I will be going into a subway station with a MS / XY combo mike that has
adjustable spread, the Studio Projects LSD2. The problem is how to adjust
the capsule spread in such a way as to obtain the best possible stereo image
while listening only on headphones.

Since headphone mixing would tend to select the best angle for binaural
listening, I'm wondering if it's possible to solve this problem by either
1. Some kind of mental adjustment. IOW, listen for some particular
characteristic other than what sounds best in the phones.
2. Some kind of matrix box that would approximate stereo speakers, at least
in arriving at the correct capsule angle.

???

I will be going into a subway station with a MS / XY combo mike that has
adjustable spread, the Studio Projects LSD2. The problem is how to adjust
the capsule spread in such a way as to obtain the best possible stereo image
while listening only on headphones.

Since headphone mixing would tend to select the best angle for binaural
listening, I'm wondering if it's possible to solve this problem by either
1. Some kind of mental adjustment. IOW, listen for some particular
characteristic other than what sounds best in the phones.
2. Some kind of matrix box that would approximate stereo speakers, at least
in arriving at the correct capsule angle.

Can't you experiment ahead of the location recording? Make test
recordings and see what configuration you like best.

"soundhaspriority" wrote in message
...
I will be going into a subway station with a MS / XY combo mike that has
adjustable spread, the Studio Projects LSD2.

Watch out for Giant green frogs and British skinheads.
Wanna bova mate? Got me bova boots on.



--
Posted via NewsDemon.com - Premium Uncensored Newsgroup Service
-------http://www.NewsDemon.com------
Unlimited Access, Anonymous Accounts, Uncensored Broadband Access

"Laurence Payne" lpayne1NOSPAM@dslDOTpipexDOTcom wrote in message
news
I will be going into a subway station with a MS / XY combo mike that has
adjustable spread, the Studio Projects LSD2. The problem is how to adjust
the capsule spread in such a way as to obtain the best possible stereo
image
while listening only on headphones.

Since headphone mixing would tend to select the best angle for binaural
listening, I'm wondering if it's possible to solve this problem by either
1. Some kind of mental adjustment. IOW, listen for some particular
characteristic other than what sounds best in the phones.
2. Some kind of matrix box that would approximate stereo speakers, at
least
in arriving at the correct capsule angle.

Can't you experiment ahead of the location recording? Make test
recordings and see what configuration you like best.

Unfortunately, no, because the artists and locations are not determined
ahead of time.

soundhaspriority wrote:
I will be going into a subway station with a MS / XY combo mike that has
adjustable spread, the Studio Projects LSD2. The problem is how to adjust
the capsule spread in such a way as to obtain the best possible stereo image
while listening only on headphones.

AKG HEARO 999 Audiosphere II

http://akg.com/products/powerslave,m...ge,E NUS.html

While I would love to, I haven't listened to this but what it does is apply
HRTF DSP to, among other things, convert an LR signal to binaural via
virtual monitors in a good room. There may have been an earlier version of
this without "Audiopsphere II" added to the name because I find that at
froogle for about half the cost. Looks like $800 might be a reasonable web
price for the thing.

I've experimented with the DSP and the measurement needed to do this kind
of thing and can attest that a _very_ plausible illusion can be created.

It may do a better job than raw 'phones at helping you find a good
placement and spread for the stereo mic. I'd only buy it, however, where
there was a good return policy. :-)


Bob
--

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

A. Einstein

Ah, here's the earlier version. Not wireless and no 'phones included:

http://akg.com/products/powerslave,m...ge,E NUS.html

Looks like it can still be purchased he

http://www.headphone.com/products/he...-hearo-999.php


Bob
--

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

A. Einstein

"Bob Cain" wrote in message
...
soundhaspriority wrote:
I will be going into a subway station with a MS / XY combo mike that has
adjustable spread, the Studio Projects LSD2. The problem is how to
adjust
the capsule spread in such a way as to obtain the best possible stereo
image
while listening only on headphones.

AKG HEARO 999 Audiosphere II

http://akg.com/products/powerslave,m...ge,E NUS.html

While I would love to, I haven't listened to this but what it does is
apply
HRTF DSP to, among other things, convert an LR signal to binaural via
virtual monitors in a good room. There may have been an earlier version
of
this without "Audiopsphere II" added to the name because I find that at
froogle for about half the cost. Looks like $800 might be a reasonable
web
price for the thing.

How could I complain? It only costs $800, with an antenna sticking out of my
head

But you've made me think. I have a Sennheiser 580 headphone sursound system,
which claims to do the above, at the more reasonable price of $100. It has
only two problems:
1. It requires a bipolar power supply, one more thing to carry around town.
2. Although it has adjustments, I was never able to create a fully
convincing stereo phantom. It required a lot of "belief" to feel it, which
makes me wonder how useful it would be in avoiding soundstage errors.

"soundhaspriority" wrote in message
...
I will be going into a subway station with a MS / XY combo mike that has
adjustable spread, the Studio Projects LSD2. The problem is how to adjust
the capsule spread in such a way as to obtain the best possible stereo
image
while listening only on headphones.

Since headphone mixing would tend to select the best angle for binaural
listening, I'm wondering if it's possible to solve this problem by either
1. Some kind of mental adjustment. IOW, listen for some particular
characteristic other than what sounds best in the phones.
2. Some kind of matrix box that would approximate stereo speakers, at
least
in arriving at the correct capsule angle.

Use the mic in MS mode with the M capsule facing straight forward and the S
capsule facing 90 degrees, positive side pointed to the left. Record the M
signal to one track and the S signal to the other. Later, during mixdown,
you can adjust the equivalent capsule angle by varying the amount of S
signal you feed into your matrix setup.

To monitor, listen to the M signal.

Peace,
Paul

On Tue, 16 May 2006 22:29:18 -0400, "Clyde Slick"
wrote:

I will be going into a subway station with a MS / XY combo mike that has
adjustable spread, the Studio Projects LSD2.

Watch out for Giant green frogs and British skinheads.
Wanna bova mate? Got me bova boots on.

Nah. It's Brazilians with rucksacks now. Watch out for the
ricochets.

On Tue, 16 May 2006 22:37:35 -0400, "soundhaspriority"
wrote:

Can't you experiment ahead of the location recording? Make test
recordings and see what configuration you like best.

Unfortunately, no, because the artists and locations are not determined
ahead of time.


So try it out in any old subway with any old voices. That will give
you a fair idea, won't it?

"soundhaspriority" wrote in message
...
I will be going into a subway station with a MS / XY combo mike that has
adjustable spread, the Studio Projects LSD2. The problem is how to adjust
the capsule spread in such a way as to obtain the best possible stereo
image while listening only on headphones.

Since headphone mixing would tend to select the best angle for binaural
listening, I'm wondering if it's possible to solve this problem by either
1. Some kind of mental adjustment. IOW, listen for some particular
characteristic other than what sounds best in the phones.
2. Some kind of matrix box that would approximate stereo speakers, at
least in arriving at the correct capsule angle.

???

(3) Recognize that X/Y and M/S are matrix systems and that you can adjust
the stereo spread during the mix.

"Paul Stamler" wrote in message
...
"soundhaspriority" wrote in message
...
I will be going into a subway station with a MS / XY combo mike that has
adjustable spread, the Studio Projects LSD2. The problem is how to
adjust
the capsule spread in such a way as to obtain the best possible stereo
image
while listening only on headphones.

Since headphone mixing would tend to select the best angle for binaural
listening, I'm wondering if it's possible to solve this problem by either
1. Some kind of mental adjustment. IOW, listen for some particular
characteristic other than what sounds best in the phones.
2. Some kind of matrix box that would approximate stereo speakers, at
least
in arriving at the correct capsule angle.

Use the mic in MS mode with the M capsule facing straight forward and the
S
capsule facing 90 degrees, positive side pointed to the left. Record the M
signal to one track and the S signal to the other. Later, during mixdown,
you can adjust the equivalent capsule angle by varying the amount of S
signal you feed into your matrix setup.

To monitor, listen to the M signal.

True as far as it goes, but ignores the fact that you can also rematrix XY.

"Arny Krueger" wrote in message
...

"soundhaspriority" wrote in message
...
I will be going into a subway station with a MS / XY combo mike that has
adjustable spread, the Studio Projects LSD2. The problem is how to adjust
the capsule spread in such a way as to obtain the best possible stereo
image while listening only on headphones.

Since headphone mixing would tend to select the best angle for binaural
listening, I'm wondering if it's possible to solve this problem by either
1. Some kind of mental adjustment. IOW, listen for some particular
characteristic other than what sounds best in the phones.
2. Some kind of matrix box that would approximate stereo speakers, at
least in arriving at the correct capsule angle.

???

(3) Recognize that X/Y and M/S are matrix systems and that you can adjust
the stereo spread during the mix.
This is literally true for MS. However, it is not literally true for XY,
because ideal capsules do not exist. Part of the problem, which I omitted,
is to decide on mike placement and whether XY or MS is to be used.

"Robert Morein" wrote in message
...

"Arny Krueger" wrote in message
...

"soundhaspriority" wrote in message
...
I will be going into a subway station with a MS / XY combo mike that has
adjustable spread, the Studio Projects LSD2. The problem is how to
adjust the capsule spread in such a way as to obtain the best possible
stereo image while listening only on headphones.

Since headphone mixing would tend to select the best angle for binaural
listening, I'm wondering if it's possible to solve this problem by
either
1. Some kind of mental adjustment. IOW, listen for some particular
characteristic other than what sounds best in the phones.
2. Some kind of matrix box that would approximate stereo speakers, at
least in arriving at the correct capsule angle.

???

(3) Recognize that X/Y and M/S are matrix systems and that you can adjust
the stereo spread during the mix.

This is literally true for MS.

No, it isn't. 2 capsules can't be coincient, and real-world capsules don't
conform to any theoretical ideal.

For example, figure-8 capsules have directivity that only approximates an
ideal figure-8, and the directivity still varies with frequency.

However, it is not literally true for XY, because ideal capsules do not
exist.

It is still a matter of dueling approximations.

Part of the problem, which I omitted, is to decide on mike placement and
whether XY or MS is to be used.

Depends on what you have on hand to work with, as much as anything.

With real-world mics, nothing is exact or in conformance with any ideal
theoretical ideal.

The mic type that is closest to any theoretical ideal would probably be the
small-diameter omni. That's still an approximiation and it is still a very
limited ideal.

"Arny Krueger" wrote in message
...

"Robert Morein" wrote in message
...

"Arny Krueger" wrote in message
...

"soundhaspriority" wrote in message
...
I will be going into a subway station with a MS / XY combo mike that has
adjustable spread, the Studio Projects LSD2. The problem is how to
adjust the capsule spread in such a way as to obtain the best possible
stereo image while listening only on headphones.

Since headphone mixing would tend to select the best angle for binaural
listening, I'm wondering if it's possible to solve this problem by
either
1. Some kind of mental adjustment. IOW, listen for some particular
characteristic other than what sounds best in the phones.
2. Some kind of matrix box that would approximate stereo speakers, at
least in arriving at the correct capsule angle.

???

(3) Recognize that X/Y and M/S are matrix systems and that you can
adjust the stereo spread during the mix.

This is literally true for MS.

No, it isn't. 2 capsules can't be coincient, and real-world capsules don't
conform to any theoretical ideal.

For example, figure-8 capsules have directivity that only approximates an
ideal figure-8, and the directivity still varies with frequency.

However, it is not literally true for XY, because ideal capsules do not
exist.

It is still a matter of dueling approximations.

Part of the problem, which I omitted, is to decide on mike placement and
whether XY or MS is to be used.

Depends on what you have on hand to work with, as much as anything.

With real-world mics, nothing is exact or in conformance with any ideal
theoretical ideal.

The mic type that is closest to any theoretical ideal would probably be
the small-diameter omni. That's still an approximiation and it is still a
very limited ideal.

Fine, I agree with this. It outlines the problem. With all the reflective
surfaces, ambient noise, etc., I can't rely on matrixing it later to get the
optimal mix. It has to be done right in the subway, a very hostile location.

"Arny Krueger" wrote in message
...

"Paul Stamler" wrote in message
...
"soundhaspriority" wrote in message
...
I will be going into a subway station with a MS / XY combo mike that
has
adjustable spread, the Studio Projects LSD2. The problem is how to
adjust
the capsule spread in such a way as to obtain the best possible stereo
image
while listening only on headphones.

Since headphone mixing would tend to select the best angle for binaural
listening, I'm wondering if it's possible to solve this problem by
either
1. Some kind of mental adjustment. IOW, listen for some particular
characteristic other than what sounds best in the phones.
2. Some kind of matrix box that would approximate stereo speakers, at
least
in arriving at the correct capsule angle.

Use the mic in MS mode with the M capsule facing straight forward and
the
S
capsule facing 90 degrees, positive side pointed to the left. Record the
M
signal to one track and the S signal to the other. Later, during
mixdown,
you can adjust the equivalent capsule angle by varying the amount of S
signal you feed into your matrix setup.

To monitor, listen to the M signal.

True as far as it goes, but ignores the fact that you can also rematrix
XY.

Of course, but why do an extra step?

Peace,
Paul

soundhaspriority wrote:

I will be going into a subway station with a MS / XY combo mike that has
adjustable spread, the Studio Projects LSD2.

Don't mix in the subway, darling,

Don't track in the pouring rain.

//Walt

"Walt" wrote in message
...
soundhaspriority wrote:

I will be going into a subway station with a MS / XY combo mike that has
adjustable spread, the Studio Projects LSD2.

Don't mix in the subway, darling,

Don't track in the pouring rain.

//Walt

Walt, this is a "production sound" problem, where hostile locations are the
norm. I probably made a mistake posting it to pro; I reposted to
rec.arts.movies.production.sound.

In article ,
Walt wrote:

soundhaspriority wrote:

I will be going into a subway station with a MS / XY combo mike that has
adjustable spread, the Studio Projects LSD2.

Don't mix in the subway, darling,

Don't track in the pouring rain.

//Walt

LOL

"Robert Morein" wrote in message
...

"Arny Krueger" wrote in message
...

"Robert Morein" wrote in message
...

"Arny Krueger" wrote in message
...

"soundhaspriority" wrote in
message ...
I will be going into a subway station with a MS / XY combo mike that
has adjustable spread, the Studio Projects LSD2. The problem is how to
adjust the capsule spread in such a way as to obtain the best possible
stereo image while listening only on headphones.

Since headphone mixing would tend to select the best angle for
binaural listening, I'm wondering if it's possible to solve this
problem by either
1. Some kind of mental adjustment. IOW, listen for some particular
characteristic other than what sounds best in the phones.
2. Some kind of matrix box that would approximate stereo speakers, at
least in arriving at the correct capsule angle.

???

(3) Recognize that X/Y and M/S are matrix systems and that you can
adjust the stereo spread during the mix.

This is literally true for MS.

No, it isn't. 2 capsules can't be coincient, and real-world capsules
don't conform to any theoretical ideal.

For example, figure-8 capsules have directivity that only approximates an
ideal figure-8, and the directivity still varies with frequency.

However, it is not literally true for XY, because ideal capsules do not
exist.

It is still a matter of dueling approximations.

Part of the problem, which I omitted, is to decide on mike placement and
whether XY or MS is to be used.

Depends on what you have on hand to work with, as much as anything.

With real-world mics, nothing is exact or in conformance with any ideal
theoretical ideal.

The mic type that is closest to any theoretical ideal would probably be
the small-diameter omni. That's still an approximiation and it is still a
very limited ideal.

Fine, I agree with this. It outlines the problem. With all the reflective
surfaces, ambient noise, etc., I can't rely on matrixing it later to get
the optimal mix. It has to be done right in the subway, a very hostile
location.

With all the reflective surfaces and ambient noise, there is considerable
lattitude for doing things as good as they are going to get. IOW, one mix
might just be just as good as another.

"Arny Krueger" wrote in message
...

"Robert Morein" wrote in message
...

"Arny Krueger" wrote in message
...

"Robert Morein" wrote in message
...

"Arny Krueger" wrote in message
...

"soundhaspriority" wrote in
message ...
I will be going into a subway station with a MS / XY combo mike that
has adjustable spread, the Studio Projects LSD2. The problem is how
to adjust the capsule spread in such a way as to obtain the best
possible stereo image while listening only on headphones.

Since headphone mixing would tend to select the best angle for
binaural listening, I'm wondering if it's possible to solve this
problem by either
1. Some kind of mental adjustment. IOW, listen for some particular
characteristic other than what sounds best in the phones.
2. Some kind of matrix box that would approximate stereo speakers, at
least in arriving at the correct capsule angle.

???

(3) Recognize that X/Y and M/S are matrix systems and that you can
adjust the stereo spread during the mix.

This is literally true for MS.

No, it isn't. 2 capsules can't be coincient, and real-world capsules
don't conform to any theoretical ideal.

For example, figure-8 capsules have directivity that only approximates
an ideal figure-8, and the directivity still varies with frequency.

However, it is not literally true for XY, because ideal capsules do not
exist.

It is still a matter of dueling approximations.

Part of the problem, which I omitted, is to decide on mike placement
and whether XY or MS is to be used.

Depends on what you have on hand to work with, as much as anything.

With real-world mics, nothing is exact or in conformance with any ideal
theoretical ideal.

The mic type that is closest to any theoretical ideal would probably be
the small-diameter omni. That's still an approximiation and it is still
a very limited ideal.

Fine, I agree with this. It outlines the problem. With all the reflective
surfaces, ambient noise, etc., I can't rely on matrixing it later to get
the optimal mix. It has to be done right in the subway, a very hostile
location.

With all the reflective surfaces and ambient noise, there is considerable
lattitude for doing things as good as they are going to get. IOW, one mix
might just be just as good as another.

I don't quite agree with you, but I do intend to try it even if there is no
way to guarantee optimality.

"Paul Stamler" wrote in message
...
"Arny Krueger" wrote in message
...

"Paul Stamler" wrote in message
...
"soundhaspriority" wrote in
message
...
I will be going into a subway station with a MS / XY combo mike that
has
adjustable spread, the Studio Projects LSD2. The problem is how to
adjust
the capsule spread in such a way as to obtain the best possible stereo
image
while listening only on headphones.

Since headphone mixing would tend to select the best angle for
binaural
listening, I'm wondering if it's possible to solve this problem by
either
1. Some kind of mental adjustment. IOW, listen for some particular
characteristic other than what sounds best in the phones.
2. Some kind of matrix box that would approximate stereo speakers, at
least
in arriving at the correct capsule angle.

Use the mic in MS mode with the M capsule facing straight forward and
the
S
capsule facing 90 degrees, positive side pointed to the left. Record
the
M
signal to one track and the S signal to the other. Later, during
mixdown,
you can adjust the equivalent capsule angle by varying the amount of S
signal you feed into your matrix setup.

To monitor, listen to the M signal.

True as far as it goes, but ignores the fact that you can also rematrix
XY.

Of course, but why do an extra step?

What extra step?

soundhaspriority wrote:
I will be going into a subway station with a MS / XY combo mike that has
adjustable spread, the Studio Projects LSD2. The problem is how to adjust
the capsule spread in such a way as to obtain the best possible stereo image
while listening only on headphones.

Just do it. In a subway, I would hazard a guess that X-Y would work
better than M-S because of the way they (in practice) handle ambience.
The LSD-2 works best at 110 degrees. Set it up that way and experiment
with positioning until it sounds OK in the headphones.

You will have the best stereo image (that you are capable of
discerning). Keep at it and you'll get better.

"Mike Rivers" wrote in message
ps.com...

soundhaspriority wrote:
I will be going into a subway station with a MS / XY combo mike that has
adjustable spread, the Studio Projects LSD2. The problem is how to
adjust
the capsule spread in such a way as to obtain the best possible stereo
image
while listening only on headphones.

Just do it. In a subway, I would hazard a guess that X-Y would work
better than M-S because of the way they (in practice) handle ambience.
The LSD-2 works best at 110 degrees. Set it up that way and experiment
with positioning until it sounds OK in the headphones.

You will have the best stereo image (that you are capable of
discerning). Keep at it and you'll get better.

I will give it a shot. The 110 degree suggestion is very significant. What
source did you experiment on?

"Arny Krueger" wrote in message
...

signal to one track and the S signal to the other. Later, during
mixdown,
you can adjust the equivalent capsule angle by varying the amount of
S
signal you feed into your matrix setup.

To monitor, listen to the M signal.

True as far as it goes, but ignores the fact that you can also rematrix
XY.

Of course, but why do an extra step?

What extra step?

Turning the L & R signals to M and S so you can adjust virtual angle. With
MS recording they're already there.

Peace,
Paul

Arny Krueger wrote:
"Robert Morein" wrote:

(3) Recognize that X/Y and M/S are matrix systems and that you can adjust
the stereo spread during the mix.

This is literally true for MS.

No, it isn't. 2 capsules can't be coincient, and real-world capsules don't
conform to any theoretical ideal.

For example, figure-8 capsules have directivity that only approximates an
ideal figure-8, and the directivity still varies with frequency.

Has anyone played with using an XY pair matrixed as the S
in an MS pair to tackle the directivity issue ?
In other words a "bent 8" for the side ?

just thinking out loud ...

rd

"RD Jones" wrote in
oups.com:

Has anyone played with using an XY pair matrixed as the S
in an MS pair to tackle the directivity issue ?
In other words a "bent 8" for the side ?

Your "bent 8" would still contain mostly Mid information. The purpose of
the side is to get the sides. Even at 180 spread, the best you can get
from a pair of cardioids is an omni mix. You need something with more
directivity, like a figure-8.

"RD Jones" wrote in

Has anyone played with using an XY pair matrixed as the S
in an MS pair to tackle the directivity issue ?
In other words a "bent 8" for the side ?

Carey Carlan wrote:

Your "bent 8" would still contain mostly Mid information. The purpose of
the side is to get the sides. Even at 180 spread, the best you can get
from a pair of cardioids is an omni mix. You need something with more
directivity, like a figure-8.

Even with one capsule inverted ? I'd would have expected
to have a null on both the front and rear halves, just a much
narrower one at front.

I've already got a Y harness wired up with one connector
inverted. Looks like I need to prove myself wrong again.

rd

"RD Jones" wrote in
oups.com:


"RD Jones" wrote in

Has anyone played with using an XY pair matrixed as the S
in an MS pair to tackle the directivity issue ?
In other words a "bent 8" for the side ?

Carey Carlan wrote:

Your "bent 8" would still contain mostly Mid information. The
purpose of the side is to get the sides. Even at 180 spread, the
best you can get from a pair of cardioids is an omni mix. You need
something with more directivity, like a figure-8.

Even with one capsule inverted ? I'd would have expected
to have a null on both the front and rear halves, just a much
narrower one at front.

I've already got a Y harness wired up with one connector
inverted. Looks like I need to prove myself wrong again.

You didn't mention inverting one capsule. That would give you a null in
the center, but all remaining signal would be off-axis (probably not
sounding very good). A poor substitute for a figure 8, but mathematically
it would work.

Carey Carlan wrote:
"RD Jones" wrote in
oups.com:

Has anyone played with using an XY pair matrixed as the S
in an MS pair to tackle the directivity issue ?
In other words a "bent 8" for the side ?

Your "bent 8" would still contain mostly Mid information. The purpose of
the side is to get the sides. Even at 180 spread, the best you can get
from a pair of cardioids is an omni mix. You need something with more
directivity, like a figure-8.

At any angle the S=L-R signal derived from a coincident, matched XY is a
figure 8 and M=L+R is an omni. What varies as you change the physical
angle is the sensitivity of the derived S relative to the derived M.

That's why changing the level of S relative to M before matrixing back to
LR is equivalent to physically varying the XY angle.

Of course the above is strictly true only if the capsules are ideal in the
symmetry of their patterns (round omni summed with a weighted cosine figure
8) but there's not much of a general nature that can be said about what
happens with real life deviations from that ideal.


Bob
--

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

A. Einstein

"Bob Cain" wrote in message
...
Carey Carlan wrote:
"RD Jones" wrote in
oups.com:

Has anyone played with using an XY pair matrixed as the S
in an MS pair to tackle the directivity issue ?
In other words a "bent 8" for the side ?

Your "bent 8" would still contain mostly Mid information. The purpose of
the side is to get the sides. Even at 180 spread, the best you can get
from a pair of cardioids is an omni mix. You need something with more
directivity, like a figure-8.

At any angle the S=L-R signal derived from a coincident, matched XY is a
figure 8 and M=L+R is an omni. What varies as you change the physical
angle is the sensitivity of the derived S relative to the derived M.

Bob, I don't see this. Reducto ad absurdium: make the angle zero. The XY sum
is still a cardioid.

Robert Morein wrote:
"Bob Cain" wrote in message
...
Carey Carlan wrote:
"RD Jones" wrote in
oups.com:

Has anyone played with using an XY pair matrixed as the S
in an MS pair to tackle the directivity issue ?
In other words a "bent 8" for the side ?
Your "bent 8" would still contain mostly Mid information. The purpose of
the side is to get the sides. Even at 180 spread, the best you can get
from a pair of cardioids is an omni mix. You need something with more
directivity, like a figure-8.
At any angle the S=L-R signal derived from a coincident, matched XY is a
figure 8 and M=L+R is an omni. What varies as you change the physical
angle is the sensitivity of the derived S relative to the derived M.

Bob, I don't see this. Reducto ad absurdium: make the angle zero. The XY sum
is still a cardioid.

You're right. Gotta go right now but will come back and figure out what
the hell I should have said (if anything) later.

r.a.o. removed. Appreciate it if you would too.


Bob
--

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

A. Einstein

Bob Cain wrote:
Robert Morein wrote:
"Bob Cain" wrote in message
...
Carey Carlan wrote:
"RD Jones" wrote in
oups.com:

Has anyone played with using an XY pair matrixed as the S
in an MS pair to tackle the directivity issue ?
In other words a "bent 8" for the side ?
Your "bent 8" would still contain mostly Mid information. The purpose of
the side is to get the sides. Even at 180 spread, the best you can get
from a pair of cardioids is an omni mix. You need something with more
directivity, like a figure-8.
At any angle the S=L-R signal derived from a coincident, matched XY is a
figure 8 and M=L+R is an omni. What varies as you change the physical
angle is the sensitivity of the derived S relative to the derived M.

Bob, I don't see this. Reducto ad absurdium: make the angle zero. The XY sum
is still a cardioid.

You're right. Gotta go right now but will come back and figure out what
the hell I should have said (if anything) later.


Seems as though mental/visual trigonometry is another of those abilities
that suffer with age. :-(

Going at it a bit more rigorously, the pattern that describes (normalized)
capsule polar sensitivity in the cardiod family is:

1 + A*cos(theta)
----------------
2

The first term is the omni component and the second is the directional
gradient component, the figure 8. Theta is the angle between the capsule
axis and the sound source, positive clockwise. If A = 1 we have a true
cardiod. If A 1 we have a sub-cardiod and if A 1 a super or hyper
cardiod. Besides the shape difference between sub and super, the sub picks
up from the rear in phase with what it picks up from the front and the
super picks up from the rear out of phase with what it picks up from the front.

For a coincident pair of cardiod mics with the angle 2*alpha between them
we have

1 + cos(theta + alpha) 1 + cos(theta - alpha)
L = ---------------------- R = ----------------------
2 2

Deriving M by summing L and R and using the trig sum identity we get:

1 + cos(alpha)*cos(theta)
M = (L + R)/2 = -------------------------
2

This derived M is a member of the cardiod family with the cosine of half
the angle between the capsule's axes acting as A in the first equation and
determining which type. If the half angle is zero, M will be a true
cardiod (as you pointed out.) As the half angle moves toward 90 degrees
the M becomes more and more sub-cardiod until, when the capsules are
opposed, the M pattern becomes an omni.

[At the common XY angle of +/- 45 degrees, the derived M pattern is the
[subcardiod:
[
[ 1 + .707*cos(theta)
[ ------------------
[ 2


Driving S by differencing L and R and using the trig sum identity we get:

sin(alpha)*sin(theta)
S = (L - R)/2 = - ---------------------
2

This derived S is always a broadside figure 8 (left side positive) scaled
by the sin of the half angle between the capsules, becoming zero when the
angle is zero and a maximum when the capsules are opposed.


On to what we were really discussing, let's re-mix M/S back to L/R with a
scaling factor B applied to the S before the re-mixing.

1 + cos(alpha)*cos(theta) - B*sin(alpha)*sin(theta)
L' = M + B*S = ---------------------------------------------------
2

1 + cos(alpha)*cos(theta) + B*sin(alpha)*sin(theta)
R' = M - B*S = ---------------------------------------------------
2

From these some trig gives:

1 + B'*cos(theta + alpha')
L' = --------------------------
2

1 + B'*cos(theta - alpha')
R' = --------------------------
2

Both are in the cardiod family with

B' = sqrt(B^2*sin(alpha)^2 + cos(alpha)^2)

being the factor by which B changes the pattern toward sub (B 1) or super
(B 1). (Assuming the capsule patterns are true cardiod.)

And with

alpha' = atan(B*tan(alpha))

being the resultant half angle between L' and R' as rotated by the S
scaling factor B (B 1 is wider and B 1 narrower).

Thus, mixing from L/R to M/S, scaling S and mixing back to L/R both changes
the pattern of the L' and R' as well as the angle between them.


I'm sure this is way more than anyone wanted to know but having first
gotten it wrong, I wanted to take the time to get it right. :-)


Bob
--

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

A. Einstein

"Bob Cain" wrote in message

Bob Cain wrote:
Robert Morein wrote:
"Bob Cain" wrote in message
...
Carey Carlan wrote:
"RD Jones" wrote in
oups.com:

Has anyone played with using an XY pair matrixed as
the S in an MS pair to tackle the directivity issue ?
In other words a "bent 8" for the side ?
Your "bent 8" would still contain mostly Mid
information. The purpose of the side is to get the
sides. Even at 180 spread, the best you can get from
a pair of cardioids is an omni mix. You need
something with more directivity, like a figure-8.
At any angle the S=L-R signal derived from a
coincident, matched XY is a figure 8 and M=L+R is an
omni. What varies as you change the physical angle is
the sensitivity of the derived S relative to the
derived M.

Bob, I don't see this. Reducto ad absurdium: make the
angle zero. The XY sum is still a cardioid.

You're right. Gotta go right now but will come back and
figure out what the hell I should have said (if
anything) later.


Seems as though mental/visual trigonometry is another of
those abilities
that suffer with age. :-(

Going at it a bit more rigorously, the pattern that
describes (normalized) capsule polar sensitivity in the
cardiod family is:

1 + A*cos(theta)
----------------
2

The first term is the omni component and the second is
the directional gradient component, the figure 8. Theta
is the angle between the capsule
axis and the sound source, positive clockwise. If A = 1
we have a true cardiod. If A 1 we have a sub-cardiod
and if A 1 a super or hyper cardiod. Besides the shape
difference between sub and super, the sub picks
up from the rear in phase with what it picks up from the
front and the
super picks up from the rear out of phase with what it
picks up from the front.

For a coincident pair of cardiod mics with the angle
2*alpha between them
we have

1 + cos(theta + alpha) 1 + cos(theta -
alpha) L = ---------------------- R =
---------------------- 2
2

Deriving M by summing L and R and using the trig sum
identity we get:

1 + cos(alpha)*cos(theta)
M = (L + R)/2 = -------------------------
2

This derived M is a member of the cardiod family with the
cosine of half
the angle between the capsule's axes acting as A in the
first equation and determining which type. If the half
angle is zero, M will be a true
cardiod (as you pointed out.) As the half angle moves
toward 90 degrees
the M becomes more and more sub-cardiod until, when the
capsules are
opposed, the M pattern becomes an omni.

IOW, matrixed cardioids can simulate an onit. This is how most if not all
multi-pattern mics work.

[At the common XY angle of +/- 45 degrees, the derived M
pattern is the [subcardiod:
[
[ 1 + .707*cos(theta)
[ ------------------
[ 2


Driving S by differencing L and R and using the trig sum
identity we get:

sin(alpha)*sin(theta)
S = (L - R)/2 = - ---------------------
2

This derived S is always a broadside figure 8 (left side
positive) scaled
by the sin of the half angle between the capsules,
becoming zero when the angle is zero and a maximum when
the capsules are opposed.

IOW, matrixed cardioids can simulate a figure-8. This is how most if not all
multi-pattern mics work.

On to what we were really discussing, let's re-mix M/S
back to L/R with a scaling factor B applied to the S
before the re-mixing.

1 + cos(alpha)*cos(theta) -
B*sin(alpha)*sin(theta) L' = M + B*S =

--------------------------------------------------- 2

1 + cos(alpha)*cos(theta) +
B*sin(alpha)*sin(theta) R' = M - B*S =

--------------------------------------------------- 2

From these some trig gives:

1 + B'*cos(theta + alpha')
L' = --------------------------
2

1 + B'*cos(theta - alpha')
R' = --------------------------
2

Both are in the cardiod family with

B' = sqrt(B^2*sin(alpha)^2 + cos(alpha)^2)

being the factor by which B changes the pattern toward
sub (B 1) or super (B 1). (Assuming the capsule
patterns are true cardiod.)

And with

alpha' = atan(B*tan(alpha))

being the resultant half angle between L' and R' as
rotated by the S
scaling factor B (B 1 is wider and B 1 narrower).

Thus, mixing from L/R to M/S, scaling S and mixing back
to L/R both changes the pattern of the L' and R' as well
as the angle between them.

Begging the question, what happens when you go X/Y with hypercardioids.

I'm sure this is way more than anyone wanted to know

Not at all. Good stuff.

Arny Krueger wrote:

Begging the question, what happens when you go X/Y with hypercardioids.

Two things:

1. The angle of acceptance of the front lobe is narrower. Therefore you
need to get farther from the source to get the same degree of ambience
but the angle between the mikes remains the same.

2. There is a rear lobe so there is some reduced pickup from the rear of
the hall. This can emphasize slap echoes occasionally but can be handy
for antiphonal organs.
--scott

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

(Scott Dorsey) wrote in
:

Arny Krueger wrote:

Begging the question, what happens when you go X/Y with
hypercardioids.

Two things:

1. The angle of acceptance of the front lobe is narrower. Therefore
you
need to get farther from the source to get the same degree of
ambience but the angle between the mikes remains the same.

The angle remains the same? Would you not need a narrower angle to
maintain the -3 dB point at center?

I flout this rule constantly and spread my hypercardioids quite wide, but I
do it with the realization that center level may suffer.

"Carey Carlan" wrote in message

(Scott Dorsey) wrote in
:

Arny Krueger wrote:

Begging the question, what happens when you go X/Y with
hypercardioids.

Two things:

1. The angle of acceptance of the front lobe is
narrower. Therefore you
need to get farther from the source to get the same
degree of ambience but the angle between the mikes
remains the same.

The angle remains the same? Would you not need a
narrower angle to maintain the -3 dB point at center?

So it would seem.

I flout this rule constantly and spread my hypercardioids
quite wide, but I do it with the realization that center
level may suffer.

I contemplate the fact that while the center level for direct sound may
suffer, there's always all of that reflected sound.

IMO hypercardioids are about trying to pry a higher-resolution sound from a
lower-resolution environment.

Carey Carlan wrote:
(Scott Dorsey) wrote in
:

Arny Krueger wrote:

Begging the question, what happens when you go X/Y with
hypercardioids.

Two things:

1. The angle of acceptance of the front lobe is narrower. Therefore
you
need to get farther from the source to get the same degree of
ambience but the angle between the mikes remains the same.

The angle remains the same? Would you not need a narrower angle to
maintain the -3 dB point at center?

You do, BUT closing that down gives you a narrower angle of acceptance
of the whole pair as a result.

I flout this rule constantly and spread my hypercardioids quite wide, but I
do it with the realization that center level may suffer.

Right. It's all a compromise. The ORTF spec is a pretty good compromise
but isn't optimal in every environment.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Arny Krueger wrote:


IOW, matrixed cardioids can simulate an onit. This is how most if not all
multi-pattern mics work.

Zackly.

IOW, matrixed cardioids can simulate a figure-8. This is how most if not all
multi-pattern mics work.

Again. :-)

Begging the question, what happens when you go X/Y with hypercardioids.

Start with

1 + A*cos(theta + alpha) 1 + A*cos(theta - alpha)
L = ------------------------ R = ------------------------
2 2

with A 1 equal to whatever defines hypercardiod and proceed to

1 + A*cos(alpha)*cos(theta)
M = (L + R)/2 = ---------------------------
2

A*sin(alpha)*sin(theta)
S = (L - R)/2 = - -----------------------
2

then mix back to

1 + A*cos(alpha)*cos(theta) - A*B*sin(alpha)*sin(theta)
L' = M + B*S = -------------------------------------------------------
2

1 + A*cos(alpha)*cos(theta) + A*B*sin(alpha)*sin(theta)
R' = M - B*S = -------------------------------------------------------
2

which simplifies to

1 + A*B'*cos(theta + alpha')
L' = ----------------------------
2

1 + A*B'*cos(theta - alpha')
R' = ----------------------------
2

with B' and alpha' defined the same as before. Simpler change than I
expected, actually.


Bob
--

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

A. Einstein

Carey Carlan wrote:
(Scott Dorsey) wrote in
:


The angle remains the same? Would you not need a narrower angle to
maintain the -3 dB point at center?

Could you explain what you mean by -3 dB point at the center and why you
want that?

Thanks,

Bob
--

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

A. Einstein