Hello,

i'm looking for an overview of the advantages of the AAC format over Vorbis (I mean what is defined by the spec, not the current state of their implementations or psymodels). On Wikipedia there's just a very general list comparing it to MP3.

I'd be grateful for any help.

Features of AAC-LC/Main that Vorbis doesn't have:

• long term prediction (backward adaptive & forward adaptive)
• adaptive M/S matrixing for each subband (Vorbis relies on channel interleaved vector quantization, so it's theoretically possible to do something similar but it's more complicated and would require bigger code books)
• perceptual noise substitution
• temporal noise shaping
• grouping short windows that share a set of scalefactors (Vorbis doesn't allow grouping short transform blocks that share floor vectors)

And then there's of course "SBR" (subband replication, HE AAC) which seems to work pretty well for very low bitrates.

Cheers,
SG

Not exactly what you asked for, but might help:
http://gabriel.mp3-tech.org/blog/index.php...ry071211-114534

Great summary IMHO!

ot, and i thought i'am the only one who believes flat-file blog is good enough

OT. This link lead me to an interesting info, Kostya’s Wild Codec World, where mentioned a new aac encoder being developed as part of ffmpeg project.

Thank you guys!