What Is Gain Staging? A Practical Guide for Mixing

Gain staging is the practice of setting the level at every point in your signal chain, from microphone to master bus, so that each stage receives the level it was designed for. Done well, it keeps your signal above the noise floor, below clipping, and in the range where analog gear and analog-emulation plugins behave the way you expect.

What is gain staging?

Gain staging means managing level from one stage of the signal chain to the next. A stage is anything that can change level or add its own sound: a preamp, a converter, a plugin, a bus.

• Noise floor: the hiss and hum a device produces with no signal present.
• Headroom: the distance between your typical operating level and the point where a device clips or distorts.

Good gain staging keeps the signal between the two.

The clearest way to see every stage is a block diagram, a drawing of a device as boxes and arrows. Andrew Scheps, award-winning mixer, producer and Campus mentor (Beyoncé, Lana Del Rey, Red Hot Chili Peppers), uses the LA-2A as an example. From the diagram alone, he says, you can tell "that gain knob is just an amplifier. It's a fader. It doesn't affect the compression at all and you can tell that by looking at the diagram." Knowing which knobs act before the processing and which act after is the heart of gain staging. Our guide to audio engineering fundamentals explains how to read block diagrams.

Why does gain staging matter?

Every analog stage adds noise, and distortion as it nears its limits. Gain staging decides how much of each you get, on purpose.

In a free Core Audio Program info session, Scheps summed up what every piece of hardware brings along:

"all analog gear, no matter what it's supposed to do and what you buy it for is a noise generator, a harmonic distortion generator, and usually high and low pass filters because that's what transformers are."

Run a stage too low and its noise becomes a larger share of the signal. Run it too hot and it distorts.

Tyler Page, the program's teaching assistant, told a story in the same session: two hardware preamps he bought for his mix bus made everything crunchy. Turning down the preamp's gain knob did nothing, because that knob sat after the stage that was overloading. The fix was level and impedance matching, and the block diagram would have shown him where the gain lived.

What is the difference between dBFS, dBu and dBV?

dBFS measures digital level relative to the maximum a fixed-point system can store. dBu and dBV measure analog voltage.

• dBFS: decibels relative to full scale. 0 dBFS is the largest value a fixed-point converter or file can hold, so levels are negative numbers below it.
• dBu: voltage referenced to 0.775 V RMS. Professional line level is +4 dBu, about 1.23 V.
• dBV: voltage referenced to 1 V RMS. Consumer line level is -10 dBV, about 0.32 V, roughly 12 dB below pro level.
• VU: a meter that shows average level over roughly 300 milliseconds. On pro gear, 0 VU is usually calibrated to +4 dBu.

Where does "-18 dBFS = 0 VU" come from?

From converter calibration. If 0 dBFS equals +22 dBu on your interface, then +4 dBu, the analog 0 VU, lands at -18 dBFS. Other alignments differ: the EBU R68 convention puts 0 dBu at -18 dBFS, and the SMPTE convention puts +4 dBu at -20 dBFS. Look up your interface's maximum level in dBu to find your own number.

Does gain staging matter in a DAW with floating-point audio?

Yes, for different reasons than on a console. Most DAWs mix in 32-bit or 64-bit floating point, which has so much range that a channel pushed over 0 dBFS inside the mixer usually does not clip, as long as it comes back down before a fixed-point stage.

Scheps makes the point when comparing plugins to hardware: "the way floating point audio works in digital mixing, which we go into in some detail, means that some of the specs we talk about for analog gear don't really apply in the same way." With hardware, by contrast, "you really do want to go look at specs in terms of headroom and noise floor and things like that when you're talking about analog gear."

The hard limits remain at the edges of the system:

• Recording: the A/D converter clips at 0 dBFS, and that clipping is printed into the file. 24-bit audio offers a theoretical 144 dB of dynamic range, more than the analog circuitry in front of it, so there is no reason to record hot.
• Playback and export: the D/A converter and fixed-point files, such as a 24-bit or 16-bit WAV, clip at 0 dBFS.
• Plugins: some clip internally, and analog emulations are designed to distort as you push them.

Why do analog-emulation plugins care about input level?

Because they are nonlinear. A clean digital EQ treats a quiet signal and a loud one the same way. A model of a console, tape machine or vintage compressor reacts to level like the hardware did, adding harmonics and squashing harder as you drive it.

Scheps, who developed a channel-strip plugin with Waves, explains what that distortion is for:

"what it's actually trying to do is to give you all the bad side effects of analog circuitry because it turns out that those things can be very pleasing to the ear."

The difference from hardware, he adds, is that "you have control instead of not having control." Gain staging is how you use that control. Many emulations are calibrated so that a specific level, often around -18 dBFS average, matches the original unit's nominal operating level, and some include a calibration control. Feed them hotter for more color and compression. The same applies to compressors without a threshold, like LA-2A- and 1176-style designs, where input level sets the amount of gain reduction. Our article on audio compression covers what happens next. Hearing how input level changes a compressor is a core exercise in Foundations of Compression.

Scheps teaches these building blocks, from block diagrams and level matching to digital audio, in the Core Audio Program.

What levels should you aim for when gain staging?

Aim for averages near your plugins' reference level, peaks with clear headroom, and a master below 0 dBFS. The digital targets are conventions.

How do you gain stage a mix in a DAW?

Set each track's level before its first plugin and use faders only for balance.

1. Check the recording for clipping; turning down cannot undo it.
2. Trim before the inserts. Use clip gain or a trim plugin in the first slot. Process.Audio's free Protility combines trim, gain, polarity and a high-pass filter for this job.
3. Meter average and peak. Set each track's loudest section to about -18 dBFS average, or your plugins' documented reference.
4. Match plugin output to input, so the next plugin sees what it expects.
5. Watch the buses. Summed tracks get louder; trim a bus input if it overdrives its processing.
6. Leave master headroom for mastering.

Practice exercise: the 20-minute level check

Use your ears and meters to find where your DAW clips and where it does not.

1. Prove floating point. Push a drum loop's fader well past 0 dBFS, then lower the master fader by the same amount. In most floating-point mixers you will hear no distortion.
2. Find your reference. With a peak and RMS or VU meter, use clip gain to set the loudest section to about -18 dBFS average. Note the peak reading.
3. Drive an emulation. Insert a tape, console or saturation plugin. Raise its input 6 dB and lower its output 6 dB, so only the drive changes.
4. Compare at matched level, since louder tends to sound better. Note changes in brightness, density and transients.
5. Repeat with an LA-2A-style compressor. Change only the clip gain by 6 dB and watch the gain reduction move while Peak Reduction stays put.
6. Check the bounce. Push the master over 0 dBFS, export a 24-bit WAV, and look at the waveform. The flat tops are clipping.

Frequently asked questions

What is gain staging in simple terms?

Gain staging is setting the right level at each point in a signal chain so every device or plugin gets the level it was designed for. The goal is a signal well above the noise floor and safely below clipping. It is not about making things loud; it is about making each stage behave predictably so the final mix sounds the way you intend.

Is -18 dBFS the right level for gain staging?

It is a useful default, not a rule. -18 dBFS comes from converter calibration, where +4 dBu, or 0 VU, lands at -18 dBFS when 0 dBFS equals +22 dBu. Other common alignments put +4 dBu anywhere from -14 to -20 dBFS. Many analog-emulation plugins use -18 dBFS as their reference, but check each plugin's manual for its own calibration.

Can you clip inside a DAW if it uses floating point?

Internal channels and buses in a floating-point mixer rarely clip, because the format has an enormous range above 0 dBFS. Clipping still happens at the A/D converter while recording, at the D/A converter, when you export to a fixed-point file such as a 24-bit WAV, and inside plugins that clip or saturate by design. Keep the master below 0 dBFS.

Does gain staging affect the sound of plugins?

It affects nonlinear plugins: analog emulations, saturators, tape and console models, and compressors whose response depends on input level. Driving them harder adds harmonics and compression. Clean linear processors, such as a transparent digital EQ, sound the same at any level in floating point. That is why you set level before the first insert, not with the fader.

Gain staging makes sense once you understand what each stage does to your signal. Andrew Scheps teaches that foundation, from the physics of sound to block diagrams, electronics and digital audio, in the Core Audio Program: four on-demand courses for all levels, with a teaching assistant, lifetime access and a 14-day money-back guarantee. You can read what students say on the reviews page.