Hi!

Wolfram Gloger wrote:

>>Actually, that's not a crime. There's no requirement that MPEG audio
>>frames are aligned inside PES packets (i.e. a frame header immediately
>>follows the PES header).
> 
> 
> Yes; I didn't state otherwise..

But your patch did ;)

>>How often does the "skipping" message appear? Does it happen while the
>>file is indexed or when it is exported?
> 
> 
> It happens whenever the file is opened (no matter whether already
> indexed or not) and repeats for some \approx 160 times, corresponding
> to the skipping of an entire TS packet until the MPA header ist
> found (in the next TS packet I assume).

In that case, the stream is probably broken.

> When it is exported, the problem with the timestamps shows up and
> there are audible "skips".

Did you try another (non-ffmpeg) muxer?

>>>However, also the old ffmpeg version included with dvbcut appears to
>>>have this "problem"; at least when the libavformat muxer is used there
>>>are error messages about "non-monotonous timestamp" when such packets
>>>are encountered.
>>>
>>>Did anyone else ever notice this?
>>
>>The "non-monotonous" message is not an indicator - the libavformat muxer
>>(at least the included version) uses a wrong formula to calculate the
>>timestamps (DTS, IIRC).
> 
> 
> Could you perhaps elaborate so the DTS computation can be fixed?

I will have to grep my mail archives for that. Stay tuned.

[...]
>>An "item" is always a PES packet, not an audio or video frame. Since
>>your patch enforces an alignment that actually need not be there, you
>>may end up with no audio at all.
> 
> 
> That is indeed possible, yes, but only if there are _never_ packets
> with an MPA header after the PES header (*).
> 
> 
>>We already had a similar problem with
>>Channel4 (UK) which transmits unaligned MPEG audio.
> 
> 
> Was that (*) the case there?

Exactly.

>>I'm not sure if the input file is the problem in the first place. It may
>>also be a bug in the version of ffmpeg you're using. Can you confirm
>>that the stream is malformed,
> 
> 
> No, I'm not in fact claiming the stream is malformed..

Well, if the stream is well-formed, then there's a bug in dvbcut that
needs to be fixed. If it's not, I may consider adding a workaround - but
only if it doesn't break anything else, including Channel4.

> I am claiming that dvbcut passes un-framed audio data to libavcodec,
> and that fails -- spectacularly with ffmpeg-current but I believe
> it was never documented to work correctly and also silently goes
> wrong with the old ffmpeg version in dvbcut.

I guess I need a piece of the stream to work out the details.

-- 
Michael "Tired" Riepe <[EMAIL PROTECTED]>
X-Tired: Each morning I get up I die a little

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
DVBCUT-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dvbcut-user

Reply via email to