A lot of jargon comes with synthesisers and their various types and uses. Oscillators, filters, envelopes, phase distortion, frequency modulation it goes on. It can be difficult to makes heads of tails of any of these terms without an underlying understanding of the types of synthesis they’re in reference to. So I thought I’d break down a few of the main ones.
This is the most popular and easy to comprehend method of synthesis. At its core it basically consists of electronically generating a raw sound and ‘subtracting’ frequencies from it - sculpting it to your taste.
Sound is generated via an oscillator, a module that outputs an oscillating waveform in a shape of your choosing i.e. a saw tooth wave. Traditionally, subtractive synths feature a sawtooth, square and triangle wave, however as synthesisers have matured, waveforms have been added - some manufacturers’ waveforms have even become distinctive characteristics of music genres i.e. Roland’s ‘super-saw’ is synonymous with early rave trance music. Different waveforms have different sound characteristics; sawtooth waveforms have a thin buzzy sound, square waves have a cleaner hollow sound.
Once you’ve chosen a waveform, you can shape it with filters and modulators like envelopes and low frequency oscillators (LFOs). Modulators act like invisible hands, changing parameters - like the filter frequency and volume - for you over time.
A more complex method is frequency modulation or FM synthesis. Rather than starting with a raw, harmonically rich waveform, you generally begin with a pure sine wave coming from an operator (in this instance an operator performs the same function as an oscillator) and route waveforms from other operators through it to create a more harmonically complex and interesting waveform, and thus sound. Changing the frequency and waveforms of the modulating operators can result in some wild sounds. Many users struggle to predict what the outcome of modifying a parameter in FM synthesis will be, but sticking to even ratios of the modulator to the carrier ie. 2:1 or 1:4 will generally always give you a musical result (as apposed to 1.48670:1 which won’t) and so the results can be learned, and predicted. The sound can be modified further by using an envelope to change the modulation depth and frequency of the modulating operator.
Characteristically, FM synthesis is known to produce clear and bright sounds. Yamaha’s DX7 introduced to FM synthesis in 1983 and went onto the becoming one of the best selling synths of all time, it’s electric piano and bell tones are heard all over the records of the 80s. These days, FM synthesis can claim responsibility for some of the gruesome growling basses heard in EDM and dubstep.
Wavetable synths stemmed from the early digital era when the industry was experimenting with the move to digital. Wavetable refers to the oscillators, which in this case don’t generate a waveform but essentially play it back from a bank of recordings. In fact every instance of a parameter’s settings and the resulting waveform is in the synth’s memory and crossfades from playing back each appropriate waveform as you modify it. It gets interesting though when manufacturers include the ability to morph from a considerably complex waveform in its ‘wavetable’ to another with a simple modulator like an LFO. In an ordinary subtractive synth, morphing between the two waveforms might require 100 parameter changes in an instant. Wavetable synthesis can be combined with subtractive or FM synthesis features and methods.
That’s three of the main go-tos. There are many more variations and offshoots - additive, phase distortion, granular, sampling and whole heaps of other ones manufacturers make up on the fly. But if you can grasp the ones I’ve mentioned, that’s a great start.