Richard Kuschel wrote:
However, as the Schoeps website says about the MK8, "directional pattern
almost completely identical at all frequencies." Every fig-8 mic can share
the exact same description, because they're all unequal front-to-rear, which
translates to unequal left-to-right in MS.



Why would that be?
If using a single diaphragm figute 8 (Schoeps, Sennheiser and Neumann) wouldn't
the response be the same front to back.

They aren't usually mechanically symmetric front and back.


Bob
--

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

A. Einstein

David Satz wrote:

Given the margin of measurement uncertainty, these are effectively the
same curve to within 1 dB--and below 3 kHz they _are_ the same curve.

That _is_ symmetric for all practical purposes. But you
knew that. :-)


Bob
--

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

A. Einstein

David Satz wrote:

Given the margin of measurement uncertainty, these are effectively the
same curve to within 1 dB--and below 3 kHz they _are_ the same curve.

That _is_ symmetric for all practical purposes. But you
knew that. :-)


Bob
--

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

A. Einstein

David Satz wrote:

So in general, I favor using a supercardioid "M" microphone. If I have
to cover so wide an area that I can't use that narrow a pattern, then a
centered, coincident stereo pickup probably isn't the best idea anyway
(unless mono compatibility is an absolute requirement).

If you think of the cardiod family M as a figure 8 summed
with an omni and if you match the gain of the figure 8
component with that of the side, then the sum and difference
that do the MS - LR encoding just create a left figure 8
pointing at -45 and a right figure 8 pointing at +45 both
superimposed with the omni component of the M mic. The
ratio of that omni component to the figure 8 is what
determines where in the cardiod spectrum the real mic and
virtual LR mics fit (hyper, super, card, sub, hypo.)

If the fig 8 sensitivities are matched between mid and side
and the M is a true card then the +-45 degree virtual LR
mics will be hyper cardiod because their fig 8 components
will have sensitivities that are the sqrt(2) times that of
the actual mics while the omni component is one times the
omni component of the mid. Thus, going somewhat
hypercardiod with the mid, as you suggest, makes the virtual
LR mics even more hypercardiod.

In practice, the M and S sensitivities are nowhere near as
carefully controled if one is setting one's own relative MS
gains and it is virtually impossible to say what virtual LR
patterns you have. For that reason, you experiment, listen
and tweak. :-)

If you further take into consideration that these patterns
are highly frequency dependant, even if symmetric front to
back, then it is _very_ hard to generalize what you get from
a coincident (intensity based) stereo configuration, MS or
XY. That's not to say, however, that they can't sound great.

One thing I have not been able to get away from with MS or
XY is that if I have a drum kit that is a fair bit to the
side then the image separation between the kick and cymbal
is very highly exagerated and the cymbal can even swap
sides. This is because of the frequency dependant patterns
and because phase matching becomes very important at the
higher frequencies due to the MS-LR summing.

The more I study this stuff the more I come to realize that
there really isn't a theoretical basis for stereo
configurations that holds any water because all the premises
of the theories are idealizations that are terribly far from
reality. It's all hueristic in the end. Try things without
much worrying about what is right or wrong and see what you get.


Bob
--

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

A. Einstein

David Satz wrote:

So in general, I favor using a supercardioid "M" microphone. If I have
to cover so wide an area that I can't use that narrow a pattern, then a
centered, coincident stereo pickup probably isn't the best idea anyway
(unless mono compatibility is an absolute requirement).

If you think of the cardiod family M as a figure 8 summed
with an omni and if you match the gain of the figure 8
component with that of the side, then the sum and difference
that do the MS - LR encoding just create a left figure 8
pointing at -45 and a right figure 8 pointing at +45 both
superimposed with the omni component of the M mic. The
ratio of that omni component to the figure 8 is what
determines where in the cardiod spectrum the real mic and
virtual LR mics fit (hyper, super, card, sub, hypo.)

If the fig 8 sensitivities are matched between mid and side
and the M is a true card then the +-45 degree virtual LR
mics will be hyper cardiod because their fig 8 components
will have sensitivities that are the sqrt(2) times that of
the actual mics while the omni component is one times the
omni component of the mid. Thus, going somewhat
hypercardiod with the mid, as you suggest, makes the virtual
LR mics even more hypercardiod.

In practice, the M and S sensitivities are nowhere near as
carefully controled if one is setting one's own relative MS
gains and it is virtually impossible to say what virtual LR
patterns you have. For that reason, you experiment, listen
and tweak. :-)

If you further take into consideration that these patterns
are highly frequency dependant, even if symmetric front to
back, then it is _very_ hard to generalize what you get from
a coincident (intensity based) stereo configuration, MS or
XY. That's not to say, however, that they can't sound great.

One thing I have not been able to get away from with MS or
XY is that if I have a drum kit that is a fair bit to the
side then the image separation between the kick and cymbal
is very highly exagerated and the cymbal can even swap
sides. This is because of the frequency dependant patterns
and because phase matching becomes very important at the
higher frequencies due to the MS-LR summing.

The more I study this stuff the more I come to realize that
there really isn't a theoretical basis for stereo
configurations that holds any water because all the premises
of the theories are idealizations that are terribly far from
reality. It's all hueristic in the end. Try things without
much worrying about what is right or wrong and see what you get.


Bob
--

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

A. Einstein

It's still a moot point though, you really need spaced cardioids to get
the
time differential aspect of natural imaging.

I strongly disagree. Coincident, intensity based methods
are in wide and sucessful use. It's a matter of taste.

The time differential that is recorded gets so munged by the
time domain mangling that speakers and rooms effect that on
reproduction it in no way resembles what was recorded.

I agree, it is a matter of taste. For example, you could take that whole
response and replace time differential issues with intensity issues, still
100% true.

There are times to say "people's stereos suck" and make comprimises, but IMO
distant micing with spaced mics for acoustic instruments never does more
harm than good if done right.

It's still a moot point though, you really need spaced cardioids to get
the
time differential aspect of natural imaging.

I strongly disagree. Coincident, intensity based methods
are in wide and sucessful use. It's a matter of taste.

The time differential that is recorded gets so munged by the
time domain mangling that speakers and rooms effect that on
reproduction it in no way resembles what was recorded.

I agree, it is a matter of taste. For example, you could take that whole
response and replace time differential issues with intensity issues, still
100% true.

There are times to say "people's stereos suck" and make comprimises, but IMO
distant micing with spaced mics for acoustic instruments never does more
harm than good if done right.

[A]s the Schoeps website says about the MK8, "directional pattern
almost completely identical at all frequencies." Every fig-8 mic
can share the exact same description, because they're all unequal
front-to-rear, which translates to unequal left-to-right in MS.

The degree of front/back discrepancy in figure-8 microphones varies
from design to design. For example, Royer has two basic models of
ribbon microphone: one for pop and one for classical (more or less).
The "classical" one is highly symmetrical front to back, while the
"pop" one is not so much, and the company is quite open about this.
The "classical" model also has the flatter frequency response of the
two models through the midrange and upper midrange. It would thus
be preferable all around as the "S" microphone for an M/S pair.

I have some special interest in your remarks since I'm the primary
German-to-English translator for Schoeps, and those are my words that
you're quoting. Small response discrepancies do show up between the
front and back lobes of a Schoeps figure-8. But discrepancies of that
magnitude show up when you run the same curve twice in a row without
changing or moving anything, so that should be taken into account, too.

I'm looking at the factory frequency response curves for my own MK 8
capsules, using the same pen and paper speeds that Schoeps' engineers
use for their own research, not some idealized version of the curves
for marketing purposes. (Or OK--I was looking at them a minute ago,
then I put them down because I can't type while holding them, but they
are the real curves.) No discrepancy between front and back is visible
on the chart recorder until you get above, say 3 kHz, where there are
occasional burbles of some fraction of a dB or so--up to maybe one
whole entire dB (horror of horrors!) above 9 kHz.

Given the margin of measurement uncertainty, these are effectively the
same curve to within 1 dB--and below 3 kHz they _are_ the same curve.

This is leading towards a debate over whether two matched cardioids are in
fact a truer match than either side of a fig-8. I'm of the opinion that the
matched card's are a truer match, mostly because you can't as easily test
the internal bipolar matching on a mic-by-mic basis like matching two
card's. In particular, when two card's achieve a null result under
opposition, they can be trusted to be matched in both mechanics and
electronics with extreme precision. It's impossible to test both sides of a
fig-8 with that level of precision. All you can do is "read curves".

It's still a moot point though, you really need spaced cardioids
to get the time differential aspect of natural imaging.

That's the direct opposite of what most people in this field believe,
and what several decades of live recording experience tell me.

Whenever you play a recording back through loudspeakers, you get the
same arrival time differences between the listeners' ears as you would
get in the original sound field. You don't need to create arrival time
differences between channels by spacing the microphones unless you're
recording for headphone ("binaural") playback. Otherwise it's an option
if you like the resulting sound, but it's not any kind of necessity.

I certainly hope that's not what most people in this field believe. You do
get time differentials from the separation of loudspeakers, but it's hardly
"one size fits all". To say spaced mic techniques are only appropriate for
binaural playback is as obtuse as saying coincident mic techniques are only
good for separated loudspeaker playback.

In particular, spaced mics can offer a sense of space that coincident mics
can't, since lateral sounds, in particular room ambience, result in greater
time differentials, producing a greater sense of width without comprimising
the coverage of the center signals, which for coincident mics means sending
the center signals further off-axis or widening the stereo recording in
mix/post.

There are times to add space between mics, times to widen the angle between
coincident mics, and times to narrow the angle and widen in post. Distant
micing of acoustic instruments generally means adding space IME. When a
sound reinforcement system is involved, coincident mics are imperative,
since spacing the mics generates very unnatural correlated differentials
from the separated sound sources putting out the same signal. When you have
a broad ensemble of acoustic sources, you end up somewhere in between.

I'm somewhat surprised at your opinion. Normally I'm trying to tell tapers
not to use spaced mics for concerts with PA's...

[A]s the Schoeps website says about the MK8, "directional pattern
almost completely identical at all frequencies." Every fig-8 mic
can share the exact same description, because they're all unequal
front-to-rear, which translates to unequal left-to-right in MS.

The degree of front/back discrepancy in figure-8 microphones varies
from design to design. For example, Royer has two basic models of
ribbon microphone: one for pop and one for classical (more or less).
The "classical" one is highly symmetrical front to back, while the
"pop" one is not so much, and the company is quite open about this.
The "classical" model also has the flatter frequency response of the
two models through the midrange and upper midrange. It would thus
be preferable all around as the "S" microphone for an M/S pair.

I have some special interest in your remarks since I'm the primary
German-to-English translator for Schoeps, and those are my words that
you're quoting. Small response discrepancies do show up between the
front and back lobes of a Schoeps figure-8. But discrepancies of that
magnitude show up when you run the same curve twice in a row without
changing or moving anything, so that should be taken into account, too.

I'm looking at the factory frequency response curves for my own MK 8
capsules, using the same pen and paper speeds that Schoeps' engineers
use for their own research, not some idealized version of the curves
for marketing purposes. (Or OK--I was looking at them a minute ago,
then I put them down because I can't type while holding them, but they
are the real curves.) No discrepancy between front and back is visible
on the chart recorder until you get above, say 3 kHz, where there are
occasional burbles of some fraction of a dB or so--up to maybe one
whole entire dB (horror of horrors!) above 9 kHz.

Given the margin of measurement uncertainty, these are effectively the
same curve to within 1 dB--and below 3 kHz they _are_ the same curve.

This is leading towards a debate over whether two matched cardioids are in
fact a truer match than either side of a fig-8. I'm of the opinion that the
matched card's are a truer match, mostly because you can't as easily test
the internal bipolar matching on a mic-by-mic basis like matching two
card's. In particular, when two card's achieve a null result under
opposition, they can be trusted to be matched in both mechanics and
electronics with extreme precision. It's impossible to test both sides of a
fig-8 with that level of precision. All you can do is "read curves".

It's still a moot point though, you really need spaced cardioids
to get the time differential aspect of natural imaging.

That's the direct opposite of what most people in this field believe,
and what several decades of live recording experience tell me.

Whenever you play a recording back through loudspeakers, you get the
same arrival time differences between the listeners' ears as you would
get in the original sound field. You don't need to create arrival time
differences between channels by spacing the microphones unless you're
recording for headphone ("binaural") playback. Otherwise it's an option
if you like the resulting sound, but it's not any kind of necessity.

I certainly hope that's not what most people in this field believe. You do
get time differentials from the separation of loudspeakers, but it's hardly
"one size fits all". To say spaced mic techniques are only appropriate for
binaural playback is as obtuse as saying coincident mic techniques are only
good for separated loudspeaker playback.

In particular, spaced mics can offer a sense of space that coincident mics
can't, since lateral sounds, in particular room ambience, result in greater
time differentials, producing a greater sense of width without comprimising
the coverage of the center signals, which for coincident mics means sending
the center signals further off-axis or widening the stereo recording in
mix/post.

There are times to add space between mics, times to widen the angle between
coincident mics, and times to narrow the angle and widen in post. Distant
micing of acoustic instruments generally means adding space IME. When a
sound reinforcement system is involved, coincident mics are imperative,
since spacing the mics generates very unnatural correlated differentials
from the separated sound sources putting out the same signal. When you have
a broad ensemble of acoustic sources, you end up somewhere in between.

I'm somewhat surprised at your opinion. Normally I'm trying to tell tapers
not to use spaced mics for concerts with PA's...

Sugarite wrote:

There are times to say "people's stereos suck" and make comprimises, but IMO
distant micing with spaced mics for acoustic instruments never does more
harm than good if done right.

I need to do more work with spaced omni's. The deeper I
investigate the operation of coincident methods the more I
see that they are fraught with difficulties due to angular
variations in frequency response (magnitude and phase.)
They aren't just theoretical either, I can hear them. I've
been at a loss for some time to understand why the cymbals
from a drumset would be so far removed from the kick thump
in an MS recording when they aren't physically removed much
at all but the reasons for that are becoming clearer.


Bob
--

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

A. Einstein

Sugarite wrote:

There are times to say "people's stereos suck" and make comprimises, but IMO
distant micing with spaced mics for acoustic instruments never does more
harm than good if done right.

I need to do more work with spaced omni's. The deeper I
investigate the operation of coincident methods the more I
see that they are fraught with difficulties due to angular
variations in frequency response (magnitude and phase.)
They aren't just theoretical either, I can hear them. I've
been at a loss for some time to understand why the cymbals
from a drumset would be so far removed from the kick thump
in an MS recording when they aren't physically removed much
at all but the reasons for that are becoming clearer.


Bob
--

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

A. Einstein

Sugarite wrote:
When a
sound reinforcement system is involved, coincident mics are imperative,
since spacing the mics generates very unnatural correlated differentials
from the separated sound sources putting out the same signal. When you have
a broad ensemble of acoustic sources, you end up somewhere in between.

But why would one be doing stereo with sound reinforcement?
The few times I've tried it the variation due to position
in the audience has made it useless unless you are really
close to center and even then it just doesn't come out right.


Bob
--

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

A. Einstein

Sugarite wrote:
When a
sound reinforcement system is involved, coincident mics are imperative,
since spacing the mics generates very unnatural correlated differentials
from the separated sound sources putting out the same signal. When you have
a broad ensemble of acoustic sources, you end up somewhere in between.

But why would one be doing stereo with sound reinforcement?
The few times I've tried it the variation due to position
in the audience has made it useless unless you are really
close to center and even then it just doesn't come out right.


Bob
--

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

A. Einstein

Bob Cain wrote:
Sugarite wrote:
When a
sound reinforcement system is involved, coincident mics are imperative,
since spacing the mics generates very unnatural correlated differentials
from the separated sound sources putting out the same signal. When you have
a broad ensemble of acoustic sources, you end up somewhere in between.

But why would one be doing stereo with sound reinforcement?
The few times I've tried it the variation due to position
in the audience has made it useless unless you are really
close to center and even then it just doesn't come out right.

It depends a lot on the room. In a narrow and long room that isn't too
bright, you can do a surprisingly good job of getting a real stereo image,
even on the sides. You may need mono front fills in order to deal with
the first couple rows on the side, though.

Also, a center cluster can do an amazing job of making stereo work in a
room where it otherwise wouldn't. It won't do miracles, though.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Bob Cain wrote:
Sugarite wrote:
When a
sound reinforcement system is involved, coincident mics are imperative,
since spacing the mics generates very unnatural correlated differentials
from the separated sound sources putting out the same signal. When you have
a broad ensemble of acoustic sources, you end up somewhere in between.

But why would one be doing stereo with sound reinforcement?
The few times I've tried it the variation due to position
in the audience has made it useless unless you are really
close to center and even then it just doesn't come out right.

It depends a lot on the room. In a narrow and long room that isn't too
bright, you can do a surprisingly good job of getting a real stereo image,
even on the sides. You may need mono front fills in order to deal with
the first couple rows on the side, though.

Also, a center cluster can do an amazing job of making stereo work in a
room where it otherwise wouldn't. It won't do miracles, though.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Bob Cain wrote:

Using three identical capsules, one for the front and two
for the side and reasonably matched is pretty near optimal
for M/S IMO.

Bob, as you know I've posted some messages here saying that matching
the sonic character of the M and S mikes is not a particular virtue,
and that a flat/smooth/accurate characteristic for S matters far more.

Then you sent me your more detailed thoughts off-list, which gave me a
chance to rethink this issue. I have to admit that there's more to it
than I thought before (i.e. there may be something to it at all).

Here's how I would express it non-mathematically. The issue is how
the stereo image will be affected by the response characteristics of
the two microphones after the M/S signals have been matrixed to L/R.

If the M microphone emphasizes certain sound components (e.g. the high
frequencies) while the S microphone doesn't, then in the eventual L/R
stereo output of the matrix, those components will tend to gather at
the image's center more than the midrange sound does. The apparent
left-right position of any given instrument or voice would thus be
blurred. Conversely, if the S microphone boosts the highs while the
M microphone doesn't, high frequency energy would tend to be shifted
more toward the left and right than the midrange sound--increasingly
so with individual voices or instruments that are already off-center.

Both effects are undesirable, but they do work in generally opposing
directions--either one would tend to compensate for the other to some
degree. So I think I see the intended point of "matching" the general
sonic characteristics of the M and S microphones now. I don't endorse
it completely, but I see a possible rational hope behind it at least.

It will take me another little while to say where I don't agree; the
main issue is the fact that most microphones (except for the best
small-diaphragm figure-8s and certain rather noisy omnis) don't have
the same frequency response at all angles of incidence. So all our
"virtual microphones" have rather complex, irregular response--that's
another reason I tend to favor a super- or hypercardioid for "M".

However, I'll also point out that the analog of using a non-flat S mike
is processing the S channel independently of the M channel--and there is
a long tradition of doing exactly that. One of the attractions of M/S
(whether the recording is made originally that way, or whether another
type of X/Y recording is matrixed to M/S) is that the M and S channels
can be equalized independently or processed in other ways independently,
then rematrixed to L/R stereo.

For example, LP disc cutters would often matrix a stereo signal to sum
and difference channels (equivalent to M and S), and limit the difference
channel and/or roll off its low-frequency energy, since the difference
energy corresponds essentially to the vertical component of the groove
to be cut, while the M channel corresponds to its horizontal component.
Excessive vertical modulation could cause skipping, and would tend to
limit the playing time of the side and the maximum modulation of the
groove--which in turn would limit the playback volume and the signal-
to-noise ratio. Cutting the loudest possible record was an obsession
with LP mastering engineers, too--not something that snuck into the
business with the diabolical introduction of digital recording.

This also addresses a point made by "Sugarite"--he's right that X/Y
(coincident) recordings often lack a sense of spaciousness, and that
spaced microphone recordings are sometimes preferred for that reason.
But one other way to increase the sense of spaciousness, or at least
to help compensate for some of the relative lack of it, is to boost
the low-frequency response of the "S" channel.

If anyone remembers Bob Carver's "Digital Time Lens" processor--a
surprisingly useful general-purpose "feel-good box" that doesn't add
distortion--that's one of the tricks he was using in that circuit.

Bob Cain wrote:

Using three identical capsules, one for the front and two
for the side and reasonably matched is pretty near optimal
for M/S IMO.

Bob, as you know I've posted some messages here saying that matching
the sonic character of the M and S mikes is not a particular virtue,
and that a flat/smooth/accurate characteristic for S matters far more.

Then you sent me your more detailed thoughts off-list, which gave me a
chance to rethink this issue. I have to admit that there's more to it
than I thought before (i.e. there may be something to it at all).

Here's how I would express it non-mathematically. The issue is how
the stereo image will be affected by the response characteristics of
the two microphones after the M/S signals have been matrixed to L/R.

If the M microphone emphasizes certain sound components (e.g. the high
frequencies) while the S microphone doesn't, then in the eventual L/R
stereo output of the matrix, those components will tend to gather at
the image's center more than the midrange sound does. The apparent
left-right position of any given instrument or voice would thus be
blurred. Conversely, if the S microphone boosts the highs while the
M microphone doesn't, high frequency energy would tend to be shifted
more toward the left and right than the midrange sound--increasingly
so with individual voices or instruments that are already off-center.

Both effects are undesirable, but they do work in generally opposing
directions--either one would tend to compensate for the other to some
degree. So I think I see the intended point of "matching" the general
sonic characteristics of the M and S microphones now. I don't endorse
it completely, but I see a possible rational hope behind it at least.

It will take me another little while to say where I don't agree; the
main issue is the fact that most microphones (except for the best
small-diaphragm figure-8s and certain rather noisy omnis) don't have
the same frequency response at all angles of incidence. So all our
"virtual microphones" have rather complex, irregular response--that's
another reason I tend to favor a super- or hypercardioid for "M".

However, I'll also point out that the analog of using a non-flat S mike
is processing the S channel independently of the M channel--and there is
a long tradition of doing exactly that. One of the attractions of M/S
(whether the recording is made originally that way, or whether another
type of X/Y recording is matrixed to M/S) is that the M and S channels
can be equalized independently or processed in other ways independently,
then rematrixed to L/R stereo.

For example, LP disc cutters would often matrix a stereo signal to sum
and difference channels (equivalent to M and S), and limit the difference
channel and/or roll off its low-frequency energy, since the difference
energy corresponds essentially to the vertical component of the groove
to be cut, while the M channel corresponds to its horizontal component.
Excessive vertical modulation could cause skipping, and would tend to
limit the playing time of the side and the maximum modulation of the
groove--which in turn would limit the playback volume and the signal-
to-noise ratio. Cutting the loudest possible record was an obsession
with LP mastering engineers, too--not something that snuck into the
business with the diabolical introduction of digital recording.

This also addresses a point made by "Sugarite"--he's right that X/Y
(coincident) recordings often lack a sense of spaciousness, and that
spaced microphone recordings are sometimes preferred for that reason.
But one other way to increase the sense of spaciousness, or at least
to help compensate for some of the relative lack of it, is to boost
the low-frequency response of the "S" channel.

If anyone remembers Bob Carver's "Digital Time Lens" processor--a
surprisingly useful general-purpose "feel-good box" that doesn't add
distortion--that's one of the tricks he was using in that circuit.

Great thread - thanks for all of the information everyone.

In general I go with ORTF or some near-coincident variant therof but I
like to use spaced-pair techniques on large ensembles when the room
permits, or if I am using two stereo pairs in conjunction - one to
accent something like a choir, in which case I use a coincident pair
on the accent to get a strong center and clear localisation to better
blend with the diffuse image of the spaced pair - usually a nice
effect. I prefer to use coicindent methods simply for the effect of a
narrower image with a strong center, to be able to control the width
of the image later without phase problems, or for sound that will end
up on video.

I have heard much about the supposed advantage of MS over XY because
of the ability to control stereo width after the fact, but I have
never used it due to the complications of rigging it. Usually my
recordings are done live in concert, so aesthetics are a concern, but
for a recording session coming up soon (a jazz trio), I was
considering combining a spaced pair in conjunction with a coincident
pair for drums and piano. I am considering MS because aesthetics are
not a consideration in this case. Investing in a Schoeps CMC6 MK8
combo is not out of the question, but as I have not often used MS, I
would rather utilise that chunk of my equipment budget in other ways.

I would specifically like to ask David Satz a question about the
Schoeps MK8 capsule due to his familiarity, of course anyone can
respond: I have seen the response charts of the MK8, and I recall that
the drop-off in the bass and the high end were rather severe, but I
know enough to take these charts with a grain of salt. Has it been
your experience that the "accuracy" of sound reproduction was
adversely affected by this?


Bob Cain wrote in message ...
David Satz wrote:

So in general, I favor using a supercardioid "M" microphone. If I have
to cover so wide an area that I can't use that narrow a pattern, then a
centered, coincident stereo pickup probably isn't the best idea anyway
(unless mono compatibility is an absolute requirement).

If you think of the cardiod family M as a figure 8 summed
with an omni and if you match the gain of the figure 8
component with that of the side, then the sum and difference
that do the MS - LR encoding just create a left figure 8
pointing at -45 and a right figure 8 pointing at +45 both
superimposed with the omni component of the M mic. The
ratio of that omni component to the figure 8 is what
determines where in the cardiod spectrum the real mic and
virtual LR mics fit (hyper, super, card, sub, hypo.)

If the fig 8 sensitivities are matched between mid and side
and the M is a true card then the +-45 degree virtual LR
mics will be hyper cardiod because their fig 8 components
will have sensitivities that are the sqrt(2) times that of
the actual mics while the omni component is one times the
omni component of the mid. Thus, going somewhat
hypercardiod with the mid, as you suggest, makes the virtual
LR mics even more hypercardiod.

In practice, the M and S sensitivities are nowhere near as
carefully controled if one is setting one's own relative MS
gains and it is virtually impossible to say what virtual LR
patterns you have. For that reason, you experiment, listen
and tweak. :-)

If you further take into consideration that these patterns
are highly frequency dependant, even if symmetric front to
back, then it is _very_ hard to generalize what you get from
a coincident (intensity based) stereo configuration, MS or
XY. That's not to say, however, that they can't sound great.

One thing I have not been able to get away from with MS or
XY is that if I have a drum kit that is a fair bit to the
side then the image separation between the kick and cymbal
is very highly exagerated and the cymbal can even swap
sides. This is because of the frequency dependant patterns
and because phase matching becomes very important at the
higher frequencies due to the MS-LR summing.

The more I study this stuff the more I come to realize that
there really isn't a theoretical basis for stereo
configurations that holds any water because all the premises
of the theories are idealizations that are terribly far from
reality. It's all hueristic in the end. Try things without
much worrying about what is right or wrong and see what you get.


Bob

Great thread - thanks for all of the information everyone.

In general I go with ORTF or some near-coincident variant therof but I
like to use spaced-pair techniques on large ensembles when the room
permits, or if I am using two stereo pairs in conjunction - one to
accent something like a choir, in which case I use a coincident pair
on the accent to get a strong center and clear localisation to better
blend with the diffuse image of the spaced pair - usually a nice
effect. I prefer to use coicindent methods simply for the effect of a
narrower image with a strong center, to be able to control the width
of the image later without phase problems, or for sound that will end
up on video.

I have heard much about the supposed advantage of MS over XY because
of the ability to control stereo width after the fact, but I have
never used it due to the complications of rigging it. Usually my
recordings are done live in concert, so aesthetics are a concern, but
for a recording session coming up soon (a jazz trio), I was
considering combining a spaced pair in conjunction with a coincident
pair for drums and piano. I am considering MS because aesthetics are
not a consideration in this case. Investing in a Schoeps CMC6 MK8
combo is not out of the question, but as I have not often used MS, I
would rather utilise that chunk of my equipment budget in other ways.

I would specifically like to ask David Satz a question about the
Schoeps MK8 capsule due to his familiarity, of course anyone can
respond: I have seen the response charts of the MK8, and I recall that
the drop-off in the bass and the high end were rather severe, but I
know enough to take these charts with a grain of salt. Has it been
your experience that the "accuracy" of sound reproduction was
adversely affected by this?


Bob Cain wrote in message ...
David Satz wrote:

So in general, I favor using a supercardioid "M" microphone. If I have
to cover so wide an area that I can't use that narrow a pattern, then a
centered, coincident stereo pickup probably isn't the best idea anyway
(unless mono compatibility is an absolute requirement).

If you think of the cardiod family M as a figure 8 summed
with an omni and if you match the gain of the figure 8
component with that of the side, then the sum and difference
that do the MS - LR encoding just create a left figure 8
pointing at -45 and a right figure 8 pointing at +45 both
superimposed with the omni component of the M mic. The
ratio of that omni component to the figure 8 is what
determines where in the cardiod spectrum the real mic and
virtual LR mics fit (hyper, super, card, sub, hypo.)

If the fig 8 sensitivities are matched between mid and side
and the M is a true card then the +-45 degree virtual LR
mics will be hyper cardiod because their fig 8 components
will have sensitivities that are the sqrt(2) times that of
the actual mics while the omni component is one times the
omni component of the mid. Thus, going somewhat
hypercardiod with the mid, as you suggest, makes the virtual
LR mics even more hypercardiod.

In practice, the M and S sensitivities are nowhere near as
carefully controled if one is setting one's own relative MS
gains and it is virtually impossible to say what virtual LR
patterns you have. For that reason, you experiment, listen
and tweak. :-)

If you further take into consideration that these patterns
are highly frequency dependant, even if symmetric front to
back, then it is _very_ hard to generalize what you get from
a coincident (intensity based) stereo configuration, MS or
XY. That's not to say, however, that they can't sound great.

One thing I have not been able to get away from with MS or
XY is that if I have a drum kit that is a fair bit to the
side then the image separation between the kick and cymbal
is very highly exagerated and the cymbal can even swap
sides. This is because of the frequency dependant patterns
and because phase matching becomes very important at the
higher frequencies due to the MS-LR summing.

The more I study this stuff the more I come to realize that
there really isn't a theoretical basis for stereo
configurations that holds any water because all the premises
of the theories are idealizations that are terribly far from
reality. It's all hueristic in the end. Try things without
much worrying about what is right or wrong and see what you get.


Bob

Eric K. Weber wrote:

The following small diaphram solutions do exist....

Sennheiser MKH30 fig-8 MKH40 card or MKH50 hypercard
http://www.sennheiser.com/sennheiser/icm.nsf/root/02872
http://www.sennheiser.com/sennheiser/icm.nsf/root/02645
http://www.sennheiser.com/sennheiser/icm.nsf/root/03109

Schoeps also has small diaphragm figure 8..
http://www.schoeps.de/E-2004/figure-8.html
http://www.schoeps.de/E-2004/ms-stereo.html#msoutdoor

It's significant that these are not only small-diapragm designs but also
_single-diaphragm_ designs. And let's not forget the Neumann KM 120:

http://www.neumann.com/infopool/mics...p?ProdID=km100

Oddly I've never heard from anyone who has ever used it--and I've
asked around repeatedly. Here goes again: If anyone here has used
this microphone, could you please comment on its sound? I sometimes
use spaced (!) figure-8s for stereo vocal recording, and can imagine
that a response characteristic as shown in Neumann's diagrams might
be useful for that application, or at least worth trying.

_Dual-diaphragm_ small-diaphragm mikes with a selectable figure-8
pattern would include the Neumann KM 56, KM 88 and KM 86. I've used
KM 86s and KM 88s as figure-8 microphones and preferred the KM 88; the
KM 86 is very bright, even harsh off axis when used as a figure-8.

But I chickened out and sold my KM 88s this past year. One accident
with their fragile nickel membranes and poof, it's all over--no original
replacement capsules are available any more, and I'm not interested in
"reskinned" capsules.

Eric K. Weber wrote:

The following small diaphram solutions do exist....

Sennheiser MKH30 fig-8 MKH40 card or MKH50 hypercard
http://www.sennheiser.com/sennheiser/icm.nsf/root/02872
http://www.sennheiser.com/sennheiser/icm.nsf/root/02645
http://www.sennheiser.com/sennheiser/icm.nsf/root/03109

Schoeps also has small diaphragm figure 8..
http://www.schoeps.de/E-2004/figure-8.html
http://www.schoeps.de/E-2004/ms-stereo.html#msoutdoor

It's significant that these are not only small-diapragm designs but also
_single-diaphragm_ designs. And let's not forget the Neumann KM 120:

http://www.neumann.com/infopool/mics...p?ProdID=km100

Oddly I've never heard from anyone who has ever used it--and I've
asked around repeatedly. Here goes again: If anyone here has used
this microphone, could you please comment on its sound? I sometimes
use spaced (!) figure-8s for stereo vocal recording, and can imagine
that a response characteristic as shown in Neumann's diagrams might
be useful for that application, or at least worth trying.

_Dual-diaphragm_ small-diaphragm mikes with a selectable figure-8
pattern would include the Neumann KM 56, KM 88 and KM 86. I've used
KM 86s and KM 88s as figure-8 microphones and preferred the KM 88; the
KM 86 is very bright, even harsh off axis when used as a figure-8.

But I chickened out and sold my KM 88s this past year. One accident
with their fragile nickel membranes and poof, it's all over--no original
replacement capsules are available any more, and I'm not interested in
"reskinned" capsules.

When a
sound reinforcement system is involved, coincident mics are imperative,
since spacing the mics generates very unnatural correlated differentials
from the separated sound sources putting out the same signal. When you
have
a broad ensemble of acoustic sources, you end up somewhere in between.

But why would one be doing stereo with sound reinforcement?
The few times I've tried it the variation due to position
in the audience has made it useless unless you are really
close to center and even then it just doesn't come out right.

"Close to the center" indeed, within a few mm of exactly dead center so each
speaker is equidistant. Even with a mono mix, I've gotten vivid stereo
results with a Rode NT4 in a rear position, widening the crap out of it in
post. Widening such a narrow field of direct sound to offer some off-center
detail from stage bleed, the room ambience is wrapped all around. What you
end up with is the best of both worlds - the PA detail of a rear position
and the impact and imaging of the front row. You really have to nail the
mic position though, and any time differentials from mic spacing are
unacceptable, even the dual card's in the typical SDC fig-8. The slightest
inequity in the front/back response of a single diaphram fig-8 would be
grossly evident as well.

For that particular purpose the NT4 is the go-to mic. For loud shows you
can pop a 9V in the chassis and use the included minijack cable directly
into the line input of a NJB3, and for ~$600 total the results are
freakishly good.

When a
sound reinforcement system is involved, coincident mics are imperative,
since spacing the mics generates very unnatural correlated differentials
from the separated sound sources putting out the same signal. When you
have
a broad ensemble of acoustic sources, you end up somewhere in between.

But why would one be doing stereo with sound reinforcement?
The few times I've tried it the variation due to position
in the audience has made it useless unless you are really
close to center and even then it just doesn't come out right.

"Close to the center" indeed, within a few mm of exactly dead center so each
speaker is equidistant. Even with a mono mix, I've gotten vivid stereo
results with a Rode NT4 in a rear position, widening the crap out of it in
post. Widening such a narrow field of direct sound to offer some off-center
detail from stage bleed, the room ambience is wrapped all around. What you
end up with is the best of both worlds - the PA detail of a rear position
and the impact and imaging of the front row. You really have to nail the
mic position though, and any time differentials from mic spacing are
unacceptable, even the dual card's in the typical SDC fig-8. The slightest
inequity in the front/back response of a single diaphram fig-8 would be
grossly evident as well.

For that particular purpose the NT4 is the go-to mic. For loud shows you
can pop a 9V in the chassis and use the included minijack cable directly
into the line input of a NJB3, and for ~$600 total the results are
freakishly good.

David Satz wrote:

If the M microphone emphasizes certain sound components (e.g. the high
frequencies) while the S microphone doesn't, then in the eventual L/R
stereo output of the matrix, those components will tend to gather at
the image's center more than the midrange sound does. The apparent
left-right position of any given instrument or voice would thus be
blurred. Conversely, if the S microphone boosts the highs while the
M microphone doesn't, high frequency energy would tend to be shifted
more toward the left and right than the midrange sound--increasingly
so with individual voices or instruments that are already off-center.

Yes. But even if they are matched blurring will occur
because frequency response is angle dependant or inversely,
the polar plot is frequency dependant. The consequence is
that the virtual mics given by the vector sum of the real
mics have axes that move about with frequency.

It will take me another little while to say where I don't agree; the
main issue is the fact that most microphones (except for the best
small-diaphragm figure-8s and certain rather noisy omnis) don't have
the same frequency response at all angles of incidence. So all our
"virtual microphones" have rather complex, irregular response--that's
another reason I tend to favor a super- or hypercardioid for "M".

I hadn't read ahead. You clearly understand the point I
made above.


However, I'll also point out that the analog of using a non-flat S mike
is processing the S channel independently of the M channel--and there is
a long tradition of doing exactly that. One of the attractions of M/S
(whether the recording is made originally that way, or whether another
type of X/Y recording is matrixed to M/S) is that the M and S channels
can be equalized independently or processed in other ways independently,
then rematrixed to L/R stereo.

You want to make sure that any such manipulation that is
frequency dependant be done with linear phase filters and
that the channel not processed always be delayed by the
group delay of the filter to keep the two channels time
coherent. Changing the relative phase of M and S outputs
also has a large effect on source placement within the image.

All of this having to do with frequency dependant polar
responses and the consequence on imaging applies as well to
the XY configuration. At one extreme, look at the
frequencies where the polar plot gets close to round and all
those frequencies will be moved to the center of the image.


Bob
--

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

A. Einstein

David Satz wrote:

If the M microphone emphasizes certain sound components (e.g. the high
frequencies) while the S microphone doesn't, then in the eventual L/R
stereo output of the matrix, those components will tend to gather at
the image's center more than the midrange sound does. The apparent
left-right position of any given instrument or voice would thus be
blurred. Conversely, if the S microphone boosts the highs while the
M microphone doesn't, high frequency energy would tend to be shifted
more toward the left and right than the midrange sound--increasingly
so with individual voices or instruments that are already off-center.

Yes. But even if they are matched blurring will occur
because frequency response is angle dependant or inversely,
the polar plot is frequency dependant. The consequence is
that the virtual mics given by the vector sum of the real
mics have axes that move about with frequency.

It will take me another little while to say where I don't agree; the
main issue is the fact that most microphones (except for the best
small-diaphragm figure-8s and certain rather noisy omnis) don't have
the same frequency response at all angles of incidence. So all our
"virtual microphones" have rather complex, irregular response--that's
another reason I tend to favor a super- or hypercardioid for "M".

I hadn't read ahead. You clearly understand the point I
made above.


However, I'll also point out that the analog of using a non-flat S mike
is processing the S channel independently of the M channel--and there is
a long tradition of doing exactly that. One of the attractions of M/S
(whether the recording is made originally that way, or whether another
type of X/Y recording is matrixed to M/S) is that the M and S channels
can be equalized independently or processed in other ways independently,
then rematrixed to L/R stereo.

You want to make sure that any such manipulation that is
frequency dependant be done with linear phase filters and
that the channel not processed always be delayed by the
group delay of the filter to keep the two channels time
coherent. Changing the relative phase of M and S outputs
also has a large effect on source placement within the image.

All of this having to do with frequency dependant polar
responses and the consequence on imaging applies as well to
the XY configuration. At one extreme, look at the
frequencies where the polar plot gets close to round and all
those frequencies will be moved to the center of the image.


Bob
--

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

A. Einstein

locosoundman wrote:
Great thread - thanks for all of the information everyone.

In general I go with ORTF or some near-coincident variant therof but I
like to use spaced-pair techniques on large ensembles when the room
permits, or if I am using two stereo pairs in conjunction - one to
accent something like a choir, in which case I use a coincident pair
on the accent to get a strong center and clear localisation to better
blend with the diffuse image of the spaced pair - usually a nice
effect. I prefer to use coicindent methods simply for the effect of a
narrower image with a strong center, to be able to control the width
of the image later without phase problems, or for sound that will end
up on video.

Sounds like a great idea to me. As I suggested in another
response, the best thing with stereo techniques is to try
things 'til you find configurations you like. The theory
doesn't go very far toward explaining what really happens.

I have heard much about the supposed advantage of MS over XY because
of the ability to control stereo width after the fact, but I have
never used it due to the complications of rigging it.

There is really very little difference between XY and MS
except for the fact that the stuff usually of greater
interest is on the axis of the M in MS. OTOH, in the plots
I've looked at, the frequency response of a mic at +/- 45
degrees, the most common XY angle, is not very different
from on axis. You generally have to go further around
toward the back to begin to see signifigant variations.

In fact, XY can be converted to MS after the fact for the
same manipulations. Just use the DAW function that
transforms from LR to MS. The result can be remixed when
going back to LR to widen or narrow the image.

I've not played at all with MS manipulations other than
remixing at different ratios going to LR but as I responded
to David, if you are doing any frequency response
manipulation of just the M or just the S, use a linear phase
filter (equalizer) and be sure to delay the other side by
the same amount as the group delay of the linear phase filter.


Bob
--

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

A. Einstein

locosoundman wrote:
Great thread - thanks for all of the information everyone.

In general I go with ORTF or some near-coincident variant therof but I
like to use spaced-pair techniques on large ensembles when the room
permits, or if I am using two stereo pairs in conjunction - one to
accent something like a choir, in which case I use a coincident pair
on the accent to get a strong center and clear localisation to better
blend with the diffuse image of the spaced pair - usually a nice
effect. I prefer to use coicindent methods simply for the effect of a
narrower image with a strong center, to be able to control the width
of the image later without phase problems, or for sound that will end
up on video.

Sounds like a great idea to me. As I suggested in another
response, the best thing with stereo techniques is to try
things 'til you find configurations you like. The theory
doesn't go very far toward explaining what really happens.

I have heard much about the supposed advantage of MS over XY because
of the ability to control stereo width after the fact, but I have
never used it due to the complications of rigging it.

There is really very little difference between XY and MS
except for the fact that the stuff usually of greater
interest is on the axis of the M in MS. OTOH, in the plots
I've looked at, the frequency response of a mic at +/- 45
degrees, the most common XY angle, is not very different
from on axis. You generally have to go further around
toward the back to begin to see signifigant variations.

In fact, XY can be converted to MS after the fact for the
same manipulations. Just use the DAW function that
transforms from LR to MS. The result can be remixed when
going back to LR to widen or narrow the image.

I've not played at all with MS manipulations other than
remixing at different ratios going to LR but as I responded
to David, if you are doing any frequency response
manipulation of just the M or just the S, use a linear phase
filter (equalizer) and be sure to delay the other side by
the same amount as the group delay of the linear phase filter.


Bob
--

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

A. Einstein

In article ,
Bob Cain wrote:

But why would one be doing stereo with sound reinforcement?
The few times I've tried it the variation due to position
in the audience has made it useless unless you are really
close to center and even then it just doesn't come out right.


Bob

I don't think that the poster was referring to stereo sound
reinforcement, but recording an event that had sound reinforcement as a
variable, as opposed to a straight acoustic event.


Edwin

In article ,
Bob Cain wrote:

But why would one be doing stereo with sound reinforcement?
The few times I've tried it the variation due to position
in the audience has made it useless unless you are really
close to center and even then it just doesn't come out right.


Bob

I don't think that the poster was referring to stereo sound
reinforcement, but recording an event that had sound reinforcement as a
variable, as opposed to a straight acoustic event.


Edwin

David Satz wrote:

It's significant that these are not only small-diapragm designs but also
_single-diaphragm_ designs. And let's not forget the Neumann KM 120:

http://www.neumann.com/infopool/mics...p?ProdID=km100

Oddly I've never heard from anyone who has ever used it--and I've
asked around repeatedly. Here goes again: If anyone here has used
this microphone, could you please comment on its sound? I sometimes
use spaced (!) figure-8s for stereo vocal recording, and can imagine
that a response characteristic as shown in Neumann's diagrams might
be useful for that application, or at least worth trying.

Since you asked: I use a Neumann KM120 with a KM140. It sounds good to
me. But then I haven't had occasion to critically compare it with
anything else in its class.

David Satz wrote:

It's significant that these are not only small-diapragm designs but also
_single-diaphragm_ designs. And let's not forget the Neumann KM 120:

http://www.neumann.com/infopool/mics...p?ProdID=km100

Oddly I've never heard from anyone who has ever used it--and I've
asked around repeatedly. Here goes again: If anyone here has used
this microphone, could you please comment on its sound? I sometimes
use spaced (!) figure-8s for stereo vocal recording, and can imagine
that a response characteristic as shown in Neumann's diagrams might
be useful for that application, or at least worth trying.

Since you asked: I use a Neumann KM120 with a KM140. It sounds good to
me. But then I haven't had occasion to critically compare it with
anything else in its class.

Edwin Hurwitz wrote:


I don't think that the poster was referring to stereo sound
reinforcement, but recording an event that had sound reinforcement as a
variable, as opposed to a straight acoustic event.

Thank you, Edwin. Confusion resolved. I was baffled by the
responses to my question and now I know why. :-)

I've been to a few large-hall surround supported shows put
on by some local Creative Labs folks who are into that and
avante garde arts and I've tried to get binaural recordings
of them. I was utterly frustrated by the extent to which
_everything_ changes with the slightest head motion.

It's not nearly so bad if the reinforcement is mono but
still, the slightest motion from left to right can
dramatically pull the apparent source to one side or the
other. I've about given up trying to get anything live in
stereo that isn't up close and cozy.


Bob
--

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

A. Einstein

Edwin Hurwitz wrote:


I don't think that the poster was referring to stereo sound
reinforcement, but recording an event that had sound reinforcement as a
variable, as opposed to a straight acoustic event.

Thank you, Edwin. Confusion resolved. I was baffled by the
responses to my question and now I know why. :-)

I've been to a few large-hall surround supported shows put
on by some local Creative Labs folks who are into that and
avante garde arts and I've tried to get binaural recordings
of them. I was utterly frustrated by the extent to which
_everything_ changes with the slightest head motion.

It's not nearly so bad if the reinforcement is mono but
still, the slightest motion from left to right can
dramatically pull the apparent source to one side or the
other. I've about given up trying to get anything live in
stereo that isn't up close and cozy.


Bob
--

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

A. Einstein

Sugarite wrote:

This is leading towards a debate over whether two matched cardioids
are in fact a truer match than either side of a fig-8. I'm of the
opinion that the matched card's are a truer match, mostly because you
can't as easily test the internal bipolar matching on a mic-by-mic
basis like matching two card's.

May I take a closer look at this rather mysterious course of logic? If
I didn't know better, I'd think you were saying that the ease of testing
which you imagine in a thought experiment (!) should take precedence over
the accuracy of competent real-world measurements.

But even granting that, I think your logical conclusion would be: "If I
had to document the front/back symmetry myself, I'd prefer two cardioids.
Then I'd have my own way to make the measurements, so I would feel more
comfortable trusting the results." Human to human, I could respect that.

Still, that isn't quite the same as the broad claim that "cardioids match
better than the front and back of a figure 8" (which is what you said).
The two statements are about different things--the first one is about
your own mindset, while the other one is about microphone capsules, no?


It's impossible to test both sides of a fig-8 with that level of
precision. All you can do is "read curves".

If I didn't know better, I'd think you were claiming to know the level
of precision of curves which you've never seen and don't know a thing
about. Also, in this case I can do a bit more than "read curves," since
the ones I described were those of a pair of capsules which I used for
nearly 25 years; I made several hundred live recordings with them. And
I've had even longer experience with frequency response curves from that
particular measurement chamber.

In 1974 I even did some blind testing--I was sent a pair of test capsules
with no indication of their characteristics. I tried them on a recording
and reported back that I had heard a response elevation of 2 dB between
a certain pair of upper midrange frequencies. I was then allowed to see
the response plots; I had estimated them exactly. I don't claim that I
could do that every time, of course. But I feel I do have a real luxury,
which is that I can indeed "read" these curves and, in relation to them,
I also know from long experience what the results sound like.

Given that situation I don't share the dismissive attitude that some
people have toward response plots--I find that they can often tell me
quite a bit about a microphone.

As you may know, ten of the manufacturers in the AES Working Group on
microphone specifications have recently completed a "round robin" test
in which each of them submitted one cardioid microphone which they and
all the other manufacturers measured, so that they could compare notes
on how they measure microphone response and how they present the results.
A paper was presented at the recent Tonmeistertagung in Germany which
described some of the results of this "round robin" test (I was the
translator for that) and I hear that there was also some discussion or
perhaps a paper about this at the most recent AES convention as well.

--best regards

Sugarite wrote:

This is leading towards a debate over whether two matched cardioids
are in fact a truer match than either side of a fig-8. I'm of the
opinion that the matched card's are a truer match, mostly because you
can't as easily test the internal bipolar matching on a mic-by-mic
basis like matching two card's.

May I take a closer look at this rather mysterious course of logic? If
I didn't know better, I'd think you were saying that the ease of testing
which you imagine in a thought experiment (!) should take precedence over
the accuracy of competent real-world measurements.

But even granting that, I think your logical conclusion would be: "If I
had to document the front/back symmetry myself, I'd prefer two cardioids.
Then I'd have my own way to make the measurements, so I would feel more
comfortable trusting the results." Human to human, I could respect that.

Still, that isn't quite the same as the broad claim that "cardioids match
better than the front and back of a figure 8" (which is what you said).
The two statements are about different things--the first one is about
your own mindset, while the other one is about microphone capsules, no?


It's impossible to test both sides of a fig-8 with that level of
precision. All you can do is "read curves".

If I didn't know better, I'd think you were claiming to know the level
of precision of curves which you've never seen and don't know a thing
about. Also, in this case I can do a bit more than "read curves," since
the ones I described were those of a pair of capsules which I used for
nearly 25 years; I made several hundred live recordings with them. And
I've had even longer experience with frequency response curves from that
particular measurement chamber.

In 1974 I even did some blind testing--I was sent a pair of test capsules
with no indication of their characteristics. I tried them on a recording
and reported back that I had heard a response elevation of 2 dB between
a certain pair of upper midrange frequencies. I was then allowed to see
the response plots; I had estimated them exactly. I don't claim that I
could do that every time, of course. But I feel I do have a real luxury,
which is that I can indeed "read" these curves and, in relation to them,
I also know from long experience what the results sound like.

Given that situation I don't share the dismissive attitude that some
people have toward response plots--I find that they can often tell me
quite a bit about a microphone.

As you may know, ten of the manufacturers in the AES Working Group on
microphone specifications have recently completed a "round robin" test
in which each of them submitted one cardioid microphone which they and
all the other manufacturers measured, so that they could compare notes
on how they measure microphone response and how they present the results.
A paper was presented at the recent Tonmeistertagung in Germany which
described some of the results of this "round robin" test (I was the
translator for that) and I hear that there was also some discussion or
perhaps a paper about this at the most recent AES convention as well.

--best regards

This is leading towards a debate over whether two matched cardioids
are in fact a truer match than either side of a fig-8. I'm of the
opinion that the matched card's are a truer match, mostly because you
can't as easily test the internal bipolar matching on a mic-by-mic
basis like matching two card's.

May I take a closer look at this rather mysterious course of logic? If
I didn't know better, I'd think you were saying that the ease of testing
which you imagine in a thought experiment (!) should take precedence over
the accuracy of competent real-world measurements.

[snip]

Reversing cardioid signals is hardly experimentation. Two card's can be
trusted as a 100% match when they produce no perceivable signal when one is
inverted against the other. That catches aspects that aren't even
understood, much less measureable. That is a far more reliable test than
any that can be applied to a fig-8, except if you had a matched pair of fig
8's that stay matched if you turn one around, something I've never heard of
happening. If you reverse one or both of the fig-8's in Blumlein format
you'll get different imaging every time. The dependence on a balanced fig-8
in MS format is equally severe.

Sugarite wrote:

Reversing cardioid signals is hardly experimentation. Two card's can be
trusted as a 100% match when they produce no perceivable signal when one is
inverted against the other. That catches aspects that aren't even
understood, much less measureable. That is a far more reliable test than
any that can be applied to a fig-8, except if you had a matched pair of fig
8's that stay matched if you turn one around, something I've never heard of
happening. If you reverse one or both of the fig-8's in Blumlein format
you'll get different imaging every time. The dependence on a balanced fig-8
in MS format is equally severe.

Since there is no way in the presently known universe to put two objects
in the same place at the same time, how can wavefronts that arrive at
different times null? Is this a fantasy test?

--
ha

hank alrich wrote:

Since there is no way in the presently known universe to put two objects
in the same place at the same time, how can wavefronts that arrive at
different times null? Is this a fantasy test?

Not at all, Hank. While they can't be at the same place,
they can be at the same distance by placing one atop the
other pointing in opposite directions with their diaphragms
in the same plane and with that plane passing through the
source.

Here's a pic of a MS employing three little Panasonic WM55A
cards that I made to experiment with:

http://www.arcanemethods.com/MS-Ball-Off.gif

Yes, things arriving from off the horizontal plane will have
some TOA difference so there is assymetry that way but for
the direct sound, it can be quite time coherent.

That there is a front to back distance between M and S on
axis is ameliorated by the M being in the null of the S
there. At 90 degrees the distance between M and S is the same.


Bob
--

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

A. Einstein

On Sat, 27 Nov 2004 03:19:54 -0500, Bob Cain wrote
(in article ):



hank alrich wrote:

Since there is no way in the presently known universe to put two objects
in the same place at the same time, how can wavefronts that arrive at
different times null? Is this a fantasy test?

Not at all, Hank. While they can't be at the same place,
they can be at the same distance by placing one atop the
other pointing in opposite directions with their diaphragms
in the same plane and with that plane passing through the
source.

Here's a pic of a MS employing three little Panasonic WM55A
cards that I made to experiment with:

http://www.arcanemethods.com/MS-Ball-Off.gif

Yes, things arriving from off the horizontal plane will have
some TOA difference so there is assymetry that way but for
the direct sound, it can be quite time coherent.

That there is a front to back distance between M and S on
axis is ameliorated by the M being in the null of the S
there. At 90 degrees the distance between M and S is the same.


Bob

Then there's the consideration of how the mic bodies might alter the results
because of their obstructional presence.

Regards,

Ty Ford



-- Ty Ford's equipment reviews, audio samples, rates and other audiocentric
stuff are at www.tyford.com

While they can't be at the same place,
they can be at the same distance by placing one atop the
other pointing in opposite directions with their diaphragms
in the same plane and with that plane passing through the
source. BRBR

Which is great for sound arriving from a single direction, i.e. an anechoic
situation. In any desirable recording venue, obviously the distance thing again
becomes slightly approximate. In the end, everything we do is slightly
approximate anyway, so what's a few millimeters this way or that?

Scott Fraser