I feel a little diffident in commenting on this in the presence of so many
experts on the Soundfield mike in theory as well as in practice,
but unless I am misunderstanding how it works, there are VERY serious
problems of other kinds with using it at the kinds of distances (fractions
of a meter less than 1/2 , much less often enough) where proximity
effect becomes really major.
Namely, as I understand it, the way the B format signals are built is
predicated upon the distances among the four capsules being quite small
compared to the distance of the source, for the following reason:
Compensation is needed for the fact that the capsules are on the faces of
a tetrahedron, not coincident and all at the center. This compensation
is based on the fact that at reasonable distances to the source, the
differences of the distances to the mikes is obtained by orthogonal
projection on the axis of arrival of the sound(to a very good
apporximation).
To make sense of this jargon, suppose a source is on the line that is
equistant from three of the capsules. Then its distance to those three
will always be the same, and if the source is reasonably far away the
distance to the fourth capsule will be a constnat. This comes from the
Pythagorean theorem limit case in effect: at large distances , the
difference between A to S and B to S is equal to the length of the
projection of the line from A to B onto the line from A to S (or B to S
these being parallel in the limit case).
If one does NOT have such large distance to the source, the variation of
distances to the capsules will be extreme and also complicated.
Just think of how the distances to the four face centers of the
tetrahedron will vary in odd ways when the source is close by!
So it seems to me(and I am prepared to be all wrong!) that
the Soundfield mike could not be expected to work at all well
except when the source is quite far away--a matter of meters, not
inches. At close distances, there will be wild phase differentials among
the four mike capsule outputs of a kind that depends on the distance
of the source from the center of the mike--something which the mike
does not "know" so that it cannot be compensated for.
Am I all wet here?
Robert
On Sat, 23 Jul 2011, dave.mal...@york.ac.uk wrote:
Hi Folks,
I have an interesting question (well, I think it's interesting). The
Soundfield microphone, like any directional microphone, has a boosted bass
response to close sounds. When listening to this through a speaker rig, we
hear this boost and tend to interpret it as meaning the sound is close
especially in a dry acoustic with a Greene-Lee head brace etc., etc.,.
However, surely (unless I am being more dense than usual tonight) this is a
learnt response based on the behaviour we have heard from directional mics?
After all, taken individually, at those sort of frequencies our ears are
essentially omnidirectional and not subject to bass boost (to anything like
the same degree).
Any thoughts, anyone?
Dave
On Jul 23 2011, Dave Hunt wrote:
Hi again,
Date: Thu, 21 Jul 2011 21:01:41 +0300 (EEST)
From: Sampo Syreeni <de...@iki.fi>
On 2011-07-21, Dave Hunt wrote:
There is certainly no consideration of values outside the unit sphere.
[...]
Correct, and we've been here before.
We certainly have.
As BLaH points out, even the first
order decoder handles distance as well as it possibly can. So does the
SoundField mic on the encoding side.
The encoding and decoding are well matched. In some ways hardly
surprising.
But the classical synthetic
encoding equation is for infinitely far away sources only, that is,
plane waves. Running the result through a proper, BLaH compliant decoder
then reconstructs a simulacrum of such a plane wave, with first order
directional blurring, spatial aliasing caused by the discrete rig, and
the purposely imposed psychoacoustic optimizations overlaid on top of
the original, extended soundfield. So in fact it's wrong to say that the
source is produced at the distance of the rig: instead it's produced
infinitely far away, modulo the above three complications. (That is
bound to be one part of why even synthetically panned sources localise
so nicely even when listening from outside the rig.)
I have already admitted the error of my original statement. You're right
that POA assumes plane waves. The encoded signals are reproduced at the
distance of the loudspeakers. The shelf filters in a BLaH compliant
decoder are (as I understand it) an attempt to compensate for the speakers
finite distance, and that they don't produce plane waves at the listener.
This is often referred to as 'distance compensation'.
If you want to synthetically encode a near-field source so to speak "by
the book", you'll have to lift the source term from Daniel, Nicol and
Moreau's NFC work. I seem to remember it amounts to a first order filter
on the first order part of the source signal in the continuous domain,
which you'll then have to discretize. (But don't take my word for it,
it's been a while since I went through DN&R.)
Me too, but as I remember it tries to build the 'distance compensation'
into the encoding, and thus is dependent on the distance of the
loudspeakers. Thus the encoding is only suitable for an identical or
similar rig, and is not transferable to other rigs. Amplitude/delay based
systems such as WFS, Delta stereophony and TiMax have similar problems.
The encoding has to be matched to the speaker rig.
Simply
manipulating the relative amplitude or even the spectral contour doesn't
in theory cut it, though it's a cheap way to get some of the
psychoacoustic effects of a nearby source.
Agreed that it is far from perfect, but this is obviously not a trivial
problem. What I'm suggesting is a fudge, though it can produce simulations
of sources both inside and outside the loudspeaker radius which can be
psychoacoustically effective, and are transferable to different rigs.
We're still left with the "40 foot high geese" problem.
The only minor nit is that synthetic
panning needs a bit more refinement for near sources that wasn't being
handled by the older literature.
The "(potentially nasty) bass boost" you refer to is obviously a problem.
You could limit it from going extremely large at very small distances, and
ensure that the output only went to 0dBFS maximum, but this would require
a huge dynamic range throughout the whole system: large bit depth, good
DACs, very quiet amplifiers etc..
If you could do the encoding assuming a given speaker distance, then
modify the decoding for a different distance it might help, though I've no
idea how to do this.
Ciao,
Dave Hunt
_______________________________________________
Sursound mailing list
Sursound@music.vt.edu
https://mail.music.vt.edu/mailman/listinfo/sursound
_______________________________________________
Sursound mailing list
Sursound@music.vt.edu
https://mail.music.vt.edu/mailman/listinfo/sursound
_______________________________________________
Sursound mailing list
Sursound@music.vt.edu
https://mail.music.vt.edu/mailman/listinfo/sursound
