[KH]

On 6/18/2012 4:46 PM, Audio Empire wrote:
On Mon, 18 Jun 2012 03:39:56 -0700, KH wrote (in
):

On 6/17/2012 6:34 PM, Audio Empire wrote:
On Sun, 17 Jun 2012 12:59:06 -0700, KH wrote (in
):

snip

How is it captured? I'm not referring to *soundstage* depth,
clearly that only requires two mics, rather I'm talking about
information density. A listener, with only minute head movements,
samples a number of different wavefronts, providing an information
density much greater than that achieved by any fixed recording
setup, whether stereo, multichannel, or binaural.

I think that the soundfield, by the time it reaches the audience,
has coalesced into a single whole that is perceived in a certain way
from each location within the audience.

I agree, relative to any specific head orientation. But not as the head
is turned to sample different incident angle information.

The human brain allows us to search within that soundfield and pick
out certain sounds upon which to concentrate, but that's part of the
human ear/intelligence interface that allows us to pick certain
sounds out of a plethora of background. I.E. it's a survival skill
that allows us to pick out the snap of a twig against a waterfall, or
for a mother to distinguish her lost child crying in a crowd. It is
not a result of the orchestra being comprised of many different
soundfields which moving our heads allows us to intersect and sample,
and which microphones miss because they are locked in a single
location.

I don't think that's quite accurate. You can think of it in simple
geometric terms. If you look directly at the center of the soundfield -
both ears equidistant from the center of the stage, you're 'sampling'
one perspective of the soundfield. If you turn your head left 10
degrees, there is now a clear separation in the arrival time of any
given sound ear to ear, a sound previously reaching each ear at the same
time (from the center of the stage), simultaneously you will hear a
higher ratio of reverberant to direct information in the left ear, and,
possibly, a lower ratio in the right ear. A situation the brain has
evolved to interpret into location clues.


That is the information that is missing; that's the information
that allows us to establish accurate positional data.

I maintain that in a properly made recording, it's not missing.

I believe the information to which I'm referring is missing from
the recording. Where, in a stereo recording, is information from
multiple wavefronts, both normal and off-angle, recorded?

It's not necessary as there aren't multiple wavefronts, or if their
are, both our microphones and our ears intersect all of them arriving
at that point in space.

Well, I believe that yes, there are multiple wavefronts in the sense
that turning your head will provide a different perspective to each ear,
and those differences in perspective allow the ear and brain to localize
sounds in three dimensions. Obviously, the ability to discriminate
different off-angle perspectives is subject to threshold values, and
sensitivity and precision constraints, all of which will vary to a
certain degree among individuals.

Again, I'm not saying there's not enough information in the recording to
create a very good reproduction. But, if we're talking about some
theory that would create a playback method/environment/setup that would
be a new paradigm of realism, and overcome all preferential effects,
then close won't do.

There is no doubt that there is sufficient information in a stereo
recording to create a left/right soundstage, as well as depth
localization, and at least an illusion of height, although I admit
I don't havee a firm geometric/visual conception of quite how that
is achieved.

Subtle phase differences that give out ears the (relative) height of
a sound source. They are captured by microphones too in a true stereo
recording.

That I get. From a visual geometry perspective, I don't have a mental
image that accounts for height differences. Left/right, front/back,
fairly clear. Height, not so much, unless frequency plays a role in
height assessment.


But the ability to sample a virtually endless number of
stereophonic (relative to listener reception) wavefronts, available
to an audience member, does not translate to a recording made from
any fixed perspective.

If you accept the premise, then your conclusion is correct. However
from my knowledge and experience, I find that your premise isn't
correct.

Ignoring threshold effects, I don't see how it could not be correct.
Not in the sense of different overall soundfields intertwined in the
venue, but rather different stereophonic interpretations of the overall
soundfield when evaluated from different incident angles. It's fairly
obvious that when you look ahead, look at the right wall, then look at
the left wall, the sound changes significantly.

Keith