[FFmpeg-user] MPEG-TS CBR?

Anjo Krank anjo at krank.net
Fri Apr 13 14:28:33 CEST 2012


So I got some more info on the rejected CBR TS. Checking the generated streams with "crosscheck" yields a long list of errors:

 http://www.krank.net/report.txt

The problem seems to the in fact the muxer, as de- and remuxing with a different muxer yielded no more errors.

How "usable" or "broadcast-ready" is the muxing in ffmpeg? Searching the web and other comments seemed to say that there is no actually working open-source muxer out there… at least in the sense that it could create broadcast-ready files. Is that true?

I also found a discussion about some patches that were written in the Google SOC 2009 which seemed to end in abandonment. Was that stuff actually working or better than the current implementation? Would it be worth the bother to clean it up and try to commit it?

 http://thread.gmane.org/gmane.comp.video.ffmpeg.devel/52773/focus=56011

Cheers, Anjo

Am 04.04.2012 um 09:43 schrieb Anjo Krank:

> 
> Am 03.04.2012 um 20:28 schrieb Michael Niedermayer:
> 
>>> It's the tiny white line at the left, where throughput drops to 3.5m. The resulting movie files looks totally normal at this location. From mucho googling I had the impression that it could maybe be a wrong setting for the  -bufsize 1000k. Is this on any way calculable or are they simply a bunch of legal settings?
>> 
>> there are simply a bunch of legal settings, try a smaller bufsize and
>> see if this fixes the issue
> 
> Can you tell me where to find those legal settings? Are they dependent on the target (TV vs DVD)? I searched high and low on google but found only very contradictory info, which is why I came up with the bufsize=bandwidth/4 scheme in the first place...
> 
> Cheers, Anjo
> _______________________________________________
> ffmpeg-user mailing list
> ffmpeg-user at ffmpeg.org
> http://ffmpeg.org/mailman/listinfo/ffmpeg-user



More information about the ffmpeg-user mailing list