Musepack playback glitch (Gigabeat F)

[ilikedirt]

Hi,

I stumbled accross a glitch with a certain Musepack file. A loud "blip" sound appers on certain times in this file when played back with Rockbox (latest build r17732 as well as older build from 2008-05-19). The same file plays back in foobar2000 just fine. I dont know if more files are affected. If someone is interested I could send the file for examination.

[saratoga]

Theres been a lot of changes to MPC in the last two months.  Trying a build from mid or early April would be interesting.

[ilikedirt]

Is there a place where I can find builds older than one month? Or can someone upload one for me? That would be somewhat easier for me than setting up the build enviroment myself - appart from me not beeing familiar with using it.

[GodEater]

Is there a place where I can find builds older than one month?

Nope - 1 month is all we keep.

Or can someone upload one for me? That would be somewhat easier for me than setting up the build environment myself - apart from me not being familiar with using it.

Possibly someone can - but it might be worth the trouble of learning how to do the building - it's not as tough and scary as it looks!

[ilikedirt]

I build r16903 from April 1st. The glitches on my file are already there.

[Buschel]

Hi,

is there any possibility get this file for further debugging? I have an idea about this issue...

Buschel

[ilikedirt]

Hi Buschel,

of course. How would you like me to send you the file, email? You could PM me your adress.

[Buschel]

hi,

should be solved with r17829. the (already present) hack never seemed to work correctly due to the wrong default scalefactor.
musepack uses differential scf-coding in time domain. if you decode from the beginning everything is fine. if you seek with in a file, the decoder may not know the reference-scf after seek -- in the file i've tested this happened with highest subband. in this case the decoder uses a much too high scf what results in burst noise. solution (or better hack) is to use the lowest scf (equals fading out of the subband) until there was a reference scf decoded (1st non-differential coded scf).

please report, if the problem is solved now.

thanks,
buschel

EDIT: The bug is reported as fixed with r17829.