Re: [FFmpeg-devel] [FFmpeg-cvslog] fftools/sync_queue: use timebase from input frames/packets

Re: [FFmpeg-devel] [FFmpeg-cvslog] fftools/sync_queue: use timebase from input frames/packets

[Michael Niedermayer]

[-- Attachment #1.1: Type: text/plain, Size: 1334 bytes --]

On Sun, Apr 09, 2023 at 01:50:08PM +0000, Anton Khirnov wrote:
> ffmpeg | branch: master | Anton Khirnov <anton@khirnov.net> | Tue Mar 21 12:36:03 2023 +0100| [090950f8327c01b1775615b1da723dc6cc8a30a0] | committer: Anton Khirnov
> 
> fftools/sync_queue: use timebase from input frames/packets
> 
> They are always properly set now. Avoid a separate timebase-setting
> call, which duplicates the knowledge of the timebase being used.

This looses bitrate information in some cases

./ffmpeg -analyzeduration 1M -y -i ~/videos/mm-short.mpg -vcodec copy -aspect 16:9 -vframes 3 tmp.avi
./ffprobe tmp.avi

Stream #0:0: Video: mpeg2video (Main) (mpg2 / 0x3267706D), yuv420p(tv, bt470bg/bt470m/bt470m, bottom first), 720x576 [SAR 16:15 DAR 4:3], 8061 kb/s, SAR 64:45 DAR 16:9, 50 fps, 50 tbr, 50 tbn
vs.
Stream #0:0: Video: mpeg2video (Main) (mpg2 / 0x3267706D), yuv420p(tv, bt470bg/bt470m/bt470m, bottom first), 720x576 [SAR 16:15 DAR 4:3], SAR 64:45 DAR 16:9, 50 fps, 50 tbr, 50 tbn


[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

The real ebay dictionary, page 1
"Used only once"    - "Some unspecified defect prevented a second use"
"In good condition" - "Can be repaird by experienced expert"
"As is" - "You wouldnt want it even if you were payed for it, if you knew ..."

[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

[-- Attachment #2: Type: text/plain, Size: 251 bytes --]

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

Re: [FFmpeg-devel] [FFmpeg-cvslog] fftools/sync_queue: use timebase from input frames/packets

[Anton Khirnov]

Quoting Michael Niedermayer (2023-04-10 21:01:40)
> On Sun, Apr 09, 2023 at 01:50:08PM +0000, Anton Khirnov wrote:
> > ffmpeg | branch: master | Anton Khirnov <anton@khirnov.net> | Tue Mar 21 12:36:03 2023 +0100| [090950f8327c01b1775615b1da723dc6cc8a30a0] | committer: Anton Khirnov
> > 
> > fftools/sync_queue: use timebase from input frames/packets
> > 
> > They are always properly set now. Avoid a separate timebase-setting
> > call, which duplicates the knowledge of the timebase being used.
> 
> This looses bitrate information in some cases
> 
> ./ffmpeg -analyzeduration 1M -y -i ~/videos/mm-short.mpg -vcodec copy -aspect 16:9 -vframes 3 tmp.avi

Cannot reproduce with a random mpeg file. Where can I find this sample?

-- 
Anton Khirnov
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

Re: [FFmpeg-devel] [FFmpeg-cvslog] fftools/sync_queue: use timebase from input frames/packets

[Michael Niedermayer]

[-- Attachment #1.1: Type: text/plain, Size: 1268 bytes --]

On Wed, Apr 12, 2023 at 10:06:19AM +0200, Anton Khirnov wrote:
> Quoting Michael Niedermayer (2023-04-10 21:01:40)
> > On Sun, Apr 09, 2023 at 01:50:08PM +0000, Anton Khirnov wrote:
> > > ffmpeg | branch: master | Anton Khirnov <anton@khirnov.net> | Tue Mar 21 12:36:03 2023 +0100| [090950f8327c01b1775615b1da723dc6cc8a30a0] | committer: Anton Khirnov
> > > 
> > > fftools/sync_queue: use timebase from input frames/packets
> > > 
> > > They are always properly set now. Avoid a separate timebase-setting
> > > call, which duplicates the knowledge of the timebase being used.
> > 
> > This looses bitrate information in some cases
> > 
> > ./ffmpeg -analyzeduration 1M -y -i ~/videos/mm-short.mpg -vcodec copy -aspect 16:9 -vframes 3 tmp.avi
> 
> Cannot reproduce with a random mpeg file. Where can I find this sample?

sorry, its here:
https://samples.ffmpeg.org/MPEG2/mm-short.mpg

[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Awnsering whenever a program halts or runs forever is
On a turing machine, in general impossible (turings halting problem).
On any real computer, always possible as a real computer has a finite number
of states N, and will either halt in less than N cycles or never halt.

[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

[-- Attachment #2: Type: text/plain, Size: 251 bytes --]

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

Re: [FFmpeg-devel] [FFmpeg-cvslog] fftools/sync_queue: use timebase from input frames/packets

[Michael Niedermayer]

[-- Attachment #1.1: Type: text/plain, Size: 1397 bytes --]

On Wed, Apr 12, 2023 at 10:41:50AM +0200, Michael Niedermayer wrote:
> On Wed, Apr 12, 2023 at 10:06:19AM +0200, Anton Khirnov wrote:
> > Quoting Michael Niedermayer (2023-04-10 21:01:40)
> > > On Sun, Apr 09, 2023 at 01:50:08PM +0000, Anton Khirnov wrote:
> > > > ffmpeg | branch: master | Anton Khirnov <anton@khirnov.net> | Tue Mar 21 12:36:03 2023 +0100| [090950f8327c01b1775615b1da723dc6cc8a30a0] | committer: Anton Khirnov
> > > > 
> > > > fftools/sync_queue: use timebase from input frames/packets
> > > > 
> > > > They are always properly set now. Avoid a separate timebase-setting
> > > > call, which duplicates the knowledge of the timebase being used.
> > > 
> > > This looses bitrate information in some cases
> > > 
> > > ./ffmpeg -analyzeduration 1M -y -i ~/videos/mm-short.mpg -vcodec copy -aspect 16:9 -vframes 3 tmp.avi
> > 
> > Cannot reproduce with a random mpeg file. Where can I find this sample?
> 
> sorry, its here:
> https://samples.ffmpeg.org/MPEG2/mm-short.mpg

btw in case anyone wonders, this is just the first 2MB of matrixbench

cmp matrixbench_mpeg2.mpg mm-short.mpg
cmp: EOF on mm-short.mpg after byte 2048000, in line 9048

thx

[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Elect your leaders based on what they did after the last election, not
based on what they say before an election.


[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

[-- Attachment #2: Type: text/plain, Size: 251 bytes --]

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

Re: [FFmpeg-devel] [FFmpeg-cvslog] fftools/sync_queue: use timebase from input frames/packets

[Anton Khirnov]

After some investigation, it seems that before this patch there was a
bug in the sync queue setup code that caused a wrong timebase to be used
for the audio stream. Because of that, many more audio frames were
output than should have been (116 frames/133632 samples/2.8 seconds).

After the patch, the correct timebase is used and only 9 audio frames
(10368 samples, 0.216 seconds) are encoded. This is much closer to the
3 video frames requested in your commandline and thus more correct.
As a result, the file is now much smaller and the
 if (lensum*9/10 > maxpos || lensum < maxpos*9/10)
check currently on line 476 of avidec.c triggers and stops bitrate
calculation. I cannot say whether that check is faulty, but this patch
seems like it's doing what it should.

-- 
Anton Khirnov
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".