ATI Avivo XCode released!

[Morpheus_xx]

Released today: http://www.chip.de/downloads/c1_downloads_18000750.html

It comes with a bulk of direct show filters that allow following formats:
MPEG 1
MPEG 2
MPEG 4 (DivX, Generic, Sony PSP)
WMV
H.264 (AVI, Generic, Sony PSP)

I'm starting the first test now...

[bond]

what does generic mpeg-4/h.264 mean? is there no special support for the ipod anymore as shown in the screenshot?

[Doom9]

too bad I don't have an ATI card, but the 7800GT simply blows ATI out of the water when it comes to games and so my choice was easy.

[Morpheus_xx]

some screenshots from the directshow filters:

video encoder
video encoder with opened dropdown
audio encoder
video mux

[Doom9]

I think there's one shot missing: the first screen without the dropdown activated , so that you can actually see what you can configure.

And an app with the standard MFC icon from a company like ATI? Why does that smell funny? And the whole thing is two weeks old already?

And isn't ATI located in Canada, not in Silicon Valley?

[bond]

thx, so it seems "generic" mpeg-4/h.264 means .mp4

cant be ati was so smart to realise that the ipod uses plain .mp4...

[Morpheus_xx]

my very first impression:

speed: ~ 25 fps for my avisynth input (also using fanfilter)
cpu load: 50% (Athlon Dual Core)

quality: less than xvid

VDub can arrive nearly the same fps but with ~100% cpu load.

PS: what happens when you try to use the tool with a non-ati board???

[Doom9]

PS: what happens when you try to use the tool with a non-ati board???It doesn't start, telling you you need a 1xxx card.
Please try a regular 3 line script (d2v source, crop and resize) and compare that to VDub.. it's so much more meaningful as filters tend to distort results and you can't quite place the results when you're not used to using those filters.

[bond]

quality: less than xvidhaha

can you upload somewhere two small .mp4 (generic) files, with h.264 and mpeg-4, created with that encoder plz

PS: what happens when you try to use the tool with a non-ati board??? the download page says it will not work

[Richard Berg]

http://anandtech.com/video/showdoc.aspx?i=2645&p=2

The decoder is hard-limited to various resolutions depending on which model card you have. Anand doesn't say, but it's probably limited to a certain profile (below HP) as well.

The encoder is not yet GPU accelerated, but this quote was nice to see:

So, is ATI getting into the software business with the Avivo Video Converter? No. In fact, ATI is providing these algorithms and hooks to partners like Nero so that regardless of what software application you're using, you will get the best performance assuming that you have ATI hardware.


Similar things were said in the Firingsquad interview last month, but nothing came of it afaik. Will be interesting to see if they are willing to work with OSS codec developers.

[acidsex]

1080p decoding of H.264 AVC sounds awesome but how much does those x1800 cards go for?

edit: just saw what they go for $400. Thats ridiculous for a video card. Might be better off saving $400 and putting it towards next generation HD-DVD player where playback shouldnt be a problem.

[mod]

I've not been able to open the .zip archive, tried latest 7zip, Winzip, Winrar..
If it's legally possible, could someone post here an uncompressed and recompressed version? I would really appreciate it.

[Yong]

It doesn't start, telling you you need a 1xxx card.

Please correct me if im wong,
Its work for me altough i only have ATI Radeon 9600XT.
I ingored the AvivoXCode.exe and run regsvr32 to register the wbocx.ocx, atidvcr.dll and atixcode.dll manually.
encoding works fine with graphedit.

[Selur]

it would also be interesting to know if different ATI card provide different encodeing/decoding support or if a small 1xxx card would be enough. ;)

@yong: what's your encoding speed, do you get any hardware support with your card?

[Doom9]

@Yong: well, I only tried running the binary, I didn't even bother installing. I guess I should try it after all, but looking at the msu codec comparison, ATI isn't exactly a performer in the areas I'm looking for.

[Morpheus_xx]

short test clip 40seconds:


LoadPlugin("C:\Programme\video\AviSynth 2.5\plugins\dgdecode.dll")

clip="d:\Capture\test\test"

MPEG2Source(clip+".d2v")
LanczosResize(720,304)



results

encoder codec bitrate(codec) bitrate(result) enc.time
avivo h264 abr1600 eff534 13s
avivo h264 abr2400 eff864 13s
avivo h264 abr800 eff231 11s
vdub x264 cbr800 eff823 50s
vdub xvid cbr1000 eff943 26s

[slavickas]

Our reference point is our test bed running the trailer in Quicktime 7, which isn't GPU accelerated, and comparing that to Windows Media Player 10 with the Cyberlink H.264 decoder

Being faster then QT isn't big deal :D, and I was thinking of getting X1300 AGP once it arrives (I'm not gamer and not willing to spend $$$ on vga)

[bond]

without knowing the used codec settings, speed and quality comparisons mean nothing, therefore:

can you upload somewhere two small .mp4 (generic) files, with h.264 and mpeg-4, created with that encoder plz

[Yong]

it would also be interesting to know if different ATI card provide different encodeing/decoding support or if a small 1xxx card would be enough. ;)

@yong: what's your encoding speed, do you get any hardware support with your card?
Just done some encoding test, i *think* its encode pretty fast :p with MPEG1 video 640x480 , 3844 frames @ 24fps.
http://img221.imageshack.us/img221/728/ati5ia.jpg (http://imageshack.us)
Hardware support? encoding or decoding? Later i will try to modify something to let my x264/MPEG4 ASP and etc etc video clips play with dshow ATI MPEG video decoder :D

BTW, the mp4(the header is different form mp4 thats created by x264) files created by ATI file writer have problem with halli media splitter but playable by quiktime player with some jerkiness(b frames problem?).

@Doom9:
It encode pretty fast but the quality is crappy @ 256kbps... compare with x264(ive done a lot of encode with x264 @ 256kbps)

[bond]

my comments:

- qt7 is a damn slow decoder, it definitely shouldnt be used as a benchmark
- apple trailers use multiple slices. a decoder can be also speeded up by making use of this, without a gpu, so as a test sample a clip without multiple slices should be used to really see the speedup get by the gpu
- the encoders speed values are obviously totally useless, as for a valid speed comparison you would need to use exactly the same settings in both encoders, which the comparison writer surely didnt do.
- "ipod video format" is plain .mp4. i dont like this product-centric naming sheme...

last but not least i am wondering if cyberlink still uses the moonlight avc decoder?

[bond]

i would be really surprised if ati's encoders would perform any good qualitywise

after all they are meant for maxing out the speed, so why offer high quality bringing the speed down?

[videomixer9]

I think their goal is too use the GPU to max the speed while keeping the quality as you know it from other encoders. Speed says nothing about quality, just because it's damn fast doesn't mean it's bad quality, it can also mean it is better written or makes better use of processors that are meant for graphic processing ... I think nvidia and ati both showed us that it is their goal to make more use of their graphic processors to accelerate video.

[bond]

you are right, and i said "i would be surprised", which means i dont know yet how things are handled

still before we can do valid speed and quality comparisons we need to know what codec settings ati uses, so gimme some damn samples! :D

[breez]

Note: at this point the encoding is done entirely in software, no help from GPU.

[bond]

Note: at this point the encoding is done entirely in software, no help from GPU.what tells you this?

[slavickas]

what tells you this?
article at anandtech states so, although maybe their verson was older

[alcuin]

what tells you this?

http://www.anandtech.com/video/showdoc.aspx?i=2645&p=3

The other item of interest that ATI briefed us on is the ATI Avivo Video Converter. We mentioned in our previous coverage that ATI was working on GPU accelerated video transcode, to speed up the conversion of videos from one format to another (e.g. MPEG-2 to H.264). Unfortunately, the GPU accelerated transcode isn't yet ready for debut, but what ATI is making available is the software front end for it. .... However, keep in mind that despite ATI's release of this tool, the video conversion itself is done entirely on the host CPU and not on the GPU. So why bother? Well, thanks to ATI's experience in dealing with video, they have optimized a number of the transcoding algorithms so that conversion using the utility is actually faster than on other software solutions.

et al.

regards,
-e

[bond]

However, keep in mind that despite ATI's release of this tool, the video conversion itself is done entirely on the host CPU and not on the GPU. So why bother? wtf? :D

[LiFe]

wtf? :D
The article actually has no usefull info. No point in quoting speed figures (even with pretty graphs) if you don't show features and quality.

[Sharktooth]

So why the hell it does work on 1xxx series only?!?

[Sulik]

Apparently, ATI's H.264 encoder is currently baseline profile only (No B-frames or CABAC), which definitely would limit its potential in quality (Would also explain the results at MSU when compared with Main profile codecs).

It would be interesting to compare the quality when using CAVLC in x264.

[CeeJay.dk]

So why the hell it does work on 1xxx series only?!?

Because the app check for the presence of a 1xxx series card and denies running if it is not found.
I read on another site that ATI did this because they wanted to add hardware-acceleration later and that it would only be possible to do on a 1xxx card.

I however believe it has more with ATI wanting people to upgrade rather than giving all their customers (or worse .. their competitors customers) something to enjoy for christmas.

Since it is done entirely on the CPU, it should be possible to either do as Yong did , or use a program to fake the make and model of the card so the driver detects a 1xxx card, or modify the driver or the program to remove this check or make it so it always passes it.

ATI is also busy at licencing this new ultrafast piece of code to thirdparty developers so versions of it might appear in other encoding applications , that do not have the 1xxx restriction.

I'm fully convinced that its only a matter of time before someone develops a hack or even reverseengineers the program.

UPDATE : That other site I remembered seeing the news at , was Elite Bastards dot com:
http://www.elitebastards.com/page.php?pageid=13022

[Doom9]

Yong, what's your filter chain to write the output? Have you managed to get a raw stream from the encoder instead? I guess that would be more useful, so we can mux using mp4box, which we know does as it should.
I ran the installation routine as well and the filters at least show up in graphedit.

[Sulik]

I can get a raw stream from the encoder:
DShowSource-> "ATI Video Scaler" -> "ATI MPEG Video Encoder" -> File Dump filter

[bond]

I can get a raw stream from the encoder:
DShowSource-> "ATI Video Scaler" -> "ATI MPEG Video Encoder" -> File Dump filterwhich filters does the tool include?

[Sulik]

It includes video scaler, video encoder, audio encoder, mpeg decoder, multiplexer and file writer.

[Doom9]

a lot of filters, all having a name starting with ATI.. just go ahead and try it.. it'll work on your box as well ;) there's also an audio encoder, playback filters, resizing filter, capturing filters.. quite a bunch.

[bond]

yeah just found it out

did anyone get things to work in graphedit? i till now was only able to connect divx5 .avi files via the m$ splitter to the ati decoder and than to the encoder

but when encoding it never finishes but produces a bigger and bigger getting file

[Doom9]

I did a short test run. AviSynth script (the same used for the codec comparison qualification) -> ati mpeg video encoder -> dump filter (from the graphedit package on my site). I trimmed the script to 2000 frames, encoded, muxed to mp4 and it plays just fine. I haven't found out how to directly create an mp4 so far. I saw no moving of the progress slider in Graphedit so that's kinda unnerving.

I can't say anything about speed because I'm currently encoding VC-1 and that occupies both cores (total load about 90%).

@edit: hmm, I actually managed to write an mp4. I connected the video encoder to the ati mpeg multiplexer, set it to output mp4 generic, and connected its output to the ati mpeg file writer, but I'm afraid the resulting mp4 will just come up with a black screen in MPC.

[bond]

here my findings with asp and avc in mp4:

ASP:

- seems to be simple profile + max 2 non adaptive b-frames
- no qpel, no gmc
- h263 quant
- places the vol together with every keyframe in mp4, which is not allowed in mp4 (seems to be a leftover of the avi workaround doing the same)
- the mandatory ctts atom carrying the b-frame timestamps is missing in mp4
- wanted filesize isnt matched correctly
- i see visual artefacts here and there when decoding
- signals a strange aspect ratio
- seems to place keyframes every ~50 frames but not exactly
- signals the average bitrate as 0
- mp4 seems to play fine in mplayer, vlc and with gabest+ffdshow
- quality is clearly not good

AVC:

- seems to be baseline profile + max 2 non adaptive b-frames
- max 1 reference frame
- no cabac, no weighted (bi)prediction, no multiple slices
- loop uses 0:0, but was never enabled in the encode
- again the strange sar signalled
- the mandatory ctts atom carrying the b-frame timestamps is missing in mp4
- wanted filesize isnt matched correctly by far
- i am unable to play the file with nero, haali and gabest mp4 splitter (jerky in qt7), mplayer and vlc handle it fine
- quality is clearly not good

[Yong]

@edit: hmm, I actually managed to write an mp4. I connected the video encoder to the ati mpeg multiplexer, set it to output mp4 generic, and connected its output to the ati mpeg file writer, but I'm afraid the resulting mp4 will just come up with a black screen in MPC.
Hmm, Halli media splitter/mp4creator unable to recognize the mp4 file created by ATI mpeg muxer, mkv file is broken if remux it with mkvtoolnix too.
but you can remux the mp4 by using mp4box :D ,im sure this time your mp4 will play with MPC.

EDIT: playback jerky problem wih qtplayer is gone if dont use b frames :rolleyes:

[Sulik]

I'm getting pretty good results using no B-frames in either MP4 or AVI formats (640x480@~2Mbps) - I've only tested a few clips, but it seems somewhat similar to DivX & Xvid.
bond: deblocking is enabled on my system when using H.264 - looks fine in Quicktime on a 2.6Ghz P4 (Without B-frames) - What's your system ?

[Doom9]

looks fine in Quicktime on a 2.6Ghz P4Please never use that POS for anything.. you can't reliably tell if what you get is what you're supposed to get.

[CeeJay.dk]

The beta 5.13 driver can now be downloaded:
http://www.warp2search.net/modules.php?name=News&file=article&sid=26836

I'm just going to wait until tuesday though , when the final version is due to be released.

BTW - With your discovery of the sub-par encoding, I'm wonder how good/fast decoding is, as this is what other sites have been reporting to have improved imensely.

[bond]

Hmm, Halli media splitter/mp4creator unable to recognize the mp4 file created by ATI mpeg muxer, mkv file is broken if remux it with mkvtoolnix too.
but you can remux the mp4 by using mp4box :D ,im sure this time your mp4 will play with MPC.you are talking about avc-in-mp4 i assume
cause with asp-in-mp4 i dont get this

EDIT: playback jerky problem wih qtplayer is gone if dont use b frames :rolleyes: qt7 seems to dont handle more than 1 b-frame correctly
additionally the .mp4 files created by ati dont include the timestamps showing when to display the b-frames, so its no surprise that a strict player expecting that info borks

bond: deblocking is enabled on my system when using H.264 - looks fine in Quicktime on a 2.6Ghz P4 (Without B-frames) - What's your system ?this doesnt depend on the system. its hardcoded in the stream on what frame to use loop and in my test clip its disabled on all frames

[MindlessOath]

ok, here is the link to an article.
http://www.elitebastards.com/page.php?pageid=13054&head=1
with a download for the tool patched to allow other users to use this... not just x1000 users.

i would have released it myself for this guy, who remains nameless for the moment being... but looks like i was sleeping :P and he got ahold of hanners :D at elitebastards.

hope you guys enjoy it!

[netchris]

;) Thanx

[acidsex]

suprisingly ipod encodes look fairly good at half the time to encode.

[acidsex]

I tried encoding to H.264 AVC (avi) but there doesnt seem to be anyway to set audio settings or to enter a manual bitrate. Also, does anyone have the graph edit filters and connections for encoding and also the filter chain for demuxing .avi to .mp4 container as the ATI muxing filter and ATI file writer leaves me with a file that plays but I see no video or audio in VLC.

Thanks.

[bond]

suprisingly ipod encodes look fairly good at half the time to encode.man, use your glasses ;)

I tried encoding to H.264 AVC (avi)why .avi?

Also, does anyone have the graph edit filters and connections for encoding and also the filter chain for demuxing .avi to .mp4 container as the ATI muxing filter and ATI file writer leaves me with a file that plays but I see no video or audio in VLC.i doubt remuxing avc from .avi to .mp4 is supposed to work (note that there is no existing solution that does this correctly)

if you want a .mp4 you should encode directly to it (doom9 already described how it works in graphedit with the ati filters)