Using the Reverb Programs
duration and complexity of the notes, and on the proper-
ties of the spaces between notes.
But human perception is more complex than just direction,
distance, and room shape. We are reacting to the music
on quite a different plane. Directional localization is not an
object in itself – it is a tool that helps us separate one musi-
cal line from another, and thus helps us grasp the intellec-
tual and emotional content of the music. The sense of
space is also not particularly interesting in itself. It is the
emotional effect of the space that makes the perception
worth the trouble. It is hard to separate the awe inspired
by the sound of a Gothic cathedral from Gregorian chant
– chant was developed to work alongside this awe, and to
enhance it. Listen to the cellos and basses play a pizzica-
to bass line in a dead opera house, and feel how each
note falls lifeless. Add an enveloping reverberation and
each note takes a life of its own – swirling around us as the
notes decay and pull on our heart with sadness or joy. Are
we aware of these effects? A trained listener can be, but
most listeners can only describe the emotional impact.
This impact can be very real.
The sense of distance has a similar emotional effect.
When the direct sound is too strong voices and instruments
seem to be stuck to the speaker – up front and in-your-
face. This perception is particularly strong when a engi-
neer tries a 5.1 mix for the first time. Put the vocals in the
center speaker and Bam! they whack you in the face. The
up-front perception is psychologically important. But you
want to use it with great caution. Sometimes it is just the
effect you want, but after a sustained period the listener
can tire, and psychologically back away. This is probably
one of the reasons many engineers raised on two channel
stereo prefer a phantom center to a hard center. The
phantom may be just as up-front, but there is no speaker
in that position, so it seems more acceptable.
There is another solution. Humans perceive distance in
large part through the strength and time behavior of
reflected sound. In a typical room the direct sound arrives
uncorrupted by reflections. If the sound has a fast rise-time
the ear is able to determine the direction of the sound by
using the interaural time delay and the interaural level dif-
ferences as cues. After 10ms or more reflections start to
arrive – contaminating the time and level differences and
making localization difficult. Our brains are able to use this
reduction in localizability of the sound as a distance cue –
the degree to which precise localization is reduced after
the onset of a note is a cue to both the size and the fur-
nishings in a room.
Reflections from almost any direction EXCEPT the direction
of the sound source can create this sense of distance, and
they can come at a great variety of times after the direct
sound. However, there are optimal directions and optimal
times. The optimal direction is frequency dependent. For
5-4
frequencies below 700Hz the optimal direction is from the
side (90 degrees from the front). At about 1500Hz the opti-
mal direction is +/- 30 degrees from the front (the standard
front speaker angle.) Reflections that arrive earlier than
about 15ms from the direct sound begin to interfere with
localization, and can cause comb filtering and timber
alteration. Reflections that arrive more than 50ms after the
direct sound can be heard as separate sound events, and
can cause serious problems with intelligibility.
Thus if we add reflections to the left and right front speak-
ers and the left and right rear speakers, and adjust the
time delay so these reflections occur between 15ms and
50ms, we can move the vocals out of our center speaker.
The vocals are pushed back into a space behind the front
speaker array. It seems magical – we are increasing the
perceived distance to a source in the center speaker by
adding reflections to all the other speakers EXCEPT the
center speaker – yet this is the result of some very simple
psychophysics.
Remember that we are not perceiving the individual
reflections separately – we are only perceiving their ability
to mess up the localization of the direct sound. Because
there is no actual perception of the reflections themselves
the distance perception (perhaps the "room" perception
would be a better description) is bound to the direction of
the direct sound. If the direct sound comes from the cen-
ter speaker, the "room" impression seems concentrated in
the front of the listening space, even though the reflec-
tions that produce it are coming equally from the front
and the rear. Switch the direct sound to the left rear, and
the "room" impression also switches to the left rear, even
without altering the reflection patterns at all.
Thus for producing the perception of "room" or "depth" it is
not necessary to have reflection patterns that mimic a
particular source position in a particular room. It is only
necessary that the reflections be primarily in loudspeakers
NOT in the same direction as the sound source, and that
the time delays of the reflections in each loudspeaker
should be different. This makes our life much simpler – we
can use a two channel echo send if we wish, and control
the distance or depth of each sound source by controlling
the amplitude of this source in the echo send.
But distance is not the only perception we need. We need
the envelopment that makes notes come alive. How can
we produce envelopment with a 5.1 system? Once again
the key is the way reflections affect horizontal localization.
Our brains have an exquisitely sensitive detector for differ-
ences in sound arrival times between our two ears. These
time differences are converted into perceived horizontal
angles, or azimuth. In the presence of reflected energy –
particularly reflections not in the direction of the source –
the time differences are not constant. As reflections come
and go the time differences (and level differences) fluctu-
Home
Next Page >>
Lexicon, Inc.
<< Go Back
Need help?
Do you have a question about the 960L and is the answer not in the manual?