so whining that this new
feature isn't supported on your 4-year old GPU is really expecting a bit
more than you should (after all, they didn't have to provide this new
feature at all).

Mike
_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


nico at youplala

Nov 16, 2008, 10:40 AM

Post #18 of 65 (7896 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

On Sun, 2008-11-16 at 13:36 -0500, Michael T. Dean wrote:
>
> I'm not saying don't use it if you've got it (when it's available and
> working in the app of your choosing, of course), I'm simply saying
> that
> it's not like NVIDIA is forcing you to go out and buy a new GPU. If
> you
> don't have and don't buy a supported GPU, you have exactly the status
> quo--you're right where you were before they announced a new feature
> that many in the community will appreciate, so whining that this new
> feature isn't supported on your 4-year old GPU is really expecting a
> bit
> more than you should (after all, they didn't have to provide this new
> feature at all).

Fully agreed.

Nico

_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


cynical at penguinness

Nov 16, 2008, 1:13 PM

Post #19 of 65 (7898 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

Nicolas Will wrote:

> You are talking about PureVideo.
>
> NVIDIA is really talking about PureVideo HD. It looks like the circuitry
> on-chip is quite different, thus needs different code.

Ah, I wasn't aware that there was a PureVideo and a PureVideo HD available.

*looks a bit more*

Hm, I think my point still stands...

> If the marketing names are about the same, it doesn't mean that the
> engineering is.


Agreed. However, the the difference between the two appears to be more
marketing than engineering.

From what I can tell at the PureVideo HD page and associated PDF's, the
only difference between the two is the HD version of PureVideo is that
the HD version is HDCP Capable. Heck, the product comparison PDF on the
PureVideo and PureVideo HD pages are the same file that I looked it
showing that every chip back to the Go 6600 support h.264 acceleration
under PureVideo.

Even if a given video card can't move everything to the GPU, which is
one of the selling points in regards to the 8xxx series cards, being
able to move even parts of the h.264 pipeline to the video card would help.

Don't get me wrong, I'm glad that NVidia is finally bringing this to the
Linux community as it's past due, but it does seem to be more of a
marketing move than a answer to something many end users have been
wanting for some time.

Regardless, this is all speculation as I don't have access to the lower
level GPU design specs, and I really couldn't understand most of it if
it was available (I am not a hardware engineer). I may be very wrong
and there may be some very large design differences that are keeping
them from implementing this newly announced API across more of their
chipsets.

However, what NVidia is stating with this announcement and PureVideo HD
seems to ignore what they have stated earlier with the 6xxx series chips
and PureVideo in regards to chipset support.
_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


daworm at comcast

Nov 16, 2008, 1:22 PM

Post #20 of 65 (7896 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

Justin The Cynical wrote:
> Even if a given video card can't move everything to the GPU, which is
> one of the selling points in regards to the 8xxx series cards, being
> able to move even parts of the h.264 pipeline to the video card would help.
>
Especially if that GPU is soldered onto the board and can't be replaced,
such as on the AppleTV.

Jeff.
--
I haven't smoked for 2 years, 2 months and 4 weeks, saving $3,702.39 and
not smoking 24,682.63 cigarettes.
_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


mtdean at thirdcontact

Nov 16, 2008, 1:36 PM

Post #21 of 65 (7885 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

On 11/16/2008 04:13 PM, Justin The Cynical wrote:
> However, what NVidia is stating with this announcement and PureVideo HD
> seems to ignore what they have stated earlier with the 6xxx series chips
> and PureVideo in regards to chipset support.

Where's the announcement of PureVideo support for the 6xxx series chips
/in Linux/?

They're not contradicting themselves. There simply starting /somewhere/.

This is new to Linux (and the other supported *nix'es).

Now go buy yourself a $30 video card ($20 after mail-in rebate) and quit
being, er, Cynical.

Mike
_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


tortise at paradise

Nov 16, 2008, 1:58 PM

Post #22 of 65 (7895 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

Re "Now go buy yourself a $30 video card ($20 after mail-in rebate) and quit being, er, Cynical."

The point is many of us have 6 and 7 series cards that are said to be HD capable, that we have already invested in, we merely await
the nix HD drivers...

Why has NVIDIA seemingly skipped drivers for these HD capable cards?

Sometimes the mere $20 cost is not that simple, for example a whole new motherboard may also be required, to also change from AGP to
PCI16.

This has a smell of buy another....and maybe we'll release the other drivers in due course, when you've invested twice in NVIDIA,
for one purpose....

Kaching. Thank you!

Perhaps it is just a few of us who do not want to invest in further kit, when we have already got kit that should do the job. (At
any price)

Somehow I don't think it is just a few of us who feel this way, indeed I think an issue should be made of this with NVIDIA.

Kind regards
David

_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


cynical at penguinness

Nov 16, 2008, 2:00 PM

Post #23 of 65 (7882 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

Michael T. Dean wrote:

> So, how does that hand taste? You know, the one that feeds you (NVIDIA
> drivers)?

Are you referring to the ones from NVidia, or the ones from XOrg?

I have a 8400 CPU in my machine. I don't have to use the NVidia driver
if I don't want to. I'm running the binary from NVidia only for the
OpenGL functions used by MythMusic visulations.


> Just be happy with the fact that they did anything for us. Besides,
> have you looked at the prices of the cards on the list of supported
> GPU's? We're talking $20 to $30--not to mention the fact that /many/ of
> the integrated GPU's are supported, too.

I am glad NVidia finally bought this functionality to Linux. However,
that doesn't stop me from pointing out that it's something that by their
own marketing material is supported by their older chipsets.

The reason for the limited chipset support may very well be due to the
design of the newer chipsets with their unified shader pool and actually
designing the hardware with video decoding in mind, vs having a 2nd
software stack with decoding API's to somewhat fake it on older hardware.

If this is the case, then it should be stated and not simply ignored.
Otherwise it would appear that this is purely market-droid driven and
done to try and steal back some of the thunder that Intel and AMD/ATI
have regarding video decoding/acceleration support.


> Oh, and there's also the whole, you really don't /need/ GPU-assisted
> decode, anyway. My system happily handles US MPEG-2 HDTV. If you're
> using an HD-PVR to record HDTV to MPEG-4 AVC from (rented)
> cable/satellite STB's (with a monthly cable bill), the extra $20 to $30
> you spend to get a supported card is really a negligible part of your TV
> cost, but you can always buy a system that can play that back even
> without the GPU assist (and the support for HD-PVR output is getting
> better every day).

Cost of the cards isn't the point. The point is that their older chips
support the functions that this new driver has. I'm not going to simply
roll over and accept whatever is handed to me by a company that has
refused to provide programming specs for their hardware[1] which would
allow the same functionality that has been available on other platforms
to be implemented on my OS of choice.

If I was going to just accept whatever is provided to me and not
complain about it's shortcomings, then I'd still be running Microsoft
software on everything with it's support of non-mandatory broadcast
flags that limit what I can and can't do with the video stream that are
sent over the airwaves.

[1] Never mind that their largest competitor has started releasing this
info.
_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


nico at youplala

Nov 16, 2008, 2:17 PM

Post #24 of 65 (7901 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

On Sun, 2008-11-16 at 13:13 -0800, Justin The Cynical wrote:
> Nicolas Will wrote:
>
> > You are talking about PureVideo.
> >
> > NVIDIA is really talking about PureVideo HD. It looks like the
> circuitry
> > on-chip is quite different, thus needs different code.
>
> Ah, I wasn't aware that there was a PureVideo and a PureVideo HD
> available.
>
> *looks a bit more*
>
> Hm, I think my point still stands...
>
> > If the marketing names are about the same, it doesn't mean that the
> > engineering is.
>
>
> Agreed. However, the the difference between the two appears to be
> more
> marketing than engineering.
>
> From what I can tell at the PureVideo HD page and associated PDF's,
> the
> only difference between the two is the HD version of PureVideo is
> that
> the HD version is HDCP Capable. Heck, the product comparison PDF on
> the
> PureVideo and PureVideo HD pages are the same file that I looked it
> showing that every chip back to the Go 6600 support h.264
> acceleration
> under PureVideo.
>
> Even if a given video card can't move everything to the GPU, which is
> one of the selling points in regards to the 8xxx series cards, being
> able to move even parts of the h.264 pipeline to the video card would
> help.
>
> Don't get me wrong, I'm glad that NVidia is finally bringing this to
> the
> Linux community as it's past due,

"past due". Shooting at the ambulance, heh?

You see, a few days ago I would certainly not have placed any bets on
NVIDIA delivering anything regarding hw accel of video codecs on Linux.

My bet would have been on Intel with an open source driver.

I was wrong. Tough luck. In the end, I'm so glad anyone shot first.


> but it does seem to be more of a
> marketing move than a answer to something many end users have been
> wanting for some time.

Ah.... end users...

Do you know who NVIDIA's end users are?

Let's say that a year ago I was in front of a tricky problem for a very
high end setup, involving 4k projectors, 2 Quadroplex units and passive
stereo. I ended up with NVIDIA's chief engineer on the phone. He said he
understood the problem. He said that it could be solved. He said that it
would require engineering time. He asked how many situations like mine
he would encounter world-wide. I answers that one hand should be enough
to count the situations, even if we are one of the 2 big guys in the
field.

I was basically told that my issue was not helping gamers at all, that
he would need to evaluate retasking engineering resources.

That was a polite "never".

I can't blame him.

How many Linux users willing to play HD does it take to retask
gaming-oriented driver developers at NVIDIA?


In the end my system worked, more out of luck than anything. Those guys
are not as bad as they are pictured most of the time.

>
> Regardless, this is all speculation as I don't have access to the
> lower
> level GPU design specs, and I really couldn't understand most of it
> if
> it was available (I am not a hardware engineer). I may be very wrong
> and there may be some very large design differences that are keeping
> them from implementing this newly announced API across more of their
> chipsets.
>
> However, what NVidia is stating with this announcement and PureVideo
> HD
> seems to ignore what they have stated earlier with the 6xxx series
> chips
> and PureVideo in regards to chipset support.


Well, I was surprised by the limitation on chips supporting VC-1, as
advertised in the announcement.

So I pulled my email client and emailed the NVIDIA guy.

The answer came quickly and nicely. It is very much in line with
hardware differences, and it is very much in line with what I know from
professional contacts with NVIDIA (much different field, advanced Oil
and Gas visualization stuff):

The only currently launched NVIDIA GPUs with support for the
entire
VC-1 pipeline are the GPUs listed in the announcement snippet
you quoted
(i.e., G98-based products). I believe most future GPUs will
also have
this support.

GPUs such as G84 and G86 could accelerate part of the VC-1
pipeline, so
we may investigate trying to expose that at some point in the
future,
based on end user feedback. But for now the priority is to
stabilize
the support in the cases that the GPU accelerates the entire
pipeline.

There are hardware differences. It may not be what you like to hear, the
marketing is not helping, but I really trust that this is true. CPU/GPU
architecture is not trivial.

Nico

_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users


spuppet at comcast

Nov 16, 2008, 2:20 PM

Post #25 of 65 (7899 views)
Permalink
Re: FYI - New HW video accellAPI from NVIDIA [In reply to]

On Fri, 2008-11-14 at 10:28 -0800, Andy Ritger wrote:
> I'm pleased to announce a new video API for Unix and Unix-like platforms,
> and a technology preview implementation of this API from NVIDIA.
>
> The API is called VDPAU (Video Decode and Presentation API for Unix).


I took some time today to test the new drivers with the h.264 samples from
the Hauppauge HD-PVR on my desktop system. The performance was impressive,
to say the least.

I tested the three samples linked on the HD-PVR wiki entry
(http://www.mythtv.org/wiki/index.php/HD-PVR) as well as a recording I had
made earlier this week (a 720p program at 9Mbps).

My test methodology was far from scientific, but the results are
encouraging. My testing was simply done by comparing the the CPU loads
displayed by top and the command line output of mplayer. My desktop system
is a Core 2 Quad (Q6600) @ 2.4GHz with 4GB of RAM and a 8800GTS card. (I
don't have a mythfrontend with a 8 series or later nVidia card.)

Running any of the samples I had with the standard mplayer without any
command line args (i.e. mplayer <file>)did a pretty good job of pegging the
CPU. mplayer also complained that the system was too slow and audio got
very far out of sync. If anyone can recommend some better command line
args, I can re-run that part of the test, as I think this comparison isn't
quite fair.

After building the patched ffmpeg and mplayer, I re-played the samples and
noted that there was no discernable CPU usage once playback got rolling. In
top, mplayer consistently showed 0% CPU usage, with a few bumps up to a few
percentage points here and there. Mplayer outputs CPU usage for the video
codec, video out, and audio codec processing. For all the samples, mplayer
eventually settled after a minute or so into 0% video codec CPU usage, 8 -
9% video out CPU usage, and 1.5% audio codec CPU usage.

I was able to run other programs (command line, firefox, etc.) at the same
time and didn't notice any stuttering.

I've got a 7 series nVidia system here that I'll test with, too. I don't
expect it to work, but maybe there's some hidden mojo.

Please let me know if you have any questions or test suggestions.

Jason

_______________________________________________
mythtv-users mailing list
mythtv-users [at] mythtv
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users

First page Previous page 1 2 3 Next page Last page  View All MythTV users RSS feed   Index | Next | Previous | View Threaded
 
 


Interested in having your list archived? Contact Gossamer Threads
 
  Web Applications & Managed Hosting Powered by Gossamer Threads Inc.