Spatial Sound for Computer Games and Virtual Reality - Game Sound Technology and Player Interaction: Concepts and Developments - page 300

sitional and directional sound sources, and also
applies angular attenuation modifications to the
amplitude of directional sound sources. When a
sound object is created it has to be assigned an
initialGain value: if this field is empty then the
value defaults to 1.0 (where 1.0 is the maximum
gain and 0.0 is the equivalent of a gain value of
-60dB). In relation to the generic Sound Node,
no distance attenuation is applied. This would
seem to be a shortcoming in the specification as
distance attenuation is one of the strongest cues
in establishing depth perception for sound and
should be accessible from the generic object.
If the developer did not want the sound to have
distance attenuation then he could simply leave
the distance field blank.
The SoundScape node (refer to Figure 9) con-
figures the acoustical properties of the listener's
environment. An unlimited number of SoundScape
nodes can be contained within a scene. The defined
SoundScape node region determines which sets
of acoustical properties are to be used. As a result
of being able to specify several SoundScape node
regions, one can generate a number of aural envi-
ronments within the scene. For instance, within
the one scene there could be three rooms, each
with a different acoustical signature. Alternatively,
with more detailed scene description, one could
set up a number of acoustical regions within a
single room using a number of SoundScape nodes.
The acoustical properties, that is, reverberation
and atmospheric attributes of the SoundScape
node, are specified in the AuralAttributes Object.
The AuralAttributes Object is a component object
of the SoundScape Node. It specifies the follow-
ing aural properties: reverberation, Doppler effect,
distance frequency filtering, and atmospheric
rolloff. Table 1 contains a list of the parameters
and their default values for when an AuralAt-
tribute Object is first constructed.
The AuralAttributes node describes rever-
beration with three components: delay time, re-
flection coefficient, and feedback loop. Delay
time is used to calculate the amount of time

taken for the sound to reach the listener having
undergone one reflection. This component is either
set explicitly or implied by the bounding regions
of the volume. Note that the bounding region is
not necessarily the same as the region specified
for the Soundscape Node. Delay time is measured
in milliseconds.
The reflection coefficient is used to determine
the attenuation factor for the sound. The reflection
coefficient(s) represent the reflective or absorp-
tion properties of the environment. A value of 1.0
represents an un-attenuated sound and a value of
0.0 represents a sound that has been fully absorbed.
These coefficients are applied as a uniform attenu-
ation across the spectrum. This is not a very refined
scheme as most reflective/absorptive materials
alter the spectrum of the sound in a non-uniform
manner. Using the specification's present imple-
mentation of reflection/absorption, there would
be very little timbral difference between a sound
that has been reflected by plaster and one that has
been reflected by a metallic object.
The final component, feedback loop, speci-
fies the number of times a sound is reflected or
the order of reflection. If the feedback loop has
a value of 0.0 no reverberation is performed, if it
is set to one then the listener hears an echo and,
if it is set to -1.0 the reverberation will continue
until the amplitude of the signal dies to -60dB.
This is known as effective zero in the Java 3D
specification. Effective zero relies upon a -6dB
drop in gain for every doubling of distance (inverse
square law). Values between 0.0 and 1.0 refer to
the number of iterations of the loop.
The parameter attributeGain is used to alter
the speed of sound in order to mimic the effects
of atmospheric change. The default value is 0.344
meters per millisecond and this refers to the speed
of sound at room temperature. This value is then
altered by the gain scale value specified by the
developer. A value greater than 1.0 will increase
the speed of sound and conversely a value less
than 1.0 will decrease the speed of sound.