A synthesizer builds a sound from scratch out of electricity instead of recording one. Underneath the hundreds of knobs is a short, fixed chain: something makes a raw tone, something shapes it, and something moves it over time. Learn those three stages and almost any synth, hardware or software, stops looking like a wall of controls.
The ideas here apply to any synth, in any DAW, or a browser instrument. Names and layouts differ, but the signal path is the same one nearly every synth is built on.
A synthesizer is an electronic musical instrument that generates sound by creating and shaping electrical signals, rather than recording or playing back audio the way a microphone or a sampler does. A guitar makes sound when a string vibrates; a synth makes sound by producing the vibration itself, electronically, and then letting you sculpt it. Nothing has to be recorded first, so a synth can make a tone that no real object could.
Synths began as large hardware instruments full of patch cables, and most of those controls survive today in software plugins and browser instruments. The look changes and the price changes, but the core design is remarkably consistent. Learn it once and you can sit down at almost any synth and find your way around.
Think of a synth as a small factory with three stations. First a source makes a raw, continuous tone. Then a shaper carves that tone into something musical. Then a set of movers change the source and the shaper over the life of each note, so the sound is not frozen. Every control on the panel belongs to one of those three jobs.
A synth makes sound by generating a raw waveform and then filtering and leveling it. The classic path is short: an oscillator produces a basic waveform, a filter shapes its tone, an amplifier sets how loud it is, and envelopes and an LFO move those settings over time. Signal flows oscillator into filter into amplifier, with the movers riding on top.
The oscillator is where it all starts. It produces a repeating waveform at the pitch of whatever note you play, and the shape of that wave decides the raw character before you touch anything else. A synth chooses its notes from you, not on its own, so which notes you play is a separate question of scales and keys. The common oscillator waveforms each have a recognisable sound:
| Waveform | Character | Often used for |
|---|---|---|
| Sine | The purest tone, a single frequency with no harmonics. Smooth and round, almost hollow. | Sub bass, soft bell tones, adding weight under other sounds |
| Triangle | Soft and mellow with only a few faint harmonics, a little brighter than a sine. | Gentle leads, flute-like and hollow tones |
| Sawtooth | Bright and buzzy, containing every harmonic, so it is the fullest raw wave to filter down. | Strings, brass, big pads, rich basses; the classic synth workhorse |
| Square | Hollow and woody, containing only odd harmonics, with a clear reedy edge. | Clarinet-like tones, retro game leads, plucky basses |
| Pulse | A square whose width you can adjust, which thins it out and makes it more nasal. | Narrow, reedy leads; movement when the width is slowly modulated |
| Noise | No pitch at all, every frequency at once, like static or wind. | Percussion, hi-hats, wind and texture, the breath in a sound |
Many synths run two or more oscillators together, tuned to different octaves or detuned a little against each other, which thickens the tone and gives it width. That raw, layered wave is only the starting material. What turns it into a finished sound is what happens next, in the controls.
The main controls shape the raw waveform into a musical sound and give it life. The filter changes the tone, the amplifier and its envelope set the level and how it evolves across a note, and an LFO adds movement. Here is what each one does.
A filter removes or emphasises frequencies in the raw wave. By far the most common is a low-pass filter, which lets the low frequencies through and turns down everything above a point called the cutoff. Lower the cutoff and the sound gets darker and rounder as the bright top is removed; raise it and the sound opens up and gets brighter. Resonance boosts a narrow band right at the cutoff, adding a vocal or whistling emphasis that grows more extreme, and eventually squelchy, as you turn it up. Sweeping the cutoff is one of the most recognisable synth gestures there is.
The amplifier sets how loud the sound is, but on its own it would only be on or off. So a synth uses an envelope to shape the level across the life of a note. The usual one has four stages, known together as ADSR: Attack is how quickly the sound fades in when you press a key, Decay is how quickly it falls to a holding level, Sustain is the level it holds while the key stays down, and Release is how long it takes to fade out after you let go. A slow attack and long release give a soft pad; a fast attack with a short decay gives a sharp pluck. The same envelope is often routed to the filter cutoff as well, so the tone brightens and darkens as the note plays.
An LFO, or low-frequency oscillator, is a wave that is too slow to hear as a pitch. Instead of making sound, it moves another control up and down on its own. Point an LFO at pitch and you get a gentle wobble, which is vibrato. Point it at the filter cutoff and the tone sweeps back and forth. Point it at the amplifier and the volume pulses, which is tremolo.
Using one control to move another automatically is called modulation, and it is where nearly all of a synth's life comes from. Envelopes and LFOs are the movers, and their targets are usually pitch, the filter cutoff and the amplifier. A still sound uses none of it; an expressive sound has two or three quiet things always moving. If a patch feels dead, the answer is almost always more modulation, gently applied.
The type of synthesis describes how a synth generates and shapes its raw sound. The oscillator into filter path above is one method, called subtractive, and it is the most common, but it is not the only one. Each type has a different strength, and this is the short version of what sets them apart:
One more distinction cuts across all of these. A synth is monophonic if it can play only one note at a time, which suits basses and single-line leads, and polyphonic if it can play several notes at once, which you need for chords and pads. Polyphony is about how many notes sound together, not which notes; choosing notes that belong together is again a matter of scales and keys.
Most modern software synths mix several of these methods in one instrument, so the labels are a guide rather than strict boxes. You do not need to pick a camp to start. Learn the subtractive path first, because its oscillator, filter and envelope turn up in almost every synth, and the rest will make more sense once those three feel familiar.
A synth clicks into place far faster when you can watch a filter open or an envelope breathe while a sound is playing. Reading the theory gets you the map; hearing it move is what makes it stick.
ARGISH is a self-playing synthesizer. The oscillator, filter and envelope controls this guide describes are right there on its panel, and because it plays itself you can hear each one move rather than only read about it. Set a mood and a key, build a progression on the chord wheel, and it composes and evolves in the background while you turn knobs and hear what they change. It runs in your browser, or as a plugin in your DAW: an Audio Unit, a VST3, an AAX for Pro Tools and a standalone app.
You can hear it free in your browser first, no install and no account, then decide.
Everything on this page also works by hand on any synth, which is rather the point of writing it down.
We write these when there is something worth writing down. One email when a new one lands or a new Tunary instrument ships. No newsletter, no schedule.
Tunary