[asterisk-users] Michael Graves post

[Michael Graves]

On Mon, 16 Feb 2009 16:05:48 +1300, Matt Riddell wrote:

>On 10/02/2009 5:08 a.m., Michael Graves wrote:
>> I unwittingly started this on Facebook, which I don't user very much.
>> Here's the gist of it.
>> 
>> A Strange Brew: VoIP/Telephony Crossed With Surround Sound
>> 
>> It couldn't be the puritanical kind of approach used in music
>> recording. It would be more a matter of using surround panning to
>> position participants in an synthetic soundfield. I wonder if this has
>> been done to any degree elsewhere?
>> 
>> Stereo is extremely limited in scope. Most of a synthetic stereo image
>> is manipulated using simplistic level based panning, not unlike an old
>> school balance control. It's coarse and two dimensional at best.
>
>Erm - excluding the use of prefade reverb it's actually one dimensional
>- moves left to right - prefade reverb allows you to move backward and
>forward - bringing it to 2 dimensional.
>
>> I'm thinking that UHJ format ambisonic encoding might prove more
>> useful. It allows for accurate, controllable three dimensional
>> positioning while only using the equivalent of a stereo stream.
>
>Surround sound is two dimensional - it just uses the room reverb/delays
>instead of added ones.  I.E. you hear the sound as being in front of you
>rather than to the side of you.

Ah yes. Lets put a little finer point on the terminology.

Stereo is essentially on dimensional.

Surround, as in the common commercial Dolby surround and the like are
two dimensional. These are also sometimes refered to as "planar
surround."

>Three dimensional would require height - which isn't really that useful.

Yes, real three dimensional surround includs a hieght component and is
properly called "periphonic surround."

>The problem is the listening environment - most people don't have
>surround for this - I do like the idea of widening the sound field -
>although it's already doable with phasing and stereo speakers.
>
>I don't think there's even a good stereo conference room.

In all surround encoding schemes backward compatability with the basic
stereo and mono presentation is crucial. The surround scheme should not
damage lesser playback situations.

Effective planar surround is possible from only two channels, allowing
for sources located anywhere in the horizontal plane around the
listener. Ambisonic UHJ format encoding is an example of this. 

OTOH, Dolby et all are not generally good examples. 

Phasing tricks using stereo speakers are good as far as generating an
effect. That is, synthesizing the perception of some image width, but
not accurately reproducing an audible scene. There are many fine
commercial examples of this. I own a old Carver "Sonic Holography"
processor which sounds really nice. Pops a stereo image open
tremendously, although being analog circuitry it's very noisy by
today's standards.

The most common basis of operation of such devices is inter-aural phase
and level differences, often called "head related transfer functions"
(HRTSs)

It's true that the playback circumstances, especially speaker placement
is crucial to surround reproduction. Only 4 speakers are required for
basic surround reproduction, but their placement may be important to
sucessful imaging.

The surround image is likely only going to be optimal in one "sweet
spot" in the listening room. If many people are gathered around a large
table it seems likely that noone will hear the optimal effect.

The question is, will anyone benefit at all? That is, is there any
merit in using existing surround processing techniques to encode
directional cues into conference calling systems?

There's also a question of scale. Does this sort of thing scale down
below full-bore telepresence suites? Or is it just the icing on a $750k
room? Simply a marketing tool?

Michael

--
Michael Graves
mgraves<at>mstvp.com
http://blog.mgraves.org
o713-861-4005
c713-201-1262
sip:mgraves at mstvp.onsip.com
skype mjgraves
fwd 54245