Howdy,
> OK, and what about the bits included in table B5 ?
Well, I was trying to avoid having to copy an entire table from the spec,
but it's small, so here goes... (I wish they would make the spec available
online in .pdf format or something - the audio part is not worth what ISO
charges for it, as it is full of errors, omissions, and poor grammar.)
Layer Protected Fields
I bits 16...31 of header
bit allocation
II bits 16...31 of header
bit allocation
scalefactor selection information
III bits 16...31 of header
side information:
- bits 0...135 of audio_data in single_channel mode
- bits 0...255 of audio_data in other modes
I note that they are counting from 0 in all cases, and that MPEG-2 LSF has
slightly different side info lengths. (No, I'm not going to look them up
for you - check the source or ISO 13818-3.)
> I saw on this list that the ISO CRC is broken. Is yours OK ?
As I may have replied to that assertion earlier, I don't know that the ISO
CRC implementation is broken so much as simply missing. I believe that the
'dist10' distribution only implements CRC checking for layers I & II. And
yes, my CRC polynomial seems to work just fine, thanks (decodes streams
encoded by commercial encoders, encodes streams decodable by commercial
decoders), or I wouldn't have posted it...
Alex
--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
