[Gary Eickmeier]

"William Sommerwerck" wrote in message
...
I wanted to respond to your remarks about surround sound, but carelessly
discarded the original e-mail. Here goes with what I remember.

Conventional two-channel stereo is fundamentally incomplete. It can
never --
ever -- sound like what you hear in a concert hall, jazz club, church,
what-have-you. The basic reason is that it isn't enough to record the hall
ambience -- it has to be recorded in a way that allows it to be
/correctly/
presented to the listener in playback. Two-channel stereo not only does
not
do this, it /cannot/ do this.

Rather than bore the group with a repeat of my blather, I would like to send
you a .pdf of my Image Model Theory and maybe one other paper. Have you seen
it? May have been in the BAS Speaker. It says that the problem is that there
still is no model, or paradigm, for the process of the facsimile
reproduction of auditory perspective. I have proposed one.

What you are saying is very true, and the reason is that the stereo
process - any process actually - changes the spatial nature of the original
to that of the reproduction system. In other words, the complex original
sound field has been reduced to, and squeezed through, those two miserable
stereo speakers, making it impossible for it to sound anything like the
original. Surround sound is a partial remedy and a step in the right
direction, but you still don't have quite enough channels to do it, and the
room is still there - and needs to be there by the way.

Floyd Toole told me an interesting experience. He was visiting a demo of
Ambisonics in an anechoic chamber. Like most of us, he wondered if this
would be an "ear opening" experience - after all, no more interference from
the room, and near perfect encoding of all directions, even periphony. What
he experienced was In Head Localization, much like with headphones! This
surprised me, because I thought that if you could turn your head you
wouldn't get this. The answer must be that without SOME real reflections to
clue you in to a real space, the perspective is still FIXED as with
binaural.

Those of you who've made live stereo recordings know that, even with the
mics close to the orchestra, there is too much reverberation. This is one
of
the reasons for multi-miking, as it suppresses hall sound. J Gordon Holt
told me that it usually took many recording sessions to find mic positions
that caught an appropriate balance -- and these were usually above the
orchestra, rather than in front of it.

By "too much," what you really are saying is that we cannot make it come
from an evenly spaced, correct set of incident angles, just from near the
speakers, so there is what is interpreted as too much reverb.


So how does one solve the problem? The ideal way is to use a recording
technology that actually captures the direct and reflected sounds
/correctly/ at a particular point. I only know of two systems, binaural
and
Ambisonics. Neither became popular.

There are still problems with binaural. Not sure why Ambisonics didn't catch
though, it is after all just another surround system, and surround has
caught on big time in home theater.

The most-important component of the ambience is the lateral sound, and in
playback, it /must/ come from the sides. (This is why the ITU standard
specifies that the rear or side speakers be located within +/- 15 degrees
of
the listener's sides.) This means that you can never correctly reproduce
lateral sound from the front speakers. It is mixed in with the direct
sound
(in a way that colors it), and the brain will never hear it as lateral
sound, because it /isn't/.

My image model attempts to array the sound in a way that models the
reproduction after the original.


As for the rear levels... In theory, they should be at the same level they
were recorded. If they are at such a low level that a logic-directed
decoder
can't handle them -- well, that's the way it is. Logic-directed decoders
are
not optimum for ambience reproduction.

I could talk more about this, but I don't have the time. I'd certainly
like
to hear the views of others who've made surround recordings, of all types.

Thank you very much for retyping all of that. I am trying to prepare some
discs for you, but I want to get it right first so you can report on how I
did. I will have one disc in stereo, one in DTS surround of the same
concert. A second disc is a few of the "soundscapes" that I recorded to test
the Zoom's surround capabilities. Might also include a third one of spaced
omni with AT-2050s.

Gary Eickmeier