Saturation and distortion basics for mixing

Every mixing forum eventually tells you to "add some saturation". It is good advice delivered badly, because saturation is not one thing and it is very easy to overdo. This guide covers what saturation and distortion actually are, why different kinds sound different, the one technical trap (aliasing) worth understanding, and the habits that keep it useful.

It sits next to our EQ basics and compression basics guides. If you have not set sensible levels yet, start with gain staging, because how hard you drive a saturator depends on what you feed it.

Distortion is any change; saturation is the gentle kind

In the broad engineering sense, Sonarworks puts it simply: distortion "describes any change made to an audio signal". An EQ, by that definition, distorts. What producers usually mean is narrower: harmonic distortion, which Sonarworks says comes from clipping, overdriving or saturating a circuit.

When a signal is pushed past what a circuit can cleanly pass, its waveform gets flattened at the peaks. That flattening adds new frequencies, called harmonics, at multiples of the original. Those added harmonics are what you hear as warmth, grit, bite or fuzz.

The difference between saturation and distortion is mostly one of degree. Sonarworks notes that soft clipping "is also commonly referred to as saturation". In practice, "saturation" tends to mean a little, used to add weight or presence, and "distortion" tends to mean enough that you clearly hear it as an effect.

Soft clipping, hard clipping and the harmonics they add

The shape of the clipping matters. Sonarworks says "soft clipping yields mostly even harmonics, while hard clipping tends to produce odd harmonic overtones", and describes odd harmonics as giving the signal "an edgier, more aggressive sound".

It also gives the usual hardware shorthand:

• Tubes and tape both produce soft clipping when slightly overdriven, and tube designs create higher levels of even-order harmonics.
• Transistor circuits typically produce a harder clipping effect.

Treat these as tendencies, not laws. Plugin emulations of each are modelled differently, and the same article stresses that every circuit has its own blend. A tape plugin and a tube plugin from two different developers may have more in common with each other than with their own labels.

What saturation does in a mix

Because harmonics sit above the original note, saturation can make a sound easier to hear without turning it up much. That is why it is so often reached for on bass that disappears on small speakers, or on a vocal that sits slightly behind the band. It also tends to make sounds feel denser, because the added harmonics fill in around the fundamental.

Sonarworks makes two points we would underline. First, "a little distortion can go a long way". Second, a full-sounding track usually comes from several subtle uses rather than "one single use of distortion". A light touch on the bass, the drum bus and a synth often does more than one heavy saturator on the master.

It also warns that distortion recorded at the microphone or preamp stage is permanent, "as you cannot remove it later". If you are unsure, record clean and saturate in the mix, where you can change your mind.

The trap: aliasing

Digital audio can only represent frequencies up to half the sample rate. Saturation creates new harmonics, and when a harmonic would land above that limit it does not simply vanish. HoRNet Plugins, a plugin developer, describes these as artefacts "folding back into the audible spectrum as unnatural content". Unlike normal harmonics, they are not musically related to the note, which is why aliasing tends to sound harsh or gritty in a way that "warmth" does not.

The fix is oversampling. Sage Audio explains that oversampling "increases the sampling rate of the signal it's processing by a fixed multiple like 2 or 4", so in a 48kHz session "a 2x oversampling setting will make the sampling rate 96kHz". The extra headroom gives harmonics room above the audible range before they are filtered out. HoRNet adds that if a plugin does not offer enough oversampling, raising the session sample rate is the other route.

The cost is processing power. Sage Audio notes oversampling "can be CPU intensive and can cause performance issues if too high of a rate is used", and suggests that the higher your native sample rate, the less you need it. For background on rates themselves, see our sample rate and bit depth explainer.

Practical habits

These are working habits rather than rules from the sources above:

1. Match levels before you judge. Most saturators make things louder, and louder usually sounds better for a moment. Use the plugin's output control so bypassing it does not change the level, then decide.
2. Use the mix knob. Blending a heavily saturated copy with the clean signal (parallel saturation) adds density while keeping the original transients.
3. Filter around it. If saturation makes a bass muddy or a vocal fizzy, an EQ before or after it often fixes more than backing off the drive.
4. Switch oversampling on for bright, heavy material. It matters most when you push hard on sounds with lots of high-frequency content. Gentle saturation on low sounds is less likely to need it.
5. Check against a reference. A reference track is the quickest way to notice that your "warm" mix has become a crunchy one.

What we think

Saturation is the most oversold tool in mixing tutorials, and also one of the most useful. The overselling comes from the loudness trick: a plugin that adds a decibel always seems to "add life" on first listen. Once you level-match, many of those improvements disappear, and the ones that survive are the ones worth keeping.

We would also give less weight to the tube-versus-tape-versus-transistor debate than the marketing does. The useful question is simpler: does this make the part easier to hear, and does it still sound good at matched level? If you can answer yes with the cheapest saturator in your DAW, you do not need another one. Save the money, learn one tool well, and turn on oversampling when the cymbals start to sound like sandpaper.