On 01/05/2011 15:28, Fons Adriaensen wrote: ..
I've tried many times to convert composers requiring spatialisation from using discrete channels to Ambisonics. It started to work when I offered 3rd order. The last example is a project I started working on just a few weeks ago. The composition was completely worked out in terms of using an ad-hoc speaker layout. On his first visit to the studio I demo'd some 3rd order mixes (both done locally and kindly provided by Joern Nettingsmeier), and let the composer play with the panning etc. He was converted in no time, and has completely rewritten the spatialisation (which will be performed 'live' in a concert later this month).
Funnily enough, we had a performance at Bath Uni a few weeks ago (Kees Tazelaar, famous among other things for digitising the original tapes of the Poeme Electronique), playing material all of which was explicitly eight discrete feeds - a different sound to each speaker. This was in a very cuboid space ( high ceiling though), built simply as a music store room for pianos, organ, etc, hard flat stone walls (one of which was used for video projection) so very live and reflective, and relatively small (audience about 20), and all we could do was put four speakers in the corners and the other four in the middle of each wall (small powered KRK somethings plus subwoofer), well out of accidental arm's reach. So, all in all, far from ideal acoustic conditions.
Nevertheless, the sounds came over very well and clearly. If anything the live environment smoothed out the "localisation" a bit, so that (insofar as it was desired) one could quite reasonably talk in terms of 'envelopment'; even though the composer had the clear goal, at least in a couple of pieces, of using the space to give a clear separation to some sometimes dense particle-like sonorities. This was certainly successful - we all "got" it. Sounds from behind were predictably less clear in direction, as they obviously reflected quite strongly from the front. This was the first outing for a freshly acquired eight-speaker set, and the event was certainly convincing for me and my composer colleagues in terms of wanting to compose for it.
Short of making the test, we will never know if the Ambisonic approach would have been "better". The concert would be perfectly well described in the above terms of being "completely worked out in terms of using an ad-hoc [well, octagonal] speaker layout". The point is that the effect was more than sufficiently engaging as is; even if the Ambisonic approach would be "better", the discrete approach was not in any meaningful sense "bad". Just, I guess, "different".
The other point I would make in this regard is that one simply does not go to a performance desperately trying to establish ~exactly~ where a sound is coming from! (well, I don't, anyway). It was clear and effective enough as it was. One just wants to relax and receive what there is to receive. My concern is that the relentless pursuit of ever-sharper localisation has become such a priority (dare I say it, an obsession), that the technical priorities have got steadily out of hand; and that as often as not a simple, minimal, positionally "dithered" outcome is just fine, and may indeed, in many cases, be artistically preferable.
Richard Dobson _______________________________________________ Sursound mailing list [email protected] https://mail.music.vt.edu/mailman/listinfo/sursound
