benh wrote: 
> I understand what you are saying, and your example is helpful.
> 
> I think perhaps I have been misunderstanding the nature of this
> particular encoding error.  I had understood it to be an error in
> encoding some audio at an arbitrary and random timecode.  What I now
> speculate may be happening is that the error is at a fixed place near
> the end of the file, perhaps in some kind of footer.  
> 
> Otherwise, I would expect some variability in when the file stops
> playing, and it is very consistent.
> 
> I haven't been able to find adequate documentation of the ALAC format
> yet to see how a file is structured, so I haven't really validated this
> idea.
> 
> Some more information:
> 
> ffmpeg gives this error when transcoding one of the example files to
> aac:
> [alac @ 0x55794a5e1c40] invalid samples per frame: 0/s speed=79.3x
> Error while decoding stream #0:0: Invalid data found when processing
> input
> 
> Of note, I can transcode the original source flac with ffmpeg with no
> issues, so the error seems to have been introduced when I used XLD to
> transcode from FLAC to ALAC.
> 
> What's **really** interesting is that I can use ffmpeg to transcode the
> original source FLAC and the XLD-transcoded (with error) ALAC both to
> aac, and end up with *identical* files, despite the error thrown by
> ffmpeg when transcoding the ALAC file.  That is to say, an md5 hash of
> each outputted file is identical.  Both play to the very last note
> outside of LMS, and I haven't yet tried them within LMS.
> 
> This suggests to me that the payload of the misencoded file is intact
> and that the error is in a container structure.

My observation on the Smokey Robinso file was an error toward the *very*
end (confirmed by both ffmpeg and my built-in decoder) which again
caused the player to raise an error (STMn) and as a result LMS
immediately stops playback for *synchronized players*, discarding all
what is in the decoded buffer, which can be a lot as decoding happens
way ahead of playback. Standalone players are not affected neither are
groups when there is transcoding happening as faad does not/cannot
return an error, simply it returns EoF and all the previously decoded
audio is already piped to flac and will be played.



LMS 8.2 on Odroid-C4 - *SqueezeAMP!*, 5xRadio, 5xBoom, 2xDuet, 1xTouch,
1xSB3. Sonos PLAY:3, PLAY:5, Marantz NR1603, Foobar2000, ShairPortW,
2xChromecast Audio, Chromecast v1 and v2, Squeezelite on Pi,  Yamaha
WX-010, AppleTV 4, Airport Express, GGMM E5, RivaArena 1 & 3
------------------------------------------------------------------------
philippe_44's Profile: http://forums.slimdevices.com/member.php?userid=17261
View this thread: http://forums.slimdevices.com/showthread.php?t=114439

_______________________________________________
plugins mailing list
plugins@lists.slimdevices.com
http://lists.slimdevices.com/mailman/listinfo/plugins

Reply via email to