[Audio Empire]

On Mon, 27 Jun 2011 08:22:00 -0700, Gary Eickmeier wrote
(in article ):

"Audio Empire" wrote in message
...
On Sun, 26 Jun 2011 06:16:26 -0700, Gary Eickmeier wrote
(in article ):

Audio Empire wrote:

Actually, you need three identical channels )including speakers to
get these three- channel recordings to sound correct. They have
spectacular soundtage even though three omnidirectional mikes
(L-C-R) are wrongheaded in my opinion. X-Y or M-S or other true
stereo techniques is always the way to go for real stereophonic sound.

Uh-oh - I think we may have argued about this one before, but....
curious,
why would you characterize X-Y or M-S as "real" stereophonic sound? What
sort of theory of stereo are you operating under? Anything deeper than
Blumlein's patent?

Gary Eickmeier


The smart-ass answer is this, Do you have three ears? Are your ears 15 ft
apart? No, to both. You have two ears and your ears are about 7 inches
apart.

Well, my shocking retort is that stereophonic sound has nothing to do with
the number of ears on your head or the spacing between them. Your M/S and
Blumlein techniques are also not spaced as the human ears are. In other
words, stereo is not a head-related system like binaural, it is a field-type
system. The goal is not ear signals, but rather reconstructing a sound field
in another room.

I said that was the smart-ass answer. From that you should have understood
that I was kidding.

Kidding aside, the more scientific answer is that spaced omnis aren't
phase
coherent.

OK stop for a minute. So what? Who says they need to be? Where does that
come from?

I think you need to study-up on how humans localize sounds. It is done by a
combination of level differences and phase differences. Spaced omnis locate
sounds within soundstage by colume differences between right and left only.
The lack of phase coherence actually confuses those cues. I can tell a spaced
omni recording every time. It's as audible to me as multi-miking (but not
nearly as annoying)

The reason why early stereo recordings by people like Fine and
Eberenz (Mercury) and Lewis Leyton (RCA) were initially three channel
spaced
omnis is so that the center mike could be used to produce the monaural
version of the record (these were the days of "dual inventory" in which
both
a separate mono and stereo record were sold of the same title). One
couldn't
just sum the L+R mikes together to get mono because the phase anomalies
would cause cancellations. It was later discovered that mixing the center
channel equally into both the right and the left channels yielded a more
stable center image.

It was known that X-Y, or A-B cardioids would yield a phase coherent
stereo
recording that would sum to mono perfectly, and result in much better
imaging
but many felt that the flatter frequency response (off-axis) of the spaced
omnis would override any advantages in soundstage and mono compatibility
afforded by using cardioids.

For some contrarian reason I don't have much sympathy for the ability to sum
to mono. What I do have is some (I believe) valid theoretical reasons that 3
or more spaced omnis are more correct for stereo.

See above. The mono compatibility is not important, per se, but it does point
out rather dramatically that spaced omnis have serious phase problems and it
shows, without question, how wrong-headed the spaced omni method is. To me
it's a serious compromise and I admit readily that I don't understand what
seems to be your blind adherence to it. When I was recording a wind
ensemble's (symphonic band) rehearsals I tried all kinds of microphone
schemes - including spaced omnis. If you could hear the difference in
soundstage and imaging between the three spaced omnis and the X-Y or M-S
stereo versions, I dare say you'd never use former again. The palpable,
almost holographic imaging of the X-Y and M-S recordings is startling while
the spaced omnis is much more vague and amorphous.

My only real point in this discussion is that one should experiment with
different mike placements and methodologies before drawing any conclusions.
It's the only way to KNOW what works best. If you try all the different
setups and still find that you prefer the three spaced omnis, then go for it
and don't let anyone tell you differently, because you'll know, from actual
experience, what kind of results they all yield.

Some record companies such as British Decca and
DGG used M-S miking and that was also phase coherent, RCA eventually, and
for
a short time, changed over to X-Y miking (before going to
multi-miking/multi-channel). I once asked Bob Fine (at a NYC Audio
Engineering Society Convention in the 1970's) why he stuck with three
spaced
ominis even after modern cardioids improved enough to make them viable
from a
flat frequency response standpoint. His answer was a practical one. He
acknowledged that while X-Y, A-B and M-S were better, his three spaced
omnis
were something he knew intimately, and was comfortable with. He also felt
that the spaced omnis picked up more hall sound and the center mike was
largely responsible for the famous "Living Presence" Mercury sound and
didn't
want to change that. He's right about the omnis and hall sound. Cardioids
have such attenuated pick-up from their backsides that if you want real
hall
sound you have to use auxiliary microphones placed out in the hall. But
this
only works if you are recording in an empty hall. If you are recording a
live
event, you want the isolation from the audience sound. Believe me you'll
still hear them applaud on the recording, you just won't hear every cough
and
program rustle.

There are lots of differing opinions about how recordings should be made.
Just because someone works in the business professionally and
successfully,
doesn't mean that they are "right". A good case for this was when I
confronted RCA producer J. David Saks about his 48-channel,
microphone-per-instrument method of recording the Philadelphia Orchestra.
I
said that I thought his recordings sounded simply awful. They had no
depth,
and the instruments didn't sound real because, in my opinion, in order for
a
group of musicians to coalesce into an orchestra, the individual
instruments
must "mix" in the air between the ensemble and the listeners (or
microphones)
and not in an electronic mixer. I also pointed out that instruments
captured
close-up did not the sound the same as they do at a distance.

Saks' answer was that he wasn't looking for "realistic" orchestra sound,
he
was aiming for "better than real". How do you argue with a wrong-headed
attitude like that?

I would agree with you there. The only way he could justify his technique
would be to play back his recording in as big a hall as it was made in, and
with as many speakers as microphones.


Two things: I wrote an entertaining article about the difference between a
head-related system and a field-type system that tries to get to the heart
of this discussion. I don't remember if I sent it to you. It is called An
Audio Fantasy: The AES Goes to Mars. I could zap it to anyone who is
interested as a PDF. Number two, I have been making some 3 spaced omni
recordings of my own to test out my audio theories from recording to
playback, and I have had some success both in my system and in critical
acclaim in three other people's systems. I made a shorter demo version that
I would be glad to send to anyone interested, if you just Email me your
address.

But you haven't compared different techniques on the same ensemble to find
which is better. I'm sure that your recordings sound fine. The old Mercury
Living Presence and RCA "Red Seals" made with three spaced omnis sound very
good. They are still among the best recorded classical performances ever
released. But they could have imaged so much better had Bob Fine and Lewis
Leyton, et al used a stereo pair of modern cardioids or an M-S setup instead
of the three spaced omnis. For contrast, listen to one of John Eargle's
Delos recordings of Gerard Schwarze and the Seattle Orchestra. Eargle used
M-S miking. Wow! What palpable, you-are-there soundstage and imaging!