Subject: Re: Request for information on multi-channel references
From: Kevin Austin (kevin.austin@videotron.ca)
Date: Thu Apr 27 2006 - 23:39:47 EDT
There is not an enormous amount written, compared to (for example)
traditional theory books.
One approach is to study the possibilities on a kind of grid.
On one side of the grid is 'mono', on the other side is a 640 (more
or less!) loudspeaker system.
Mono can be "multi-channel" (see my recent posting regarding Phil Spector).
The next step is to two loudspeakers. Do they each have totally
different signals, a kind of dual-mono situation? (The so-called
'point source' approach.)
Or is there an attempt to create a 'field' with sound appearing to
come from between the speakers. Both of these models can be extended
to 3, 4, 5, 7, 8, 12 ... speakers ... 'point source' (mono channels),
or 'matrix' (panned).
And another part of the question is how the composer conceptualizes
the individual 'stems' that go into the piece.
Consider taking a piece like the Bach Prelude in C major from WTC I,
which is a line of notes played one at a time (mono?), but
representing five-voice polyphony. Take the five 'voices', and place
one in each speaker in the corner of a (five-sided) room.
When played in mono, the piece tends to be heard as the arpeggiation
of chords, when played from five speakers in five corners, there is a
much stronger tendency to stream the individual lines. (Henry Brant
commented on this when he had Bach chorales performed by four
players, each in the corner of a room. He noted that he could no
longer hear the "harmony". In modern terminology, he was no longer
able to complete the (vertical) integration of the four parts into
chords.)
Some composers hear their sounds as 'complete' in themselves, so
called 'sound objects', and some composers de-compose sound objects
into sonic shards. These fragments can be handled as 'non-integrated'
elements. To help in keeping the component elements segregated, it
would be a matter of putting each in its own speaker. (See Brant
above.)
In varying degrees, for some pieces, multi-speakers allow a more
refined 'grid' for the projection matrix -- it is possible to have
greater locational definition when panning between 16 speakers than
when panning between 4.
In some cases, the 16 speakers may each contain a single (type of)
sound, the 'point source' approach. And combinations of these two.
One thing you may want to examine is the relationship between the
'basic duration' of the sonic object, and whether the composer tends
towards the matrix or the point-source approach. As I think back, I
feel (unsupported) that pieces with longer gestural entities tend
more towards the matrix (where positional blurring is less
important), whereas sections with short gestural units (many short
sounds), tend to be more clearly articulated when the short sounds
are not 'panned' across speakers, but rather are assigned to a unique
speaker.
There seems to be little formal discussion of this wider topic as so
few people have regular access to systems in the 8.1 (8.2) and up
category for their compositions. (This used to be a technological
limitation of sorts.)
Carry on
Best
Kevin
At 6:23 PM +0100 06 04 27, mopani@tiscali.co.uk wrote:
>I'm trying to put together a reading list on multi-channel composition, on
>techniques, composer's working methods and so on. I'm doing this to learn
>more about how other composers work and out of general interest in
>the techniques used. Any help in the form of references and links
>would be most appreciated.
>Thanks.
>
>James
>www.jameswyness.org.uk
>
>___________________________________________________________
>
>Tiscali Broadband from 14.99 with free setup!
>http://www.tiscali.co.uk/products/broadband/
This archive was generated by hypermail 2b27 : Sat Dec 22 2007 - 01:46:18 EST