[Eric Scheirer writes]
> I wonder if some of the folks (John, Giorgio) that have been
> working on compiled SAOL implementations could comment on this?
It would be great if someone read this thread and decided to implement a
simple MPEG 2 decoder, along with a transcoder that took MPEG 2 audio
bitstreams and converted them into class sample's. Not only could we
get a quantitative datapoint, it would also be a classic benchmark program
that we could all optimize our SA decoders to do well on ...
I'd recommend David Pan's review article as a starting point:
PAN D. A TUTORIAL ON MPEG AUDIO COMPRESSION.
IEEE MULTIMEDIA, SUM, 1995, V2(N2):60-74.
It assumes you don't know very much about the topic, and gives you
enough details to start coding sample SAOL programs for the inner
loops (probably not enough detail to do the whole thing, but there
is a fair amount of bitstream info in the article as I recall ...).
This archive was generated by hypermail 2b29 : Mon Jan 28 2002 - 11:46:39 EST