Studio Mix · concept guide

Gain Staging: Why Every Stage of a Mix Wants Headroom

Where levels should sit at tracking, on the channel, on the buses and into mastering, why "it is digital, it does not clip" is only half true, and the drill that trains level matching by ear.

Play it first · Level Match

Free drill, no account

Level Match

Balance two sources by ear, in decibels.

6 rounds, about two minutes. Headphones help. Nothing is saved until you choose to.

Gain staging is the least glamorous concept in mixing and the one that quietly decides whether everything after it works. A channel that arrives hot pushes every plugin on it into a range it was not designed for, a bus that is already at the ceiling leaves the bus compressor nothing to do but grab, and a mix printed with no headroom hands the mastering engineer a problem instead of a song. None of it sounds like distortion. It sounds like a mix that never quite opens up.

The drill above trains the underlying ear skill: matching the level of two sources so that you are comparing tone rather than loudness. The cards then walk the chain: recording levels, clip gain to a sensible nominal on every channel, headroom when mixing entirely in the box, feeding the mix bus at a level that leaves the compressor a job, and the headroom a master needs on delivery.

The concepts

8 lessons from the Studio Mix deck

  1. 1

    Gain Staging Before Mixing

    Gain staging means setting each track's level before you process anything, so its AVERAGE lands around -18 dBFS rather than slamming near full scale. On normal programme material that puts peaks somewhere around -10 to -6 dBFS, which leaves headroom for transients to breathe and for the mix bus to sum many tracks without clipping. Average, not peak, and the difference is about 12 to 18 dB. Analog-modeled plugins are calibrated so that -18 dBFS corresponds to 0 VU on the hardware they model, and a VU meter is deliberately slow: it reports an average and barely responds to transients at all. Feed a plugin a signal that is too hot and its modeled circuitry is overdriven, which is why a compressor can sound harsh for no obvious reason. Set the peaks to -18 instead and you make the opposite mistake, running about 14 dB under the level the plugin was built for, where it does almost nothing.

    In practice: A vocal recorded hot, peaking near 0 dBFS, makes an analog-modeled compressor sound clippy. Pull the clip gain down until the average reads -18 (about 15 dB here, leaving peaks near -8) and the same compressor turns smooth and musical.

    Gain stage by the AVERAGE: -18 dBFS average, which leaves peaks around -10 to -6.

  2. 2

    Gain Staging: Clip Gain to -18 dBFS

    Before inserting any plugins, gain-stage every track: use clip gain to set the AVERAGE level around -18 dBFS, set the channel trim to unity, and leave the fader at 0 dB. Average, not peak. -18 dBFS is the analog nominal level plugin models expect, and nominal has always meant average: 0 VU on the hardware being modelled is a slow, averaged reading. On normal material an average of -18 leaves peaks somewhere around -10 to -6 dBFS, which is the headroom that lets all the tracks sum on the bus without clipping.

    In practice: A session with a guitar peaking at -6, a vocal at -24, and a clipped bass at 0 dBFS: put an RMS meter on each and clip-gain until the average reads about -18 before any plugin goes on.

    Clip-gain to a -18 dBFS AVERAGE, not -18 dBFS peaks: peaks land around -10 to -6.

  3. 3

    Recording Levels & Gain Staging

    In 24-bit digital, aim for an average recording level around -18 dBFS, with peaks landing well short of 0. The "record as hot as possible" habit is a holdover from analog tape, where higher levels improved the signal-to-noise ratio. Digital does not work that way: 24-bit has about 144 dB of dynamic range so noise is a non-issue, and digital clipping at 0 dBFS is harsh and unrecoverable. Loudness is set later, not at tracking.

    In practice: A player wanting the meters near 0 "because louder is better": pull the preamp back so peaks sit around -12 to -10, the quality is identical and you keep headroom.

    Track around -18 dBFS average, hot levels are an analog-tape habit digital does not need.

  4. 4

    Headroom Management in ITB Mixing

    DAWs use floating-point math with effectively infinite internal headroom, so why does headroom still matter? Because plugins, especially analog-modeled ones, are calibrated for specific input levels and distort unmusically when driven too hot. And the D/A converter at your interface output is fixed-point, it hard-clips above 0 dBFS. Proper gain staging keeps every stage of the chain in its sweet spot.

    In practice: A mix where analog-modeled plugins sound harsh despite the internal bus never clipping: the plugins are being fed too hot, gain-stage so they see their designed level.

    Floating-point won't clip internally, but plugins distort when overdriven and the D/A output hard-clips.

  5. 5

    Gain Staging the Mix Bus

    Analog-modelled bus plugins are calibrated so that -18 dBFS corresponds to 0 VU on the hardware they model. That is an AVERAGE level, measured on a steady tone, not a peak level. So aim for an average of about -18 dBFS on the mix bus. Music with a normal crest factor runs 12 to 18 dB above its average, so that leaves peaks somewhere around -10 to -6 dBFS with plenty of headroom left. Both errors are common. Too hot and the compressor and saturator overreact and sound harsh. Too quiet, which is what happens if you set the PEAKS to -18, and they barely engage: the average lands near -32 dBFS and the plugins do almost nothing while you turn their knobs up wondering why.

    In practice: An SSL bus comp and tape plugin sounding harsh at gentle settings: the bus averages around -9 dBFS. Pull the faders down collectively until the average reads -18 and the peaks land near -8, and they smooth out.

    The -18 dBFS calibration is an AVERAGE, not a peak. Average about -18 dBFS, peaks near -10 to -6.

  6. 6

    Headroom & Gain Staging for Mastering

    A mix handed off for mastering should leave room to work: peaks around -6 dBTP, and no limiter or maximizer on the mix bus, loudness is the mastering engineer's job, not the mixer's. The reason this matters so much: if a mix is delivered clipped, peaks slammed into 0 dBFS with flat-topped waveforms, that distortion is permanent. Turning it down later just gives you quieter distortion.

    In practice: A "normalized" mix with flat-topped transient peaks cannot be rescued in mastering, ask for a fresh export with the mix-bus limiter removed and peaks near -6 dBTP.

    Deliver mixes with headroom and no bus limiter, clipping is permanent.

  7. 7

    Faders, Pans, and dB

    The fader sets how loud a track is, measured in decibels. Its "0" or "U" mark is unity, signal passes through with no change in level, and that is your reference starting point, with room to push up or pull down from there. The pan knob sets where the track sits between your two speakers: hard left, hard right, or anywhere in between. Decibels are logarithmic, not linear, and the two numbers people mix up are worth keeping straight. 6 dB doubles the signal's amplitude, so it is a big, obvious jump on a meter. But PERCEIVED loudness, how much louder it actually sounds to you, takes about 10 dB to double. That gap is why the bottom of a fader's travel barely does anything while the top inch covers a huge range: small moves up high matter far more than they look.

    In practice: A lead vocal almost always sits panned dead center, where both ears hear it equally and it anchors the listener. Supporting parts get nudged off-center so they don't fight it for the same spot.

    Unity is your reference; pan creates space.

  8. 8

    The Volume Trick: Always A/B at Matched Level

    Parallel processing always raises the overall level, because you are adding signal to the dry source. And our ears reliably hear louder as "better", so without level-matching, you will mistake the volume bump for an improvement. The discipline: measure the output with and without the parallel signal, trim them to match, then A/B. If it does not sound better at matched level, it is not helping, you were hearing volume.

    In practice: A mandolin that sounds "bigger and warmer" with parallel compression: measure the level rise, turn the dry signal up to match, then compare: keep it only if it still wins.

    Parallel processing raises level, and louder fools the ear, always A/B at matched loudness.

These lessons are the teach side of knowledge cards from the Studio Mix deck. In the daily plan each one is followed by a quiz, spaced over weeks, so it stays known rather than read once.

The plan

Reading this once is not the same as hearing it.

A daily session schedules the drill above at your level, quizzes these cards until they stick, and moves you on when the numbers say you are ready. Twenty minutes a day, in the browser, no plugins.

Related concept guides