sound possibilities for the future?

[Eric Pepke]

In article <8912142322.AA25307 at chem.chem.ucsd.edu> sd%chem at ucsd.edu (Steve Dempsey) writes:
>In keeping with the spirit of things, the high end 4D's should have DOLBY
>multi-channel surround sound capabilities.  After all, a machine capable of
>sophisticated 3D video imaging ought to support 3D AUDIO imaging as well.

DOLBY Schmolby.  Do it like this.

Include a DSP which is capable of generating at least two channels of CD
quality sound at 44 KHz.  This should be trivial; the NeXT can do this now,
and you can buy better DSP's dirt cheap.

Then, allow the user to specify sound sources in the same coordinate space
as the graphics.  Transform these points the same way the graphics coordinates
are transformed.  Then, provide some hearing transformations in the same
spirit as the viewing transformations perspective and ortho.  The most
interesting would be

binaural(headwidth);

where headwidth is the width of the virtual head that follows around according
to lookat and friends.  This would cause the DSP automatically to calculate
1) the volume based on the distance of the sound source from the head, 
2) the phase difference based on the differential beween the ears, and 
3) slight frequency correcions based on a simplified model of sound
absorption of the ears and the rest of the head.

Next, wire the sucker into a virtual reality helmet, remove all breakable
objects from the room, and provide rehabilitation for the survivors. :-)

Eric Pepke                                     INTERNET: pepke at gw.scri.fsu.edu
Supercomputer Computations Research Institute  MFENET:   pepke at fsu
Florida State University                       SPAN:     scri::pepke
Tallahassee, FL 32306-4052                     BITNET:   pepke at fsu

Disclaimer: My employers seldom even LISTEN to my opinions.
Meta-disclaimer: Any society that needs disclaimers has too many lawyers.