On Synthesizers: Answers to Readers' Questions

Synth pioneer Bob Moog's original column for Keyboard explored issues affecting musicians and technology developers.
Publish date:
Social count:
Synth pioneer Bob Moog's original column for Keyboard explored issues affecting musicians and technology developers.
Image placeholder title

Continuing our celebration of 40 years of Keyboard, we are presenting Bob Moog's original "On Synthesizers" columns in their entirety.


As announced in my column in the March/April issue, I'm devoting this column to answering some reader questions. Next issue I'll be returning to more general topics.

What makes a keyboard instrument polyphonic? What are the different modes of polyphony? What is "assignment priority"?

Most currently available synthesizers have monophonic keyboards; only one key at a time has effect. A monophonic synthesizer has one independent sound generator rank, and its keyboard produces only one pitch control signal and one trigger. It is a solo instrument. Depressing more than one key will still give only one tone, just as depressing more than one valve on a trumpet or more than one point on a single guitar string will still give only one tone.

A polyphonic keyboard is one on which it is possible to play two or more independent tones. The simplest commonly available type of polyphonic keyboard is the duophonic, or two-note keyboard. Keyboards of instruments such as the Moog Sonic Six and the ARP Odyssey produce two pitch control signals, one determined by the lowest key depressed and the other determined by the highest. These two pitch control signals are usually routed to two independent sound generators. Some duophonic synthesizers have keyboard logic that prevents the second sound generator from changing pitch when only one key is depressed.

Multiplexed keyboards are an important new class of polyphonic instruments (example: the Oberheim Polyphonic Synthesizer). Such a keyboard contains circuitry that electrically scans the keys one at a time, and produces a pulse if the scanned key is down. Digital circuitry then stores information as to which keys are down. The keyboard circuitry is coupled to a group of sound generating ranks through switching circuitry that assigns the pitch of a newly-depressed key to a sound generator rank, according to predetermined assignment priority rules. The simplest assignment priority scheme is one in which each sound generator rank has a number, and a newly depressed key is assigned (electrically switched) to "play" the lowest number uncommitted generator. Another assignment priority scheme assigns a newly depressed key to play the generator that has been uncommitted for the longest time.

Multiplexed keyboards offer limited polyphony, with the timbral richness and flexibility of monophonic synthesizers. Such instruments produce only as many tones as there are generator ranks (which themselves are virtually complete monophonic synthesizers). Instruments that produce any number of tones simultaneously, up to the total number of keys, are called fully polyphonic (example: the Polymoog). Rather than being multiplexed, the keys of these keyboards are hard-wired (permanently connected) to a bank of synthesizer-type sound generator ranks, one per key. Full polyphonic instruments are musically similar to acoustic polyphonic keyboard instruments such as the piano in that they are not limited by assignment priority rules and are thus capable of sounding any number of keys simultaneously.

How are microprocessors being used in conjunction with synthesizers?

Digital (number-processing) circuitry is now used extensively in several synthesizers and accessories. Digital sequencers provide the means for storing series of numbers representing note values (pitches) and durations. Multiplexed keyboards employ economical means for deriving signals that tell which keys are down. And some synthesizers actually employ digital circuitry to construct audio waveforms on a point-by-point basis. In these devices, the mode of operation of the digital circuitry is generally fixed by the way the circuit elements are wired together. Digital engineers say that these devices are hardware-programmed.

A microprocessor is a device consisting of a collection of digital elements that function together according to a program of numbers that are generally stored in a memory. Thus, a microprocessor is software-programmed. Its mode of operation can be changed merely by being fed a new program (instruction sequence). What's more, a portion of the program may instruct the microprocessor to change its mode of operation periodically, or upon receipt of a command trigger!

Microprocessors, then, may be used in digital systems wherever the mode of operation of the system is to be frequently changed. For instance, a digital sequencer which stores and plays back sequences of notes is usually hardware-programmed, using single-function digital devices. However, if the sequencer is to be called upon to rearrange, edit, and similarly process long sequences in a number of ways, then the use of a microprocessor would be appropriate.

The difference between hardware-programmed and software-programmed digital systems is similar to the difference between hardware-programmed analog instruments (like most electronic organs) and software-programmed (i.e. voltage-controlled) analog instruments (like most synthesizers). Hardware-programmed systems are characterized by simplicity and low cost; software-programmed systems are characterized by versatility.

What are fc and Q contours?

Contour is shape or form. When a musician thinks of a sound, he generally things of how it changes as it builds up, sustains, and decays. That is, he considers the time-domain contour of the sound. Synthesizers often enable the musician to impart different time-domain contours to individual properties of the sound. For instance, the loudness contour (or envelope) describes how the overall strength of a sound is shaped in time, while the filter contour, which may have a different shape, determines how the harmonic content of a sound changes in time.

fc and Q are center frequency and quality, two properties of resonant (bandpass) filters. fc is the position of the resonance in the frequency spectrum, whereas Q is a measure of how pronounced the resonance is. If fc is swept up, then down during the course of a tone, we would hear the characteristic "wow" sound imparted by a wah-wah pedal or voltage-controlled resonant filter. This is one common shape of an fc contour. Q contours are much less common, since the sharpness of a resonant filter is generally not voltage-controlled on synthesizers.

Image placeholder title