[vorbis-dev] Spectral band replication

[Erik Kruus]

Philip Cowles wrote:
> Then, there is the fundamental razor -- good ol' Nyquist.  If you sample at a
> certain rate, anything you bung in afterwards is just noise.  You invented it
> chum.  But -- do we want spurious musical inventions?  Is this the point of OGG?

   The point of OGG is sound "quality" achieved per bit sent.  Any tool that achieves
this is not to be poo-pooed.  IF one is willing to consider that supplying "nothing"
above Nyquist is an equally spurious musical invention as providing a remotely
reasonable guess [optimally for zero or few bits:-)] THEN one is prepared to be
less critical of people's SBR experiments.

HOWEVER, IF one's goal is to supply everything above Nyquist highly accurately, THEN
you probably have less cares about bit rate today.  In this case if high bit rate
OGG satisfies your subjective listening critique, and it falls within your
acceptable price range, by all means use OGG for that too, and ignore the SBR
folks as technical tricksters at this stage --- although they may just be able
to lower your high-quality bit rate some day if they pull enough tricks out
of their bag :)

--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'vorbis-dev-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.