Synthesis

Generating sound electronically — the foundation of electronic music and modern production.

synthesissynthsubtractivefmwavetableproduction

Synthesis is creating sound from electronic signals rather than recording an existing sound. The first synthesisers in the 1960s used analogue circuits to generate raw waveforms; modern software synths can do the same in code, plus dozens of other things impossible in hardware.

Almost every popular record made in the last forty years has at least some synthesis on it.

The Core Components

Most synths share a common architecture, regardless of the synthesis method:

Oscillator(s). The sound source. Generates raw waveforms (sine, saw, square, triangle, noise).
Filter. Shapes the timbre. The most common is a low-pass filter, which removes high frequencies above a chosen cutoff.
Envelope(s). Shapes how the sound evolves over time. Standard ADSR: Attack, Decay, Sustain, Release.
LFO (Low-Frequency Oscillator). A slow oscillator used to modulate other parameters — vibrato, tremolo, filter sweeps.
Amplifier. Final volume stage, usually controlled by an envelope.

The Major Synthesis Methods

Subtractive

Start with a harmonically rich waveform (saw or square), then subtract frequencies using filters. The classic synth sound — Moog basslines, Prophet pads, every 80s synthesiser hit. Instruments: Moog Subsequent 37, Sequential Prophet-6, Roland Juno-106 (and software versions thereof).

Additive

Build a sound by adding individual sine waves at different frequencies and amplitudes. Mathematically pure but tedious to program by hand. Used heavily in resynthesis tools and church-organ emulations. Examples: Image-Line Harmor, Camel Audio Alchemy.

FM (Frequency Modulation)

One oscillator (the modulator) modulates the frequency of another (the carrier). The result: complex, metallic, evolving timbres. The Yamaha DX7 made FM famous in the 80s — every electric piano in pop music between 1985 and 1992 was a DX7 preset. Examples: Native Instruments FM8, Yamaha DX7 (and reissues like the Reface DX).

Wavetable

A wavetable is a sequence of single-cycle waveforms. The synth scans through them over time, producing rich, evolving timbres. Modern wavetable synths can be edited graphically and modulated heavily. Examples: Xfer Serum, Steinberg Padshop, Native Instruments Massive X.

Granular

Audio is broken into tiny grains (5–100 ms each) and rearranged, pitched, and overlapped. Useful for textures, time-stretching, and sound design. Examples: Output Portal, Tonsturm Whoosh, the Granular module in Bitwig.

Physical Modelling

Mathematically simulates physical instruments — strings, brass, mallets, plucked strings. No samples; just equations. Examples: Applied Acoustics Strum, AAS Chromaphone, Modartt Pianoteq.

Where to Start

If you're new to synthesis, start with subtractive. It maps cleanly to your intuition: oscillator generates sound, filter removes some of it, envelope shapes it over time. Once you've internalised that loop, every other method becomes easier to understand.

A free subtractive synth like Surge XT or u-he Tyrell N6 will teach you more than reading about synthesis.

Common Mistakes

Loading a preset and using it without changes. Presets are starting points. Most professional sound design happens in the 5–10 minutes after loading a preset, not in the preset itself.

Using the same synth for everything. Different synths sound different for reasons rooted in their architecture. A Moog plugin and a wavetable plugin will give you different colours of "bass" — not better or worse, just different.

Skipping the manual. Synths reward learning their architecture. The 30 minutes you spend reading the modulation matrix manual saves you hours of "I can't get the sound I want."