New submission from Kieran K kie...@kunhya.com:
Libav should play the backwards compatible AVC layer and ignore the MVC
layer. The file is not being detected as an H.264 file.
http://dl.dropbox.com/u/2701213/x264/MVCSPS-1.264
--
messages: 13945
priority: normal
status: new
substatus
New submission from Kieran K kie...@kunhya.com:
(not really worth using /incoming at this stage)
Encode any file to H.264 intra with x264. Decoding will complain about:
./ffmpeg -i ~/x264/out.264
FFmpeg version git-d1c5fdf, Copyright (c) 2000-2011 the Libav developers
built on Apr 10 2011 18
Kieran K kie...@kunhya.com added the comment:
The SPS is corrupt in the first frame.
__
Libav issue tracker iss...@roundup.libav.org
https://roundup.libav.org/issue2672
__
Kieran K kie...@kunhya.com added the comment:
This happens because ffmpeg's decoder ignores a new PCE in the stream.
FFmpeg issue tracker iss...@roundup.ffmpeg.org
https://roundup.ffmpeg.org/issue2631
New submission from Kieran K kie...@kunhya.com:
Sonic Foundry Soft Encode produces AC-3 files with a 16-byte header.
File 'odd_ac3.ac3' not attached - you can download it from
https://roundup.ffmpeg.org/file1361.
--
files: odd_ac3.ac3
messages: 13818
priority: normal
status: new
Kieran K kie...@kunhya.com added the comment:
Patches welcome I guess.
FFmpeg issue tracker iss...@roundup.ffmpeg.org
https://roundup.ffmpeg.org/issue2582
Kieran K kie...@kunhya.com added the comment:
Investigated this further and this has nothing to do with LATM. It writes
a new CPE as the stream changes from 2.0 to 5.1 which the AAC decoder
rejects. Removal of the code which stops parsing of a new CPE, allows
decoding. Almost certain breaks
New submission from Kieran K kie...@kunhya.com:
http://www.aktiv-verlag.com/~webvideos/videos/2010/Surfcup.mp4
--
files: typescript
messages: 12681
priority: normal
status: new
substatus: new
title: h264_mp4toannexb filter crashes
type: bug
Kieran K kie...@kunhya.com added the comment:
There is a mostly complete copy in the private ffmpeg repo, which a few
developers have looked at.
Because of the commercial savings from this decoder and the time it took
to write, Merzbt and I are looking for a sponsor to fund it's completion
Kieran K kie...@kunhya.com added the comment:
As discussed on doom10 (http://doom10.org/index.php?topic=653.0) patch
welcome to pass through AFDs to libx264.
FFmpeg issue tracker iss...@roundup.ffmpeg.org
https://roundup.ffmpeg.org/issue2255
New submission from Kieran K kie...@kunhya.com:
Artefacts in BBC HD sample which do not appear with JM nor DXVA decoding.
See bbc_hd_artefacts.264
--
files: errors.txt
messages: 12112
priority: normal
status: new
substatus: new
title: H.264 decoding artefacts
type: bug
Kieran K kie...@kunhya.com added the comment:
implemented but not committed.
--
substatus: open - implemented
FFmpeg issue tracker iss...@roundup.ffmpeg.org
https://roundup.ffmpeg.org/issue937
New submission from Kieran K kie...@highdefnow.com:
My router doesn't play well with /incoming's ftp for some reason.
http://dl.dropbox.com/u/2701213/ffmpeg/japan_aac_switching.ts
--
files: aac.txt
messages: 10772
priority: normal
status: new
substatus: new
title: AAC channel switch
Kieran K kie...@highdefnow.com added the comment:
Looks like the H.264 decoder can't handle ANY sps_id that has the value
31.
FFmpeg issue tracker iss...@roundup.ffmpeg.org
https://roundup.ffmpeg.org/issue1930
Kieran K kie...@highdefnow.com added the comment:
get_ue_golomb_31 seems to be broken and is causing this problem.
FFmpeg issue tracker iss...@roundup.ffmpeg.org
https://roundup.ffmpeg.org/issue1930
Kieran K kie...@highdefnow.com added the comment:
Samples added.
/DOLBY-E/DOLBY-E-1.ts (57MB)
/DOLBY-E-2/DOLBY-E-2.ts (34MB)
(ftp client reported upload problems with second stream so re-uploaded)
FFmpeg issue tracker iss
Kieran K kie...@highdefnow.com added the comment:
More info:
http://www.aes.org/e-lib/browse.cfm?elib=8112
http://www.aes.org/e-lib/browse.cfm?elib=8147
FFmpeg issue tracker iss...@roundup.ffmpeg.org
https://roundup.ffmpeg.org/roundup/ffmpeg
Kieran K kie...@highdefnow.com added the comment:
MPEG2Repair (A windows program for showing receiving errors which works
with H.264 transport streams as well) shows many errors:
Sequence Frame 477(0-X) / Time 0:00:06 :
VideoWarning: Discontinuity of (1+) packet(s). First packet ending
Kieran K kie...@highdefnow.com added the comment:
Uploaded as:
consumed_only_bytes_error.mp4
_
FFmpeg issue tracker iss...@roundup.ffmpeg.org
https://roundup.ffmpeg.org/roundup/ffmpeg/issue1061
_
Kieran K kie...@highdefnow.com added the comment:
ffmpeg -i dolby_e_sample.ts only sees the first two PIDs of this
transport stream.
In fact there is a third PID (0x101) that is disguised as an AES3 PCM
stream but is in fact up to 8 channels of Dolby E (http://www.dolby.com/
professional
New submission from Kieran K kie...@highdefnow.com:
Following the discussion on #ffmpeg a few days ago I attach the Dolby E
sample in directory Dolby_e_stream and filename Dolby_e_sample.ts
(Dolby E Stream has PID 0x101)
ffmpeg -i dolby_e_sample.ts
FFmpeg version SVN-r16573, Copyright (c
21 matches
Mail list logo
