Symetrix 601 - 2. Basics; 2.1 What Does the 601 Do?; 2.2 Digital and Analog Differences; 2.3 Gain Setting

2-1
Rev 2.2, 10/31/94
2. Basics
If the particular combination of processors in the 601 is strange or foreign to you, then we
suggest that you read and digest this section of the manual. If you should find some of the
terminology strange, you'll find a glossary of terms at the end of the manual. A very good
dictionary-style reference is also listed in the Bibliography.
2.1 What Does the 601 Do?
The 601 is a unique combination of four digital signal processors in one box: a versatile three-
band parametric equalizer, a dynamic filter, a dynamics processor, and a digital delay. All of
these processors are implemented in the digital domain and the 601 can accept (or output)
signals in either the analog or digital domains.
One way to think about the particular combination of processors in the 601 is in terms of a
modern mixing console. Today, most mixing consoles have microphone and line inputs, some
sort of equalization, effects sends and returns, and (occasionally) on-board dynamics
processing. For a typical voice-over session, you would probably have a compressor/limiter and
a digital delay patched in as outboard processors. The 601 provides each of these processors
wrapped into one tidy one rack-space package.
2.2 Digital and Analog Differences
A large difference between the 601 and a mixing console is that the processing functions in the
601 are implemented totally within the digital domain whereas those within the console are
most likely implemented in the analog domain.
Outwardly there is no difference between an analog and a digital processor. A digital parametric
equalizer has the same controls that you're familiar with in the analog world. Granted, the way
that you access these controls may be different, but how much difference is there in seeing +9
dB on an LED display or in reading it off of a knob against a scale on the front panel?
2.3 Gain Setting
Wire is probably the only component of a sound system where we don't need to take signal
levels into account (usually). Any other active component of a sound system that passes signal
has a finite dynamic range. This means that our old friends dynamic range, headroom, and
noise floor are present and must be accounted for.
Tackling these terms in reverse order, noise floor represents the signal level of the device's
residual noise level. Realistically, this is somewhat lower than the lowest signal level that you'd
want to process (unless you want the output to sound noisy).
Headroom is the difference between the average signal level and peak clipping. Peak clipping
occurs because the processor can't increase its output to follow the signal. When this occurs,
the output signal simply flat-tops over the period that it can't follow its input (sort of like
clipping the tip of the peak off with a hedge-trimmer to level it off). Audibly speaking, clipping
and using the hedge-trimmer are about equivalent.
Dynamic range is the difference between the highest signal that may pass (limited by peak
clipping) and the lowest signal that will pass (limited by the noise floor). In a digital processor, a
0 dB signal may output a -120 dB noise floor but the smallest signal that may be represented
by 18-bits is a -108 dB square wave (because there is only one bit to toggle on and off).
Somewhere between these two points is the average level of your signal, as set by the
processor's level control. Set the average level too high and peak clipping will smash your peaks
flat, set it too low and suddenly the noise floor becomes audible (but you've got lots of
headroom!).