Stereo Microphone Placement

[Fred McKenzie]

I am attempting to set up a couple of microphones to record an Orchestra
performance. In order to achieve a stereo effect, it was my impression
that the microphones should be spaced several feet apart, one on each
side of the center of the stage.

I came across an article that placed the two microphones together with
front apertures almost touching in a "cross-eyed" configuration. The
right microphone was pointed to the left and vice versa.

I can see where it might be bad if microphones were too far apart. Our
ears are fairly close together and hear things in stereo. Does it make
sense to locate the microphones closer together than our ears are spaced?

Fred

[Ron C]

There's a lot of art that goes into microphone selection
and placement. There are hundreds of microphones
to choose from as well as dozens of common microphone
setups. You might want to post this question to the
rec.audio.pro newsgroup. There are a lot of people
there with years of live and studio recording experience.

==
Later...
Ron Capik
--

[isw]

In article <fmmck-61D296....@5ad64b5e.bb.sky.com>,

Google on "blumlein configuration".

Isaac

[Fred McKenzie]

In article <isw-7F1E16.21300119082014@[216.168.3.50]>,

isw <i...@witzend.com> wrote:

> > I came across an article that placed the two microphones together with
> > front apertures almost touching in a "cross-eyed" configuration. The
> > right microphone was pointed to the left and vice versa.
> >
> > I can see where it might be bad if microphones were too far apart. Our
> > ears are fairly close together and hear things in stereo. Does it make
> > sense to locate the microphones closer together than our ears are spaced?
>
> Google on "blumlein configuration".

Isaac-

Thanks for that. I found a couple of interesting articles on the
Blumlein configuration. I was not familiar with "figure 8" microphones.

In my situation, I already have some decent cardioid microphones. The
equivalent configuration for directional microphones was called "X-Y" in
one of the articles.

There was also mention of spacing microphones seven inches apart, left
pointing left and right pointing right at a 90 degree angle. This
agrees with my thinking that they should approximate human hearing.

Both configurations are worth trying. I can do my experimenting during
rehearsals.

Ron - Thanks for mentioning rec.audio.pro. I have added that group to
my list, and will check it out.

Fred

[Scott Dorsey]

Fred McKenzie <fm...@aol.com> wrote:
>I am attempting to set up a couple of microphones to record an Orchestra
>performance. In order to achieve a stereo effect, it was my impression
>that the microphones should be spaced several feet apart, one on each
>side of the center of the stage.

This is called A-B stereo. It gives you some intensity imaging but no phase
imaging because the phase differences between the channels are too great for
the brain to make sense of them. It was very popular back in the 1950s and
1960s when good directional microphones did not exist and omnis were the
order of the day.

>I came across an article that placed the two microphones together with
>front apertures almost touching in a "cross-eyed" configuration. The
>right microphone was pointed to the left and vice versa.

This is an X-Y configuration, which is used with directional microphones.
Again, there is no phase imaging, only intensity imaging. However, the
phase differences between channels are eliminated completely which makes
FM broadcast and LP cutting much easier.

>I can see where it might be bad if microphones were too far apart. Our
>ears are fairly close together and hear things in stereo. Does it make
>sense to locate the microphones closer together than our ears are spaced?

Go right now and purchase Streicher's book _The New Stereo Soundbook_ which
is a good basic introduction to stereophony.

There are a lot of different ways to get good stereo imaging. The one that
is appropriate depends a lot on the room and the kind of microphones that you
have available.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Scott Dorsey]

Fred McKenzie <fm...@aol.com> wrote:
>In my situation, I already have some decent cardioid microphones. The
>equivalent configuration for directional microphones was called "X-Y" in
>one of the articles.

X-Y is also called 'coincident cardioids.'

>There was also mention of spacing microphones seven inches apart, left
>pointing left and right pointing right at a 90 degree angle. This
>agrees with my thinking that they should approximate human hearing.

This is a typical near-coincident method. This gives you some phase
differences between channels and therefore better low end imaging than
you would get from X-Y. Don't get hung up on the exact settings, though,
because the room affects what may be optimal. The standard ORTF and NOS
methods are good starting points but do not take them as gospel.

The bad news is that these methods are a bit less mono-compatible. NPR
doesn't like near-coincident stuff.

>Both configurations are worth trying. I can do my experimenting during
>rehearsals.

Yes, but realize that if the rehearsal isn't in the same hall, things will
be different.

[Fred McKenzie]

In article <lti7pd$o28$1...@panix2.panix.com>,

klu...@panix.com (Scott Dorsey) wrote:

> >Both configurations are worth trying. I can do my experimenting during
> >rehearsals.
>
> Yes, but realize that if the rehearsal isn't in the same hall, things will
> be different.

Scott-

I am trying to record what a person would hear if they were at the same
spot as the microphones, realizing that there are differences.

At a recent concert, I had the two microphones about 10 inches apart,
but pointing in almost the same direction. They were about 7 feet above
the stage floor, and about 5 feet left of center. I am very happy with
the results as far as frequency response is concerned. I am not happy
with some instruments not being heard as loudly as expected, and the
lack of stereo effect.

From all I've read, the ORTF system looks like it will be the best
approach. I plan to space the microphones about 6 inches, pointing 90
degrees apart. To keep them from bumping into each other, it looks like
one will need to mounted "upside down" on the bracket, with the bracket
tilted to keep them at the same height. I'm still working on mounting
details.

I have acquired a taller microphone stand with a boom attachment that
will take the height to about 8 feet, and will place it directly behind
the conductor.

Fred

[Scott Dorsey]

Fred McKenzie <fm...@aol.com> wrote:
>
>I am trying to record what a person would hear if they were at the same
>spot as the microphones, realizing that there are differences.

Stop that. That's not how stereophony works.

That _is_ how binaural recording works. But stereophony is a trick, and
the microphone does not pick up sound in the same way the ear does. As a
consequence, the placement of the microphones to get a given sound is going
to be different than where you would get that same sound in the room.

And, the placement will depend on the miking technique itself. If you want
to get a given sense of space with a given ratio of direct to ambient sound,
a baffled omni pair will have to be placed a lot closer to the source than
a coincident cardioid pair.

This can be to your advantage if you are limited in the locations where you
can place mikes or you have audience noise issues or you have particular
bad spots in the room due to flutter echoes, etc.

>At a recent concert, I had the two microphones about 10 inches apart,
>but pointing in almost the same direction. They were about 7 feet above
>the stage floor, and about 5 feet left of center. I am very happy with
>the results as far as frequency response is concerned. I am not happy
>with some instruments not being heard as loudly as expected, and the
>lack of stereo effect.

What kind of microphones were they, with what pattern? This would indeed
give you little to no stereo effect above 1 Khz but some stereo imaging
at low frequencies due to the phase difference between channels.

7 feet is not very high, so you are apt to get mostly the instruments at
the front of the orchestra. As you raise the pair up more and more, you
will get more of the instruments at the rear of the orchestra. There is
some angle at which the fiddles will become prominent and then as it increases
or decreases they will be reduced, since they are comparatively directional.
By adjusting the height you can adjust these balances to some extent. By
moving closer or farther back you can adjust the ratio of direct to ambient
sound and change the stereo image and sense of space.

>From all I've read, the ORTF system looks like it will be the best
>approach. I plan to space the microphones about 6 inches, pointing 90
>degrees apart. To keep them from bumping into each other, it looks like
>one will need to mounted "upside down" on the bracket, with the bracket
>tilted to keep them at the same height. I'm still working on mounting
>details.

If you have good cardioids, ORTF is a good way to go and it is a good
starting point to learn mike placement. You can buy a standard stereo
bar from any pro audio place that will allow you to set up various near
coincident configurations.

>I have acquired a taller microphone stand with a boom attachment that
>will take the height to about 8 feet, and will place it directly behind
>the conductor.

That's not very high. You will probably want to get farther back and
higher with an ORTF pair to get a good sense of space. It is not unusual
to crank microphone pairs up 30 or 40 feet in a big hall or to suspend them
from a ceiling or catwalk.

If you are forced to place microphones that closely you might consider
baffled omnis.

[Fred McKenzie]

In article <ltt1ka$ivq$1...@panix2.panix.com>,

klu...@panix.com (Scott Dorsey) wrote:

> If you have good cardioids, ORTF is a good way to go and it is a good
> starting point to learn mike placement. You can buy a standard stereo
> bar from any pro audio place that will allow you to set up various near
> coincident configurations.
>
> >I have acquired a taller microphone stand with a boom attachment that
> >will take the height to about 8 feet, and will place it directly behind
> >the conductor.
>
> That's not very high. You will probably want to get farther back and
> higher with an ORTF pair to get a good sense of space. It is not unusual
> to crank microphone pairs up 30 or 40 feet in a big hall or to suspend them
> from a ceiling or catwalk.

Scott-

The two microphones are Shure model SM81 cardioid.

I have a bracket that you might call a "stereo bar", purchased from a
music store. It mounts on a microphone stand, and has mounting places
for five microphones spaced over nine inches. (I tried upside-down
mounting one of the microphones, but it did not work out.)

Clearly I have a lot to learn. I did not realize the microphone height
was so important. I was trying to just get higher than the conductor's
head. I am trying to set up a relatively simple system, and probably
would not be able to go as high as you suggest.

I appreciate your comments and suggestions.

Fred

[Scott Dorsey]

Fred McKenzie <fm...@aol.com> wrote:
>
>The two microphones are Shure model SM81 cardioid.

You could do much worse, though you could surely do better too. The SM81
came out in 1972 and it was probably the first reliable back-electret mike
sold. It's not the flattest thing ever off-axis, but it is amazingly rugged.
You could run a truck over it.

>I have a bracket that you might call a "stereo bar", purchased from a
>music store. It mounts on a microphone stand, and has mounting places
>for five microphones spaced over nine inches. (I tried upside-down
>mounting one of the microphones, but it did not work out.)

That'll be fine. You can get nicer calibrated ones, but that will do you
perfectly well.

>Clearly I have a lot to learn. I did not realize the microphone height
>was so important. I was trying to just get higher than the conductor's
>head. I am trying to set up a relatively simple system, and probably
>would not be able to go as high as you suggest.

Stick a finger in one ear and listen only with the other ear, and you can
get some basic sense of what the mike pair is getting. Walk around the
hall listening like that.... raise your head and lower it... climb up on
a ladder, get a sense of how the sound changes as you move around the room.

You'll find that a change of only a foot or so can make a dramatic tonal
change sometimes. And sometimes in a great room you can move halfway across
the hall without much tonal change.

>I appreciate your comments and suggestions.

Get the New Stereo Soundbook! Get an assistant to move a mike pair around
while you listen in an improvised sound booth over a pair of speakers. Don't
try to judge the direct/ambient ratio on headphones! Keep listening and
have fun!

[Fred McKenzie]

In article <lu75ik$d0j$1...@panix2.panix.com>,

klu...@panix.com (Scott Dorsey) wrote:

> Stick a finger in one ear and listen only with the other ear, and you can
> get some basic sense of what the mike pair is getting. Walk around the
> hall listening like that.... raise your head and lower it... climb up on
> a ladder, get a sense of how the sound changes as you move around the room.
>
> You'll find that a change of only a foot or so can make a dramatic tonal
> change sometimes. And sometimes in a great room you can move halfway across
> the hall without much tonal change.
>
> >I appreciate your comments and suggestions.
>
> Get the New Stereo Soundbook! Get an assistant to move a mike pair around
> while you listen in an improvised sound booth over a pair of speakers. Don't
> try to judge the direct/ambient ratio on headphones! Keep listening and
> have fun!

Scott-

At this week's rehearsal the microphones were configured as close to
ORTF as I could manage. There was a fairly good stereo effect, with
instruments and voices coming from the correct directions. (I am
listening on a stereo system with speakers.)

But the microphones were only about 4 feet from the wall. The drums
seemed to dominate the sound in a different way than in the concert
recording a few weeks ago. I'm hoping the sound will be a little more
balanced in the auditorium.

The New Stereo Soundbook seems expensive, unless you compare it with the
price of microphones!

Fred

[Scott Dorsey]

Fred McKenzie <fm...@aol.com> wrote:
>At this week's rehearsal the microphones were configured as close to
>ORTF as I could manage. There was a fairly good stereo effect, with
>instruments and voices coming from the correct directions. (I am
>listening on a stereo system with speakers.)

So far so good.

>But the microphones were only about 4 feet from the wall. The drums
>seemed to dominate the sound in a different way than in the concert
>recording a few weeks ago. I'm hoping the sound will be a little more
>balanced in the auditorium.

What wall? The rear wall? A side wall? In general if you get close to
walls you're going to get low end changes due to reflections from the wall
forming a comb filter. You can use this to your advantage sometimes, but
sometimes it's a problem.

>The New Stereo Soundbook seems expensive, unless you compare it with the
>price of microphones!

$68 is a lot of money for a book, but it's a lot of book, and it's worth
every penny.

[Walter Bushell]

In article <lti7pd$o28$1...@panix2.panix.com>,

And microphone placement is an art not a science. And good
microphones for music are not cheap.

--
Never attribute to stupidity that which can be explained by greed. Me.

[Fred McKenzie]

In article <proto-A61F45....@news.panix.com>,

Walter-

I did some research on microphones before purchasing the Shure SM81
microphones. They had a good reputation, with frequency response
relatively flat from 20 Hz to 20 KHz. I thought they were expensive,
until finding others with prices into the thousands of dollars.

The first recordings sounded much like the Radio Shack dynamic
microphones until I heard the bass drum being caressed by a soft mallet!
One drawback is that the bass drum seems to dominate.

The "boomy" bass drum appeared during a test recording in rehearsal. At
the time I attributed it to the room, but it also shows up in concert
recordings. The effect may not be due to microphone placement. It is
not noticeable when listening to a system with small speakers, so may
actually be a problem with my amplifier/speaker system!

Fred

[Scott Dorsey]

Walter Bushell <pr...@panix.com> wrote:
>
>And microphone placement is an art not a science. And good
>microphones for music are not cheap.

This is true, although I will say that good microphones are far, far cheaper
than they were 25 years ago.

As far as it being an art, that's what makes it fun. And you know, you
can always hire me... I am very reasonable and come with a remote truck
and a nice mike closet...

[Scott Dorsey]

Fred McKenzie <fm...@aol.com> wrote:
>
>I did some research on microphones before purchasing the Shure SM81
>microphones. They had a good reputation, with frequency response
>relatively flat from 20 Hz to 20 KHz. I thought they were expensive,
>until finding others with prices into the thousands of dollars.

They are very good entry level condensers. Shure designed them well and
has production down very consistent. Unfortunately the SM80 omni has been
discontinued; I would have recommended them over the SM81 except that Shure
never could figure out how to sell the things.

If you look in the Jan 2007 issue of Recording magazine, I published a
modification for the SM-81 that you can perform which will improve the top
end (both frequency and phase response as well as pattern) at the expense
of making the mike more delicate and easier to pop with close sources. It
is fairly delicate work and it can't be undone once you have done it, though.

>The first recordings sounded much like the Radio Shack dynamic
>microphones until I heard the bass drum being caressed by a soft mallet!
>One drawback is that the bass drum seems to dominate.
>
>The "boomy" bass drum appeared during a test recording in rehearsal. At
>the time I attributed it to the room, but it also shows up in concert
>recordings. The effect may not be due to microphone placement. It is
>not noticeable when listening to a system with small speakers, so may
>actually be a problem with my amplifier/speaker system!

If you don't have a good accurate playback system, you don't have any way of
telling if your recording system is any good or not. Play back a tone ladder,
walk around the room, and just listen to hear if the low end tones stay at
about the same level. If they don't, you may be in for some room treatement.

[Anton Shepelev]

Scott Dorsey to Fred McKenzie:

> > I am attempting to set up a couple of micro-

> > phones to record an Orchestra performance. In

> > order to achieve a stereo effect, it was my im-

> > pression that the microphones should be spaced

> > several feet apart, one on each side of the cen-

> > ter of the stage.
>
> This is called A-B stereo.

Only if the microphones are omnidirectional and
placed closer together, usually about 20-30 cm,
which is about the ear-to-ear distance. The musi-
cians are positioned within a 40-degree angle. One
may achieve it by varying the distance.

Other combinations of distance and microphone spac-
ing are possible, but then you risk to get nasty
comb filtering or poor spatial impression:

S = L * sqrt( 1 + ( 2d/w )^2 )

where S is microphone spacing, d the distance to the
source, w the width of the source, and L the
wavepath that crates a phase difference perceived as
the location of the source directly to the left or
right. It may be estimted from the recommendation
above:
25 * sin 20 = 8.55 (cm)

> It gives you some intensity imaging but no phase
> imaging because the phase differences between the
> channels are too great for the brain to make sense
> of them. It was very popular back in the 1950s
> and 1960s when good directional microphones did
> not exist and omnis were the order of the day.

True A-B technique (with smaller spacing) is purely
phase-based stereo, which is the only right kind of
stereo because it imitates human hearing. Another
great phase-based technique is SASS, which does not
suffer from the A-B limitations mentioned above.
But unfortunatly the overwhelming majority of
mordern recordings are made using polymicrophone
technique and consequenty intensity-based stereo.

--
() ascii ribbon campaign - against html e-mail
/\ http://preview.tinyurl.com/qcy6mjc [archived]

[Fred McKenzie]

In article <20170111162354.f377...@gmail.com>,

Anton Shepelev <anto...@gmail.com> wrote:

> True A-B technique (with smaller spacing) is purely
> phase-based stereo, which is the only right kind of
> stereo because it imitates human hearing.

Anton-

Since posting my earlier request, I settled on an approximation of this
"ORTF" method. I am using a pair of Shure SM81 Cardioid microphones. I
probably do not have the angles and separation exactly right, but have
been satisfied with the results.

One thing I found was that the SM81 microphones are extremely sensitive
to mechanical disturbance. Mounted on an extended tall stand, motion
induced by either flexing of the wooden floor or a breeze from the air
handler, resulted in a loud rumble. Apparently the microphones or
connectors were rubbing against each other! Fortunately there is a
Shure vibration isolator that fits the SM81.

The loud rumble was improved using a 55 Hz High Pass filter, but at the
expense of Tympani and Bass Drum levels.

Fred

[Scott Dorsey]

Fred McKenzie <fm...@aol.com> wrote:
>
>Since posting my earlier request, I settled on an approximation of this
>"ORTF" method. I am using a pair of Shure SM81 Cardioid microphones. I
>probably do not have the angles and separation exactly right, but have
>been satisfied with the results.

So try altering them to get a sense of what it does to the sound.

>One thing I found was that the SM81 microphones are extremely sensitive
>to mechanical disturbance. Mounted on an extended tall stand, motion
>induced by either flexing of the wooden floor or a breeze from the air
>handler, resulted in a loud rumble. Apparently the microphones or
>connectors were rubbing against each other! Fortunately there is a
>Shure vibration isolator that fits the SM81.

The vibration mount is a very good thing, also if possible it is wise to
use a very flexible cable going into the microphone to reduce vibration
transmitted through the cable. Olson will sell you a very good windscreen
for wind issues.

If you think the SM81 is sensitive to this kind of thing, you should see
what the DPA omnis are like.

>The loud rumble was improved using a 55 Hz High Pass filter, but at the
>expense of Tympani and Bass Drum levels.

Find out where it's coming from and fix it. If it's an air handler issue,
it's possible moving a few feet will deal with it.

[Anton Shepelev]

Fred McKenzie to Anton Shepelev:

> > True A-B technique (with smaller spacing) is
> > purely phase-based stereo, which is the only
> > right kind of stereo because it imitates human
> > hearing.
>

> Since posting my earlier request, I settled on an
> approximation of this "ORTF" method. I am using a

> pair of Shure SM81 Cardioid microphones. I proba-

> bly do not have the angles and separation exactly
> right, but have been satisfied with the results.

Glad to know that. It will be intensity-based
stereo. May I listen to a fragment of your record-
ing, unprocessed and preferably losseless?

[Fred McKenzie]

In article <o55vi1$bk1$1...@panix2.panix.com>,

It took several sessions to realize that the problem was the result of
mechanical motion. In the worst case, the microphone connectors were
definitely touching each other. Vibration isolators along with making
sure the connectors and cables do not touch, seems to have completely
eliminated the problem.

Fred

[Anton Shepelev]

Fred McKenzie:

> I am trying to record what a person would hear if

> they were at the same spot as the microphones, re-
> alizing that there are differences.

A very commendable goal, which deplorably few sound
engineers strive for. Binaural (dummy head) record-
ing is the best way to achieve it, but the recordigs
must be listened to either via headphones or a con-
ventional stereo system supplied with a binaural
processor, which removes cross-feeding (right speak-
er to left ear and vice versa).

> At a recent concert, I had the two microphones
> about 10 inches apart, but pointing in almost the
> same direction.

Were they the cardioid SM81s?

> They were about 7 feet above the stage floor, and
> about 5 feet left of center. I am very happy with

> the results as far as frequency response is con-
> cerned.

Note that cardioid will attenuate the lower frequen-
cies when placed far from the sound source.

> I am not happy with some instruments not being

> heard as loudly as expected ->

Do you mean as the listener would percieve their
loudness at the microphone position? That may have
to do with directivity and be amendable by orienting
directional mics or using omnis.

> -> and the lack of stereo effect

How far away was the (actual) scene and how wide?
Try increasing the spacing to 20 inches. It may
help.

Try to compare your recording with the sound of ei-
ther channel in mono using headphones. This con-
trast helps to perceive even a small stereo effect
if it is there, but only for AB stereo, which in
your case means parallel microphones.

One advantage of time-based stereo is that it will
make individual instruments more discernible by bin-
aural demasking.

[Fred McKenzie]

In article <20170113001914.2e0a...@gmail.com>,

Anton Shepelev <anto...@gmail.com> wrote:

> > At a recent concert, I had the two microphones
> > about 10 inches apart, but pointing in almost the
> > same direction.
>
> Were they the cardioid SM81s?

I forget how long it has been since this thread started. My memory is
getting old, but I believe I was using the SM81s then. They are now
arranged at an angle of about 90 degrees.

I use an adapter that would allow up to five microphones to be mounted
on one stand. Two SM-81s are on the outside, with a Zoom H4N recorder
in the center. The SM-81s are now each mounted in a Shure vibration
isolator, with an 18 inch cable connecting between each microphone and
the recorder.

Needless to say, I am not a pro! My recordings are intended to be used
by the Conductor and a few key Musicians to analyze the performance.

In article <20170112193528.971a...@gmail.com>,

Anton Shepelev <anto...@gmail.com> wrote:

> Since posting my earlier request, I settled on an
> > approximation of this "ORTF" method. I am using a

> > pair of Shure SM81 Cardioid microphones. I proba-

> > bly do not have the angles and separation exactly
> > right, but have been satisfied with the results.
>

> Glad to know that. It will be intensity-based
> stereo. May I listen to a fragment of your record-
> ing, unprocessed and preferably losseless?

The H4N recorder is configured to automatically set the recording level.
As the concert progresses, the level gradually gets lower. I use
"Audacity" to separate each piece, and then use it to amplify each to
approximately the same level, sometimes more than 10 DB.

Before separating pieces, the .WAV recordings run about 2 GB per hour.
So unprocessed and/or lossless fragments would be too big for my limited
internet access!

Fred

[Anton Shepelev]

Fred McKenzie:

> Needless to say, I am not a pro!

Neither am I.

> My recordings are intended to be used by the Con-
> ductor and a few key Musicians to analyze the per-
> formance.

I should say that the conductor and performers are
more critical audience than the layman listener and
usually have a more wholesome taste and criterii.

If they are satisfied with your work it must be
good.

> The H4N recorder is configured to automatically

> set the recording level. As the concert progress-
> es, the level gradually gets lower. I use "Audac-

> ity" to separate each piece, and then use it to
> amplify each to approximately the same level,
> sometimes more than 10 DB.

So the level is set automatically at the beginning
only? I simply ask the musician to play the loudest
part and adjust my controls for that. Does the H4N
use analog or digital attenuation? The latter is a
bad idea.

I too have used Audacity for simple editing. If you
have to amplify by 10 db then you are using about
30% of the recorder's dynamic range, but if the con-
cert is one indivisible programme it is all right.
Otherwise, I should adjust the levels for every com-
position.

> Before separating pieces, the .WAV recordings run

> about 2 GB per hour. So unprocessed and/or loss-
> less fragments would be too big for my limited in-
> ternet access!

That must be a high-definition format. Three min-
utes of CD audio take about 30 Mb in WAV and about
half that space in the lossless format FLAC.

Here's a recording that made using the A-B tech-
nique:

https://soundcloud.com/anton-shepelev/honey-dont-cover

It lasts 2:14 and weights only 12 Mb in FLAC.

[Anton Shepelev]

I miswrote:

> ...and usually have a more wholesome taste and
> criterii.

Should be: "and usually have more wholesome taste
and criteria."

[Fred McKenzie]

In article <20170113165222.2196...@gmail.com>,

Anton Shepelev <anto...@gmail.com> wrote:

> Fred McKenzie:
>
> > Needless to say, I am not a pro!
>
> Neither am I.
>
> > My recordings are intended to be used by the Con-
> > ductor and a few key Musicians to analyze the per-
> > formance.
>
> I should say that the conductor and performers are
> more critical audience than the layman listener and
> usually have a more wholesome taste and criterii.
>
> If they are satisfied with your work it must be
> good.

They take what they can get! One of the members used to record using a
video cam in the auditorium projection booth.

> > The H4N recorder is configured to automatically
> > set the recording level. As the concert progress-
> > es, the level gradually gets lower. I use "Audac-
> > ity" to separate each piece, and then use it to
> > amplify each to approximately the same level,
> > sometimes more than 10 DB.
>
> So the level is set automatically at the beginning
> only? I simply ask the musician to play the loudest
> part and adjust my controls for that. Does the H4N
> use analog or digital attenuation? The latter is a
> bad idea.

A concert starts with a tuning note followed by the Presentation of
Colors and National Anthem. H4N recording level is set by the loudest
sound, often the Bass Drum. By the time the first concert piece is
started, level has been set. Over the course of a 90 minute concert,
occasional loud sounds will reduce level further.

I believe the recording level is digitally controlled. Whether it is
gain or attenuation may be a matter of point of view!

> I too have used Audacity for simple editing. If you
> have to amplify by 10 db then you are using about
> 30% of the recorder's dynamic range, but if the con-
> cert is one indivisible programme it is all right.
> Otherwise, I should adjust the levels for every com-
> position.
>
> > Before separating pieces, the .WAV recordings run
> > about 2 GB per hour. So unprocessed and/or loss-
> > less fragments would be too big for my limited in-
> > ternet access!
>
> That must be a high-definition format. Three min-
> utes of CD audio take about 30 Mb in WAV and about
> half that space in the lossless format FLAC.

I was not familiar with .FLAC format. I see that Audacity can export
it. If I do not forget, I'll check to see if the H4N can use it. There
is a 2 GB limit to the size of a file the SD card format can accept, so
a 90 minute concert is broken into two files. Perhaps .FLAC would help.

> Here's a recording that made using the A-B tech-
> nique:
>
> https://soundcloud.com/anton-shepelev/honey-dont-cover
>
> It lasts 2:14 and weights only 12 Mb in FLAC.

Listening on my laptop, your recording is very clean. It sounds like
you have a nice studio with absolutely no background noise. In addition
to extraneous audience noises, my recordings range from the solo Flute
in a quiet passage to the blaring Brass in a loud passage. I have been
tempted to try compression, but have resisted the temptation.

Fred

[Scott Dorsey]

Fred McKenzie:

>
> I am trying to record what a person would hear if

> they were at the same spot as the microphones, re-
> alizing that there are differences.

Would you like rainbow-pooping unicorns with that?

This is not only not possible, but it is a dangerous attitude to have because
it will distract you from the actual goal of making an accurate recording.

Things will _never_ sound the same at the microphone position, if only
because when you listen on stereo playback you have the playback room
acoustics superimposed on the original recording. So the original recording
needs to be made drier, and specifically lacking in the short term reflections
that will be dominant in the playback environment. Combine that with the
microphones never having the same pickup pattern as the ears and you will find
that the best position for the microphones is never going to be the bes
position for the listener.

The only exception to this is for headphone playback, either with a
binaural recording or with a mono recording. But that is a totally
different animal indeed.

> I am not happy with some instruments not being

> heard as loudly as expected and the lack of stereo effect.

Which ones? If you want more strings, raise the mikes. If you want more
brass, drop them down.

If the microphones are parallel, you will have zero intensity stereo.

[Anton Shepelev]

Fred McKenzie:

> A concert starts with a tuning note followed by
> the Presentation of Colors and National Anthem.
> H4N recording level is set by the loudest sound,

> often the Bass Drum. By the time the first con-

> cert piece is started, level has been set. Over
> the course of a 90 minute concert, occasional loud
> sounds will reduce level further.

Understood.

> I believe the recording level is digitally con-
> trolled.

Well, there must be access to the level of analog
attenuation, because the DAC assumes the signal is
normalized to a certain fixed level...

> Whether it is gain or attenuation may be a matter
> of point of view!

Indeed. My viewpoint is that the maximum digital
level is unity, so the rest is more or less attenu-
ated.

> I was not familiar with .FLAC format. I see that
> Audacity can export it. If I do not forget, I'll
> check to see if the H4N can use it. There is a 2
> GB limit to the size of a file the SD card format
> can accept, so a 90 minute concert is broken into
> two files. Perhaps .FLAC would help.

Yes, or you might try decreasing the resolution down
to the CDDA standard, 16/44.

> > Here's a recording that made using the A-B

> > technique:
> >
> > https://soundcloud.com/anton-shepelev/honey-dont-cover

> >
> Listening on my laptop, your recording is very
> clean.

Hopefully via headphones rather than the built-in
speakers?

> It sounds like you have a nice studio with abso-
> lutely no background noise.

To me, clarity is not the absence of background
noises but, rather, the absense distortion in the
recording-playback tract.

That recording was made in terrible conditions -- in
a tiny closet generously treated with sound-absorb-
ing material -- an acoustically dead space.

> In addition to extraneous audience noises, ->

That ambience or "atmosphere" is an essential part
of any live recording, and I think it is a responsi-
bility of the sound engineer faithfully to convey it
while keeping it unobtrusive.

> -> my recordings range from the solo Flute in a
> quiet passage to the blaring Brass in a loud pas-

> sage. I have been tempted to try compression, but
> have resisted the temptation.

Good for you. There is a less harmful tool in Au-
dacity that can help you. It is called Envelope:

http://manual.audacityteam.org/man/envelope_tool.html

With you can gradually increase the volume before a
quiet solo and go back to unity gain afterwards.

[Ange]

On Tuesday, August 19, 2014 at 1:04:34 PM UTC-4, Fred McKenzie wrote:
> I am attempting to set up a couple of microphones to record an Orchestra
> performance. In order to achieve a stereo effect, it was my impression

> that the microphones should be spaced several feet apart, one on each

> side of the center of the stage.

>
> I came across an article that placed the two microphones together with
> front apertures almost touching in a "cross-eyed" configuration. The
> right microphone was pointed to the left and vice versa.
>

> I can see where it might be bad if microphones were too far apart. Our
> ears are fairly close together and hear things in stereo. Does it make
> sense to locate the microphones closer together than our ears are spaced?
>

> Fred

My first esoteric encounter with this craft came one evening when my son was doing some guitar rendition with friends. I had a portable Nakamichi tape deck with headphones to hear what was being recorded, and a stereo mic pair that I was obliged to hold in both hands. I did hold then less than a foot apart, and the stereo action was clear. Having a necessary scratch, I temporarily put the two mics together in one hand (to scratch with the other). The stereo held until the mics were less than an inch apart. I then tested that phenomenon and concluded that stereo is achieved when the mics are 2" or more apart. Hypothetically, when they are about 5" apart, they should perform as do our ears naturally. That settles the basic issue.
So why wide mic separations? IMHO, it is to sample two separate orchestra sound fields. On another occasion I witnessed that the 2 kHz etc sound radiation from the violin section (viewed as left of orchestra center) radiates upward to the right (as seen from the audience)so that the left mic will intercept that more than the right mic (as seen from the audience). The right side of the orchestra contains largely horns and brass so it is feasible to procure rich strings via the left mic and strong brass from the right mic. To that extent, the wider separation will provide that separation service. How you craft your mic setup is up to you.

[Fred McKenzie]

In article <2f3197b5-f2df-4f90...@googlegroups.com>,

Ange-

Since my original posting, I have used a pair of microphones spaced
about Six inches, pointing about 90 degrees apart. Results have been
acceptable.

One thing I have noticed is that my microphones may be too close for
weaker instruments in the front on each side. If they were located
several feet apart, they would each be closer to the weak instruments.
However the Stereo effect might be artificial sounding.

Fred

[Scott Dorsey]

Ange <a.camp...@att.net> wrote:
>My first esoteric encounter with this craft came one evening when my son wa=
>s doing some guitar rendition with friends. I had a portable Nakamichi tape=
> deck with headphones to hear what was being recorded, and a stereo mic pa=
>ir that I was obliged to hold in both hands. I did hold then less than a fo=
>ot apart, and the stereo action was clear. Having a necessary scratch, I te=
>mporarily put the two mics together in one hand (to scratch with the other)=
>. The stereo held until the mics were less than an inch apart. I then teste=
>d that phenomenon and concluded that stereo is achieved when the mics are 2=
>" or more apart. Hypothetically, when they are about 5" apart, they should =
>perform as do our ears naturally. That settles the basic issue.=20

The problem is that listening on headphones totally changes all of the imaging
in every possible way. So it's hard to translate anything you're hearing on
headphones into anything going on with speakers.

>So why wide mic separations? IMHO, it is to sample two separate orchestra s=
>ound fields. On another occasion I witnessed that the 2 kHz etc sound radia=
>tion from the violin section (viewed as left of orchestra center) radiates =
>upward to the right (as seen from the audience)so that the left mic will in=
>tercept that more than the right mic (as seen from the audience). The right=
> side of the orchestra contains largely horns and brass so it is feasible t=
>o procure rich strings via the left mic and strong brass from the right mic=
>. To that extent, the wider separation will provide that separation service=

>. How you craft your mic setup is up to you.

Traditionally the reason why wide separation was popular was because the only
microphones that were any good were omnis, and getting clean and accurate
directionality with a big baffle between omni mikes is problematic. Putting
a widely spaced triad of microphones up above the orchestra gives you the
ability to control balances by moving up and down and side to side as you
note, and it allows you to use omnis.

The bad part about that technique is that the imaging is only from amplitude
differences between channels. The phase differences are so wide that the
ear can't correlate them. The overall effect is not bad but has a weird sense
of depth that you don't hear live. The Mercury Living Presence recordings
made by Bob Fine are the classic examples of the spaced triad technique.