hank alrich wrote:

David Satz wrote:


The exactness of the "S" microphone's figure-8 pattern, on the other hand,
matters a whole lot--it should be as even and symmetrical as possible. It
is also important for the frequency response to be as smooth as possible.


From this does it follow that the Royer SF1 is appropriate but the R121
is not, due to the latter's difference in front/back response?


Yes, unless you don't mind a different response on the left
channel than the right channel. The effect is the same as
using non-matched cards in XY.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

David Satz wrote:

Please believe me, you do _not_ understand M/S yet,

(Over and over and again he has stated this)

even though you seem
to feel certain that you do.

To which I reply I know exactly how it works. Beyond this it is pointless
to argue since you are David Satz and I am just me.

However, you never seem to see my point. I find that the lack of
interest from sound engineers in truly experimenting with stereophonic
recording techniques is very disheartening. All people ever do around
here is implement techniques that have been around for 50 years or more.
These are excellent techniques based on very imaginative people who
did a lot of experimenting. Of course, most people never make the effort
to try to go beyond following recipes (and that's what they are) and
do a little experimenting themselves.

I prefer to use these great old techniques, try them over and over, do
my best to understand them, go beyond the very simple math, and figure
out why certain aspects work out well and others don't. Then, after
I understand them a little more, I like to modify them, combine them
and even reach something very different.

It is in the application.

Obviously, as you know, there is no "best" technique.

MS isn't the best technique. It isn't the worst technique.

Like every other technique, it usually comes down to an engineer having
to quickly understand the constraints of the recording environment,
what the music is, what tools are available, and what result are we
trying to achieve.

Sadly, for most engineers (and I am not implying that you are one of them
because I dont' really know anything about you), they take the following
approach to recording a musical event:

Question 1
You are recording music A, in the location B, with C musicians,
you have D equipment available and E will be the listener, which
technique should you use?

1)XY
2)Baffled omnis
3)MS
4)Blumlein
(of course there are a bunch of others NOS, AB, Decca Tree etc).

And then they choose one.

I am suggesting that there should be more awareness of answers 5)
and 6) which unfortunately are often not considered.

I will leave it to you guess what they are.

Rob R.

In article writes:

You guys are great! I've learned more in the last couple of days about M/S
than I ever could have reading a book.

You'll learn even more once you start setting up microphones and
playing with them. Use three cardioids if that's all you have.


--
I'm really Mike Rivers )
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me he double-m-eleven-double-zero at yahoo

In article writes:

is it possible to avoid the "matrix" of
the figure 8 (which, I gather, is just an XLR Y-cable with polarity reversed
on one side) by just duplicating the figure 8 track and reversing its
polarity?

Yes. You can record the mid and side mics separately, then with a DAW,
copy the side mic track, invert its polarity, and mix each one with
the mid mic to get the left and right channels. The problem with this
is that you can't do it in real time, so you can't hear what you're
getting until you've got it.


--
I'm really Mike Rivers )
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me he double-m-eleven-double-zero at yahoo

i think we're thinking about two different nulls here. i'm talking
about the *rear* null. if you take a cardiod (or a hyper or super)
aim it sideways and then phase reverse, whatever rear null pickup was
there is going to overlap the polar plot of the rear response of the
original signal. so it's just another reason not to do the gimmick
he's talking about. i'm probably not even explaining myself well
anyway. but in any event, figure 8 is definitely the pattern you want
for the mid-mic.

Scott Dorsey wrote:

I don't much like M-S for the same reason that I don't like X-Y... I
think all the coincident systems sound too flat to me because there
are no phase differences between channels and so low frequency imaging
is poor.

Hi, Scott--

There will be no low-frequency difference information for direct sound,
but there certainly can be difference information due to reverberant
sound--and with an M/S pickup you are free to boost the low-frequency
energy of the "S" channel to increase the sense of spaciousness. Up to
a point, that trick actually works but beyond that point, it can start to
"feel" funny in playback. (It's more of a feel than a sound, strangely.)

--I have two other complaints with M/S, one of which is solved only by a
SoundField-type approach and the other one of which is solved only with
A/B stereo (spaced mikes)--thus I never get everything I want all at once.

Complaint #1 is that when you're trying to set the right amount of "S" to
send in to the matrix in playback, the more "S" you choose the wider the
stereo stage seems to be AND at the same time the more reverberation will
be heard. There's no way to vary those two parameters independently with
M/S recording, unfortunately--and the result is that at least in my own
experience, each recording has only a fairly narrow range of "S" gain
that sounds at all plausible. It's not a wide palette of possibilities.

Complaint #2 is that with an ensemble or orchestra/chorus that is set up
across a wide stage, the distance from the microphones to the nearest
performers is a small fraction of the distance to the farthest performers.
Thus the performers closest to the microphones are picked up louder AND
much more clearly than the farthest performers. Some amount of that is
good, but too much of it is not--and it is too easy to make recordings
that have too much of it.

Rob Reedijk wrote:

I am suggesting that there should be more awareness of answers 5)
and 6) which unfortunately are often not considered.

I will leave it to you guess what they are.

Why do that if you kinow what they are? Why are you unwilling to share?

--
ha

David Satz wrote:
Scott Dorsey wrote:

I don't much like M-S for the same reason that I don't like X-Y... I
think all the coincident systems sound too flat to me because there
are no phase differences between channels and so low frequency imaging
is poor.

There will be no low-frequency difference information for direct sound,
but there certainly can be difference information due to reverberant
sound--and with an M/S pickup you are free to boost the low-frequency
energy of the "S" channel to increase the sense of spaciousness. Up to
a point, that trick actually works but beyond that point, it can start to
"feel" funny in playback. (It's more of a feel than a sound, strangely.)

This is absolutely true, and I have heard some really fine recordings done
this way.

--I have two other complaints with M/S, one of which is solved only by a
SoundField-type approach and the other one of which is solved only with
A/B stereo (spaced mikes)--thus I never get everything I want all at once.

Complaint #1 is that when you're trying to set the right amount of "S" to
send in to the matrix in playback, the more "S" you choose the wider the
stereo stage seems to be AND at the same time the more reverberation will
be heard. There's no way to vary those two parameters independently with
M/S recording, unfortunately--and the result is that at least in my own
experience, each recording has only a fairly narrow range of "S" gain
that sounds at all plausible. It's not a wide palette of possibilities.

Agreed. The Soundfield still doesn't fix my first complaint, though,
so I don't normally get this far in the argument.

Complaint #2 is that with an ensemble or orchestra/chorus that is set up
across a wide stage, the distance from the microphones to the nearest
performers is a small fraction of the distance to the farthest performers.
Thus the performers closest to the microphones are picked up louder AND
much more clearly than the farthest performers. Some amount of that is
good, but too much of it is not--and it is too easy to make recordings
that have too much of it.

This can sometimes be solved by cranking the microphone way up in the
hall to equalize the distances a bit. It can also be helped by using
hypercardioids and allowing you to pull the whole mike pair way back.
Sometimes you can't, though. If one configuration worked for every
situation they'd only need to make one mike.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

xy wrote:
i'm probably not even explaining myself well
anyway.

Agreed. Want to give it another shot?

but in any event, figure 8 is definitely the pattern you want
for the mid-mic.

Especially if you believe that. That's a Blumlein.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

On 7 Apr 2004 18:09:36 -0400, (Mike Rivers) wrote:


In article writes:

is it possible to avoid the "matrix" of
the figure 8 (which, I gather, is just an XLR Y-cable with polarity reversed
on one side) by just duplicating the figure 8 track and reversing its
polarity?

Yes. You can record the mid and side mics separately, then with a DAW,
copy the side mic track, invert its polarity, and mix each one with
the mid mic to get the left and right channels. The problem with this
is that you can't do it in real time, so you can't hear what you're
getting until you've got it.


I have been playing with this a little recently and in Adobe Audition
there is a preset option that provides the M/S mix. It ends up with
four sliders to select the amount of each track and phase to be mixed
to the new tracks. I have been able to adjust while listening to
create the sound field I like, and then mix to new tracks when I am
ready.

Dave

In Article znr1081352508k@trad, (Mike Rivers) wrote:

In article writes:

First of all, with two cardioids, you have pretty much escaped the idea of
M/S. M/S was conceived with the idea of using two mics, not three. Once you
go to three, all bets are off as I outlined in my previouos post.

It works for illustrative purposes but an array of three microphones
can't really be set up to work well for coincident pickup. Too many
"shadows" and capsules a bit too far separated.


Jeeze MIke, do you HAVE to tell EVERYTHING!? Now there are no more secrets!!

Ty

For equipment reviews and other audio delights,
try http://www.jagunet.com/~tford

Bob Cain wrote:

but in any event, figure 8 is definitely the pattern you want
for the mid-mic.


Especially if you believe that. That's a Blumlein.

I'd refer to that as M-S Blumlein, which can be dematrixed into Blumlein.

Heck, I was already doing that, Mike. That's how I knew that the side
cardioids I was using were problematic because of the lengthy bodies on
them: trying to get the diaphragms coincident and pointed in opposite
directions looked ugly.

But it sounded...okay. I at least got the idea. And I have borrowed a 414
to experiment more. But, so far, I am liking the flexibility of this
technique in that it doesn't require me to make firm decisions about how
that instrument will be used in a final mix. I can get some stereo spread,
but I can also have a good mono signal without any phasing funnies.

Again, thanks to all.

George Reiswig
Song of the River Music

"Mike Rivers" wrote in message
news:znr1081369588k@trad...

In article
writes:

You guys are great! I've learned more in the last couple of days about
M/S
than I ever could have reading a book.

You'll learn even more once you start setting up microphones and
playing with them. Use three cardioids if that's all you have.


--
I'm really Mike Rivers )
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me he double-m-eleven-double-zero at yahoo

Rob Reedijk wrote:
Yes, the left channel gets the M summed with the left lobe in
+polarity. But, hidden in there, you also have the right lobe being
subtracted. And
I know that that subtraction in combination with the mid
"mathematically" leads to an end result of an X on the left side,
that doesn't really happen. Your capsules are not truly coincident,
and because one is an 8 and the other is a card, they are not equal
enough, and for sure, their response patterns are very different
from the mathematical ideal. What you really have, is
a forward facing caridioid with +polarity left lobe and -polarity
right lobe. Similar stuff on the right side.

Wouldn't you actually have 'forward facing cardioid with +polarity left
lobe and *without* +polarity right lobe? Due to the fact that the -polarity
stuff from the figure 8 mic would be ( mostly ) subracted from, rather than
added to, the dematrixed left channel? Isn't this the basic premise for why
this works in the first place?

Now, your question about using two back-to-back cardioids instead of
a figure 8---that's an interesting idea! It's a question of how will
you get three cardioids in that coincident array. I hope you can
figure it out (or should I say "figure eight out"!?)

But in order to get an accurate sound field out of that ( meaning something
besides an XY with an exaggerated center ), even supposing you could get
the mics into a coincident array, wouldn't you first need to run one
cardioid
polarity reversed, then sum that with the other side ( to create the null in
the
center, as opposed to reinforcing center information ), and then use that as
your S channel, and matrix appropriately? Which would have the same end
result as using a figure 8 for the S channel anyway? It seems to me that
the
reason M-S works at all is the deep null and polarity relationships inherent
in
the figure 8 pattern. I think you'd end up with a pretty narrow and not
very
realistic stereo image using three cardioids with + polarity. Why not just
stick
to XY, narrow the angle of incidence, and remove one mic from the
equation?

What I like about this is that you won't have opposite lobes in
inverted polarity imbedded in the left and right sides.

No, but you'll have too much center information, unless you attenuate the
center
mic, in which case you'll have a fair bit more rear information than with
M-S, due
to the wider pickup of the cardioid pattern than the front or back of a
figure 8.

Next up for me is Blumlein (actually I will be doing more of the
2 omnis + cardioids all forward facing since the client really liked
it
for some reason). But Blumlein disturbs me since the rear lobe
thing is once again highly counterintuitive. The left channel will
have the
right rear imbedded in it (polarity being inverted doesn't seem to
matter). It's the fact that the left front + rear back will go the
left side.
It seems to me that the ambience becomes completely distorted by
this.

I assume you meant "right rear"? Anyway, I don't think the problem is so
much the ambience being 'distorted' ( distorted how, exactly? ) as it is the
ambience being too prominent. This is a well known drawback of a Blumlein
array, and the main reason why it works really well only in certain
situations.

HTH,
-Joe.

In article writes:

However, you never seem to see my point. I find that the lack of
interest from sound engineers in truly experimenting with stereophonic
recording techniques is very disheartening.

Actually there are a few. However it doesn't pay very well, so you
read more about techniques that allow for more flexibility in
production and less commitment to getting a good recording up front.

Question 1
You are recording music A, in the location B, with C musicians,
you have D equipment available and E will be the listener, which
technique should you use?

Answer 1: It depends. Sometimes experience with the same location and
type of music (assuming this isn't a repeat gig) will scream out the
starting point. But an engineer who has only worked in a studio with
multiple mics won't have a clue and will probalby go for X-Y because
that's the easiest to set up.


--
I'm really Mike Rivers )
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me he double-m-eleven-double-zero at yahoo

Kurt Albershardt wrote in
:

Bob Cain wrote:

but in any event, figure 8 is definitely the pattern you want
for the mid-mic.

Especially if you believe that. That's a Blumlein.

I'd refer to that as M-S Blumlein, which can be dematrixed into Blumlein.

Same M+S, M-S as standard? Or does the figure 8 pattern complicate the
dematrixing?

Carey Carlan wrote:

Kurt Albershardt wrote in
:


Bob Cain wrote:

but in any event, figure 8 is definitely the pattern you want
for the mid-mic.

Especially if you believe that. That's a Blumlein.

I'd refer to that as M-S Blumlein, which can be dematrixed into Blumlein.


Same M+S, M-S as standard? Or does the figure 8 pattern complicate the
dematrixing?


Same matrix & coefficients but the result is Blumlein rather than X-Y.

Carey Carlan wrote:
Kurt Albershardt wrote in
:

Bob Cain wrote:

but in any event, figure 8 is definitely the pattern you want
for the mid-mic.

Especially if you believe that. That's a Blumlein.

I'd refer to that as M-S Blumlein, which can be dematrixed into Blumlein.

Same M+S, M-S as standard? Or does the figure 8 pattern complicate the
dematrixing?

I believe it is easier to grasp when you realise that MS-Blumlien and
Blumlein are just 45 degrees in difference.

Blumlein gets more intersting when it is considered that if you were to
put a Blumlein in the centre of a room of musicians, say if you were to
record a group jazz players off the floor, you could, after the fact,
rotate the axis of stereo sound any way you like.

But there is nothing new to that since that is just a 2 dimensional version
of a Soundfield mic.

Rob R.

Carey Carlan wrote:

Kurt Albershardt wrote in
:


Bob Cain wrote:

but in any event, figure 8 is definitely the pattern you want
for the mid-mic.

Especially if you believe that. That's a Blumlein.

I'd refer to that as M-S Blumlein, which can be dematrixed into Blumlein.


Same M+S, M-S as standard? Or does the figure 8 pattern complicate the
dematrixing?

Yes, if you do M+S and M-S on a forward facing and side
facing pair you get the same directivity as a Blumlein at
+/- 45 degrees. Because of variation in frequency response
within the pattern, however it can sound different. If your
material is mostly in the center I can see a justifiable
preference for the matrixed MS setup.

From a crossed pair of figure 8's you can synthesize, via
the correctly weighted matrix, a logical figure 8 pointing
in any direction. Or any number of them.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

Rob Reedijk wrote:

Blumlein gets more intersting when it is considered that if you were to
put a Blumlein in the centre of a room of musicians, say if you were to
record a group jazz players off the floor, you could, after the fact,
rotate the axis of stereo sound any way you like.

Agreed.




But there is nothing new to that since that is just a 2 dimensional version
of a Soundfield mic.

Not quite, since it lacks the W component. Add an omni and you're there...

Bob Cain wrote in
:

Yes, if you do M+S and M-S on a forward facing and side
facing pair you get the same directivity as a Blumlein at
+/- 45 degrees. Because of variation in frequency response
within the pattern, however it can sound different. If your
material is mostly in the center I can see a justifiable
preference for the matrixed MS setup.

From a crossed pair of figure 8's you can synthesize, via
the correctly weighted matrix, a logical figure 8 pointing
in any direction. Or any number of them.

How intriguing! But it makes sense. Crossed figure-eights represent a
symmetrical pattern in the horizontal plain. The proper ratios of + and -
can point you in any direction.

Carey Carlan wrote:


From a crossed pair of figure 8's you can synthesize, via
the correctly weighted matrix, a logical figure 8 pointing
in any direction. Or any number of them.


How intriguing! But it makes sense. Crossed figure-eights represent a
symmetrical pattern in the horizontal plain. The proper ratios of + and -
can point you in any direction.

Right, and add an omni at the center of that and you can mix
it in to make that virtual mic have any 1st order pattern.


Bob

--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

Bob Cain wrote in
:

Carey Carlan wrote:


From a crossed pair of figure 8's you can synthesize, via
the correctly weighted matrix, a logical figure 8 pointing
in any direction. Or any number of them.


How intriguing! But it makes sense. Crossed figure-eights represent
a symmetrical pattern in the horizontal plain. The proper ratios of
+ and - can point you in any direction.

Right, and add an omni at the center of that and you can mix
it in to make that virtual mic have any 1st order pattern.

A scenario I have faced in the past:

Recording a choral group on stage. During the final number they walk off
stage to a finale standing in the aisles of the audience. With a proper
setup I could rotate the pattern as the group moves and maintain some
semblance of a "forward" image to the very end.

I had never considered using such a pattern, nor matrixing it for effect in
this situation. as the light dawns

Carey Carlan wrote:

A scenario I have faced in the past:

Recording a choral group on stage. During the final number they walk off
stage to a finale standing in the aisles of the audience. With a proper
setup I could rotate the pattern as the group moves and maintain some
semblance of a "forward" image to the very end.


That would be cool, especially if there were an automatable
matrix plugin. I'm not sure that exists. It might just be
doable as a "compiled" SynthEdit patch. How much is it
worth to you? :-)


Bob

--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

Bob Cain wrote in
:

Carey Carlan wrote:

A scenario I have faced in the past:

Recording a choral group on stage. During the final number they walk
off stage to a finale standing in the aisles of the audience. With a
proper setup I could rotate the pattern as the group moves and
maintain some semblance of a "forward" image to the very end.

That would be cool, especially if there were an automatable
matrix plugin. I'm not sure that exists. It might just be
doable as a "compiled" SynthEdit patch. How much is it
worth to you? :-)

You're making it 'way too hard, Bob. Just created + and - versions of each
channel and crossfade as needed.

Carey Carlan wrote:

Bob Cain wrote in
:

That would be cool, especially if there were an automatable
matrix plugin. I'm not sure that exists. It might just be
doable as a "compiled" SynthEdit patch. How much is it
worth to you? :-)


You're making it 'way too hard, Bob. Just created + and - versions of each
channel and crossfade as needed.

Not quite. To maintain the sensitivity as you pan, the fade
gains must follow a law that keeps (Gm^2 + Gs^2) a constant
where Gm and Gs are the linear (not logarithmic) gains
applied to the mid and side figure 8's.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

Bob Cain wrote in message ...
A. & G. Reiswig wrote:

Hi, all,
I've got theory questions I'm hoping you can help with. I'm having a
hard time understanding *why* the mid-side technique works. I know that it
is typically done with a figure 8 for the side, and a cardioid or omni for
the middle.

Many will argue this but I insist that it be a cardiod for
reasons below.

SNIP

The reason an omni as a mid is a bad idea should be
apparent. What you get after the L=M+S and R=M-S is a left
facing cardiod and a right facing caridiod, a coincident
configuration that is never used.

This is true only if it is a true omni at all frequencies, and most
real-world omni mics do not exhibit this behavior. In practice, omnis
are "omni" in the low frequency range and more directional in the
higher frequencies (as a function of the size of the diaprhagm and mic
head assembly).

The real advantage of using a true omni in this configuration is
distance-indipendant low frequency response, since the omni mic will
not exhibit proximity effect. It's a nice way to record solo classical
piano, for instance.

Also, omni's usually don't have a tendency to "reach" on percussive
transients the way directional mics often do.

Just my .02

Karl Winkler
Sennheiser
http://www.sennheiserusa.com

Okay, another dumb question perhaps: is it possible to avoid the "matrix" of
the figure 8 (which, I gather, is just an XLR Y-cable with polarity reversed
on one side) by just duplicating the figure 8 track and reversing its
polarity? I realize that wouldn't work for using one cardioid, but how is
it different from a "real" M/S setup?

George Reiswig
Song of the River Music


That would work just fine and I have done that.

Richard H. Kuschel
"I canna change the law of physics."-----Scotty

Kurt Albershardt wrote:

Rob Reedijk wrote:

Blumlein gets more intersting when it is considered that if you were to
put a Blumlein in the centre of a room of musicians, say if you were to
record a group jazz players off the floor, you could, after the fact,
rotate the axis of stereo sound any way you like.

Agreed.

Disagreed...or at least with what I had read Rob to be saying - that
weighting in different amounts of Fig-8 S into the sum 'n difference
matrix allows you to swivel (pan) your virtual mic pair.

It's true that the Fig-8 pattern is maintained for a virtual mic
derived from matrixing of two Fig-8's in any weighting, but the
angling of this virtual pair still stays symmetrically distributed
astride the original straight-ahead M axis. All that changes is the
total included angle.

For example, some weightings:
0.7M + 0.3S - included angle 46.4 deg (+/- 23.2 deg)
0.5M + 0.5S - included angle 90 deg (+/- 45 deg)
0.7M + 0.3S - included angle 133.6 deg (+/- 66.8 deg)

In short, its more a "zoom" than a "pan'. It can't be deployed, as I
believe was suggested elsewhere in the thread, to pan along with
musicians trooping sideways off stage.

Tom McCreadie
(remove 'spamjam.' for my real address)

Tom McCreadie wrote:

Kurt Albershardt wrote:


Rob Reedijk wrote:

Blumlein gets more intersting when it is considered that if you were to
put a Blumlein in the centre of a room of musicians, say if you were to
record a group jazz players off the floor, you could, after the fact,
rotate the axis of stereo sound any way you like.

Agreed.


Disagreed...or at least with what I had read Rob to be saying - that
weighting in different amounts of Fig-8 S into the sum 'n difference
matrix allows you to swivel (pan) your virtual mic pair.

It's true that the Fig-8 pattern is maintained for a virtual mic
derived from matrixing of two Fig-8's in any weighting, but the
angling of this virtual pair still stays symmetrically distributed
astride the original straight-ahead M axis. All that changes is the
total included angle.


I'm not sure what you are saying. From a crossed pair of
figure 8's you can find a weighting that will create a
logical figure 8 pointing in any arbitrary direction. From
that it follows that you can derive two logical figure 8's
pointing in any two arbitrary directions or any N figure 8's
point in N arbitrary directions.

Adding an appropriately scaled omni to any of those N
logical figure 8's creates N logical mics with any arbitrary
1st order patterns pointing in the same directions.


Bob
--

"Things should be described as simply as possible, but no
simpler."

A. Einstein

Bob Cain wrote:

Tom McCreadie wrote:

Kurt Albershardt wrote:


Rob Reedijk wrote:

Blumlein gets more intersting when it is considered that if you were to
put a Blumlein in the centre of a room of musicians, say if you were to
record a group jazz players off the floor, you could, after the fact,
rotate the axis of stereo sound any way you like.

Agreed.


Disagreed...or at least with what I had read Rob to be saying - that
weighting in different amounts of Fig-8 S into the sum 'n difference
matrix allows you to swivel (pan) your virtual mic pair.

It's true that the Fig-8 pattern is maintained for a virtual mic
derived from matrixing of two Fig-8's in any weighting, but the
angling of this virtual pair still stays symmetrically distributed
astride the original straight-ahead M axis. All that changes is the
total included angle.


I'm not sure what you are saying. From a crossed pair of
figure 8's you can find a weighting that will create a
logical figure 8 pointing in any arbitrary direction. From
that it follows that you can derive two logical figure 8's
pointing in any two arbitrary directions or any N figure 8's
point in N arbitrary directions.

Yes, I agree completely with what you are saying, and summarizing very
lucidly too, but that requires at least _two_ stages of
combinatorial work. I'd mistakenly read too much into Rob's statement
- thought he was contending that you could get the pan-rotation of the
virtual mic assembly in a _one_ stage operation, with only the matrix
unit at your disposal, i.e. by, say, a one-knob gain change of the S
or M Fig-8 mic before the signals enter the matrix module.

Apologies, particularly to Rob, for any confusion sewn. As penance I
offer the group the correct spelling of "cardioid" g.

Tom McCreadie
(remove "spamjam." for my real address)