On 2011-09-19, Fons Adriaensen wrote:
Though perhaps more surround than most would think. Come with me,
here: if you take a look at the compensation circuitry which turns
A-format into B-format within the conventional SoundField microphone,
isn't it pretty much *exactly* a 3D Blumlein shuffler?
It's not the same. The shuffler is designed to process signals that
have no significant level differences in the frequency range it
handles, just the phase differences.
Like the soundfield above some 10kHz? No?
It is normally used with omni mics, or directional ones pointing in
the same direction.
Yes. But the entirety of the soundfield approaches an omni mic at the
very lowest frequencies. Between that 10-12kHz high range and the more
usual 2-4kHz well-optimized range (or so), it's so'n'so. At the very
highest frequencies it again approaches a monopole, because of the
stupendeous amount of spatial aliasing, which then leads to Gerzon's
idea that the energy analysis applies even off-centre, and in fact in
many cases beyond that.
The capsules in an A-format mic are directional, also at those
frequencies where phase difference become significant, and the basic
A-B processing is based on that directionality while the phase
differences are just a nuisance.
My problem is that I came to process/understand this acoustical problem
from above/outside. I don't see it the way or many other on-list do.
What I see is that there is no such thing as a directional capsule. It's
just one more imperfect electromechanical thingy out there. I'm for
example perfectly sure that given the right apparatus, I could excite a
*nasty* resonance in any of the capsules used to build a SoundField Mic.
Even an "impossible" resonance within the SF itself, given some time and
proper measurements.
The point with the A-format usual mic isn't about four separate,
directional mics. It's about what Gerzon already said in one of his
papers: Gaussian quadrature against *all* of the mic array, and then
inversion into local soundfield.
Gerzon never talked about what to do with the stuff that goes against
the local diffraction limit. What I suggest is that Blumlein Shuffling
might have something to tell about that, perceptually. At the same time,
we should be *very* careful about where to apply that sort of perceptual
stuff, because we now know much more about both a) the physical stuff,
and b) when, where and how the physical turns into the psychoacoustical.
For example, why don't you try a relative-far-field-detonation wave
within an anachoic chamber. It's going to give you rather funky results
towards on SF mic, when the explosion is held constant and the mic is
slowly turned, wrt the source of the detonation.This
OTOH, the shuffler algorithm is exactly the one used to extract the
Ambisonic X,Y,Z signals by mics using omni capsules on a rigid spere.
This is something totally new to me. Rigid sphere? Originally it was
about an open 1D sphere (thus, "line segment", so how did it mutate into
a rigid sphere in between, in the process?
I can get how it might have something to do with the rigid sphere as
well, but I don't see the direct, formal, mathematical connection.
Especially since since the basic topology of the problem seems to be
somewhat off.
--
Sampo Syreeni, aka decoy - [email protected], http://decoy.iki.fi/front
+358-50-5756111, 025E D175 ABE5 027C 9494 EEB0 E090 8BA9 0509 85C2
_______________________________________________
Sursound mailing list
[email protected]
https://mail.music.vt.edu/mailman/listinfo/sursound
