On 06/18/2013 12:22 AM, Sampo Syreeni wrote:
Agreed. But how precisely does that affect the definition of the
signal set so that it suddenly isn't at your given order?
Suppose that you'll puncture the perceptual noise floor with
amplified mic noise at 3rd order at, say, 400Hz. Why can't you just
shelf it down to 2nd order on the encoding side and/or even cut out
some of the material altogether, to taste?
say you have a zeroth order over 10 octaves, fourth order over 4 if
you're lucky, and the other orders with a bandwidth somewhere in between.
let's also assume we're in the rE domain for all those higher orders,
which means the perceptual result is governed by adding the energies,
not the amplitudes.
now if the POA subset is spectrally flat, any further order set will add
more HF energy, shifting the spectral balance. this can be compensated
by using more of the lower orders at LF, with a very messy filter
network, to yield a flat response, for any order N.
but then, if you truncate to order N-1, you lose more HF than LF enery,
hence the spectral balance will be destroyed.
i suspect this problem is far more fundamental than the comparably minor
issues you get from sub-optimal psychoacoustic decoding of normal
mixed-order sets.
usual disclaimer: severely math-challenged poster here, who'd love to be
disproved and become a happy eigenmike user ever after.
--
Jörn Nettingsmeier
Lortzingstr. 11, 45128 Essen, Tel. +49 177 7937487
Meister für Veranstaltungstechnik (Bühne/Studio)
Tonmeister VDT
http://stackingdwarves.net
_______________________________________________
Sursound mailing list
Sursound@music.vt.edu
https://mail.music.vt.edu/mailman/listinfo/sursound
