Oscillator Waveforms Explained: Sine, Triangle, Saw, Square

What are the basic oscillator waveforms?

The basic oscillator waveforms are sine, triangle, square (with its pulse variants) and sawtooth, and each one is defined by which harmonics it contains and how loud they are. Apple's Logic Pro guide notes that the level relationship between the fundamental and the harmonics of the chosen waveform is what sets the basic timbre.^[1]

Gordon Reid's Synth Secrets series makes the key point plainly: harmonics and waveforms are two ways of expressing the same thing, so a waveform name is shorthand for a particular set of harmonics at particular levels.^[2] Learn the harmonic recipe and you can predict how each wave will react to a filter before you touch one.

Sine wave: why does it sound so pure?

A sine wave sounds pure because it contains only the first harmonic, the fundamental, with nothing above it. Apple compares it to whistles, tuning forks and a wet finger on the rim of a glass.^[1]

That purity makes the sine the default for sub bass and the body of an 808: all its energy sits at the note you play, so it gives weight without clutter. It also means a filter has nothing to remove, which is why a sine patch barely changes when you sweep cutoff. The catch is small speakers. A 40 Hz sine can nearly vanish on a phone, so producers add saturation or layer a brighter wave above it to create harmonics the ear can follow; the 808 design guide covers that layering.

Sawtooth wave: every harmonic at once

A sawtooth contains every harmonic, odd and even, and the amplitude of the nth harmonic is 1/n of the fundamental: the second harmonic is at half level, the third at a third, and so on.^[2] That full, gently tapering series is why it sounds bright and buzzy.

Because nothing is missing, the saw is the most flexible starting point for subtractive synthesis. Close a low-pass filter and it becomes a warm bass; open it and it becomes a brass stab; stack detuned copies and you get the supersaw. Apple lists it as ideal for string, pad, bass and brass sounds.^[1] For the stacked version, see the supersaw sound design tutorial.

• 2nd harmonic One octave above the note, at half level.
• 3rd harmonic An octave and a fifth above, at one-third level.
• 4th harmonic Two octaves above, at one-quarter level.
• Higher harmonics Crowd closer together and get quieter; this is the fizz a low-pass filter removes first.

Square and pulse waves: what does pulse width do?

A square wave contains only the odd-numbered harmonics, again falling in proportion to harmonic number, which gives the hollow tone people often compare to a clarinet.^[3] Pulse width changes that recipe: it is the length of the high part of the cycle compared with the whole cycle, so 50% is a square and 10% is a narrow pulse.

Moving away from 50% weakens the fundamental and lower harmonics, and the tone gets thinner and more nasal. Paul Wiffen's Synth School in Sound On Sound adds that a 30% pulse sounds the same as a 70% one, and that some analog synths limit width to 5-95% or 10-90% because extreme settings can silence the oscillator.^[3] Ableton's Analog describes the same behavior: low width values sound tinny or pinched, and 100% gives a perfect square with only odd harmonics.^[4]

Pulse width modulation (PWM)

Point a slow LFO at pulse width and the harmonics keep shifting, which gives the classic lush, chorus-like PWM pad and bass. Synth School suggests an LFO of no more than one cycle per second on low notes, and backing off the depth if the sound drops out at the narrowest point of each cycle.^[3]

Triangle wave: soft odd harmonics

A triangle wave contains the fundamental plus odd harmonics only, like a square, but its higher harmonics roll off much faster, so it sounds softer. Apple recommends it for flutes, pads and vocal "ooh" sounds.^[1]

In beat production the triangle has two quiet jobs. It makes a gentle lead that sits under a vocal without fighting it, and it works as a sub that is slightly more audible than a sine on small speakers because it carries a little upper content. It is also the go-to LFO shape for smooth sweeps, a separate job covered in the LFO guide.

Which waveform should you use for sub, bass, lead and pad?

Pick the waveform whose harmonics already resemble the finished sound, then shape it. The table gives starting choices for common parts in trap, drill, house and R&B; treat them as defaults, not rules.

Many synths bundle a sub oscillator for the low-end job. Ableton's Wavetable, for example, has a sub oscillator whose Tone control produces a pure sine at 0% and adds harmonics as you turn it up, and it can drop one or two octaves below the played note.^[4] For the wide growl version of a saw bass, the Reese bass tutorial goes deeper.

How to pick a starting waveform for any patch

Choose the waveform by listening for three qualities in your reference: how bright it is, whether it sounds hollow and how much low end it carries. These steps work in any synth.

1. Open everything
   Load an init patch, open the filter fully, set the amp envelope to 0 attack and full sustain, and turn off effects so you hear the raw wave.
2. Describe the target
   Write one word for the sound you want: pure, soft, hollow, nasal or bright. That word maps almost directly to sine, triangle, square, pulse or saw.
3. A/B the candidates
   Play the same two-bar phrase on each candidate wave at matched volume. Louder always sounds better, so level-match before judging.
4. Test the extremes
   Play the lowest and highest notes of the part. A saw that works in the middle can turn harsh up high, and a triangle can disappear down low.
5. Shape, then decide
   Add the filter and envelope you plan to use. If two waves now sound close, keep the one with fewer harmonics; it will be easier to mix.
6. Layer when needed
   If one wave cannot cover both weight and brightness, split the jobs: sine for the sub, saw or square above it, each on its own oscillator.

Once the wave is chosen, the filter decides how much of it the listener hears. Synth filter types explained covers that next stage, and the subtractive synthesis tutorial builds four full patches from a single saw.

Beyond the basic four: noise, modeled shapes and aliasing

Modern synths add shapes that fall outside the classic four. Noise contains energy across the whole spectrum and is used for snares, wind and texture; Apple notes that pink noise drops 3 dB per octave and red noise 6 dB per octave.^[1] Ableton's Drift includes a Shark Tooth wave based on a classic Moog analog shape and a Saturated wave that works well for bass.^[4] Wavetable synths go further and let you scan between hundreds of shapes; see what wavetable synthesis is.

Harmonically rich waves also bring aliasing, a digital side effect in which harmonics above the sample-rate limit fold back as unrelated pitches. Ableton's Operator manual notes that complex waveforms tend to be more sensitive to aliasing than pure sines and that it mostly occurs at high frequencies.^[4] Vital's developer states its oscillators have a sharp cutoff at Nyquist for almost no aliasing.^[5] If high saw notes sound grainy or out of tune, raise the synth's oversampling or quality setting before blaming the waveform.