Metering, Levels, and Phase

Types of Meters

Peak Meters

Show the instantaneous maximum level of the signal. Fast response — they catch transients and brief spikes that your ears might not register as loud. Most DAW meters are peak meters. When the meter hits 0 dBFS (decibels full scale), you’re clipping. In digital audio, clipping is absolute — there’s nothing above 0 dBFS, and anything that tries to go there gets chopped flat. It sounds harsh and it’s permanent in the recording.

RMS Meters

Show the average power of the signal over a short time window. Closer to how your ears actually perceive loudness — a sustained chord feels louder than a sharp snare hit at the same peak level, and an RMS meter reflects that. RMS meters move more slowly and smoothly than peak meters. They’re useful for understanding how “loud” something feels, as opposed to how high its peaks spike.

LUFS Meters

Loudness Units Full Scale — the current standard for measuring perceived loudness. LUFS accounting weights the frequency spectrum to match human hearing sensitivity (we’re most sensitive to midrange), then averages over time. Streaming platforms use LUFS targets: roughly -14 LUFS for Spotify, -16 for Apple Music. Essential for mastering, less relevant during tracking and recording.

VU Meters

The analog-era meter. Slow response time, designed to approximate how loud something sounds rather than how high it peaks. Calibrated so that 0 VU corresponds to +4 dBu (professional line level). Still found on outboard hardware and loved by engineers who came up on analog consoles. Not common in software, but useful to understand because the concept — metering that tracks perceived loudness — is the ancestor of LUFS.

Gain Structure Through the Chain

Every stage of the signal chain has an ideal operating level. Getting it right at each stage prevents noise from accumulating and distortion from creeping in:

1. Mic → Preamp: Set gain so the loudest peaks hit -12 to -6 dBFS. This leaves headroom for unexpected transients.
2. A/D converter: Already handled if the preamp gain is set correctly — the preamp feeds the converter directly.
3. DAW channels → plugins: Most plugins are modeled to behave best with input levels around -18 dBFS (roughly equivalent to 0 VU in the analog world). Hitting them much harder can cause them to clip or behave unpredictably. Too quiet and you’re wasting their usable range.
4. Mix bus → master output: Leave headroom for mastering. Peaks around -6 to -3 dBFS on your mix bus give a mastering engineer room to work.

The principle is simple: keep the signal comfortably within the operating range at every point. Don’t let it get too quiet (noise floor rises relative to the signal) or too loud (distortion).

Phase

Phase is the timing relationship between two copies of a signal. When two waveforms are perfectly in phase, their peaks and troughs line up — they reinforce each other and the combined signal is louder. When one is inverted (180° out of phase), the peaks of one align with the troughs of the other — they cancel.

Complete cancellation is rare in practice. More common is partial phase cancellation — certain frequencies cancel while others reinforce, producing a hollow, thin, flanged character. This happens whenever two mics pick up the same source from different distances.

Polarity vs. Phase

These get used interchangeably, but they’re different:

Polarity is a binary flip — the signal is inverted. Positive becomes negative, negative becomes positive. The polarity button on your preamp or DAW does this. It’s either flipped or it’s not.

Phase is a continuous relationship — how far one signal is shifted in time relative to another. Moving a microphone one inch changes the phase relationship differently at every frequency. Low frequencies (long wavelengths) are barely affected; high frequencies (short wavelengths) shift dramatically.

Polarity problems are easy to fix — press a button. Phase problems require either repositioning microphones or using time-alignment tools in the DAW.

Comb Filtering

When two signals of the same source arrive with a small time offset, they create alternating bands of cancellation and reinforcement across the frequency spectrum. Plot the resulting frequency response and it looks like the teeth of a comb — hence the name. Comb filtering is the most common phase artifact in multi-mic recordings.

The 3:1 rule is a guideline for recording two instruments with two microphones: the mics should be at least three times farther from each other than they are from their respective sources. The objective is to minimize phase problems from off-axis pickup — when one source bleeds into the other’s microphone. If you talk to enough people, you’ll hear at least three or four variations on this rule — sometimes it’s two mics on one source, sometimes the distance formula changes — but the concept is always the same: minimize interference by maximizing distance. Always verify with your ears and the polarity check from Chapter 10.

What to Practice

1. Watch your meters while recording. Pay attention to peak levels as you track. Get comfortable with signals peaking around -12 to -6 dBFS. Resist the urge to push them higher — headroom is free in 24-bit.
2. Compare peak and RMS. Pull up a peak meter and an RMS meter on the same track. Play a snare-heavy drum loop — notice the peak meter jumps much higher than the RMS. Now play a sustained pad — the two meters read closer together. Understanding this difference is understanding the difference between peak level and loudness.
3. Hear comb filtering. Set up two mics on the same source (two mics on a guitar amp, or two mics on a voice). Record both simultaneously. In the DAW, nudge one track forward or backward in time by small increments — 1 ms, 2 ms, 5 ms. Listen to how the tone changes. That hollow, phased sound is comb filtering.
4. Use the polarity button. With two mics on the same source summed to one output, flip the polarity on one channel. If it sounds fuller, leave it flipped — the mics were partially out of phase. If it sounds thinner, flip it back. This should be automatic on every multi-mic session.