The most basic generator module in the synthesis process is the oscillator. The oscillator can output the waveforms discussed earlier, and it is often used as the beginning point of the synthesis process.
Diagram 16-1: Outputs periodic waveforms at a specified frequency.
The oscillator, however, may vary from one synthesis system to another. For instance, in one system the oscillator can produce only a few basic waves, while in another system the oscillator has virtually unlimited capacity for producing any type of waveform.
The terms that denote an oscillator-type function may also vary. Voltage-controlled oscillator (VCO), digitally-controlled oscillator (DCO), partial, wavesample, harmonic, and harmonic generator may all indicate oscillator modules. While this may seem confusing at first, the key is to be able to recognize an oscillator module by its function. Despite differences in nomenclature, an oscillator can be identified simply because it outputs periodic waveforms that begin the process of synthesizing sound.
One of the most fundamental steps in the synthesis process is selecting which waveform will be generated by the oscillator. In most systems, the oscillator is capable of producing at least several different types of waveforms. For example, an oscillator might be able to output a sine wave, a square wave, a triangle wave, and a sawtooth wave. In other cases a user could have an oscillator with an output capability of ten, twenty, a hundred, even a thousand different waveforms. Some systems allow for users to design their own waveforms. In any case, the user of a synthesis system will be given some means of defining which waveform the oscillator will produce.
In most systems the output from an oscillator will be a periodic waveform; in other words, it will produce the same waveshape continuously. The precise way it does this varies from system to system, and even from oscillator to oscillator. In each system, however, there is always a means for waveform selection.
How does the oscillator produce such a wide variety of waveforms? In digital synthesis systems a single cycle of waveform is stored in the waveform memory. The oscillator accesses this memory, reads the stored cycle, and outputs it. By repeating this process over and over, the oscillator produces a continuously sounding waveform.
Example 16-2: Waveform memory allows the oscillator to output a continuous periodic waveform.
Besides providing for a choice of waveforms, all oscillators also allow the user to determine the frequency of the output waveform. Often this capability is determined by two or three separate controls: one control for a particular octave; another control to specify the frequency upward or downward by semitones; and a fine-tuning control to move the oscillator upward or downward by cents (one hundredth of a semitone).
This tuning is very similar to pipe or electric organs where the organist can get the instrument to reach different octaves from the same key by the use of different organ stops. In older synthesizers these controls took the forms of rotary knobs or switches. More recent systems enter a number called a variable to determine the octave, semitone, and fine tuning. A variable might also be used to define a certain frequency; for example, 440 Hz might be entered.
One common way to control an oscillator is with a piano-like keyboard often called a keyboard controller. An important function of the keyboard control is to turn the oscillator on and off. In digital synthesizers the keyboard turns the oscillator on when the key is depressed, and off when the key is released. When the oscillator is turned on, it outputs its waveform; when the oscillator is turned off, it stops generating a signal.
In addition to turning an oscillator on and off, a keyboard controller may be used to control the frequency that an oscillator produces. This means that scales can be played from the keyboard controlling a synthesizer in the same way a pianist plays scales on a piano.
If the oscillator is instructed to play middle C from a keyboard, depending upon which octave the oscillator is tuned to, it will sound a note that will be either middle C, or higher or lower than middle C. This means that if two oscillators were tuned with the first a semitone higher than the second, the first oscillator would always sound a half-step higher than the second no matter what note both oscillators were instructed to produce. Of course, there are many more ways, too numerous to list here, that a keyboard can be used to control oscillators.