Symetrix 601, Руководство пользователя

Страница: 66 / 142

7-12
Rev 2.2, 10/31/94
between
L
EVEL
and
B
ANDWIDTH
to find the magic setting. Toggling the band-switch between in
and out can help too.
It is much easier to hear changes in amplitude (level) than it is to hear bandwidth changes. It is
also easier to hear the abundance of something rather than the absence of the same thing.
Even if you intend to apply cut (negative level) to a particular frequency, it is still easier to find
that frequency by boosting first, tuning second, and resetting the boost/cut last according to
taste or need.
It's generally easier to apply boost to a sound for shaping (and that's how many engineers.
start). Many times, however, you may want to experiment with removing an offending sound (as
opposed to drowning it out with something else). In a complex mix, this may work better
because it may require less overall EQ to remove the offending sound; the end result will sound
more natural.
7.5.5 Metering and the Dynamics Block
Each component of the dynamics block uses a concept called "gain-reduction." Gain-reduction
is the degree to which the overall gain has been lowered in response to some signal condition.
When adjusting any of the dynamics block components (dynamic noise reduction, de-esser,
expander, compressor or AGC), the right-hand LED meter changes to a gain-reduction meter.
Use the lower scale to translate the meter indication into numbers. The meter reverts to
displaying level whenever you leave any of the dynamics block.
7.5.6 Dynamic Noise Reduction
The dynamic noise reducer (
NR
) uses a sliding lowpass filter controlled by the relative level of
the signal rejected by the filter. This topology makes a filter that responds more to the content
of the signal than its absolute level; it is easier to adjust.
There are three front-panel adjustable parameters:
F
REQ
and
T
HRESH
. The
F
REQ
parameter
sets the resting frequency of the sliding filter and has a range of 1 kHz to 21.11 kHz. The
T
HRESH
parameter sets the relative threshold ( r ) of the onset of filter activity. Pressing on the
T
HRESH
switch again accesses the absolute threshold (
A
), which governs the transition between
spectral content and signal level as the basis for the filter's action.
In general, use lower resting frequencies to remove excess noise. Higher resting frequencies
result in a more subtle action. To set the NR, with signal applied, set the resting
F
REQ
uency at
1 kHz. Vary the
T
HRESH
old setting until you see activity on the right LED display. Listening,
you should hear the noise reduction removing the noise and more than likely your signal. Set
the threshold at 0 (zero). Raise the filter
F
REQ
uency until you hear onset of the noise. Lower
the filter
F
REQ
uency until you hear the noise disappear. Now lower the
T
HRESH
old setting until
you find the magic compromise between the noise, the music, and the audibility of the filter
working. Higher
T
HRESH
old settings (closer to zero) make it more difficult to "open up" the
dynamic filter and lower settings (closer to -35) cause the filter to almost always run "wide
open."
Finally, use the absolute
T
HRESH
old ( A ) to determine the signal level at which you want the
filters action to become level dependent. Usually, this is at a fairly low level, and it is probably
more important to eliminate the noise, even at the expense of the signal. The useful range for
this parameter runs from -80 dB to -50 dB.
7.5.7 De-Esser
The de-esser uses a limiter controlled by a mildly peaked highpass filter in its sidechain. In
sibilant speech, the dominant frequency component is the sibilance itself. Reducing the overall
gain during periods of sibilance reduces the level of the sibilant.