>
> struct scalefac_struct sfBandIndex[6] =
> {
> ...
> { /* Table B.2.c: 24 kHz */
> {0,6,12,18,24,30,36,44,54,66,80,96,114,136,162,194,232,278,330,394,464,540,576},
> 332 !!!
> {0,4,8,12,18,26,36,48,62,80,104,136,180,192}
> },
> { /* Table B.2.a: 16 kHz */
> {0,6,12,18,24,30,36,44,54,66,80,96,116,140,168,200,238,248,336,396,464,522,576},
> 284 !!!
> {0,4,8,12,18,26,36,48,62,80,104,134,174,192}
> },
> ...
> };
>
> Regards,
>
> -Leonid
Looks like another bug :-). I checked with the MPEG2 standard and
according to the printed tables, Leonid is correct. Have you
tested this? I'll bet the bug causes some noticable MPEG2
artifacts.
I found a similar bug a while ago: the '54' in the 16kHz table is
a 45 in the ISO code. But this value was so different that it would
crash on almost all 16kHz input files. This is why the old 'screamer'
code would not work at 16kHz.
Mark
--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
