LUFS, True Peak & Loudness Standards Explained

If you learned to mix between 1995 and 2015, you learned to watch the VU meter. You learned that 0 VU was "nominal level" and that pushing into the red was bad. Then streaming happened, and everything you knew about loudness became obsolete overnight. The VU meter measures voltage — a physical electrical property that has nothing to do with how humans perceive loudness. LUFS measures perceived loudness — how loud something actually sounds to a human ear. They are not the same thing, and confusing them is why so many independent releases sound crushed and lifeless on Spotify.

LUFS: A Human-Centric Measurement

LUFS stands for Loudness Units relative to Full Scale. It was standardized by the ITU (International Telecommunication Union) in recommendation BS.1770 and has gone through four revisions — the current standard is BS.1770-4. Unlike peak or RMS meters, LUFS applies a frequency weighting filter (called the K-weighting filter) that mimics human hearing sensitivity. We hear midrange frequencies (2-4kHz) as louder than bass frequencies at the same actual SPL. LUFS accounts for this. A 50Hz sine wave at -10dBFS and a 3kHz sine wave at -10dBFS measure identically on a peak meter. On a LUFS meter, the 3kHz tone reads about 6-8 LU higher.

Integrated vs. Short-Term vs. Momentary

LUFS comes in three flavors, and they serve different purposes. Integrated LUFS is the average loudness of the entire track from start to finish. This is what streaming platforms use for normalization. If Spotify's target is -14 LUFS and your track measures -10 LUFS Integrated, Spotify turns it down by 4dB. You didn't make it louder — you just made it quieter with less dynamic range.

Short-Term LUFS measures a 3-second sliding window. This tells you how loud individual sections are — verses vs. choruses, drops vs. breakdowns. A track with a -14 LUFS Integrated measurement might have verses at -16 LUFS Short-Term and choruses at -11 LUFS Short-Term. That 5 LU swing is your dynamic range, and it's what makes a chorus feel like a chorus.

True Peak: The Hidden Clipping

A regular peak meter samples the digital audio waveform at specific points. True Peak sampling oversamples by 4x and reconstructs the analog waveform shape between samples. This matters because digital-to-analog converters (the thing that turns your WAV file into actual sound coming out of speakers) reconstruct the continuous waveform between sample points. If the reconstructed waveform exceeds 0dBFS, the DAC clips — even though your digital meter showed everything under 0dBFS.

Lossy encoding (MP3, AAC, Ogg Vorbis — the formats streaming platforms actually serve to listeners) makes this worse. The encoding process alters the waveform shape and can create new peaks that didn't exist in the original WAV. A master that shows -0.3dBFS True Peak in WAV format might hit +1.2dBFS after MP3 encoding. That's audible distortion on Spotify, Apple Music, and YouTube. The fix: leave at least -1.0dBFS True Peak headroom. Many mastering engineers now target -1.5dBFS to be safe across all encoders.

Platform Targets Cheat Sheet

Spotify normalizes to -14 LUFS Integrated with a -1.0dBFS True Peak limit. Apple Music uses -16 LUFS (Sound Check enabled). YouTube normalizes to approximately -14 LUFS. Tidal and Amazon Music both target -14 LUFS. But here's the critical nuance: these are normalization targets, not mastering targets. Most professionally mastered commercial releases measure between -10 and -8 LUFS Integrated. They get turned down by the platform — but they retain their dynamics, transient impact, and perceived energy. A master at -14 LUFS that gets turned down to -14 LUFS sounds the same. A master at -8 LUFS that gets turned down to -14 LUFS still sounds like a dynamic, punchy -8 LUFS master — it's just quieter.