Daniel Verkamp added the comment:
Unfortunately, this only works on the cut 10M sample, not the actual
(nearly 4 GB) file; stream 0 has an index (at the end of the file).
avi->non_interleaved is getting set by:
avi->non_interleaved |= get_le32(pb) & AVIF_MUSTUSEINDEX;
The index s
Daniel Verkamp added the comment:
This could very well be a FRAPS bug; the 3.2.2 changelog notes:
"Fixed trouble opening movies with VLC, Avidemux, AVI synth, Virtualdub
and some other applications"
It's possible FRAPS just write
New submission from Daniel Verkamp :
An AVI produced by FRAPS version 3.2.1 (build 11425) does not decode
with ffmpeg.
First 10 MB of file:
http://drv.nu/temp/fraps-broken-ni.avi
ffmpeg output for this file:
$ ./ffmpeg -loglevel debug -i ~/temp/wowvids/WoW\ 2010-05-19\ 22-03-40-
49.avi
Daniel Verkamp added the comment:
Would a patch moving all of those 4-bit ADPCM codecs to the generic
av_get_bits_per_sample() code be ok?
--
nosy: +DrV
_
FFmpeg issue tracker
<https://roundup.ffmpeg.org/roundup/ffmpeg/issue1
Daniel Verkamp added the comment:
Applied, thanks.
--
status: open -> closed
substatus: approved -> fixed
_
FFmpeg issue tracker
<https://roundup.ffmpeg.org/roundup/ffmpeg/issue1569>
_
Daniel Verkamp added the comment:
Add me to nosy
--
nosy: +DrV
_
FFmpeg issue tracker
<https://roundup.ffmpeg.org/roundup/ffmpeg/issue1569>
_
Daniel Verkamp added the comment:
Sample created by ffmpeg with patch applied
_
FFmpeg issue tracker
<https://roundup.ffmpeg.org/roundup/ffmpeg/issue1569>
__
Daniel Verkamp added the comment:
Attached patch fixes the issue in the FLV muxer and the equivalent
problem in the demuxer.
Tested with the Windows Flash browser plugin - unsigned sounds correct,
signed does not.
_
FFmpeg issue tracker
Daniel Verkamp added the comment:
With the addition of error messages in r20757, this now prints:
[...]
Output #0, avi, to '/home/daniel/temp/fit-roll.avi':
Stream #0.0(eng), 3/125: Audio: mp2, 48000 Hz, 2 channels, s16, 64 kb/s
Stream mapping:
Stream #0.0 -> #0.0
Pres
Daniel Verkamp added the comment:
Duplicate of issue 1503
--
substatus: reproduced -> duplicate
superseder: +Crash decoding wmav2
_
FFmpeg issue tracker
<https://roundup.ffmpeg.org/roundup/ffmpeg/issue1532>
_
Daniel Verkamp added the comment:
Tried setting s->nb_block_sizes = 8 both inside if block (line 150) and after
the if block (line 152), still crashes (now at a different location):
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7ffa168626f0 (LWP 25
Daniel Verkamp added the comment:
Also note that sample plays fine in Windows Media Player.
_
FFmpeg issue tracker
<https://roundup.ffmpeg.org/roundup/ffmpeg/issue1503>
_
New submission from Daniel Verkamp :
Decoding the audio of this file crashes ffmpeg: http://drv.nu/temp/fit-roll.wmv
(~5 MB)
Input audio is wmav2. The file was created by the Xbox360 game Forza 3.
This is the first problem encountered: if (v >= s->nb_block_sizes) is triggered:
St
Daniel Verkamp added the comment:
Re-tested with latest ffmpeg sources, and I no longer get the errors (previous
test was with 0.5 release).
I also get the same md5sum as you for the decoded audio, and it sounds okay.
So I think this is fixed...
--
nosy: +DrV
Daniel Verkamp added the comment:
(Sorry, missed responses, as I assumed they'd be emailed...)
I am using the Gentoo-provided libspeex on x86-64 (speex-1.2_rc1).
With the same sample as before:
$ ffmpeg -i noah-2sec-1600.spx /tmp/dec.wav
FFmpeg version 0.5, Copyright (c) 2000-2009 Fa
Daniel Verkamp added the comment:
This is probably the same as issue 731...
_
FFmpeg issue tracker
<https://roundup.ffmpeg.org/roundup/ffmpeg/issue1113>
_
New submission from Daniel Verkamp :
Decoding any .spx file with ffmpeg results in error messages like the following:
notification: Invalid mode encountered. The stream is corrupted.
[libspeex @ 0xdafaa0]Error decoding speex frame
Error while decoding stream #0.0
notification: Invalid mode
Daniel Verkamp added the comment:
Looks like the problem is that the file uses 48-bit RGB (16 bits per channel),
which isn't handled.
FFmpeg issue tracker
<https://roundup.ffmpeg.org/roundup/ffmpeg/issue813>
18 matches
Mail list logo
