From: Ray Lambert <listlizard_at_interthingy.net>
Date: Fri, 18 Feb 2005 11:58:39 -0500

Thanks to everyone (especially Andreas) for the recent information about
gapless playback.

I tried the lame --nogap method yesterday and it worked very well! My
live recordings no longer have gaps.

Unfortunately, a new problem has arisen. When you use --nogap with lame
(using VBR) it does not output the Xiph VBR table and, as a result,
rockbox appears to not be able to determine the correct track length.
Everything still plays well, but the total time display is wrong and the
track time status bar fills before the track actually finishes.

So, is there a way to have my cake and eat it too? :)

I've looked on the lame web site but can't even find --nogap documented
anywhere! (I'll search the mailing list archive later.)

Also, BlueChip provided us with this:

> http://www.gamingforce.com/forums/archive/index.php/t-29618.html
>
> "LAME's --nogap switch is only useful for multiple MP3 encodes on the
> same commandline. Instead of leaving the trailing silence, whatever
> unfilled time that remains on the end of the MP3 is used to encode the
> beginning of the next track. Because of this, --nogap is only useful
> for listening to such songs strictly in order... or you'll find that
> some of the MP3s are missing a small clip of the beginning & contain a
> small bit of the next track at the end."

I was thinking yesterday about how --nogap might work and I came up with
something very similar to this. The one difference is that I would have
expected them to take the partial frame at the end of the current track
and insert before the NEXT track. This would seem (to me) to avoid the
problem described above in the vast majority of cases. As this seems to
make more sense to me, I wonder if this isn't really what they're
doing. Perhaps I'll mention this to the lame guys.

~ray

_______________________________________________
http://cool.haxx.se/mailman/listinfo/rockbox
Received on 2005-02-18