ï~~S C DP, I-IT0 r
Figure 1. Fixed source channel mapping common in traditional stereo diffusion systems.
channel to speaker 1, inverting the normal left-right axis.
As with all of the previous examples, this mapping can be
controlled with a single fader and implemented alongside
other mappings.
7. ADDITIVE SPATIALISATION
As a simple scenario let's say that fader a controls the level
of a stereo source mapped to the main pair of loudspeakers, while fader b controls the level of the same source
mapped to all of the loudspeakers within the array. If we
use both faders together, their effects will be mixed together: fader b diffuses the stereo source 'everywhere,'
and we can boost the level of the main pair of loudspeakers further by additionally raising fader a. This can be
referred to as additive spatialisation.
8. SUBTRACTIVE SPATIALISATION
It is also possible for a fader to have a subtractive influence, such that raising the fader will progressively reduce
the level of a given set of loudspeakers. If, for instance,
fader c is configured to act subtractively on loudspeakers
3 through 8, then raising fader c will reduce the levels
of these six loudspeakers without affecting the levels of
loudspeakers 1 and 2. We might describe this particular
subtractive routine as an 'all to mains' transition. It is the
ability to perform manipulations additively and subtractively that differentiates Resound from other matrix based
systems such as the DM-8 [10].
9. SEMI-AUTOMATED SPATIALISATION
BEHAVIOURS
In all of the examples so far there has been a simple proportionality - either direct (additive) or inverse (subtractive) - between fader position and resulting output levels. Resound also provides a framework for defining semiautomated spatialisation 'behaviours' whereby the faders
are used to control parameters other than loudness. Some
examples are as follows.
9.1. Mexican Wave
The Mexican Wave behaviour cycles through a sequence
of loudspeakers, or a sequence of groups of two or more
loudspeakers. The user can control the rate at which the
sequence is iterated (including the direction of iteration)
and the overall amplitude level by assigning these parameters to faders. The Mexican Wave behaviour can be used,
for instance, to automate front-to-back or back-to-front
movement within a venue; rotational movement is also a
possibility.
9.2. Random
The Random behaviour also acts upon a group of loudspeakers, but iterates through the group in a random order
as opposed to sequentially. Again, the user can control
the rate at which loudspeakers are selected, and overall
amplitude level of the behaviour. This behaviour can be
particularly effective for the spatialisation of textural or
granular material where an erratic spatial effect is sought.
9.3. Multi-point Crossfade
The Multi-point Crossfade behaviour is similar to the Mexican Wave, however in this case the behaviour does not
iterate automatically through the sequence. Instead, the
active loudspeaker(s) at any given time is governed by the
position of a fader. This behaviour might be useful, for
example, as a way of adjusting the width of stereo images. Traditionally, this would be done by cross-fading the
source between 'main' and 'wide' pairs of loudspeakers,
necessitating the use of four faders. With the Multi-point
Crossfade the same operation can be carried out with one
fader.
10. PUT IT ALL TOGETHER...
All of the above techniques can be combined within a
single performance. A thirty-two fader OSC controller
might comprise, for instance, eight faders that operate on
specific loudspeakers individually, eight that operate on
groups of two, eight that operate on groups of four, and a
further eight configured with semi-automated behaviours.
Because all interactions within the Resound system can
be applied additively or subtractively, there is no conflict
between faders that act upon the same loudspeakers: the
results are simply summed (positively or negatively). This
is particularly useful in the context of semi-automated behaviours, which can be used to dramatic effect on their
own or subtly mixed in, as appropriate to the musical context.
11. COLLECTIVES
Systems based around mixing desk hardware tend to pose
problems with multi-channel sources, owing to the limitations of the 'one fader to one loudspeaker' paradigm. As
Harrison notes:
