Technology (1)
Technology (1)
Technology (2)
Technology (3)
Modules (1)
Modules (2)
Modules (3)
Modular Ads (I)
Modular Ads (II)
The Technology (Published in SOS October 2003)

So what were these enthusiastic buyers getting for their money? If
you open up the boxes that make up a Moog modular, you'll find
they're mostly empty space. This was partly deliberate. Ease of
access to patch jacks and panel controls dictated the physical
dimensions of the modules. As a result, it remains true that once
you've worked out how to drive it, a large modular is far more
controllable and accessible than almost any other synthesizer.

But it wasn't all good news. The earliest modules were built to a
very tight budget. The use of the then-recently invented transistor
contributed significantly to the price, and Moog attempted to make up
for this by relying on the cheapest and least accurate resistors then
available. At today's prices, you can buy all of the components
needed to clone the electronics inside one of the early modulars for
perhaps 250 pounds. The panels, cases, circuit boards and metalwork
cost far more than the electronics.

Apart from making the sound and performance more variable than it
could have been, this resistor problem also contributed to the tuning
disaster that was the first series of oscillators. The
technologically understated Mark I 901 VCO modules were almost
entirely unmusical. Not only was keyboard tracking limited to a small
range — the specs claimed five octaves, but two-and-a-half was more
realistic — but a design flaw meant that even the tiniest temperature
variations would send the oscillators drifting away from each other
at random. Larger temperature shifts caused more drastic drifts. The
stability of the system's power supply was also something that didn't
quite match professional expectations. If the mains voltage drifted,
so would the oscillators. This made it nigh-on impossible to use a
Moog in the run-up to meal times. The cooking habits of your nearest
and dearest could send the supply voltage dipping by a few percent
and the oscillators drifting by a few semitones.

It's part of the charm of acoustic instruments that they drift
slightly, but by acoustic standards this was ridiculous. Dedicated
exponents such as Wendy Carlos had to resort to tuning before a take,
hitting record, playing, checking the tuning after a take, and re-
recording any line that had drifted. This was made worse by what
would otherwise have been a good idea. The VCOs were designed to work
in banks, with a single oscillator driver module, called the 901A,
feeding up to four 901B VCOs which produced the waveforms. In theory,
you could patch keyboard and other control signals into the driver,
and then all the VCOs would track together. Except they didn't — at
least not reliably. Because tuning linearity was less than solid,
tuning the oscillators to intervals was also asking for trouble. Even
the octave range switches didn't work without retuning.

On the plus side, these first-generation oscillators were
deliberately designed to work as both VCOs and VCLFOs. If you wanted
the latter you could switch the octave setting to 'Lo' mode, patch
the output to modulate something else, and listen to the slow
variations this produced. Although the output levels were too low for
more extreme filter sweep effects, this made experimentation easy.
Many users discovered that unique new timbres could be produced by
modulating one oscillator with the output of another. This was the
beginnings of Yamaha-style Frequency Modulation (FM) years before a
keyboard called the DX7 made it famous.

But most people created raw sounds with simple analogue waveform
mixing. The 901 saw the first use of the now-standard selection of
ramp, triangle, sine and pulse waveforms. There was also a single
rather minimal pulse-width control on the driver, which worked on all
the oscillators in a bank. In spite of the limitations and problems,
Moog fans still swear by the sheer hugeness of the raw tones sounds
produced by the 901s — although anyone who tries to make music with
them is more likely to swear at them.

Elsewhere, the circuitry was more consistent. The sound of the famous
Moog filter module, labelled the 904, was an instant hit with almost
everyone who heard it. Moog's inclusion of resonant feedback was
perhaps the first example of the definitive squelchy filtered
synthesizer sound. The 904 came in two variants. The 904A model was
the famous low-pass filter, with its famous fat sound. The 904B
version was a high-pass filter, designed to remove low-end grunge and
bloat, but unfortunately, it lacked resonance — Moog's ingenious
transistor ladder design wasn't well suited to adding resonance when
used in a high-pass configuration. There was also a 904C filter
coupler which allowed the filters to be combined in series to create
a band-pass response, and in parallel for a rather fine and whooshy
notch effect.

The 911 Ussachevsky-inspired ADSRs had a snappy 1ms attack time,
which contributed further to the Moog's punchy, ballsy sound. But
they lacked voltage control of timing — an omission that was never
made good on any Moog modular. The 902 VCA allowed voltage control of
volume in both linear and exponential modes, and also had a manual
offset knob, so you could get a fixed-level sound from the VCA
without patching any control signals to it. The VCA tended to distort
in interesting ways, and was another major contributor to the fat
sound. All of today's digital synths sound shimmery and lightweight
in comparison to the gut-kick sonics a modular running at full tilt
can produce. Users were often confused by the VCA's differential
inputs, where one input was added to the output, and the other
subtracted from it; this was only occasionally useful for audio, but
made far more sense when adding control signals together.