[ lame-Bugs-3555476 ] Resampling ruins end of audio loop

Bugs item #3555476, was opened at 2012-08-08 17:41
Message generated for change (Comment added) made by sobukus
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100290&aid=3555476&group_id=290

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Unknown
Group: Quality
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Steven Vachon (pr0m3th)
Assigned to: Nobody/Anonymous (nobody)
Summary: Resampling ruins end of audio loop

Initial Comment:
Before encoding, the loop plays back smoothly; after, there seems to be a half-second gap at the end.

The command I'm using is:
lame -q 0 -V 8 -b 32 -B 320 --resample 44.1 -m j music.wav music.mp3

I've tried all sorts of variations of the above. The output file must be 44100hz so that Adobe Flash will accept it for import.

WAV and MP3 files:
http://www.svachon.com/lame-before-after.zip

----------------------------------------------------------------------

Comment By: Thomas Orgis (sobukus)
Date: 2012-08-16 13:45

Message:
mpg123 should play the loop encoded by LAME just fine (or any program using
libmpg123). I spent quite some effort on getting that right (skipping
samples at beginning and end; sample-accurate seeking).

You can actually check using mpg123's --loop parameter.

(Disclaimer about strange bugs applies;-)


----------------------------------------------------------------------

Comment By: Steven Vachon (pr0m3th)
Date: 2012-08-16 09:40

Message:
sf says I can't send you a message 'cuz you haven't provided an email
address.. ?

----------------------------------------------------------------------

Comment By: Michael (mjbshaws)
Date: 2012-08-16 09:12

Message:
It skips the gap (the delay added to the beginning and end). It doesn't
remove it. My goal is sample exactness though, so it's a bit more
complicated than just play(someTime, 0).

Anyway, I hate to keep having a full discussion on a bug tracker about
this. Send me a message if you want more details, as I don't want to keep
spamming people with Flash talk.

----------------------------------------------------------------------

Comment By: Steven Vachon (pr0m3th)
Date: 2012-08-16 07:31

Message:
Hmm, well, I just did sound.play(124, 999999) and it skips over the gap
every time. Is this basically what your SWF file would do, or does it
remove the silence from the MP3 itself?

----------------------------------------------------------------------

Comment By: Michael (mjbshaws)
Date: 2012-08-16 07:21

Message:
As far as I know, this cannot be avoided. There's a popular page people
point to: http://www.compuphase.com/mp3/mp3loops.htm that talks about
making loop-able MP3 tracks, but it's quite involved, the program didn't
work for me, and it sacrifices audio quality, from my understanding.

Sorry to do a shameless plug, but I'm actually working on a program that'll
let you compile your MP3s into an SWF file and it'll handle take care of
this gap for you (SWF files just skip these silent samples... long story).
If you're interested, I can let you know when I have a working pre-alpha
version (hopefully by the end of this weekend). I'd be happy to have a
tester :)

----------------------------------------------------------------------

Comment By: Steven Vachon (pr0m3th)
Date: 2012-08-15 18:11

Message:
Oh, my bad. I just read through the part of the technical FAQ that pertains
to my issue, and I'm not really sure what the solution would be. Can
ENCDELAY have a value of 0? If so, why doesn't it already to avoid this
issue?

----------------------------------------------------------------------

Comment By: Michael (mjbshaws)
Date: 2012-08-15 13:07

Message:
This should be expected. The technical FAQ explains this:
http://lame.sourceforge.net/tech-FAQ.txt

----------------------------------------------------------------------

Comment By: Steven Vachon (pr0m3th)
Date: 2012-08-08 17:50

Message:
Actually, it adds the silence to the BEGINNING of the mp3 file, not the
end.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100290&aid=3555476&group_id=290

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/