What is granular synthesis?

>Like wavetable synthesis, granular has samples at its heart. Granular synthesis is, broadly, the opposite of wavetable synthesis though – where a wavetable synth builds on the capabilities of a sampler by stacking multiple waveforms and morphing between them, granular synthesis breaks a sample into pieces, creating additional scope from a sampled sound by dividing it, not multiplying.

It uses the same principle used to power the timestretching capabilities found in most DAWs by looping miniscule elements of an audio file, known as grains, allowing each to be sustained without altering the pitch.

What granular and wavetable synthesis have in common though is that both come to life with a little modulation and variation. In the case of granular synthesis, this usually involves modulating the position, length or pitch of grains to add variety to a patch. The results created will, naturally, depend on the source material used, but with a sample with plenty of timbral variety, the results can be ethereal or glitchy. One common technique is to play back multiple grains of a source sample at once, creating a ‘grain cloud’ effect, so called because of the ‘fluffy’, ambient tones that can be achieved.

Randomisation is a common tool found in granular synths. Whereas wavetable synths tend to work best with smooth modulation that ‘morphs’ the sound from one table to another, granular synths can create earcatching effects using both subtle and overt randomisation. This can be a great way to create sounds that change in an unpredictable and unusual manner.