MS vs XY Figure 8

[Gary Eickmeier]

I just made some test recordings of MS with figure 8 pattern vs Blumlein 90°
Figure 8, and it is close. Is the one mathematically equivalent to the
other? The XY was very spacious and very good on localization - better than
cardioids in XY coincident, probably because the cardioid pattern is a lot
fatter than the Fig 8 pattern.

Perhaps the main advantage of MS is the adjustability of the pattern in
post. I had shied away from it because of an uncertainty about that
processing, but I think I have cracked the code. I download the S and the M
channels, invert one copy of S, and place them on the multitrack timeline.
Track 1 is S, Track 2 is M, Track 3 is -S. I assign Track 1 to the left
channel, Track 2 to center, and Track 3 to the right channel. If the gains
are equal in all 3, I believe I get a 45/45 frontal pattern. But I can take
down the M channel and get greater separation if desired. I tried some test
tracks taking it down by 1 dB at a time, and the difference is audible. Even
at same levels for all 3 I am getting a full frontal stage width as wide as
my room. But by the time I get to M at -3 dB it is more like a 180° arc in
front of me - a bit too much, even for me!

The XY Fig 8 at 90° pattern is also very wide and very pinpoint - very much
like the MS with all levels equal. For the next concert I could see it
either way. Both of these patterns, or techniques, are recording all around,
right? The concert will have audience participation, so I wouldn't mind
hearing them as well as the band. Wish I knew how to encode them so that
they image to the rear in Dolby Pro Logic II.

Any ideas?

Gary Eickmeier

[William Sommerwerck]

"Gary Eickmeier" wrote in message ...

> I just made some test recordings of MS with figure 8 pattern
> vs Blumlein 90° figure 8, and it is close. Is the one mathematically
> equivalent to the other?

I think you'll find they are, though MS gives room for post-recording
manipulation.

Didn't you take trig in school?

[Gary Eickmeier]

"William Sommerwerck" > wrote in message
...
> "Gary Eickmeier" wrote in message
> ...
>
>> I just made some test recordings of MS with figure 8 pattern
>> vs Blumlein 90° figure 8, and it is close. Is the one mathematically
>> equivalent to the other?
>
> I think you'll find they are, though MS gives room for post-recording
> manipulation.
>
> Didn't you take trig in school?

Wow, there's a leap! But anyway, Mike Rivers sent me a nice Email about this
well-known subject. If you take a coincident XY recording and extract X+Y
and X-Y from it, you have yersself an MS recording. Now invert X-Y or just
do Y-X and you have the -S part and you can then remix these for different
ratios of M and S to vary the width of the soundstage. Nice to know, but I
would rather decide beforehand whether I want MS or XY. I sincerely think I
will do MS

But hey - how do you do X-Y? Invert Y and then add them by exporting to a
single track?

Gary Eickmeier

[Mike Rivers[_2_]]

On 12/3/2013 11:06 PM, Gary Eickmeier wrote:

> I would rather decide beforehand whether I want MS or XY. I sincerely think I
> will do MS

When you get really experienced at single point stereo recording you'll
develop a sense of which method (including spaced mics - ORTF or wide
spaced) will work best. It's a combination of the size of the source
you're recording, the acoustics of the room, and what you want the
recording to sound like.

But you're just getting started at this. Just pick one or the other and
go for it. There's not clear right or wrong, just different. Set up your
mics and listen to what you're getting. If you have more than one
opportunity to record the same group in the same location, or even
another group in the same location, try another setup on another
session. Listen to what you're getting and this may help you to decide
what's best for that room or that group.

> But hey - how do you do X-Y? Invert Y and then add them by exporting to a
> single track?

Huh? If you want mono from an X-Y mic setup, just add the two channels.
If you want stereo, record one mic to one channel and the other mic to
the other channel. The one pointing to the left goes to the left
channel, the one pointing to the right goes to the right channel.

If you want to play with M-S processing of a stereo track, most DAWs
have a button that splits a stereo track into two mono tracks, so you
have one for left and one for right. You can then add and subtract those
to get your mid and side tracks, play with them, and then put them back
to a single stereo left/right track.




--
For a good time, visit http://mikeriversaudio.wordpress.com

[Tom McCreadie]

Gary Eickmeier wrote:
>
>Perhaps the main advantage of MS is the adjustability of the pattern in
>post. I had shied away from it because of an uncertainty about that
>processing, but I think I have cracked the code.

MS at 1:1 and XY at 90° are mathematically equiivalent when the Fig8's have
ideal patterns, with perfect off-axis response etc..

XY is also adjustable in post, for you just need to run things _twice_ through
the matrix:
XY -> MS -> M'S' -> X'Y'

MS, though, does have that extra degree of flexibility in that one can select a
wider range of patterns for the M mic: omni, diffuse field omni, subcardiod,
widish cardiod...etc. - and also a mic from other manufacturing brands. This
allows access some virtual mic patterns and angles that are denied to the XY
route.

But why still hanging onto this this convoluted decoding procedure?
Just save your MS tracks as a stereo file, then in Edit mode of Audition 2 or 3,
either:

a) Use the free Voxengo MSED plugin.
or
b) Menu > Effects > Stereo Imagery > Channel Mixer
New Left Channel = +50L, +50R
New Right Channel = +50L, -50R
(this example. for 1:1, but adjustable to taste)
__
Tom McCreadie

"Ah, where would we be without humour." "Germany?"

[Scott Dorsey]

Gary Eickmeier > wrote:
>I just made some test recordings of MS with figure 8 pattern vs Blumlein 90°
>Figure 8, and it is close. Is the one mathematically equivalent to the
>other?

Yes. And because the patterns on most figure-8 microphones are pretty
good, the differences between them are less than they would be with most
other matrix vs. non-matrix coincident arrangement comparisons.

>The XY was very spacious and very good on localization - better than
>cardioids in XY coincident, probably because the cardioid pattern is a lot
>fatter than the Fig 8 pattern.

That's something that you can alter with positioning. What you can't
alter is the degree to which the directivity changes with frequency.

>Perhaps the main advantage of MS is the adjustability of the pattern in
>post. I had shied away from it because of an uncertainty about that
>processing, but I think I have cracked the code. I download the S and the M
>channels, invert one copy of S, and place them on the multitrack timeline.
>Track 1 is S, Track 2 is M, Track 3 is -S. I assign Track 1 to the left
>channel, Track 2 to center, and Track 3 to the right channel. If the gains
>are equal in all 3, I believe I get a 45/45 frontal pattern. But I can take
>down the M channel and get greater separation if desired. I tried some test
>tracks taking it down by 1 dB at a time, and the difference is audible. Even
>at same levels for all 3 I am getting a full frontal stage width as wide as
>my room. But by the time I get to M at -3 dB it is more like a 180° arc in
>front of me - a bit too much, even for me!

You can matrix an X-Y recording after the fact and make the same changes
just as easily.

>The XY Fig 8 at 90° pattern is also very wide and very pinpoint - very much
>like the MS with all levels equal. For the next concert I could see it
>either way. Both of these patterns, or techniques, are recording all around,
>right? The concert will have audience participation, so I wouldn't mind
>hearing them as well as the band. Wish I knew how to encode them so that
>they image to the rear in Dolby Pro Logic II.

The problem here is that the actual stereo field is narrow.... you have to
be a good distance away from the band or most of them falls off the edge of
the stereo field because of the narrow angle of acceptance. But in some rooms
that's a huge advantage.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Tom McCreadie]

Jeff Henig wrote:

>> "Ah, where would we be without humour." "Germany?"
>
>Now THAT'S funny.
>
>What movie is that from?

I heard it as a quip from the comedian Jimmy Carr, a couple of weeks ago on the
BBC's "The Graham Norton Show".
--
Tom McCreadie

Live at The London Palindrome - ABBA

[Tom McCreadie]

Gary Eickmeier wrote:

>....... but I think I have cracked the code. I download the S and the M
>channels, invert one copy of S, and place them on the multitrack timeline.
>Track 1 is S, Track 2 is M, Track 3 is -S. I assign Track 1 to the left
>channel, Track 2 to center, and Track 3 to the right channel. If the gains
>are equal in all 3, I believe I get a 45/45 frontal pattern. But I can take
>down the M channel and get greater separation if desired. I tried some test
>tracks taking it down by 1 dB at a time, and the difference is audible. Even
>at same levels for all 3 I am getting a full frontal stage width as wide as
>my room. But by the time I get to M at -3 dB it is more like a 180° arc in
>front of me - a bit too much, even for me!

You may have overlooked the "Panning Law" setting of -3dB that is typically set
in Audition and other DAWs (Menu > Edit > Settings > Multitrack > Stereo Panning
Mode = -3dB)

So this could result in you actually having processed the M at -3dB w.r.t. S,
when your intention was 0dB (1:1), thus getting a puzzling premature widening in
soundstage changes. :-)

Some theoretical values for Fig8 MS => Fig8 XY:
S at 0dB => XY incl. angle 90°, SRA 76° (Sengpiel data)
S at -1dB => XY incl. angle 97°, SRA 69°
S at -2dB => XY incl. angle 103°, SRA 63°
S at -3dB => XY incl. angle 109°, SRA 58°
S at -4dB => XY incl. angle 116°, SRA 53°
S at -5dB => XY incl. angle 121°, SRA 47°
S at -6dB => XY incl. angle 127°, SRA 43°

Sources beyond the SRA soon fall into a large so-called ambiophonic sector -
think of "that as the wild 'n woozy West (and East) :-) - where the L and R
channel voltage values are not of the same algebraic sign. Further to the array
rear is a sector of again sharp focus, and its width identical to the front SRA
sector, but now with the both L & R signals of inverted polarity...and more
significantly, now with a lateral inversion of imaging (i.e. a sound source in
the audience at position "5 o'clock" will image in the left half of your two
channel stereo playback, as if were coming from 7 o'clock).

Apologies if all the foregoing is familiar to you. I'm afraid I can't contribute
much to your multispeaker, surround sound queries.
--
Tom McCreadie

Sudden Death Syndrome? Sounds serious - what are the symptoms?"

[John Williamson]

On 04/12/2013 18:26, Tom McCreadie wrote:
> Jeff Henig wrote:
>
>>> "Ah, where would we be without humour." "Germany?"
>>
>> Now THAT'S funny.
>>
>> What movie is that from?
>
> I heard it as a quip from the comedian Jimmy Carr, a couple of weeks ago on the
> BBC's "The Graham Norton Show".
>
And unattributed as I don't remember the name of the Austrian girl that
told me:-

"What do you call a German with no sense of humour?" "An Austrian."

--
Tciao for Now!

John.

[Peter Larsen[_3_]]

Tom McCreadie wrote:

> Gary Eickmeier wrote:

>> ....... but I think I have cracked the code.

Use the channel mixer, it used to have a preset for stereo 2 ms and the
other way in 1.5 that it inherited from the ancestor Cool Edit, then
incompetence struck @ the adobe programming farm and it got removed from the
x-works settings. Yell if you want me to look it up.

Kind regards

Peter Larsen

[Tom McCreadie]

Peter Larsen wrote:

>
>Use the channel mixer, it used to have a preset for stereo 2 ms and the
>other way in 1.5 that it inherited from the ancestor Cool Edit, then
>incompetence struck @ the adobe programming farm and it got removed from the
>x-works settings. Yell if you want me to look it up.

Hi Peter, I'd recommended the Channel Mixer to Gary earlier in the thread.

Th Channel Mixer of AA-2 and AA-3 didn't come with a supplied preset for MS to
XY - but it's trivial to create one's own preset. CEP-2.1 and AA-1 did have
one, and it was restored in AA-CS6.
Curiously, the latter three adopted differing normalization scaling:
(L = +100%M/+100%S, R = +100%M/-100%S) vs (+50/+50, +50/-50) vs (+70.1/+70.1,
+70.1, /-70.1), respectively.

[Peter Larsen[_3_]]

Tom McCreadie wrote:

> Peter Larsen wrote:

>> Use the channel mixer, it used to have a preset for stereo 2 ms and
>> the other way in 1.5 that it inherited from the ancestor Cool Edit,
>> then incompetence struck @ the adobe programming farm and it got
>> removed from the x-works settings. Yell if you want me to look it up.

> Hi Peter, I'd recommended the Channel Mixer to Gary earlier in the
> thread.

Sorry, overlooked it.

> Th Channel Mixer of AA-2 and AA-3 didn't come with a supplied preset
> for MS to XY - but it's trivial to create one's own preset. CEP-2.1
> and AA-1 did have one, and it was restored in AA-CS6.

> Curiously, the latter three adopted differing normalization scaling:
> (L = +100%M/+100%S, R = +100%M/-100%S) vs (+50/+50, +50/-50) vs
> (+70.1/+70.1, +70.1, /-70.1), respectively.

AA 1.5 does LR to MS at

New left channel "50/50" and New right channel "50/-50"

and MS to LR at

New left channel "100/100" and New right channel "100/-100".

Which happens to look completely correct on the screen.

Have they implemented easy settings export and import and the ability to
load settings from floppy/USB - required for easy backup, easy project
transfer, easy system replacement and easy working on other computers so
that you only bring the settings and not the entire computer if working
elsewhere.

Kind regards

Peter Larsen

[Gary Eickmeier]

"Mike Rivers" > wrote in message
...
> On 12/3/2013 11:06 PM, Gary Eickmeier wrote:


> When you get really experienced at single point stereo recording you'll
> develop a sense of which method (including spaced mics - ORTF or wide
> spaced) will work best. It's a combination of the size of the source
> you're recording, the acoustics of the room, and what you want the
> recording to sound like.
>
> But you're just getting started at this. Just pick one or the other and go
> for it. There's not clear right or wrong, just different. Set up your mics
> and listen to what you're getting. If you have more than one opportunity
> to record the same group in the same location, or even another group in
> the same location, try another setup on another session. Listen to what
> you're getting and this may help you to decide what's best for that room
> or that group.

Agreed!

>
>> But hey - how do you do X-Y? Invert Y and then add them by exporting to a
>> single track?
>
> Huh? If you want mono from an X-Y mic setup, just add the two channels. If
> you want stereo, record one mic to one channel and the other mic to the
> other channel. The one pointing to the left goes to the left channel, the
> one pointing to the right goes to the right channel.

A misread Mike. The question was how do you do X minus Y.

> If you want to play with M-S processing of a stereo track, most DAWs have
> a button that splits a stereo track into two mono tracks, so you have one
> for left and one for right. You can then add and subtract those to get
> your mid and side tracks, play with them, and then put them back to a
> single stereo left/right track.

Yes, I got that. Starting with a Blumlein XY coincident figure 8 recording,
take X+Y for the M channel and X-Y for the S channel. Remix as an MS
recording. Which brings us back to the above question. But I must be correct
that to subtract Y from X you just invert the polarity of Y and then add
them. Right?

Gary Eickmeier

[Gary Eickmeier]

"Scott Dorsey" > wrote in message
...
> Gary Eickmeier > wrote:

>>The XY Fig 8 at 90° pattern is also very wide and very pinpoint - very
>>much
>>like the MS with all levels equal. For the next concert I could see it
>>either way. Both of these patterns, or techniques, are recording all
>>around,
>>right? The concert will have audience participation, so I wouldn't mind
>>hearing them as well as the band. Wish I knew how to encode them so that
>>they image to the rear in Dolby Pro Logic II.
>
> The problem here is that the actual stereo field is narrow.... you have to
> be a good distance away from the band or most of them falls off the edge
> of
> the stereo field because of the narrow angle of acceptance. But in some
> rooms
> that's a huge advantage.
> --scott

Yes, agreed, but I am thinking of the backside of the Fig 8 pattern picking
up the audience just as loud as the frontside picks up the orchestra. Can
the back half of the pattern be somehow decoded with some matrix and placed
in the rear channels of some system such as DD or DPL II?

Gary Eickmeier

[Mike Rivers[_2_]]

On 12/5/2013 12:40 AM, Gary Eickmeier wrote:
> I must be correct
> that to subtract Y from X you just invert the polarity of Y and then add
> them. Right?

Right. Same way you subtract 2 from 4. The way you accomplish this in a
DAW is to invert the polarity of the right track and mix left and
inverted right tracks to a new mono track which will be left minus right.

--
For a good time, visit http://mikeriversaudio.wordpress.com

[Tom McCreadie]

Gary Eickmeier wrote:

>Yes, agreed, but I am thinking of the backside of the Fig 8 pattern picking
>up the audience just as loud as the frontside picks up the orchestra. Can
>the back half of the pattern be somehow decoded with some matrix and placed
>in the rear channels of some system such as DD or DPL II?

It's alas mathematically impossible to split out the + lobe and the - lobe of a
fig-8 into two separate entities, for further independent manipulation.
--
Tom McCreadie

Tinnitus is a pain in the neck

[Scott Dorsey]

Gary Eickmeier > wrote:
>
>Yes, agreed, but I am thinking of the backside of the Fig 8 pattern picking
>up the audience just as loud as the frontside picks up the orchestra. Can
>the back half of the pattern be somehow decoded with some matrix and placed
>in the rear channels of some system such as DD or DPL II?

No, because the two rear lobes have the same polarity.

In order to image to the rear with Dolby Stereo (which is NOT Dolby Digital
OR Pro-Logic), a signal needs to be out of phase in the front channels.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Tom McCreadie]

>It's alas mathematically impossible to split out the + lobe and the - lobe of a
>fig-8 into two separate entities, for further independent manipulation.

Edit: Of course, if that fig8 was not a 'real fig8' but had been synthesized
from a back-to-back arrangement of two cardioid mics (or any unipolar mics),
you would probably have also recorded those individual cardioid tracks...so you
are not hen so straightjacketed with your lobe juggling experimentation.

[Gary Eickmeier]

"Tom McCreadie" > wrote in message
...
> >It's alas mathematically impossible to split out the + lobe and the -
> >lobe of a
>>fig-8 into two separate entities, for further independent manipulation.
>
> Edit: Of course, if that fig8 was not a 'real fig8' but had been
> synthesized
> from a back-to-back arrangement of two cardioid mics (or any unipolar
> mics),
> you would probably have also recorded those individual cardioid
> tracks...so you
> are not hen so straightjacketed with your lobe juggling experimentation.
>

Well, but, of course if you just turn the whole scene around 180° then you
could envision this as an MS recording of the audience. No?

Gary

[Gary Eickmeier]

"Scott Dorsey" > wrote in message
...
> Gary Eickmeier > wrote:
>>
>>Yes, agreed, but I am thinking of the backside of the Fig 8 pattern
>>picking
>>up the audience just as loud as the frontside picks up the orchestra. Can
>>the back half of the pattern be somehow decoded with some matrix and
>>placed
>>in the rear channels of some system such as DD or DPL II?
>
> No, because the two rear lobes have the same polarity.
>
> In order to image to the rear with Dolby Stereo (which is NOT Dolby
> Digital
> OR Pro-Logic), a signal needs to be out of phase in the front channels.
> --scott

If you invert the polarity of M, then you are miking the audience with M.
Then you can handle S as always. However, if you try to mix this in with the
front pattern, something just might cancel out (like, if you put M and
invert of M in the same recording, you get a big fat ZERO).

Back to the drawing boards.

Gary Eickmeier

[Gary Eickmeier]

Ooh I guess I better study all this a little. But does mixing a channel in
at -50 mean inverted at 50%? Because there is also an "Invert" checkbox down
below in that screen.

Gary


"Peter Larsen" > wrote in message
...
> Tom McCreadie wrote:
>
>> Peter Larsen wrote:
>
>>> Use the channel mixer, it used to have a preset for stereo 2 ms and
>>> the other way in 1.5 that it inherited from the ancestor Cool Edit,
>>> then incompetence struck @ the adobe programming farm and it got
>>> removed from the x-works settings. Yell if you want me to look it up.
>
>> Hi Peter, I'd recommended the Channel Mixer to Gary earlier in the
>> thread.
>
> Sorry, overlooked it.
>
>> Th Channel Mixer of AA-2 and AA-3 didn't come with a supplied preset
>> for MS to XY - but it's trivial to create one's own preset. CEP-2.1
>> and AA-1 did have one, and it was restored in AA-CS6.
>
>> Curiously, the latter three adopted differing normalization scaling:
>> (L = +100%M/+100%S, R = +100%M/-100%S) vs (+50/+50, +50/-50) vs
>> (+70.1/+70.1, +70.1, /-70.1), respectively.
>
> AA 1.5 does LR to MS at
>
> New left channel "50/50" and New right channel "50/-50"
>
> and MS to LR at
>
> New left channel "100/100" and New right channel "100/-100".
>
> Which happens to look completely correct on the screen.
>
> Have they implemented easy settings export and import and the ability to
> load settings from floppy/USB - required for easy backup, easy project
> transfer, easy system replacement and easy working on other computers so
> that you only bring the settings and not the entire computer if working
> elsewhere.
>
> Kind regards
>
> Peter Larsen
>
>
>
>
>

[Tom McCreadie]

Gary Eickmeier wrote:

>> >It's alas mathematically impossible to split out the + lobe and the -
>> >lobe of a fig-8 into two separate entities, for further independent manipulation.
>>
>> Edit: Of course, if that fig8 was not a 'real fig8' but had been
>> synthesized from a back-to-back arrangement of two cardioid mics (or any unipolar
>> mics), you would probably have also recorded those individual cardioid
>> tracks...so you are not hen so straightjacketed with your lobe juggling experimentation.
>
>Well, but, of course if you just turn the whole scene around 180° then you
>could envision this as an MS recording of the audience. No?

We're still talking abiout MS with two Fig8's, right?

With two Fig8's you have complete symmetry...there's equal weighting to what's
going on in front of or behind the array. You record a 2-track composite of the
band (probably sounding stronger, closer and more direct) _ and_ the audience
(probably sounding fainter, more distant, more diffuse). Rotating the MS array
180° before doing the recording, or flipping the recording's channel polarities
in post, does nothing to alter or disentangle that already baked-in
relationship.

Ever managed to buy a hand mirror that, singly, lets you see the front and the
back of your head? :-)

[jason]

On Thu, 5 Dec 2013 06:30:54 +0100 "Peter Larsen" >
wrote in article >
>

>
> Have they implemented easy settings export and import and the ability to
> load settings from floppy/USB - required for easy backup, easy project
> transfer, easy system replacement and easy working on other computers so
> that you only bring the settings and not the entire computer if working
> elsewhere.
>
> Kind regards
>
> Peter Larsen

The File menu has Export & Import of application settings as XML files.

[Gary Eickmeier]

"Tom McCreadie" > wrote in message
...
> Gary Eickmeier wrote:

>>Well, but, of course if you just turn the whole scene around 180° then you
>>could envision this as an MS recording of the audience. No?
>
> We're still talking abiout MS with two Fig8's, right?
>
> With two Fig8's you have complete symmetry...there's equal weighting to
> what's
> going on in front of or behind the array. You record a 2-track composite
> of the
> band (probably sounding stronger, closer and more direct) _ and_ the
> audience
> (probably sounding fainter, more distant, more diffuse). Rotating the MS
> array
> 180° before doing the recording, or flipping the recording's channel
> polarities
> in post, does nothing to alter or disentangle that already baked-in
> relationship.

Not rotating the microphones 180°, rotating the whole scene. You would still
have two fig 8 mikes, and if you chose to mix them as audience R and L you
could. But I fear that you cannot do this in the same recording, or else M
will drop out and maybe even S. Got to think that one over.

> Ever managed to buy a hand mirror that, singly, lets you see the front and
> the
> back of your head? :-)
>

Not real pertinent. An omni or a fig 8 mike can record both front and back
at equal volumes. We just choose to mix MS for the frontal sound, not the
back half. But even so, wouldn't M record both front and rear at the same
volumes? It won't decode to the rear, but it is still hearing behind. It's
bidirectional.

I have a Sony single point stereo mike that they call MS and it seems to be
able to do surround sound in DPL. You swear you can hear the audience all
around.

Gary Eickmeier

[Peter Larsen[_3_]]

Jason wrote:

> The File menu has Export & Import of application settings as XML
> files.

Thank you Jason!

Kind regards

Peter Larsen

[Tom McCreadie]

Gary Eickmeier wrote:

>
>"Tom McCreadie" > wrote
>> With two Fig8's you have complete symmetry...there's equal weighting to
>> what's going on in front of or behind the array. You record a 2-track composite
>> of the band (probably sounding stronger, closer and more direct) _ and_ the
>> audience (probably sounding fainter, more distant, more diffuse). Rotating the MS
>> array 180° before doing the recording, or flipping the recording's channel
>> polarities in post, does nothing to alter or disentangle that already baked-in
>> relationship.
>
>Not rotating the microphones 180°, rotating the whole scene. You would still
>have two fig 8 mikes, and if you chose to mix them as audience R and L you
>could. But I fear that you cannot do this in the same recording, or else M
>will drop out and maybe even S. Got to think that one over.
>
>> Ever managed to buy a hand mirror that, singly, lets you see the front and
>> the back of your head? :-)
>
>Not real pertinent. An omni or a fig 8 mike can record both front and back
>at equal volumes. We just choose to mix MS for the frontal sound, not the
>back half. But even so, wouldn't M record both front and rear at the same
>volumes? It won't decode to the rear, but it is still hearing behind. It's
>bidirectional.
>
>I have a Sony single point stereo mike that they call MS and it seems to be
>able to do surround sound in DPL. You swear you can hear the audience all
>around.

Ach, my corny mirror quip was an inexact analogy, more confusing than
illuminating.

Yes of course, Fig8 records from both front and back with equal weighting. I'd
already said that. But my point was just that there's no way that you can
manipulate two channels of a 'real' Fig8 recording so that you end up with:
- two channels of "front-enriched" info (say, more emphasis on the band) going
to your front speakers
- two channels of "rear-enriched" info (less band, more audience and ambient)
going to your rear speakers.
(Even then, you still have to reconcile that issue of lateral inversion of the
rear images)

If you can pull all this off, you could get a gig helping mathematicians to
square circles and trisect angles. :-)

[Gary Eickmeier]

"Tom McCreadie" > wrote in message
...


> Yes of course, Fig8 records from both front and back with equal weighting.
> I'd
> already said that. But my point was just that there's no way that you can
> manipulate two channels of a 'real' Fig8 recording so that you end up
> with:
> - two channels of "front-enriched" info (say, more emphasis on the band)
> going
> to your front speakers
> - two channels of "rear-enriched" info (less band, more audience and
> ambient)
> going to your rear speakers.
> (Even then, you still have to reconcile that issue of lateral inversion of
> the
> rear images)
>
> If you can pull all this off, you could get a gig helping mathematicians
> to
> square circles and trisect angles. :-)

OK does this mean that in general, you don't want to do MS when it is a live
event with an audience? Maybe choose XY cardioid instead? In this event
there will be audience participation, like sing alongs, so I might choose
the MS anyway.

Mike Rivers said that after a while I will get a feel for when to use MS and
when to use XY or NOS and similar. I am thinking that some of the criteria
would be is it a really wide group such as a symphony or a smaller one such
as a Jazz group. We don't want a jazz trio to be 40 ft wide, so we would use
something other than MS.

Thaniks,
Gary

[Peter Larsen[_3_]]

Gary Eickmeier wrote:

> OK does this mean that in general, you don't want to do MS when it is
> a live event with an audience? Maybe choose XY cardioid instead? In
> this event there will be audience participation, like sing alongs, so
> I might choose the MS anyway.

There are many flavors of MS, but general for all of them is that they offer
optimum microphone frequency response for the center.

> Mike Rivers said that after a while I will get a feel for when to use
> MS and when to use XY or NOS and similar. I am thinking that some of
> the criteria would be is it a really wide group such as a symphony or
> a smaller one such as a Jazz group.

"Wide or small", not good words to use, think opening angle or rather angle
of inclusion and remember that for XY - coincident or spaced on a stereo
bar - the wider the angle of inclusion you need the less the microphones
should be angled apart if you want to avoid a glaring hole in the middle.
No, it is NOT contraintuitive, to avoid hole in the middle the mics need to
hear more of the other channel.

I have only used the ORTF inclusion angle of 115 degrees once, and count it
three times, the first, the only and one too many. Not that it can not have
relevance, I just do not want to be as far from the ensemble as it
necessitates and it puts the center image sound at a great frequency
response disadvantage. But it was modern improvised jazz (Marc Levin) and so
durn loud that it was impossible to hear anything in headphones.

> We don't want a jazz trio to be
> 40 ft wide, so we would use something other than MS.

IF drummer in center THEN MS, IF drummer at one side THEN XY. The parameter
I weigh most important here is smooth frequency response on cymbals.

> Thanks,
> Gary

Kind regards

Peter Larsen

[Frank Stearns]

"Gary Eickmeier" > writes:

-snips-

>OK does this mean that in general, you don't want to do MS when it is a live
>event with an audience? Maybe choose XY cardioid instead? In this event
>there will be audience participation, like sing alongs, so I might choose
>the MS anyway.

>Mike Rivers said that after a while I will get a feel for when to use MS and
>when to use XY or NOS and similar. I am thinking that some of the criteria
>would be is it a really wide group such as a symphony or a smaller one such
>as a Jazz group. We don't want a jazz trio to be 40 ft wide, so we would use
>something other than MS.

You can use all those, and move closer or farther... you'll typically get a
reasonably good left-to-right image.

While I still occasionally use ORTF and MS, and once in a while NOS, my complaint
with all of these techniques is that they don't seem to capture depth quite as well
as one might hope.

50 cm spaced omnis, with diffraction spheres and splayed up and out, do
left-to-right and front-to-back with an eerie accuracy. Scott has indicated that
Jeklin disc rigs do something similar, but that's one technique I've not yet tried.

I also like this technique because you need to be pretty close, which can give a
little more definition, but the omnis also do a nice catpure of the room sound.
(And, because they're omnis, the low end is typically better.)

The one caveat would be to use with care (if at all) in a small or lesser quality
room. If you're in that situation (low ceiling or excessive side or rear slap), use
one of the more "directed" techniques, such as a closer ORTF, or MS with not so
much of the S mixed in. YMMV

Frank
Mobile Audio
--

[Peter Larsen[_3_]]

Frank Stearns wrote:

> The one caveat would be to use with care (if at all) in a small or
> lesser quality room. If you're in that situation (low ceiling or
> excessive side or rear slap), use one of the more "directed"
> techniques, such as a closer ORTF, or MS with not so much of the S
> mixed in. YMMV

Directional microphones are indeed not about what you want to record, but
about what you do not want to record. And if there is something you really
do not want, a fig. 8 comes in real handy.

> Frank
> Mobile Audio

Kind regards

Peter Larsen

[Tom McCreadie]

"Gary Eickmeier wrote:

>
>"Tom McCreadie" wrote >
>
>> Yes of course, Fig8 records from both front and back with equal weighting.
>> I'd already said that. But my point was just that there's no way that you can
>> manipulate two channels of a 'real' Fig8 recording so that you end up with:
>> - two channels of "front-enriched" info (say, more emphasis on the band)
>> going to your front speakers
>> - two channels of "rear-enriched" info (less band, more audience and
>> ambient) going to your rear speakers.
>> (Even then, you still have to reconcile that issue of lateral inversion of
>> the rear images)
>>
>> If you can pull all this off, you could get a gig helping mathematicians
>> to square circles and trisect angles. :-)
>
>OK does this mean that in general, you don't want to do MS when it is a live
>event with an audience? Maybe choose XY cardioid instead? In this event
>there will be audience participation, like sing alongs, so I might choose
>the MS anyway.
>
>Mike Rivers said that after a while I will get a feel for when to use MS and
>when to use XY or NOS and similar. I am thinking that some of the criteria
>would be is it a really wide group such as a symphony or a smaller one such
>as a Jazz group. We don't want a jazz trio to be 40 ft wide, so we would use
>something other than MS.
>
Mike's right: No sense to be too prescriptive or dogmatic on criteria for array
choice. By trial 'n error in rehearsals and concerts, you soon amass a minstore
of what was successful, promising, or plain disappointing - this for the halls
and circumstances you work in, and for the audio properties that you cherish the
most.

Cardioids as main-pair is for me usually ORTF, then NOS; almost never XY (though
I've used narrow-angled XY as low profile spot for 4 frontstage oratorio
soloists). When going with two Fig8's. I prefer MS ("MS Blumlein") over straight
Blumlein XY.

But what exactly are you striving to pick up for your rear channel feed? A
diffuse, lush ambience?... or specific, discernible audience sounds such as sing
along, banter, whoops, hollers, catcalls etc.? In my situation, often live
classical concerts in churches, Fig8's can, despite their strong rear response,
actually be a Godsend - no pun intended :-). This arises from my the mics being
typically sited amongst and above the audience, the mics aiming slightly down,
with rear lobes pointing to the far roof/wall. In this way, their off-axis
response drop-off does a fine job in suppressing the nearby yakkers, coughers
and peppemint munchers

And there's no logic to the notion that using MS gives a wide soundstage. For
with MS - and XY, for that matter - you can dial in any width you want. Of
course, changing the width in that fashion has a concomitant effect on the
direct/indirect sound ratio. That is, 'width' and 'direct/reverbant ratios' are
alas not _independent_ variables...you only get that freedom when you also
adjust the mic stand distance.

[Tom McCreadie]

>Cardioids as main-pair is for me usually ORTF, then NOS; almost never XY

To clarify, my "XY" throuighout above post means "" coincident XY".

[hank alrich]

Tom McCreadie > wrote:

> >Cardioids as main-pair is for me usually ORTF, then NOS; almost never XY
>
> To clarify, my "XY" throuighout above post means "" coincident XY".

It's chromosomal setup.

--
shut up and play your guitar * HankAlrich.Com
HankandShaidriMusic.Com
YouTube.Com/WalkinayMusic

[Adrian Tuddenham[_2_]]

Peter Larsen > wrote:

> Frank Stearns wrote:
>
> > The one caveat would be to use with care (if at all) in a small or
> > lesser quality room. If you're in that situation (low ceiling or
> > excessive side or rear slap), use one of the more "directed"
> > techniques, such as a closer ORTF, or MS with not so much of the S
> > mixed in. YMMV
>
> Directional microphones are indeed not about what you want to record, but
> about what you do not want to record. And if there is something you really
> do not want, a fig. 8 comes in real handy.

It is worth noting that the null planes of a crossed pair of Fig.8
microphones coincide along one axis. If the mics were mounted for
conventional Blumlein stereo and truly concident and well-matched, there
woud be dead spots directly above and below the mics. This can be
useful, for example, in reducing audience noise below a high-flown mic
at an orchestral concert

Even if the unwanted noise source is not located exactly above or below
the microphone, it is possible to tilt the mic so that it rejects the
unwanted noise but still picks up a good stereo sound image. I was able
to reduce the noise of the bellows pedals when recording an
harmonium-type instrument by using this method (and a lot of blankets
around the player's feet).

In theory, any combination of microphones and signal summing systems
which emulate a Blumlein stereo pair should have the same effect; but
the mikes must be close-spaced and the two channels very well matched if
a good null is to be achieved over the whole audio spectrum.

There is no equivalent common null in the stereo pattern generated by a
pair of cardioid mics - also, any nulls which might be achieved with
spaced mics, no matter what type or what arrangement they are in, will
be highly frequency-dependent.


--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

[Gary Eickmeier]

"Peter Larsen" > wrote in message
...
> Gary Eickmeier wrote:
>
>> OK does this mean that in general, you don't want to do MS when it is
>> a live event with an audience? Maybe choose XY cardioid instead? In
>> this event there will be audience participation, like sing alongs, so
>> I might choose the MS anyway.
>
> There are many flavors of MS, but general for all of them is that they
> offer optimum microphone frequency response for the center.

I have noticed that there is less Bass response with MS than other mike
patterns and configurations. Does that have something to do with the Fig 8
back half being out of polarity with the front?

I have been fixating on another curious property of MS - very audible and
very true technically - that the two S channels are mutually exclusive, i.e.
the resultant L channel has nothing in it from the other side of the
pattern, and so on. This makes for a very spacious recording, no matter how
M is mixed in, and in effect it covers the entire frontal hemisphere when
recording, even though you may be making the equivalent XY pattern something
less than that. Example, I am recording our rehearsals in a room with the
air conditioner from hell. It sits on the right side, in a wall directly
right of the microphones. On playback, it is heard unmistakably coming from
the right wall of my listening room. The right side of the band is heard
right of the right speaker, nearer the corner of the room, and same for the
left. Complete spread of instruments across the front of my room, with great
positioning etc.

This is fascinating to me right now, and my only question might be what
exactly are we doing with the mixing of M and S besides maybe distorting the
center field a little to simulate other than a 90° XY pattern? This question
might be equivalent to the SRA of the XY technique. If I set the two Fig 8
XY mikes to be pointing directly at the R and L sides of the band, are they
supposed to image exactly between the speakers? If I set them narrower, does
that spread out the band? If I set them wider, does that scrunch up the
band?

I know this info has been discussed before and I am probably just "iggerant"
or haven't run it all through my understanding yet, but anyway that is the
basis for the question about mike patterns and how they affect your approach
to the job etc.

The MS results are very different from my beloved three cardioids that I
discovered and was enamored of for a while. The coincident microphones are a
lot less intrusive on the stage as well.

>
>> Mike Rivers said that after a while I will get a feel for when to use
>> MS and when to use XY or NOS and similar. I am thinking that some of
>> the criteria would be is it a really wide group such as a symphony or
>> a smaller one such as a Jazz group.
>
> "Wide or small", not good words to use, think opening angle or rather
> angle of inclusion and remember that for XY - coincident or spaced on a
> stereo bar - the wider the angle of inclusion you need the less the
> microphones should be angled apart if you want to avoid a glaring hole in
> the middle. No, it is NOT contraintuitive, to avoid hole in the middle the
> mics need to hear more of the other channel.

I think this jibes with what I was saying above, and thanks for this.

> I have only used the ORTF inclusion angle of 115 degrees once, and count
> it three times, the first, the only and one too many. Not that it can not
> have relevance, I just do not want to be as far from the ensemble as it
> necessitates and it puts the center image sound at a great frequency
> response disadvantage. But it was modern improvised jazz (Marc Levin) and
> so durn loud that it was impossible to hear anything in headphones.

So OK, I know you can use other patterns for M but also that those other
patterns will change the equivalent XY pattern effect. But an omni pattern
would have better bass response. I think my next rehearsal will compare M
patterns and see what happens!

Thanks to all for putting up with all this! I hope a journey thru various
techniques will be of interest to someone else lurking in!

Gary Eickmeier

[Gary Eickmeier]

Fantastic Adrian and Frank! This is very useful in deciding on a pattern for
a live event. I never thought about the nulls directly above and below a
Blumlein pair. And I might just tilt the mikes down toward the band a
little - so the result would be some audience participation, but not
obnoxiously close or loud. Unless I use an omni M pattern, which would
change things to a degree I am not prepared to contend with yet in my
experience, so....

Gary


"Adrian Tuddenham" > wrote in message
nvalid.invalid...
> Peter Larsen > wrote:
>
>> Frank Stearns wrote:
>>
>> > The one caveat would be to use with care (if at all) in a small or
>> > lesser quality room. If you're in that situation (low ceiling or
>> > excessive side or rear slap), use one of the more "directed"
>> > techniques, such as a closer ORTF, or MS with not so much of the S
>> > mixed in. YMMV
>>
>> Directional microphones are indeed not about what you want to record, but
>> about what you do not want to record. And if there is something you
>> really
>> do not want, a fig. 8 comes in real handy.
>
> It is worth noting that the null planes of a crossed pair of Fig.8
> microphones coincide along one axis. If the mics were mounted for
> conventional Blumlein stereo and truly concident and well-matched, there
> woud be dead spots directly above and below the mics. This can be
> useful, for example, in reducing audience noise below a high-flown mic
> at an orchestral concert
>
> Even if the unwanted noise source is not located exactly above or below
> the microphone, it is possible to tilt the mic so that it rejects the
> unwanted noise but still picks up a good stereo sound image. I was able
> to reduce the noise of the bellows pedals when recording an
> harmonium-type instrument by using this method (and a lot of blankets
> around the player's feet).
>
> In theory, any combination of microphones and signal summing systems
> which emulate a Blumlein stereo pair should have the same effect; but
> the mikes must be close-spaced and the two channels very well matched if
> a good null is to be achieved over the whole audio spectrum.
>
> There is no equivalent common null in the stereo pattern generated by a
> pair of cardioid mics - also, any nulls which might be achieved with
> spaced mics, no matter what type or what arrangement they are in, will
> be highly frequency-dependent.
>
>
> --
> ~ Adrian Tuddenham ~
> (Remove the ".invalid"s and add ".co.uk" to reply)
> www.poppyrecords.co.uk
>

[Gary Eickmeier]

"Tom McCreadie" > wrote in message
...


> And there's no logic to the notion that using MS gives a wide soundstage.
> For
> with MS - and XY, for that matter - you can dial in any width you want. Of
> course, changing the width in that fashion has a concomitant effect on the
> direct/indirect sound ratio. That is, 'width' and 'direct/reverbant
> ratios' are
> alas not _independent_ variables...you only get that freedom when you also
> adjust the mic stand distance.

It's just that with the S channels we are miking directly to the two sides
and neither side contains any sound from the other channel, almost as if the
two elements (or sides of an element) were on different sides of a brick
wall. They will mix with M and get some of that info back, but I am
intrigued by the basic wide-ness of the MS technique that others don't seem
to have.

Gary

[Les Cargill[_4_]]

Tom McCreadie wrote:
>> Cardioids as main-pair is for me usually ORTF, then NOS; almost never XY
>
> To clarify, my "XY" throuighout above post means "" coincident XY".
>


There is a *non* coincident X/Y?


--
Les Cargill

[Scott Dorsey]

Gary Eickmeier > wrote:
>
>I have noticed that there is less Bass response with MS than other mike
>patterns and configurations. Does that have something to do with the Fig 8
>back half being out of polarity with the front?

No, that has to do with the specific microphones you are using, and is not
inherent in the technique.

>I have been fixating on another curious property of MS - very audible and
>very true technically - that the two S channels are mutually exclusive, i.e.
>the resultant L channel has nothing in it from the other side of the
>pattern, and so on.

Ahh, but it does, just as much as an X-Y recording.

You can artifically crank the separation way up on either one, if you want,
although the effect is to create a hole in the center and you lose all
of the image beyond the speakers.
--scott

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

[Scott Dorsey]

Les Cargill > wrote:
>Tom McCreadie wrote:
>>> Cardioids as main-pair is for me usually ORTF, then NOS; almost never XY
>>
>> To clarify, my "XY" throuighout above post means "" coincident XY".
>
>There is a *non* coincident X/Y?

ORTF and other near-coincident methods might be called that, although
nobody really does call them that in practice.
--scott

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

[Peter Larsen[_3_]]

Gary Eickmeier wrote:

> "Peter Larsen" > wrote in message
> ...
>> Gary Eickmeier wrote:

>>> OK does this mean that in general, you don't want to do MS when it
>>> is a live event with an audience? Maybe choose XY cardioid instead?
>>> In this event there will be audience participation, like sing
>>> alongs, so I might choose the MS anyway.

>> There are many flavors of MS, but general for all of them is that
>> they offer optimum microphone frequency response for the center.

> I have noticed that there is less Bass response with MS than other
> mike patterns and configurations. Does that have something to do with
> the Fig 8 back half being out of polarity with the front?

The bass roll-off from directional mics is "known and described in the
literature", however it is not just about level, if you compare with a pair
of spaced omnis you compare with a setup that spreads the bass in time.

> I have been fixating on another curious property of MS - very audible
> and very true technically - that the two S channels are mutually
> exclusive, i.e. the resultant L channel has nothing in it from the
> other side of the pattern, and so on.

There is only one S channel.

> The MS results are very different from my beloved three cardioids
> that I discovered and was enamored of for a while. The coincident
> microphones are a lot less intrusive on the stage as well.

Yes.

> So OK, I know you can use other patterns for M but also that those
> other patterns will change the equivalent XY pattern effect. But an
> omni pattern would have better bass response. I think my next
> rehearsal will compare M patterns and see what happens!

Learning what which setup does is wise.

> Thanks to all for putting up with all this! I hope a journey thru
> various techniques will be of interest to someone else lurking in!

Thanks also to you for coming back with reports of how things work out.

> Gary Eickmeier

Kind regards

Peter Larsen

[Peter Larsen[_3_]]

Les Cargill wrote:

> Tom McCreadie wrote:

>>> Cardioids as main-pair is for me usually ORTF, then NOS; almost
>>> never XY

>> To clarify, my "XY" throuighout above post means "" coincident XY".

> There is a *non* coincident X/Y?

Yes, in the german tech language zone two cards on a K&M short bar are also
XY, it is coincident if the word coincident is used, otherwise not. Also
ORTF means ORTF, with is a 115 degree angle between cards and only that, XY
is used for any pair of cards with a short transverse stick and any angle
between them. See also "the stereophonic zoom" and compare with the use of
the wording "decorrelate" in 5.1 recording.

Kind regards

Peter Larsen

[Peter Larsen[_3_]]

Scott Dorsey wrote:

> Les Cargill > wrote:
>> Tom McCreadie wrote:

>>>> Cardioids as main-pair is for me usually ORTF, then NOS; almost
>>>> never XY

>>> To clarify, my "XY" throuighout above post means "" coincident XY".

>> There is a *non* coincident X/Y?

> ORTF and other near-coincident methods might be called that, although
> nobody really does call them that in practice.

I do and usually toss in a bit of extra ascii.

> --scott

Kind regards

Peter Larsen

[Tom McCreadie]

OPeter Larse" wrote:

>Les Cargill wrote:
>
>> Tom McCreadie wrote:
>
>>>> Cardioids as main-pair is for me usually ORTF, then NOS; almost
>>>> never XY
>
>>> To clarify, my "XY" throuighout above post means "" coincident XY".
>
>> There is a *non* coincident X/Y?
>
>Yes, in the german tech language zone two cards on a K&M short bar are also
>XY, it is coincident if the word coincident is used, otherwise not. Also
>ORTF means ORTF, with is a 115 degree angle between cards and only that, XY
>is used for any pair of cards with a short transverse stick and any angle
>between them. See also "the stereophonic zoom" and compare with the use of
>the wording "decorrelate" in 5.1 recording.
>
I'm of the same school as Les, in always having taken bare 'XY" to imply
"coincidence". But having tearfully observed others going with different
definitions, I thought it prudent to remove any conceivable ambiguityin in my
prose by including some more, extra, additional, superfluous, supplemental
redundancy. :-)

Incidentally, Peter, ORTF is actually 110°. Was perhaps that inadvertent, extra,
additional, suppplemental superfluous five degree arc so critical, that it
scarred you for life, concerning ORTF as a decent technique :-)

[hank alrich]

Les Cargill > wrote:

> Tom McCreadie wrote:
> >> Cardioids as main-pair is for me usually ORTF, then NOS; almost never XY
> >
> > To clarify, my "XY" throuighout above post means "" coincident XY".
> >
>
>
> There is a *non* coincident X/Y?

People achieve alignment in the vertical axis with varying degrees of
success, sometimes derived from varying degrees of concern with it.

I like to be able to imagine a vertical line through the center of the
capsules, one mic over the other.

Some pals of mine merely bring the caps together as closely as possible,
which fails to achieve my preferred alignment.


--
shut up and play your guitar * HankAlrich.Com
HankandShaidriMusic.Com
YouTube.Com/WalkinayMusic

[Don Pearce[_3_]]

On Sun, 8 Dec 2013 14:06:43 -0600, (hank alrich)
wrote:

>Les Cargill > wrote:
>
>> Tom McCreadie wrote:
>> >> Cardioids as main-pair is for me usually ORTF, then NOS; almost never XY
>> >
>> > To clarify, my "XY" throuighout above post means "" coincident XY".
>> >
>>
>>
>> There is a *non* coincident X/Y?
>
>People achieve alignment in the vertical axis with varying degrees of
>success, sometimes derived from varying degrees of concern with it.
>
>I like to be able to imagine a vertical line through the center of the
>capsules, one mic over the other.
>
>Some pals of mine merely bring the caps together as closely as possible,
>which fails to achieve my preferred alignment.

20kHz has a wavelength of about half an inch. You'd certainly want
alignment within about one tenth of a wavelength, so you should aim
for an accuracy of about 50 thou.

In practice you don't need to be this good, but to avoid comb
filtering you certainly don't want to be too much worse.

d

[Tom McCreadie]

Gary Eickmeier wrote:

>
>I have noticed that there is less Bass response with MS than other mike
>patterns and configurations. Does that have something to do with the Fig 8
>back half being out of polarity with the front?
>
Welcome to the world of compromises. Yes, fig8 has in principle a wimpier bass
response than an omni. There's a solid theoretical basis for this.
For that and other reasons, omni is often the preference of many performers, who
tend to listen out for their voice/instrument sound, rather than fixate on
holographic imaging etc.
Cardioid, lying between fig8 and omni - generally also has a low end drop, but

less extreme.

For a ballpark indication, the "Marketing response curves" from the Schoeps
literature gives:
50Hz 100Hz 200Hz
Fig-8 (MK8) -3dB -1dB 0dB
Cardioid (MK4) -6dB -3dB -1dB

Often, I'll perform a little bass re-equalization in post, particularly of the
MK8's, by applying a milder, inverted version of the published response curves.


>...... Example, I am recording our rehearsals in a room with the
>air conditioner from hell. It sits on the right side, in a wall directly
>right of the microphones. On playback, it is heard unmistakably coming from
>the right wall of my listening room. The right side of the band is heard
>right of the right speaker, nearer the corner of the room, and same for the
>left. Complete spread of instruments across the front of my room, with great
>positioning etc.
>
>This is fascinating to me right now, and my only question might be what
>exactly are we doing with the mixing of M and S besides maybe distorting the
>center field a little to simulate other than a 90° XY pattern? This question
>might be equivalent to the SRA of the XY technique. If I set the two Fig 8
>XY mikes to be pointing directly at the R and L sides of the band, are they
>supposed to image exactly between the speakers? If I set them narrower, does
>that spread out the band? If I set them wider, does that scrunch up the
>band?
>

To get a handle on all this:

1. You first need to establish what angle the L- and R-edges of your Band are
subtending to your mic array. For this, you could just cobble up a letter-size
sheet of white cardboard, marked with protractor angle lines (and distances.)

2. If using an XY array, also use this same 'protractor' to establish the
included angle set for the XY mics.

3. Calculate the SRA (Stereophonic Recording Angle) of the array. Recall that
the SRA is that angular arc, in the hall in front of the array, that will
stretch itself to fill the space between your two front playback speakers. The
SRA is a property of the mic array, not the band.

4. Now check the band / SRA "match-up": You normally want the arc occupied by
the band to fit inside that SRA. If the band is set up too wide, the images of
edge most instruments start to clutter up in the speakers..or, even worse, go
outside the speakers, with an uneasy vague imaging and at the same time leaving
a "hole in the middle".

5. Correct any glaring disparity by one or more of:
a) reseating the band
b) adjust mic-stand distance
c) adjust SRA of array - via XY mic angling or M/S gain ratios

6. As a rough guide:
- Fig8 XY at 90° incl. angle, or Fig8 MS at 1:1 M/S, gives an SRA of ca. 76° -
the SRA decreases ca. 10° per 10° of XY incl. angle increase
- the SRA decreases ca. 7° per dB S increase

In these early trials, I suspect that your band had ben spilling significantly
outside their SRA cage. :-)
--
Tom McCreadie

Live at The London Palindrome - ABBA

[Les Cargill[_4_]]

Peter Larsen wrote:
> Les Cargill wrote:
>
>> Tom McCreadie wrote:
>
>>>> Cardioids as main-pair is for me usually ORTF, then NOS; almost
>>>> never XY
>
>>> To clarify, my "XY" throuighout above post means "" coincident XY".
>
>> There is a *non* coincident X/Y?
>
> Yes, in the german tech language zone two cards on a K&M short bar are also
> XY, it is coincident if the word coincident is used, otherwise not. Also
> ORTF means ORTF, with is a 115 degree angle between cards and only that, XY
> is used for any pair of cards with a short transverse stick

Right.

> and any angle
> between them.

http://www.dpamicrophones.com/en/Mic-University/Stereo-Techniques/~/media/Images/Content/MicUni/Stereo%20Techniques/DM00580.jpg

> See also "the stereophonic zoom" and compare with the use of
> the wording "decorrelate" in 5.1 recording.
>
> Kind regards
>
> Peter Larsen
--
Les Cargill

[Les Cargill[_4_]]

hank alrich wrote:
> Les Cargill > wrote:
>
>> Tom McCreadie wrote:
>>>> Cardioids as main-pair is for me usually ORTF, then NOS; almost never XY
>>>
>>> To clarify, my "XY" throuighout above post means "" coincident XY".
>>>
>>
>>
>> There is a *non* coincident X/Y?
>
> People achieve alignment in the vertical axis with varying degrees of
> success, sometimes derived from varying degrees of concern with it.
>
> I like to be able to imagine a vertical line through the center of the
> capsules, one mic over the other.
>

Yep.

> Some pals of mine merely bring the caps together as closely as possible,
> which fails to achieve my preferred alignment.
>
>

That works pretty well. I like it a bit better for drum overheads.

--
Les Cargill

[Gary Eickmeier]

Thanks Tom - printing it out and taking it with me tomorrow to rehearsal.

Gary Eickmeier


"Tom McCreadie" > wrote in message
...
> Gary Eickmeier wrote:
>
>>
>>I have noticed that there is less Bass response with MS than other mike
>>patterns and configurations. Does that have something to do with the Fig 8
>>back half being out of polarity with the front?
>>
> Welcome to the world of compromises. Yes, fig8 has in principle a wimpier
> bass
> response than an omni. There's a solid theoretical basis for this.
> For that and other reasons, omni is often the preference of many
> performers, who
> tend to listen out for their voice/instrument sound, rather than fixate on
> holographic imaging etc.
> Cardioid, lying between fig8 and omni - generally also has a low end drop,
> but
>
> less extreme.
>
> For a ballpark indication, the "Marketing response curves" from the
> Schoeps
> literature gives:
> 50Hz 100Hz 200Hz
> Fig-8 (MK8) -3dB -1dB 0dB
> Cardioid (MK4) -6dB -3dB -1dB
>
> Often, I'll perform a little bass re-equalization in post, particularly of
> the
> MK8's, by applying a milder, inverted version of the published response
> curves.
>
>
>>...... Example, I am recording our rehearsals in a room with the
>>air conditioner from hell. It sits on the right side, in a wall directly
>>right of the microphones. On playback, it is heard unmistakably coming
>>from
>>the right wall of my listening room. The right side of the band is heard
>>right of the right speaker, nearer the corner of the room, and same for
>>the
>>left. Complete spread of instruments across the front of my room, with
>>great
>>positioning etc.
>>
>>This is fascinating to me right now, and my only question might be what
>>exactly are we doing with the mixing of M and S besides maybe distorting
>>the
>>center field a little to simulate other than a 90° XY pattern? This
>>question
>>might be equivalent to the SRA of the XY technique. If I set the two Fig 8
>>XY mikes to be pointing directly at the R and L sides of the band, are
>>they
>>supposed to image exactly between the speakers? If I set them narrower,
>>does
>>that spread out the band? If I set them wider, does that scrunch up the
>>band?
>>
>
> To get a handle on all this:
>
> 1. You first need to establish what angle the L- and R-edges of your Band
> are
> subtending to your mic array. For this, you could just cobble up a
> letter-size
> sheet of white cardboard, marked with protractor angle lines (and
> distances.)
>
> 2. If using an XY array, also use this same 'protractor' to establish the
> included angle set for the XY mics.
>
> 3. Calculate the SRA (Stereophonic Recording Angle) of the array. Recall
> that
> the SRA is that angular arc, in the hall in front of the array, that will
> stretch itself to fill the space between your two front playback speakers.
> The
> SRA is a property of the mic array, not the band.
>
> 4. Now check the band / SRA "match-up": You normally want the arc occupied
> by
> the band to fit inside that SRA. If the band is set up too wide, the
> images of
> edge most instruments start to clutter up in the speakers..or, even worse,
> go
> outside the speakers, with an uneasy vague imaging and at the same time
> leaving
> a "hole in the middle".
>
> 5. Correct any glaring disparity by one or more of:
> a) reseating the band
> b) adjust mic-stand distance
> c) adjust SRA of array - via XY mic angling or M/S gain ratios
>
> 6. As a rough guide:
> - Fig8 XY at 90° incl. angle, or Fig8 MS at 1:1 M/S, gives an SRA of ca.
> 76° -
> the SRA decreases ca. 10° per 10° of XY incl. angle increase
> - the SRA decreases ca. 7° per dB S increase
>
> In these early trials, I suspect that your band had ben spilling
> significantly
> outside their SRA cage. :-)
> --
> Tom McCreadie
>
> Live at The London Palindrome - ABBA

[Gary Eickmeier]

I think from the few tests I was able to do between patterns that the M set
to omni is going to be the winner. I didn't realize that the FR was so
different for all of the patterns. The Fig 8 pattern is wimpy on the bass,
but when I put it (the Mid) to omni or cardioid, transformed to rich, deep,
wide frequency response that affects all instruments and still has great
imaging.

Main problem now will be tactical. I am using a bracket that suspends the
mikes one above the other and it is a little rickety in holding them in
place together. I am attempting some Velcro to collar them together and keep
them in exact coincidence. If I fix the mikes in place properly the bracket
tells me the aiming of the stereo pattern so I can turn the high stand in
the right direction toward the band. Must hoist the mike stand onto the
stage between the pre-show and the main concert without disturbing the peace
or something falling over. Glad there are only two cables to secure now.
Will wear my black Ninja recordist outfit to not be too obvious to whatever
cameras.

Gary Eickmeier

[Tom McCreadie]

Gary Eickmeier wrote:

>I think from the few tests I was able to do between patterns that the M set
>to omni is going to be the winner. I didn't realize that the FR was so
>different for all of the patterns. The Fig 8 pattern is wimpy on the bass,
>but when I put it (the Mid) to omni or cardioid, transformed to rich, deep,
>wide frequency response that affects all instruments and still has great
>imaging.

MS with 1:1 omni:fig8 equates to back-to-back (180°) XY cardioids. The SRA of
ca. 102° is more generous than that from MS with two fig8's, allowing you to get
the mics closer on stage. If you're obliged to go really close with your stand,
you can always boost the M 6dB w.r.t. the S, essentially giving back-to-back
subcardioids (plus a beefier bass), with an even more forgiving SRA of 180°.
Such gain adjustments can be put off until post...where you might also consider
a mild tweak to the wimpy bass of the fig8 channel.

With your rather large AT 2050's, the MS route is less visually obtrusive. Go
get 'em tiger!

[Mike Rivers[_2_]]

On 12/12/2013 3:35 AM, Gary Eickmeier wrote:

> Main problem now will be tactical. I am using a bracket that suspends the
> mikes one above the other and it is a little rickety in holding them in
> place together.

You need a Shure A27M
http://www.shure.com/americas/products/accessories/microphones/microphone-cases-adapters/a27m

With some clever arranging of the parts, you can get just about any
two-mic setup you want, even using two large capsule mics with their
shock mounts. It's expensive, but not as expensive as some of the AEA or
Schoeps stereo brackets. And when the show turns into a riot, it makes a
good billy club.

Another approach is to get familiar with the Atlas Sound catalog of mic
stand parts and adapters. Many of these parts are sold in music stores
on the rack of blister packs from On Stage Stands. With a few flanges,
pieces of straight and right-angle-bent threaded tubing, and some 1x2
wood strips, you can make up a sturdy stereo bracket for your favorite
configuration that will cost you less than $10.


--
For a good time, visit http://mikeriversaudio.wordpress.com

[Gary Eickmeier]

"Tom McCreadie" > wrote in message
...
> Gary Eickmeier wrote:
>
>>I think from the few tests I was able to do between patterns that the M
>>set
>>to omni is going to be the winner. I didn't realize that the FR was so
>>different for all of the patterns. The Fig 8 pattern is wimpy on the bass,
>>but when I put it (the Mid) to omni or cardioid, transformed to rich,
>>deep,
>>wide frequency response that affects all instruments and still has great
>>imaging.
>
> MS with 1:1 omni:fig8 equates to back-to-back (180°) XY cardioids. The SRA
> of
> ca. 102° is more generous than that from MS with two fig8's, allowing you
> to get
> the mics closer on stage. If you're obliged to go really close with your
> stand,
> you can always boost the M 6dB w.r.t. the S, essentially giving
> back-to-back
> subcardioids (plus a beefier bass), with an even more forgiving SRA of
> 180°.
> Such gain adjustments can be put off until post...where you might also
> consider
> a mild tweak to the wimpy bass of the fig8 channel.
>
> With your rather large AT 2050's, the MS route is less visually
> obtrusive. Go
> get 'em tiger!
>

Freaky observation #1: With the S aimed sideways and the M set to omni,
there is no way for it to differentiate between front and back. I am
literally recording in surround sound - only not quite. This could be
interesting.

Gary

[Scott Dorsey]

Gary Eickmeier > wrote:
>
>
>Freaky observation #1: With the S aimed sideways and the M set to omni,
>there is no way for it to differentiate between front and back. I am
>literally recording in surround sound - only not quite. This could be
>interesting.

Yes. Same thing with Blumlein. Same thing with spaced omni. This is
where some of your sense of space is coming from... the rear of the hall
reflected into the front.
--scott

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

[Gary Eickmeier]

Thanks Mike - I will look into Atlas Sound, but until then I will have to
make do with my home made bracket. The Shure A27M isn't quite right for this
particular microphone.

Gary


"Mike Rivers" > wrote in message
...
> On 12/12/2013 3:35 AM, Gary Eickmeier wrote:
>
>> Main problem now will be tactical. I am using a bracket that suspends the
>> mikes one above the other and it is a little rickety in holding them in
>> place together.
>
> You need a Shure A27M
> http://www.shure.com/americas/products/accessories/microphones/microphone-cases-adapters/a27m
>
> With some clever arranging of the parts, you can get just about any
> two-mic setup you want, even using two large capsule mics with their shock
> mounts. It's expensive, but not as expensive as some of the AEA or Schoeps
> stereo brackets. And when the show turns into a riot, it makes a good
> billy club.
>
> Another approach is to get familiar with the Atlas Sound catalog of mic
> stand parts and adapters. Many of these parts are sold in music stores on
> the rack of blister packs from On Stage Stands. With a few flanges, pieces
> of straight and right-angle-bent threaded tubing, and some 1x2 wood
> strips, you can make up a sturdy stereo bracket for your favorite
> configuration that will cost you less than $10.
>
>
> --
> For a good time, visit http://mikeriversaudio.wordpress.com
>

[Scott Dorsey]

Gary Eickmeier > wrote:
>I think from the few tests I was able to do between patterns that the M set
>to omni is going to be the winner. I didn't realize that the FR was so
>different for all of the patterns. The Fig 8 pattern is wimpy on the bass,
>but when I put it (the Mid) to omni or cardioid, transformed to rich, deep,
>wide frequency response that affects all instruments and still has great
>imaging.

Some of this is just the nature of the dual-diaphragm capsule. You'll
notice also that your omni isn't really very omni at all.
--scott

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

[Gary Eickmeier]

"Scott Dorsey" > wrote in message
...
> Gary Eickmeier > wrote:
>>
>>
>>Freaky observation #1: With the S aimed sideways and the M set to omni,
>>there is no way for it to differentiate between front and back. I am
>>literally recording in surround sound - only not quite. This could be
>>interesting.
>
> Yes. Same thing with Blumlein. Same thing with spaced omni. This is
> where some of your sense of space is coming from... the rear of the hall
> reflected into the front.
> --scott

So then the only thing that places the band in front on playback is the
front speakers. So if I listen in DPL II, with rear and side speakers
separated from the front by only a little more delay in the receiver, that
is why I am sensing a surround effect, with the A/C unit off to the side
wall of my room, and some audience effects all around.(?)

So if XY Blumlein or even cardioid isn't much better (hypercardioid might be
OK), then what is the best miking pattern? Super sharp frontal imaging, good
FR?

Gary Eickmeier

PS I see they have the Zoom H6 in Recording magazine, but they do no
technical testing on it whatsoever. What use is that? We can see all of its
features at the web site.

[Mike Rivers[_2_]]

On 12/12/2013 11:46 AM, Gary Eickmeier wrote:
> The Shure A27M isn't quite right for this
> particular microphone.

I forgot what microphone(s) you're using. I can cross two U87s
grill-to-grill with it, which will get me X-Y or M-S depending on how
the mics are set and aimed. It makes more sense when you realize that
the centerline of the coincident diaphragms doesn't need to be on the
line between the mic stand and the point where you're aiming them. You
aim the mic and let the stand stand where it needs to stand.

--
For a good time, visit http://mikeriversaudio.wordpress.com

[Scott Dorsey]

Gary Eickmeier > wrote:
>
>So then the only thing that places the band in front on playback is the
>front speakers. So if I listen in DPL II, with rear and side speakers
>separated from the front by only a little more delay in the receiver, that
>is why I am sensing a surround effect, with the A/C unit off to the side
>wall of my room, and some audience effects all around.(?)

If you listen in Dolby Stereo, then the output of the S microphone is
low-passed and sent to the surround speakers, modulo a little phase
shift.

That is, everything not in the center of the stereo image is spuriously
appearing in the rear channels.

If you are using a Pro Logic decoder, the same thing is happening except
that the added steering logic will cause the front image to waver in
position.

>So if XY Blumlein or even cardioid isn't much better (hypercardioid might be
>OK), then what is the best miking pattern? Super sharp frontal imaging, good
>FR?

If you're running it through a Pro-Logic decoder? None. Pro-Logic isn't
for that.

>PS I see they have the Zoom H6 in Recording magazine, but they do no
>technical testing on it whatsoever. What use is that? We can see all of its
>features at the web site.

Write a letter of complaint to the editors, tell them you want real
measurements in reviews.
--scott

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

[Mike Rivers[_2_]]

On 12/12/2013 5:36 PM, Scott Dorsey wrote:
>> >PS I see they have the Zoom H6 in Recording magazine, but they do no
>> >technical testing on it whatsoever. What use is that? We can see all of its
>> >features at the web site.

> Write a letter of complaint to the editors, tell them you want real
> measurements in reviews.

I put technical measurements in my reviews, usually with some
explanation of why the measurements are significant. Unfortunately the
editors who pay won't take them because they're too long. I'm thinking
about putting a virtual tip jar on my web site.


--
For a good time, visit http://mikeriversaudio.wordpress.com

[Gary Eickmeier]

"Mike Rivers" > wrote in message
...
> On 12/12/2013 11:46 AM, Gary Eickmeier wrote:
>> The Shure A27M isn't quite right for this
>> particular microphone.
>
> I forgot what microphone(s) you're using. I can cross two U87s
> grill-to-grill with it, which will get me X-Y or M-S depending on how the
> mics are set and aimed. It makes more sense when you realize that the
> centerline of the coincident diaphragms doesn't need to be on the line
> between the mic stand and the point where you're aiming them. You aim the
> mic and let the stand stand where it needs to stand.

My bracket is basically two table top mike stands with the base replaced by
my bracket, on which the two posts are screwed in about 6 inches apart. The
AT 2050 shockmount mike holders are mounted to the posts, and the mikes are
held one over the other in coincidence. The positioning is adjustable, but a
little rickety in holding position, so some velcro would be marvie in
solving that.

The ideal would be a high quality single case MS or XY mike on a tiltable
base. I can't tilt mine down more toward the band when I hoist the stand up
to 9 feet. Maybe don't need it so high.

Gary

[Gary Eickmeier]

"Scott Dorsey" > wrote in message
...
> Gary Eickmeier > wrote:
>>I think from the few tests I was able to do between patterns that the M
>>set
>>to omni is going to be the winner. I didn't realize that the FR was so
>>different for all of the patterns. The Fig 8 pattern is wimpy on the bass,
>>but when I put it (the Mid) to omni or cardioid, transformed to rich,
>>deep,
>>wide frequency response that affects all instruments and still has great
>>imaging.
>
> Some of this is just the nature of the dual-diaphragm capsule. You'll
> notice also that your omni isn't really very omni at all.
> --scott

After examining the FR graphs of the omni vs the cardioid, I think I will
change my mind to cardioid for the M channel.

Gary

[Scott Dorsey]

Gary Eickmeier > wrote:
>"Scott Dorsey" > wrote in message
...
>> Gary Eickmeier > wrote:
>>>I think from the few tests I was able to do between patterns that the M
>>>set
>>>to omni is going to be the winner. I didn't realize that the FR was so
>>>different for all of the patterns. The Fig 8 pattern is wimpy on the bass,
>>>but when I put it (the Mid) to omni or cardioid, transformed to rich,
>>>deep,
>>>wide frequency response that affects all instruments and still has great
>>>imaging.
>>
>> Some of this is just the nature of the dual-diaphragm capsule. You'll
>> notice also that your omni isn't really very omni at all.
>
>After examining the FR graphs of the omni vs the cardioid, I think I will
>change my mind to cardioid for the M channel.

Don't worry, the manufacturer's response plots are totally useless anyway
since they're only done on-axis. They really tell you very close to nothing
about the far-field performance of the mike.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Gary Eickmeier]

"Scott Dorsey" > wrote in message
...
> Gary Eickmeier > wrote:

>>After examining the FR graphs of the omni vs the cardioid, I think I will
>>change my mind to cardioid for the M channel.
>
> Don't worry, the manufacturer's response plots are totally useless anyway
> since they're only done on-axis. They really tell you very close to
> nothing
> about the far-field performance of the mike.
> --scott

The session went well. I have no complaints about getting my way in mike
positioning and I did the MS with M set to cardioid. Editing looks favorable
and sounds pretty good. Can't wait to play it on the main system.

My makeshift mike bracket is beginning to annoy me. Guess I will look thru
the Atlas catalogue soon.

Gary Eickmeier

[Gary Eickmeier]

I have done the deed now - MS with the M channel cardioid. It wasn't totally
what I expected. I did everything right, but the imaging isn't quite as
precise as I expected. I wonder if this is because the cardioid pattern is
just too fat. I am beginning to realize that the shape of the pattern with
coincident miking is very importantl. Like, with the figure 8 pattern the
imaging was more precise because the pattern is narrower, but the FR is not
quite as hefty in the low end. So why not try MS fig 8 or XY fig 8 and just
EQ the low end up?

Are my observations valid from your experience? What was your conclusion?

Gary Eickmeier

[Gary Eickmeier]

I have done the deed now - MS with the M channel cardioid. It wasn't totally
what I expected. I did everything right, but the imaging isn't quite as
precise as I expected. I wonder if this is because the cardioid pattern is
just too fat. I am beginning to realize that the shape of the pattern with
coincident miking is very importantl. Like, with the figure 8 pattern the
imaging was more precise because the pattern is narrower, but the FR is not
quite as hefty in the low end. So why not try MS fig 8 or XY fig 8 and just
EQ the low end up?

Are my observations valid from your experience? What was your conclusion?

Gary Eickmeier

[Peter Larsen[_3_]]

Gary Eickmeier wrote:

> I have done the deed now - MS with the M channel cardioid. It wasn't
> totally what I expected. I did everything right, but the imaging
> isn't quite as precise as I expected. I wonder if this is because the
> cardioid pattern is just too fat. I am beginning to realize that the
> shape of the pattern with coincident miking is very importantl. Like,
> with the figure 8 pattern the imaging was more precise because the
> pattern is narrower, but the FR is not quite as hefty in the low end.
> So why not try MS fig 8 or XY fig 8 and just EQ the low end up?

> Are my observations valid from your experience? What was your
> conclusion?

I can not comment, I wasn't there. It is always a trade-off, so in setting
up mics you need to know what deals the actual equipment can give you as you
deploy it.

You can fix the tonal balance when you come home, you can not alter the
patter of a single mic capsule after the recording.

It comes across as tempting to try a subcardioid as M, it has 6 dB room
rejection instead of the 10 dB room rejection of a cardioid, but you get a
neater low frequency response.

If the tonal balance is wrong, then I EQ when I come home. If it is the room
I use a FFT type EQ because it is not minimum phase and if it is the mic I
use minimum phase EQ and tend to gravitate towards a few standard presets
for each mic I use regularly depending on context.

> Gary Eickmeier

Kind regards

Peter Larsen

[Marc Wielage[_2_]]

On Wed, 18 Dec 2013 22:45:24 -0800, Gary Eickmeier wrote
(in article >):

> I have done the deed now - MS with the M channel cardioid. It wasn't totally
> what I expected. I did everything right, but the imaging isn't quite as
> precise as I expected.
>------------------------------<snip>------------------------------<

Are you aware you can vary the balance between M & S with a lot of M/S
decoding plug-ins?

http://www.soundonsound.com/sos/sep10/articles/pt-0910.htm

Waves' S1 is one way of varying M/S level, but there's lots of ways to do it
after the fact:

http://www.waves.com/plugins/s1-stereo-imager

Specific imaging precision I think is going to depend more on mic position
and room acoustics than anything else.

--MFW

[Gary Eickmeier]

"Marc Wielage" > wrote in message
.com...
> On Wed, 18 Dec 2013 22:45:24 -0800, Gary Eickmeier wrote
> (in article >):
>
>> I have done the deed now - MS with the M channel cardioid. It wasn't
>> totally
>> what I expected. I did everything right, but the imaging isn't quite as
>> precise as I expected.
>>------------------------------<snip>------------------------------<
>
> Are you aware you can vary the balance between M & S with a lot of M/S
> decoding plug-ins?
>
> http://www.soundonsound.com/sos/sep10/articles/pt-0910.htm
>
> Waves' S1 is one way of varying M/S level, but there's lots of ways to do
> it
> after the fact:
>
> http://www.waves.com/plugins/s1-stereo-imager
>
> Specific imaging precision I think is going to depend more on mic position
> and room acoustics than anything else.
>
> --MFW

Thanks Marc but I have covered my MS mixing technique in previous posts. I
can mix in any amount of M that I want, but I find that if there is too much
S you get too wide a soundstage with hole in the middle. I do about equal M
and S and that seems to be fhe best..

Gary

[Gary Eickmeier]

"Marc Wielage" > wrote in message
.com...
> On Wed, 18 Dec 2013 22:45:24 -0800, Gary Eickmeier wrote
> (in article >):
>
>> I have done the deed now - MS with the M channel cardioid. It wasn't
>> totally
>> what I expected. I did everything right, but the imaging isn't quite as
>> precise as I expected.
>>------------------------------<snip>------------------------------<
>
> Are you aware you can vary the balance between M & S with a lot of M/S
> decoding plug-ins?
>
> http://www.soundonsound.com/sos/sep10/articles/pt-0910.htm
>
> Waves' S1 is one way of varying M/S level, but there's lots of ways to do
> it
> after the fact:
>
> http://www.waves.com/plugins/s1-stereo-imager
>
> Specific imaging precision I think is going to depend more on mic position
> and room acoustics than anything else.
>
> --MFW

Thanks Marc but I have covered my MS mixing technique in previous posts. I
can mix in any amount of M that I want, but I find that if there is too much
S you get too wide a soundstage with hole in the middle. I do about equal M
and S and that seems to be fhe best..

Gary

[Peter Larsen[_3_]]

Gary Eickmeier wrote:

> Thanks Marc but I have covered my MS mixing technique in previous
> posts. I can mix in any amount of M that I want, but I find that if
> there is too much S you get too wide a soundstage with hole in the
> middle. I do about equal M and S and that seems to be fhe best..

Yes, the probable range is S equal to or no more than 5 dB less than M.

> Gary

Kind regards

Peter Larsen

[Tom McCreadie]

Gary Eickmeier wrote:

>I have done the deed now - MS with the M channel cardioid. It wasn't totally
>what I expected. I did everything right, but the imaging isn't quite as
>precise as I expected.

Did the imaging actually seem precise to you while listening live at the venue,
sitting under the mics (ideally with eyes closed)? Wwas your home evaluation
via two- or multi-speaker playback? [Bounce it Over nearby Surfaces Equipment,
right? :-)] And how sharp was that home playback via headphones? Got any
commercial cd's or recordings from other people that imaged significantly
sharper?

Maybe you're just coming up against the limitations of those mics...say, a
ragged off-axis response...or modest capsule matching, particularly between the
two back-to-back cardioids that get used to emulate the Fig-8... or mics in the
array insufficiently 'coincident' w.r.t. the main sound arrival direction, etc.
But there's no reason for not trying to tweak the wimpy low end a little in
post... provided you're not then exacerbating any unwanted stuff like
air-conditioning growl, stand-borne resonance's etc.

[Frank Stearns]

"Peter Larsen" > writes:

>Gary Eickmeier wrote:

>> Thanks Marc but I have covered my MS mixing technique in previous
>> posts. I can mix in any amount of M that I want, but I find that if
>> there is too much S you get too wide a soundstage with hole in the
>> middle. I do about equal M and S and that seems to be fhe best..

>Yes, the probable range is S equal to or no more than 5 dB less than M.

>> Gary

Careful of doing M-S mixing by specific "formulas." I say this because S will
often wind up being a somewhat different dynamic envelope. S consists of more
diffused, reverberant room sound rather than direct sound. (That is, S will often
be lower than M in overall level.)

These suggested ratios from Peter and others are probably fine -- once you take into
account what might be different record levels between M and S.

But setting a good M-S balance is easy. Just listen! Take it as wide as you need for
realism, but avoid that out-of-phase speaker sound when S is too high.

BTW, assuming you're doing a two-fader control of S (S+ and S-), are you doing an S
"calibration"?

Assuming you're using two faders for S (S+ and S-), you can calibrate them like
this:

1. Temporarily pan S+ and S- to exactly dead center.

2. Play your track. Adjust level, EQ, and anything else in the two signal paths so
that you get maximum null. Crank up the monitors so you can really hear and get to a
complete null. If you're doing this on an analog console, don't rely on the markings
of any control. They're all going to be slightly off. You have to null by ear. Even
with the monitors all the way up, you shouldn't hear a thing. Careful, the
slightest mis-alignment can suddenly cause a lot of level to appear!

3. Turn down the monitors to a normal level, spead out the S+ and S- pans to hard
left and right again.

If you'd been out of calibration before, you might notice now that the whole thing
sounds "snappier" and more real, with a much better image.

I typically decoded S with three faders: raw S into the first, with an aux send post
fader/post EQ (or a direct out), unity gain, split into two additional channel
strips accepting the input.

The raw S channel was *not* assigned to the mix bus, only the aux send. The S+ and
S- faders were unity gain, panned hard left and right, with the right channel
flipped 180. After the calibration step, I now had single fader control of S (the
raw S channel), with no need to re-cal everytime I changed the level of S
in the mix.

You can also do this "asymmetrically" with that send from the raw S channel going to
the S- strip and assigning the raw S channel to the mix bus as S+, panned hard left.
This can be a little trickier to calibrate, depending on the gear.

If you're doing this in a DAW, you might need to adjust for any time mis-alignments
that crop up. The advantage of the three-fader S decode is that you'd only
then need to look at time misalignments between M and S, not S+, S-, *and* M.

Frank
Mobile Audio


--

[Gary Eickmeier]

Frank - this is very interesting and I will play with it and see if I get
it. But obviously you are talking about this nulling procedure with M muted,
right? Also, can't you see the muting on the resulting file that their
output produces if you mix them down?

I have wondered about the difference in FR between the M and S signals when
you aren't using them both on the figure 8 pattern. On this particular show
I am hearing a screechier high end that I didn't hear in the testing phase.
Could be my hearing with a cold and sinuses stopped up, but I am not
grooving in it yet.

Gary Eickmeier


"Frank Stearns" > wrote in message
acquisition...
> "Peter Larsen" > writes:
>
>>Gary Eickmeier wrote:
>
>>> Thanks Marc but I have covered my MS mixing technique in previous
>>> posts. I can mix in any amount of M that I want, but I find that if
>>> there is too much S you get too wide a soundstage with hole in the
>>> middle. I do about equal M and S and that seems to be fhe best..
>
>>Yes, the probable range is S equal to or no more than 5 dB less than M.
>
>>> Gary
>
> Careful of doing M-S mixing by specific "formulas." I say this because S
> will
> often wind up being a somewhat different dynamic envelope. S consists of
> more
> diffused, reverberant room sound rather than direct sound. (That is, S
> will often
> be lower than M in overall level.)
>
> These suggested ratios from Peter and others are probably fine -- once you
> take into
> account what might be different record levels between M and S.
>
> But setting a good M-S balance is easy. Just listen! Take it as wide as
> you need for
> realism, but avoid that out-of-phase speaker sound when S is too high.
>
> BTW, assuming you're doing a two-fader control of S (S+ and S-), are you
> doing an S
> "calibration"?
>
> Assuming you're using two faders for S (S+ and S-), you can calibrate them
> like
> this:
>
> 1. Temporarily pan S+ and S- to exactly dead center.
>
> 2. Play your track. Adjust level, EQ, and anything else in the two signal
> paths so
> that you get maximum null. Crank up the monitors so you can really hear
> and get to a
> complete null. If you're doing this on an analog console, don't rely on
> the markings
> of any control. They're all going to be slightly off. You have to null by
> ear. Even
> with the monitors all the way up, you shouldn't hear a thing. Careful,
> the
> slightest mis-alignment can suddenly cause a lot of level to appear!
>
> 3. Turn down the monitors to a normal level, spead out the S+ and S- pans
> to hard
> left and right again.
>
> If you'd been out of calibration before, you might notice now that the
> whole thing
> sounds "snappier" and more real, with a much better image.
>
> I typically decoded S with three faders: raw S into the first, with an aux
> send post
> fader/post EQ (or a direct out), unity gain, split into two additional
> channel
> strips accepting the input.
>
> The raw S channel was *not* assigned to the mix bus, only the aux send.
> The S+ and
> S- faders were unity gain, panned hard left and right, with the right
> channel
> flipped 180. After the calibration step, I now had single fader control of
> S (the
> raw S channel), with no need to re-cal everytime I changed the level of S
> in the mix.
>
> You can also do this "asymmetrically" with that send from the raw S
> channel going to
> the S- strip and assigning the raw S channel to the mix bus as S+, panned
> hard left.
> This can be a little trickier to calibrate, depending on the gear.
>
> If you're doing this in a DAW, you might need to adjust for any time
> mis-alignments
> that crop up. The advantage of the three-fader S decode is that you'd only
> then need to look at time misalignments between M and S, not S+, S-, *and*
> M.
>
> Frank
> Mobile Audio
>
>
> --
> .

[Gary Eickmeier]

"Tom McCreadie" > wrote in message
...
> Gary Eickmeier wrote:
>
>>I have done the deed now - MS with the M channel cardioid. It wasn't
>>totally
>>what I expected. I did everything right, but the imaging isn't quite as
>>precise as I expected.
>
> Did the imaging actually seem precise to you while listening live at the
> venue,
> sitting under the mics (ideally with eyes closed)? Wwas your home
> evaluation
> via two- or multi-speaker playback? [Bounce it Over nearby Surfaces
> Equipment,
> right? :-)] And how sharp was that home playback via headphones? Got any
> commercial cd's or recordings from other people that imaged significantly
> sharper?
>
> Maybe you're just coming up against the limitations of those mics...say, a
> ragged off-axis response...or modest capsule matching, particularly
> between the
> two back-to-back cardioids that get used to emulate the Fig-8... or mics
> in the
> array insufficiently 'coincident' w.r.t. the main sound arrival direction,
> etc.
> But there's no reason for not trying to tweak the wimpy low end a little
> in
> post... provided you're not then exacerbating any unwanted stuff like
> air-conditioning growl, stand-borne resonance's etc.

Well, the imaging was fine enough live in the front row center. I am playing
back in DPL II on my surround system. It also sounds more screechy than
before, but I have a head cold right now so waiting final judgement til U
get better. But wonder if it could be a lousy combination of the FR of the
cardioid and the fig 8 patterns for this microphone. Audio Technica 2050.

Gary Eickmeier

[Scott Dorsey]

Gary Eickmeier > wrote:
>
>I have wondered about the difference in FR between the M and S signals when
>you aren't using them both on the figure 8 pattern. On this particular show
>I am hearing a screechier high end that I didn't hear in the testing phase.
>Could be my hearing with a cold and sinuses stopped up, but I am not
>grooving in it yet.

When the responses of the two microphones do not match, the end result is
a difference in response across the sound field. That is, if your S mike
is reduced in top end, then high frequency sounds seem to be pulled more
toward the center of the soundstage than they normally would be.

However, I suggest that your experiments may not translate in any way to
anyone else's playback system.
--scott

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

[Scott Dorsey]

Gary Eickmeier > wrote:
>Well, the imaging was fine enough live in the front row center. I am playing
>back in DPL II on my surround system. It also sounds more screechy than
>before, but I have a head cold right now so waiting final judgement til U
>get better. But wonder if it could be a lousy combination of the FR of the
>cardioid and the fig 8 patterns for this microphone. Audio Technica 2050.

If you are taking a minimalist recording and playing it back through a
Dolby decoder, you are basically mutilating your stereo image and all bets
are off about what the end result will sound like but I guarantee it won't
be natural and I guarantee that you're going to be hearing stuff in the
surrounds that you would never hear coming from behind you in a real hall.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Tom McCreadie]

Frank Stearns wrote:

>BTW, assuming you're doing a two-fader control of S (S+ and S-), are you doing an S
>"calibration"?
>
>Assuming you're using two faders for S (S+ and S-), you can calibrate them like
>this:
>
>1. Temporarily pan S+ and S- to exactly dead center.
>
>2. Play your track. Adjust level, EQ, and anything else in the two signal paths so
>that you get maximum null. Crank up the monitors so you can really hear and get to a
>complete null. If you're doing this on an analog console, don't rely on the markings
>of any control. They're all going to be slightly off. You have to null by ear. Even
>with the monitors all the way up, you shouldn't hear a thing. Careful, the
>slightest mis-alignment can suddenly cause a lot of level to appear!
>
>3. Turn down the monitors to a normal level, spead out the S+ and S- pans to hard
>left and right again.

Frank, unless one needs to work in real time with a hardware mixer and perhaps
no computer, this S+, S- procedure seems imho unnecessarily convoluted
nowadays. For the simple Sum & Difference decode arithmetic can be quickly and
precisely done on many software apps or plugins.

Also do I see a slight circular illogic in your above description?: In step 2
you caution: "don't rely on the markings of any control", yet in step 1 you say:
"pan S+ and S- to exactly dead center". But how then are we to determine with
certainty when we have hit this exact dead center - surely not merely the
detente position? Moreover, given the tolerances of typical pan pots, do these
balanced S+ and S- signals retain their hard-won level balance when subsequently
panned full L and R?

[Mike Rivers[_2_]]

On 12/20/2013 11:36 AM, Gary Eickmeier wrote:
> But obviously you are talking about this nulling procedure with M muted,
> right? Also, can't you see the muting on the resulting file that their
> output produces if you mix them down?

The only time I've done "nulling" (and I do it every time) is when I'm
recording left/right using an M-S mic setup, a splitter on the side mic,
and a mixer. You can usually trust the faders to track close enough for
jazz. Pan both the S and -S to the same side, set the two channel faders
to the same position, usually "unity gain," and use the channel trims to
null the output. Then pan the side channels to full left and right and
adjust the faders so it sounds right.

Take into account that you're going to hear more separation in the
headphones than you will in speakers, so don't worry if it sounds too
wide, just be sure that you don't hear a hole in the middle (too little
mid) or essentially mono (too much mid).

If you're recording mid and side channels separately and then mixing
them in a DAW, you don't need to fiddle with nulling anything. Trust the
math to do the inversion correctly, set up the +S and -S faders as a
stereo pair, and trust the software to adjust them together accurately.

Does the Zoom H6n have a feature that lets you record M and S separately
but monitor in left/right using a preset M/S ratio? If so, use it.
--
For a good time, visit http://mikeriversaudio.wordpress.com

[Frank Stearns]

Tom McCreadie > writes:

>Frank Stearns wrote:

>>BTW, assuming you're doing a two-fader control of S (S+ and S-), are you doing an S
>>"calibration"?
>>
>>Assuming you're using two faders for S (S+ and S-), you can calibrate them like
>>this:
>>
>>1. Temporarily pan S+ and S- to exactly dead center.
>>
>>2. Play your track. Adjust level, EQ, and anything else in the two signal paths so
>>that you get maximum null. Crank up the monitors so you can really hear and get to a
>>complete null. If you're doing this on an analog console, don't rely on the markings
>>of any control. They're all going to be slightly off. You have to null by ear. Even
>>with the monitors all the way up, you shouldn't hear a thing. Careful, the
>>slightest mis-alignment can suddenly cause a lot of level to appear!
>>
>>3. Turn down the monitors to a normal level, spead out the S+ and S- pans to hard
>>left and right again.

>Frank, unless one needs to work in real time with a hardware mixer and perhaps
>no computer, this S+, S- procedure seems imho unnecessarily convoluted
>nowadays. For the simple Sum & Difference decode arithmetic can be quickly and
>precisely done on many software apps or plugins.

Correct -- if you have that available on location. I probably should have expanded
the explanation of my "conditions" for all this.

When I was using M-S on a regular basis (before switching to AB), I always tracked
the M and S mics "raw", undecoded, but had a special balanced Y cable to feed the S
channel from the recorder output to the location monitor mixer. (I never use a
computer for location tracking -- it's always a dedicated recorder and separate
monitor mixer. Actually, it's a set of twin recorders so that I am laying down data
on two different drives from the get-go.)

One of the Y legs did the polarity flip for S- right in the connector. I brought
those in on a stereo input of the monitor mixer (balanced inputs, obviously). I
didn't worry too much about the "quality" of the decode there, it was just for
monitoring.

But during post is when the more exacting calibration took place.

>Also do I see a slight circular illogic in your above description?: In step 2
>you caution: "don't rely on the markings of any control", yet in step 1 you say:
>"pan S+ and S- to exactly dead center". But how then are we to determine with
>certainty when we have hit this exact dead center - surely not merely the
>detente position? Moreover, given the tolerances of typical pan pots, do these
>balanced S+ and S- signals retain their hard-won level balance when subsequently
>panned full L and R?

Correct again. It was always a bit of a guess with hardware; you went for the best
null you could, tapping a control here, barely touching another other there. Not
sure why, but minor deviations in pan pots seemed slightly more forgiving than, say,
errors in the channel gain (trim or faders) or EQ. Plus, you could always pre-check
pan pot accuracy on the S+ and S- inputs. One at a time, feed a signal, pan center,
and make sure your metering was exactly equal. Adjust the pan pot position if
needed. I was always lucky with the little Yamahas and Soundcraft; center pan
accuracy was pretty good.

As far as spreading back out to hard L/R following the nulling... at those extreme
L/R positions I would expect the least amount of error -- the wiper is all the way
"against the metal". Should be zero (or near zero) ohms from the wiper to either the
L or R output bus.

Thanks for pointing out those items; glad to clarify.

Frank
Mobile Audio

--

[Frank Stearns]

"Gary Eickmeier" > writes:

>Frank - this is very interesting and I will play with it and see if I get
>it. But obviously you are talking about this nulling procedure with M muted,

Yes, mute M while checking S+ and S-. Sorry for the omission.

>right? Also, can't you see the muting on the resulting file that their
>output produces if you mix them down?

You mean meter the nulling rather than listen? Sure, depends on how your meters are
scaled. If -20 to +3, that's not good enough. I'd want to meter down to at least
-60 before trying this.

But, it really is useful to listen, especially if you have the same EQ applied to S+
and S- and are trying to get the EQ matched. You can hear where the null is failing,
and tweak the appropriate EQ. (And yes, it's a little tricky to do that,
especially when the Q is a bit off.)


>I have wondered about the difference in FR between the M and S signals when
>you aren't using them both on the figure 8 pattern. On this particular show
>I am hearing a screechier high end that I didn't hear in the testing phase.
>Could be my hearing with a cold and sinuses stopped up, but I am not
>grooving in it yet.

Indeed. Plugged sinuses can really mess up ear response!

Frank
Mobile Audio
--

[Gary Eickmeier]

"Scott Dorsey" > wrote in message
...
> Gary Eickmeier > wrote:
>>Well, the imaging was fine enough live in the front row center. I am
>>playing
>>back in DPL II on my surround system. It also sounds more screechy than
>>before, but I have a head cold right now so waiting final judgement til U
>>get better. But wonder if it could be a lousy combination of the FR of the
>>cardioid and the fig 8 patterns for this microphone. Audio Technica 2050.
>
> If you are taking a minimalist recording and playing it back through a
> Dolby decoder, you are basically mutilating your stereo image and all bets
> are off about what the end result will sound like but I guarantee it won't
> be natural and I guarantee that you're going to be hearing stuff in the
> surrounds that you would never hear coming from behind you in a real hall.
> --scott

Yes, probably correct. Especially with these MS recordings with a lot of
channel difference. Anyway, my adventure for the Christmas concert turned
out fine, I just had a bad head cold when I first played it back. Now the
only aspect that I question is the loudness, because it is fine at home but
I can hardly get it loud enough in the car. Need to learn compression big
time. I have the article (series of 3 articles) from Recording magazine, but
he was talking about only outboard analog compressors and I need to learn it
on a DAW (Audition 2). I have a book on Audition, but it isn't big on
compression.

Gary Eickmeier

[Scott Dorsey]

Ty Ford > wrote:
>On Thu, 12 Dec 2013 20:29:38 -0500, Mike Rivers wrote
>(in article >):
>
>> On 12/12/2013 5:36 PM, Scott Dorsey wrote:
>>>>> PS I see they have the Zoom H6 in Recording magazine, but they do no
>>>>> technical testing on it whatsoever. What use is that? We can see all of
>>>>> its
>>>>> features at the web site.
>>
>>> Write a letter of complaint to the editors, tell them you want real
>>> measurements in reviews.
>>
>> I put technical measurements in my reviews, usually with some
>> explanation of why the measurements are significant. Unfortunately the
>> editors who pay won't take them because they're too long. I'm thinking
>> about putting a virtual tip jar on my web site.
>
>
>I haven't gotten rich off my tip jar yet. :)
>
>YMMV.
>
>I do publish specs on my blog. btw, the trade mag folks I talk to say ditch
>the specs, readers can always find them on line.

Specs are not measurements.

Measurements are what reviewers do to explain and verify the specs.
--scott

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