Sound & signal
LUFS
LUFS means loudness units relative to full scale — a loudness measurement weighted to match how ears hear rather than what a peak meter reads. Streaming platforms and DJ software both normalise to a LUFS target.
Measured in our catalog
Records first released before 2000 sit 3.33 dB quieter than the 2020s across our catalog — -13.47 against -10.14 — which is the loudness war, measured.
The climb is monotonic across all four eras and the population is not small: 110,225 tracks carry both an analysed level and a researched first-release year. It is also an understatement rather than an exaggeration, because a reissue is read at its remaster's level rather than at the original pressing's. One caveat travels with the unit: what we store is RMS dBFS, not LUFS. LUFS is K-weighted and gated, so the two agree on direction and differ in value — this is the shape of the trend, in our own measurement.
-13.47 dB
Median, before 2000
-10.14 dB
Median, 2020s
3.33 dB
The climb
110,225
Tracks with a first-release year
| First released | Tracks | Median | Quietest tenth | Loudest tenth |
|---|---|---|---|---|
| Before 2000 | 5,531 | -13.47 dB | -18.79 dB | -9.94 dB |
| 2000s | 7,513 | -11.96 dB | -15.48 dB | -9.53 dB |
| 2010s | 48,602 | -10.43 dB | -12.85 dB | -8.32 dB |
| 2020s | 48,579 | -10.14 dB | -12.59 dB | -7.85 dB |
Measured on 11 August 2026 across public tracks carrying an analysed level and a researched first-release year (110,225 of 443,460), grouped by first release rather than by store date, and measured as RMS dBFS rather than K-weighted LUFS. Method: the full loudness distribution.
Why it matters
LUFS is the unit behind every "why did my track get turned down" conversation. A peak meter tells you how close a file came to the ceiling; a LUFS reading tells you how loud it will actually feel, which is the thing a normaliser acts on and the thing a floor responds to.
For a DJ the practical consequence is that a crate spanning two decades is a crate spanning several decibels. Records mastered in different eras arrive at genuinely different levels, and that difference is not a mistake anyone made — it is the loudness war, and it is measurable in any catalog deep enough to contain both ends of it.
Frequently asked
What is a good LUFS level?+
For streaming, around −14 LUFS integrated is the usual target because that is where the big platforms normalise. For a club file the answer is whatever the master needs, since the player is not normalising anything unless you turned it on.
Is LUFS the same as dBFS?+
No. dBFS measures a level against digital full scale with no weighting. LUFS weights the measurement to match hearing and averages it over time, so two files at the same peak dBFS can read several LUFS apart.
Why are older records quieter?+
Because mastering got louder. Across our catalog the median level climbs 3.33 dB from records first released before 2000 to those from the 2020s, measured over 110,225 tracks.
Related terms
Loudness
Loudness is a track’s average level in decibels below full scale, so every value is negative and closer to zero is louder. It decides how far you move the gain, not how hard the record hits.
dBFS
dBFS means decibels relative to full scale — the digital measurement where 0 is the loudest a file can be and everything else is negative. A track at −10 dBFS sits ten decibels below the ceiling.
Gain normalisation
Gain normalisation sets every track to the same measured loudness automatically, so a quiet record and a loud one arrive at the channel matched. It corrects for mastering, not for how hard a record hits.
True peak
True peak is the highest level a waveform actually reaches between the samples that describe it. A file whose every sample sits below the ceiling can still overshoot it once a player reconstructs the sound.
Explore: How the analysis works · Energy arc tool · Browse the catalog · All DJ terms