On 01/05/2011 17:25, Marc Lavallée wrote:
I have a naive question for experts: would it be possible to recreate
the acoustics of the Philips Pavillon using room simulation techniques
and ambisonics spatialization?
That is what they/we did for the "Virtual Electronic Poem" Project:
http://www.edu.vrmmp.it/vep
Sadly I never got to hear the final result. My contribution was strictly
compositional (composing the "sound routes" in the almost complete
absence of original data - the original 30-channel perforated control
tape which controlled both the sound movements and the visual elements
exists physically but is unplayable).
The acoustic reconstruction was handled by the Berlin team. The project
is described in CMJ 33 Vol 2, andd presetned at ICMC 2005; I don't know
offhand if the CMJ paper is downloadable externally anywhere.
As is the way of such things, it is rare indeed to get any funding etc
for follow-up work, so the reconstruction software is probably stowed
away somewhere obscure, never to see the light of day again. You would
need to contact members of the team to see if any sort of access is
possible. We always hoped to be able to create a publicly usable model
of the space that could be used e.g. in Csound, so composers could
explore their music as it might sound in that space.
For the acoustic modelling they created a huge amount (GB-worth) of hrtf
impulse responses for every speaker (350 of them), for a particular
central listener position. These were cross-faded according to the
head-tracked motions of the listener. The modelling was pretty
comprehensive, even taking into account the properties of the interior
surfaces. Resolution was 1deg horizontal and 5deg vertical.
The binaural rendering was programmed in SuperCollider, and the newly
published SuperCollider Book (MIT Press) includes a chapter on this aspect.
Richard Dobson
_______________________________________________
Sursound mailing list
Sursound@music.vt.edu
https://mail.music.vt.edu/mailman/listinfo/sursound
