On 2013-06-03, Fons Adriaensen wrote:
No surprise, a lot of issues got mixed up in this thread:
Agreed.
2) Will using some filters help to achieve this ?
Here I'm not too sure yet. If you look at e.g.
dafx04.na.infn.it/WebProc/Proc/P_250.pdf , on the WFS side they seem to
be including frequency dependent terms for the correction as well. Those
Hankel functions are closely related to NFC-HOA's distance correction
terms, and they become about in WFS because even if the emitter is
linear, in the vertical direction their propagation is still
cylindrical, leading to "spherical looking terms". The point is, I don't
think those terms go away when you bend the line array into a
pantophonic rig, so that the amplitude correction might become frequency
dependent there as well, while at the same time being of a form which
could be absorbed into the existing near field compensation and monopole
synthesis filters.
3) Is there a fundamental difference between B-format signals produced
by a tetra mic and A/B processing and 'native' B-format obtained
directly from capsules?
What Fons said. If you want to go to a system which is "genuinely
pantophonic", your mic needs to have a geometry with very steep
attenuation off the horizontal plane, and your speakers need to be
vertical line sources. In such a cylindrical setup those weird extra
terms go away entirely, but then it's a given that the system has very
little to do with conventional periphony or pantophony, and it's not at
all certain you'd ever want to go there.
All I was trying to say is that pantophony now inherits some of the
characteristics of such a system and some of those of full periphony.
That first part is what threw Faller off, and is something that while it
certainly doesn't cause ambisonic to not work, might not have been
analysed in full before the WFS folks came about and started throwing
around their correction terms.
Or has it? It's been a while since I went through Gerzon's work, and we
all know how it has the tendency to foresee this kind of stuff.
4) Given W,X,Y,Z, and when using a horizontal decode, is it possible
to replace W by something that would match the vertical pattern of X
and Y ?
The answers are no, no, no, and no. (4) can be approximated
by using higher order signals.
Fons's is a much more elegant way of looking at this than mine was.
--
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
