[MPlayer-users] [BUGREPORT] Audio sample-rate regression
The Wanderer
inverseparadox at comcast.net
Fri Jan 6 17:20:15 CET 2006
RC wrote:
> On Sun, 27 Nov 2005 11:44:02 -0800 RC <rcooley at spamcop.net> wrote:
>
>> As of 1.0pre7, mplayer can no longer properly play back some avi
>> files created by 1.0pre6 (or earlier). The audio is played much
>> faster, the pitch is much higher to match, and audio is completely
>> out-of-sync with video. I've only seen this problem on AVI files
>> with MP3 audio at 24/32KHz, and only a few files show this problem
>> at all...
>>
>> I've uploaded a small sample that plays back properly with 1.0pre6,
>> but with the above problems if played with 1.0pre7 or CVS
>> (051105). If needed, I can provide a few more.
>>
>> http://illhostit.com/files/9728744859146377/32k.avi
>
> 7 weeks, and no response at all.
>
> Anybody care at all?
Well, I care, just in principle (I am decidedly a perfectionist) - but
I'm not exactly in a position to do much about it.
It occurs to me that I may, just perhaps, have another file with a
comparable problem. It identifies itself as having been created with
MEncoder 0.92 (!), which was probably current at the time. I vaguely
seem to recall its having played back correctly at one point, but for
some time now it has played much too fast - I can get it back to only
slightly too fast by using '-speed 0.9', but it still seems odd.
However, I do not notice the "pitch too high" or "A/V out of sync"
problems you describe[1], so it may not be quite the same thing.
We should both, perhaps, do regression testing to see what commit caused
the change... you've done more in that direction than I have, but
'1.0pre6' to '1.0pre7' is still quite a jump.
[1] Rather, A/V desync does occur, but not at all for about the first
third of the file; then suddenly it desyncs at a jump, and varies
slightly thereafter.
--
The Wanderer
Warning: Simply because I argue an issue does not mean I agree with any
side of it.
Secrecy is the beginning of tyranny.
More information about the MPlayer-users
mailing list