[Openal] multichannel output

[Artem Baguinski]

hello openal

My lab is going to work with a sound artist on an interactive
soundscape installation, where the audience controls 3 dimensional
audio synthesizer (indirectly - they are being tracked in space and
their movements / location etc control the composition to some limited
extent).

Fot the spatialization part I would like to use openal, if possible,
because I know the API, it is integrated in our tools and we'd like to
reuse the software we'd write for this installation later. But I'm not
sure which combination of OS, hardware and driver/openal
implementation can do what we want:

- the physical space is at least 10x10 meters room
- the sound output gets amplified and output from largish speakers at
80-100 dB, well at least I hope it's these levels in the middle of the
room, not harder.
- how many speakers and where they are hung remains to be determined.
- the virtual sound sources are "hung" in the room
- the primary participant's position and head orientation are tracked
with a number of sensor systems

we'd like to adjust the sound scape for the primary visitor in
realtime, based on relative positions of the visitor and ther
speakers, to maintain the illusion of sound sources located inside the
room - between the speakers. the rest of the audience sort of hears
the soundscape from wrong perspective - not unlike secondary
(untracked) viewers in VR Cave see the images distorted, while the
primary (tracked) viewer sees it just right.

it is my understanding that this isn't something openal is designed to
do out of the box, but may be there are some tricks to at least
approximate this?

the fallback scenario is to give tracked visitors headphones and
combine the "average" sound scape in the physical space with a
"personalized" one in the headphones. still the question remains for
the loud part - which platforms / hardware / drivers support output to
how many channels and how accurate the physical properties of the
attached speakers can be configured to localize the virtual sources in
physical space.

or am i completely dellusional and should look at some other rig for
what we have in mind?

-- 
cheers,
artm

http://lab.v2.nl/