Hi Fons
On 31/05/2012 14:42, Fons Adriaensen wrote:
I did a small experiment a few weeks ago, and was quite surprised by the result. In a concert we
did at the CdS there were three pieces for solo flute and 'tape'. We got the 'tapes' as CDs of
course. The artistic director of the festival asked me if I could somehow 'spatialize' the tapes
instead of just playing them via two speakers. There wasn't much time to do anything fancy, so I
created six filters, one for each of X,Y,U,V,P,Q which would distribute a mono source in function
of frequency, with one full cycle in azimuth for each octave. In fact I made two sets, one going
clockwise and the second in the opposite sense. The tapes were all electronic noises, nothing you
could recognise as a natural sound, and it worked quite well. Afterwards I took the filter set to
the studio at the CdM which also has 3rd order monitoring, and used it on some non-electronic
music recordings. Of course this produces quite unnatural effects, sounds which you know as from a
single source are split in direction (but 2nd and 4th harmonic would coincide with the
fundamental). Each of the speaker signals separately is extremely coloured, and when you solo them
in the right order you get the 'infinitely ascending pitch' effect. What surprised me is that
nobody could associate pitch and speakers. I somehow expected it would be obvious that e.g. all
'C' notes would come from the same direction, and that one would be able to identify the pitch of
each speaker when listening to the complete signal. But that was not the case, and you had to go
quite close to any speaker in order to notice there was something strange with the sound it produced.
That's interesting - it kind of chimes with some experiments I have been doing recently with digital
recreations of Gerzon's spreaders, which used phase shift based processing. Although technically
they are doing what is described in MAG's original hand written reports, the way they sound doesn't
really correspond very closely to description of how they should sound in the same report. Whilst a
broadband sound processed through one of my pluggins (imaginatively named 'mgspreadpan') does lose
the sense of being a point source as you turn the spread up, there's no clear feeling that sounds of
particular frequencies are coming from a particular direction. Have you compared the results of
having separate X,Y,U,V,P,Q filters to generate the "panning" (which is how I interpret what you say
above) with pre-filtering the sounds then panning the filter outputs?
PS - I gather you guys in Parma might be getting pretty shaken up by the
earthquakes in Northern Italy - hope all is well there.
Here in Parma there's no major damage so far, but in the region
just NE of Modena (60..70 km from here) it's dire misery. 16 people
died on monday, mostly employees who were just resuming work a week
after the first shock. And very probably it's not finished, there are
lots of small tremors all day and night, and some more big ones can
be expected.
Glad to hear that you are ok and hope that it stays that way,
All the best
Dave
--
These are my own views and may or may not be shared by my employer
/*********************************************************************/
/* Dave Malham http://music.york.ac.uk/staff/research/dave-malham/ */
/* Music Research Centre */
/* Department of Music "http://music.york.ac.uk/"; */
/* The University of York Phone 01904 322448 */
/* Heslington Fax 01904 322450 */
/* York YO10 5DD */
/* UK 'Ambisonics - Component Imaging for Audio' */
/* "http://www.york.ac.uk/inst/mustech/3d_audio/"; */
/*********************************************************************/
_______________________________________________
Sursound mailing list
Sursound@music.vt.edu
https://mail.music.vt.edu/mailman/listinfo/sursound
