[Paranoia] --sample-offset option

[Артём Попов]

2011/3/21 Thomas Vander Stichele <thomas at apestaart.org>:
> Hi everyone,
>
> in morituri, my cd ripper, I use AccurateRip to determine the correct
> sample offset for a drive.
>
> However, cdparanoia does something I don't understand for negative
> offsets.
>
> Consider this output:
>
> $ cdparanoia --sample-offset=0
> 1[00:00:00.00]-1[00:00:46.15] /tmp/offset0.wav
> cdparanoia III release 10.2 (September 11, 2008)
>
>
> Ripping from sector       0 (track  1 [0:00.00])
>          to sector    3465 (track  1 [0:46.15])
>
> outputting to /tmp/offset0.wav
>
>  (== PROGRESS == [                              | 003465 00 ] == :^D * ==)
>
> Done.
>
>
> [morituri-trunk] [thomas at otto trunk]$ cdparanoia --sample-offset=-1 1[00:00:00.00]-1[00:00:46.15] /tmp/offset-1.wav
> cdparanoia III release 10.2 (September 11, 2008)
>
>
> Ripping from sector       0 (track  1 [0:00.01])
>          to sector    3464 (track  1 [0:46.15])
>
> outputting to /tmp/offset-1.wav
>
>  (== PROGRESS == [                              | 003464 00 ] == :^D * ==)
>
> Done.
>
>
> [morituri-trunk] [thomas at otto trunk]$
> [morituri-trunk] [thomas at otto trunk]$ cdparanoia --sample-offset=1 1[00:00:00.00]-1[00:00:46.15] /tmp/offset1.wav
> cdparanoia III release 10.2 (September 11, 2008)
>
>
> Ripping from sector       0 (track  1 [0:00.00])
>          to sector    3465 (track  1 [0:46.15])
>
> outputting to /tmp/offset1.wav
>
>  (== PROGRESS == [                              | 003465 00 ] == :^D * ==)
>
> Done.
>
>
>
> and the resulting file sizes:
>
> $ ls -l /tmp/offset*wav
> -rw-r--r--. 1 thomas sbox 8152076 Mar 20 19:00 /tmp/offset0.wav
> -rw-r--r--. 1 thomas sbox 8149724 Mar 20 19:01 /tmp/offset-1.wav
> -rw-r--r--. 1 thomas sbox 8152076 Mar 20 19:03 /tmp/offset1.wav
>
> Why is the negative sample offset provoking cdparanoia to read one sector less ?
>
> How can I make it read all 3466 sectors at this negative sample offset ?
>
> Thanks
> Thomas
>

Hello Thomas,

I may be wrong, but a sample offset of -1 would require overreading
into the lead-in and it's a feature that should be supported by your
drive.

On the other hand, cdparanoia's overreading feature is said to be
broken. There's a patch for fixing it, but I haven't tested it at all:
https://bugzilla.redhat.com/show_bug.cgi?id=635925

--Artem