Single threaded H.264 decoding performance

Blue_MiSfit · 2nd July 2010, 12:34

[EDIT] This was all updated quite a bit. The interesting graph I came up with a few hours later has been added to this OP just to grab eyes
Consider my thoughts in this OP somewhat misguided...

tl;dr - ffmpeg-mt is faster (for these sources anyway) than CoreAVC or DivX H.264!

Hey folks,

So we all know that CoreAVC is a great place to start if you're looking for FAST H.264 decoding on Windows. That's all well and good.

However, I recently realized I'd made the assumption that it would be very fast in all scenarios. As my numbers below show, this is not the case!

All tests were performed on a 2.4 GHz Intel Q6600 quad-core CPU running Windows 7 x64. Benchmarks were made using avs2avi. DSS2 provided DirectShow interaction. All sources were of identical content, and were 1080p24.

Source 1 - ~50mbps x264 encode, "fastdecode" tune (no CABAC, B-Frames, or Deblocking)
CoreAVC - 103.7 fps
FFMS2 - 31.5 fps
Fake CoreAVC "single threaded" ~ 25.93 fps

Source 2 - ~50mbps x264 encode, no CABAC or B-Frames, but has deblocking
CoreAVC - 86.9 fps
FFMS2 - 27.6 fps
Fake CoreAVC "single threaded" ~ 21.73 fps

Source 3 - ~50mbps x264 encode, with CABAC, B-frames, and deblocking
CoreAVC - 56.6 fps
FFMS2 - 13.7 fps
Fake CoreAVC "single threaded" ~ 14.2 fps

My conclusion? In typical scenarios for this community, CoreAVC or similar high performance multi-threaded decoders are the way to go.

However, in MY CASE

, things are a little different. I usually work with H.264 sources that have no B-Frames, and use CAVLC instead of CABAC. Both of these sacrifices are made to facilitate real-time capture, which is a huge ugly animal all on its own!

Given my typical sources, and the fact that I'm doing a LOT of transcoding at once (anywhere from 3-6 1080p encodes on a system at once, depending on how many cores I have), FFMS2, single threaded, seems like it will actually give me better throughput - provided it never bottlenecks x264!

Sure, the numbers are always higher with CoreAVC, but the numbers are always with it eating up almost 100% of my CPU! That's fine for playback, but for high volume transcoding, I want the most efficient decoder for my type of files, correct?

I THINK I'm seeing this correctly. What do you guys think?

The best possible solution is to do all the decoding on a couple GPUs, but that's another story.

Derek

Blue_MiSfit · 2nd July 2010, 12:59

Some more numbers, this time with timeCodec.exe. They're higher overall, but I assume that's because it avoids AviSynth etc. The advantage is, I can restrict CPU affinity reliably this way

Also, since FFMS2 isn't available in this framework, I'm just using libavcodec from ffdshow-tryouts rev 3463 (may 29 2010)

Source 1 (--tune fastdecode, zerolatency)
CoreAVC (4 threads) - 126.3
libavcodec (1 thread) - 35.9
CoreAVC (1 thread) - 34

Source 2 (--no-cabac, --tune zerolatency)
CoreAVC (4 threads) - 101.4
libavcodec (1 thread) - 30.4
CoreAVC (1 thread) - 27.7

Source 3 (all on)
CoreAVC (4 threads) - 62.5
libavcodec (1 thread) - 17.6
CoreAVC (1 thread) - 16.5

Interesting. The trend is basically the same, though the differences here are less pronounced (less than 10% in most cases).

Interestingly, good old MPEG-2 at 80mbps, with typical GOP structure decodes at ~44fps single threaded with FFMS2, which is almost 2x faster than DGDecode

Derek

clsid · 2nd July 2010, 13:15

ffdshow also has multi-threading capability if you select ffmpeg-mt as H.264 decoder.

Blue_MiSfit · 2nd July 2010, 13:19

Indeed. In my testing it's always been slower than CoreAVC, but it's worth a look!

Derek

Blue_MiSfit · 2nd July 2010, 13:58

Here's more results, this time using ffmpeg-mt (and including the CoreAVC results for reference

Source 1 (--tune fastdecode, zerolatency)
ffmpeg-mt (1 thread) - 35.5 fps
ffmpeg-mt (4 threads) - 130.4 fps
CoreAVC (1 thread) - 34 fps
CoreAVC (4 threads) - 126.3 fps

Source 2 (--no-cabac, --tune zerolatency)
ffmpeg-mt (1 thread) - 28 fps
ffmpeg-mt (4 threads) - 110.4 fps
CoreAVC (1 thread) - 27.7 fps
CoreAVC (4 threads) - 101.4 fps

Source 3 (all on)
ffmpeg-mt (1 thread) - 17.3 fps
ffmpeg-mt (4 threads) - 66.2 fps
CoreAVC (1 thread) - 16.5 fps
CoreAVC (4 threads) - 62.5 fps

WOW. I was NOT expecting that! ffmpeg-mt is faster than CoreAVC in all cases for my usual files! It looks like the ffmpeg devs have been working hard

I'll make a pretty little graph to illustrate, and maybe put in DivX H.264 as well, I don't have DiAVC and don't feel like spending $10 right now...

Derek

kieranrk · 2nd July 2010, 14:26

Have you tried DiAVC?

Blue_MiSfit · 2nd July 2010, 14:58

Quote:

I don't have DiAVC and don't feel like spending $10 right now...

Derek

Underground78 · 2nd July 2010, 22:49

Quote:

Originally Posted by Blue_MiSfit

I don't have DiAVC and don't feel like spending $10 right now...

I thought a trial version was available but I may be mistaken.

Dark Shikari · 2nd July 2010, 23:18

Performance may depend heavily on the source material -- e.g. whether the bitrate is high or low, etc.

sneaker_ger · 3rd July 2010, 01:01

Quote:

Originally Posted by Underground78

I thought a trial version was available but I may be mistaken.

The current trial seems to be expired, but schweinsz gladly distributes trials on request.

Blue_MiSfit · 3rd July 2010, 02:12

Indeed. The bitrate was very high in my case, usually around 60mbps (1080p24)

Derek

Dark Shikari · 3rd July 2010, 02:15

Quote:

Originally Posted by Blue_MiSfit

Indeed. The bitrate was very high in my case, usually around 60mbps (1080p24)

Derek

Try a much lower bitrate sample. At that bitrate, you're benching the CABAC decoder.

Blue_MiSfit · 3rd July 2010, 02:28

Hence why 2/3 of my tests have CABAC disabled

Dark Shikari · 3rd July 2010, 02:34

Quote:

Originally Posted by Blue_MiSfit

Hence why 2/3 of my tests have CABAC disabled

Then you're benching the CAVLC decoder

Blue_MiSfit · 3rd July 2010, 03:12

Fair enough.

Still, I'm trying to determine the optimal way to decode high bitrate H.264 for high volume transcode. In my mind, this means the decoder that gives the best per-thread performance. Is that logical?

This seems like a valid test, no?

As I said, MPEG-2 decodes very quickly as well

Derek

STaRGaZeR · 3rd July 2010, 11:12

If you feel like it, it'll be really nice to have a comparison chart with more samples, with a complete description of the used settings. That way we'll see where each decoder shines or fails

iwod · 3rd July 2010, 17:29

Um... So we come to the conclusion that FFmpeg - Mt is fastest.

No loner a reason to use CoreAVC, Divx, or Diva decoder?

Blue_MiSfit · 3rd July 2010, 18:27

Well, for these (unusual) test cases anyway. I haven't bothered to do anything with standard BluRay streams or typical 4-15mbps backup type streams.

Derek

CeeJay.dk · 4th July 2010, 11:16

@Dark Shikari : You wrote on your blog about some new deblocking optimization you made to x264. Could this optimization to an encoder deblocker be applied to a decoder deblocker , like the one in FFmpeg ?
Have it already been applied - Is that why the new FFmpeg 0.6 release decodes h264 faster ?

Dark Shikari · 4th July 2010, 19:53

Quote:

Originally Posted by CeeJay.dk

@Dark Shikari : You wrote on your blog about some new deblocking optimization you made to x264. Could this optimization to an encoder deblocker be applied to a decoder deblocker , like the one in FFmpeg ?
Have it already been applied - Is that why the new FFmpeg 0.6 release decodes h264 faster ?

It could be, and no, it hasn't been. It would help a lot more in a decoder than in an encoder, in fact.

I haven't done it because the second stage of the process -- "fix things up if the neighbors are different for deblocking than for normal decoding -- is rather hard when you have to consider all possible options (PAFF, MBAFF, etc, etc).

2nd July 2010, 12:34	#1 \| Link
Blue_MiSfit Derek Prestegard IRL Join Date: Nov 2003 Location: Los Angeles Posts: 5,989	Single threaded H.264 decoding performance [EDIT] This was all updated quite a bit. The interesting graph I came up with a few hours later has been added to this OP just to grab eyes Consider my thoughts in this OP somewhat misguided... tl;dr - ffmpeg-mt is faster (for these sources anyway) than CoreAVC or DivX H.264! Hey folks, So we all know that CoreAVC is a great place to start if you're looking for FAST H.264 decoding on Windows. That's all well and good. However, I recently realized I'd made the assumption that it would be very fast in all scenarios. As my numbers below show, this is not the case! All tests were performed on a 2.4 GHz Intel Q6600 quad-core CPU running Windows 7 x64. Benchmarks were made using avs2avi. DSS2 provided DirectShow interaction. All sources were of identical content, and were 1080p24. Source 1 - ~50mbps x264 encode, "fastdecode" tune (no CABAC, B-Frames, or Deblocking) CoreAVC - 103.7 fps FFMS2 - 31.5 fps Fake CoreAVC "single threaded" ~ 25.93 fps Source 2 - ~50mbps x264 encode, no CABAC or B-Frames, but has deblocking CoreAVC - 86.9 fps FFMS2 - 27.6 fps Fake CoreAVC "single threaded" ~ 21.73 fps Source 3 - ~50mbps x264 encode, with CABAC, B-frames, and deblocking CoreAVC - 56.6 fps FFMS2 - 13.7 fps Fake CoreAVC "single threaded" ~ 14.2 fps My conclusion? In typical scenarios for this community, CoreAVC or similar high performance multi-threaded decoders are the way to go. However, in MY CASE , things are a little different. I usually work with H.264 sources that have no B-Frames, and use CAVLC instead of CABAC. Both of these sacrifices are made to facilitate real-time capture, which is a huge ugly animal all on its own! Given my typical sources, and the fact that I'm doing a LOT of transcoding at once (anywhere from 3-6 1080p encodes on a system at once, depending on how many cores I have), FFMS2, single threaded, seems like it will actually give me better throughput - provided it never bottlenecks x264! Sure, the numbers are always higher with CoreAVC, but the numbers are always with it eating up almost 100% of my CPU! That's fine for playback, but for high volume transcoding, I want the most efficient decoder for my type of files, correct? I THINK I'm seeing this correctly. What do you guys think? The best possible solution is to do all the decoding on a couple GPUs, but that's another story. Derek __________________ These are all my personal statements, not those of my employer :) Last edited by Blue_MiSfit; 2nd July 2010 at 17:53.

2nd July 2010, 12:59	#2 \| Link
Blue_MiSfit Derek Prestegard IRL Join Date: Nov 2003 Location: Los Angeles Posts: 5,989	Some more numbers, this time with timeCodec.exe. They're higher overall, but I assume that's because it avoids AviSynth etc. The advantage is, I can restrict CPU affinity reliably this way Also, since FFMS2 isn't available in this framework, I'm just using libavcodec from ffdshow-tryouts rev 3463 (may 29 2010) Source 1 (--tune fastdecode, zerolatency) CoreAVC (4 threads) - 126.3 libavcodec (1 thread) - 35.9 CoreAVC (1 thread) - 34 Source 2 (--no-cabac, --tune zerolatency) CoreAVC (4 threads) - 101.4 libavcodec (1 thread) - 30.4 CoreAVC (1 thread) - 27.7 Source 3 (all on) CoreAVC (4 threads) - 62.5 libavcodec (1 thread) - 17.6 CoreAVC (1 thread) - 16.5 Interesting. The trend is basically the same, though the differences here are less pronounced (less than 10% in most cases). Interestingly, good old MPEG-2 at 80mbps, with typical GOP structure decodes at ~44fps single threaded with FFMS2, which is almost 2x faster than DGDecode Derek __________________ These are all my personal statements, not those of my employer :)

2nd July 2010, 13:15	#3 \| Link
clsid ***** Join Date: Feb 2005 Posts: 5,647	ffdshow also has multi-threading capability if you select ffmpeg-mt as H.264 decoder. __________________ MPC-HC 2.2.1

2nd July 2010, 13:19	#4 \| Link
Blue_MiSfit Derek Prestegard IRL Join Date: Nov 2003 Location: Los Angeles Posts: 5,989	Indeed. In my testing it's always been slower than CoreAVC, but it's worth a look! Derek __________________ These are all my personal statements, not those of my employer :)

2nd July 2010, 13:58	#5 \| Link
Blue_MiSfit Derek Prestegard IRL Join Date: Nov 2003 Location: Los Angeles Posts: 5,989	Here's more results, this time using ffmpeg-mt (and including the CoreAVC results for reference Source 1 (--tune fastdecode, zerolatency) ffmpeg-mt (1 thread) - 35.5 fps ffmpeg-mt (4 threads) - 130.4 fps CoreAVC (1 thread) - 34 fps CoreAVC (4 threads) - 126.3 fps Source 2 (--no-cabac, --tune zerolatency) ffmpeg-mt (1 thread) - 28 fps ffmpeg-mt (4 threads) - 110.4 fps CoreAVC (1 thread) - 27.7 fps CoreAVC (4 threads) - 101.4 fps Source 3 (all on) ffmpeg-mt (1 thread) - 17.3 fps ffmpeg-mt (4 threads) - 66.2 fps CoreAVC (1 thread) - 16.5 fps CoreAVC (4 threads) - 62.5 fps WOW. I was NOT expecting that! ffmpeg-mt is faster than CoreAVC in all cases for my usual files! It looks like the ffmpeg devs have been working hard I'll make a pretty little graph to illustrate, and maybe put in DivX H.264 as well, I don't have DiAVC and don't feel like spending $10 right now... Derek __________________ These are all my personal statements, not those of my employer :) Last edited by Blue_MiSfit; 2nd July 2010 at 14:18.

2nd July 2010, 14:26	#6 \| Link
kieranrk Registered User Join Date: Jun 2009 Location: London, United Kingdom Posts: 707	Have you tried DiAVC? __________________ Authoring Professional Blu-rays with x264 - Open Broadcast Encoder

2nd July 2010, 23:18	#9 \| Link
Dark Shikari x264 developer Join Date: Sep 2005 Posts: 8,666	Performance may depend heavily on the source material -- e.g. whether the bitrate is high or low, etc. __________________ Follow x264 development progress \| akupenguin quotes \| x264 git status ffmpeg and x264-related consulting/coding contracts \| Doom10

3rd July 2010, 02:12	#11 \| Link
Blue_MiSfit Derek Prestegard IRL Join Date: Nov 2003 Location: Los Angeles Posts: 5,989	Indeed. The bitrate was very high in my case, usually around 60mbps (1080p24) Derek __________________ These are all my personal statements, not those of my employer :)

3rd July 2010, 02:28	#13 \| Link
Blue_MiSfit Derek Prestegard IRL Join Date: Nov 2003 Location: Los Angeles Posts: 5,989	Hence why 2/3 of my tests have CABAC disabled __________________ These are all my personal statements, not those of my employer :)

3rd July 2010, 03:12	#15 \| Link
Blue_MiSfit Derek Prestegard IRL Join Date: Nov 2003 Location: Los Angeles Posts: 5,989	Fair enough. Still, I'm trying to determine the optimal way to decode high bitrate H.264 for high volume transcode. In my mind, this means the decoder that gives the best per-thread performance. Is that logical? This seems like a valid test, no? As I said, MPEG-2 decodes very quickly as well Derek __________________ These are all my personal statements, not those of my employer :) Last edited by Blue_MiSfit; 3rd July 2010 at 03:16.

3rd July 2010, 17:29	#17 \| Link
iwod Registered User Join Date: Apr 2002 Posts: 756	Um... So we come to the conclusion that FFmpeg - Mt is fastest. No loner a reason to use CoreAVC, Divx, or Diva decoder?

3rd July 2010, 18:27	#18 \| Link
Blue_MiSfit Derek Prestegard IRL Join Date: Nov 2003 Location: Los Angeles Posts: 5,989	Well, for these (unusual) test cases anyway. I haven't bothered to do anything with standard BluRay streams or typical 4-15mbps backup type streams. Derek __________________ These are all my personal statements, not those of my employer :)

4th July 2010, 11:16	#19 \| Link
CeeJay.dk Registered User Join Date: Dec 2003 Location: Denmark Posts: 122	@Dark Shikari : You wrote on your blog about some new deblocking optimization you made to x264. Could this optimization to an encoder deblocker be applied to a decoder deblocker , like the one in FFmpeg ? Have it already been applied - Is that why the new FFmpeg 0.6 release decodes h264 faster ?

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode