Loudness and true peak by destination
Streaming services turn loud tracks down to a reference level, so the loudness that matters depends on where the track is played. Only some of these numbers are published by the platform itself; the last column says which.
| Destination | Loudness | True peak | Who says so |
|---|---|---|---|
| Spotify | -14 LUFS integrated (normalization reference, ITU 1770) | below -1 dBTP; below -2 dBTP if the master is louder than -14 LUFS | Spotify for Artists |
| Apple Music | No published LUFS target. Sound Check plays songs at about the same loudness; about -16 LUFS is a common working reference | at least 1 dB of headroom | Apple Digital Masters (headroom only) |
| YouTube | No published target. Loud uploads are turned down, quiet ones are not turned up; about -14 LUFS is a common working reference | below -1 dBTP (our recommendation: audio is re-encoded) | YouTube Help (behaviour only) |
| Apple Podcasts | about -16 LKFS, ยฑ1 dB | not above -1 dBFS | Apple Podcasts for Creators |
| Broadcast (EBU R 128) | -23.0 LUFS (ยฑ1.0 LU where the target is not practicable, e.g. live) | not above -1 dBTP | EBU R 128 v5.0 |
| Club and DJ files | No normalization; club masters commonly sit around -9 to -7 LUFS | true-peak safe | Practice, not a standard |
LUFS and LKFS are the same unit (EBU R 128). A difference of 1 LU is 1 dB (EBU Tech 3341). The full explanation is in LUFS vs dB and how many LUFS a master should be.
Frequency ranges at a glance
The bands below are conventions, not a standard: engineers draw the boundaries in slightly different places. They are a map for finding a problem, not settings to copy.
| Range | Approx. frequencies | What lives there | Typical problem |
|---|---|---|---|
| Sub | 20โ60 Hz | kick fundamental, sub bass | rumble that uses up headroom |
| Bass | 60โ250 Hz | bass line, body of the kick | boom, kick and bass masking each other |
| Low mids | 250โ500 Hz | warmth and body of most instruments | mud when many parts stack up |
| Mids | 500 Hzโ2 kHz | the core of vocals, guitars, keys | boxy or honky tone |
| Upper mids | 2โ5 kHz | presence and intelligibility | harshness and listening fatigue |
| Highs | 5โ10 kHz | attack, consonants, cymbals | sibilance |
| Air | 10โ20 kHz | openness, sparkle | brittle or hissy top |
On a master the moves are broad and small. If one instrument is the problem, fix it in the mix: an EQ on the whole master changes every instrument in that range. See why mixes sound muddy.
Sample rate and bit depth
| Destination | Sample rate | Bit depth | Note |
|---|---|---|---|
| CD (Red Book, IEC 60908) | 44.1 kHz | 16-bit | stereo PCM; reduce to 16-bit with dither |
| Spotify delivery | 44.1 kHz or higher, keep the native rate | 24-bit if that is the native master; 16-bit only if no higher master exists | FLAC preferred, WAV accepted |
| Apple Digital Masters | highest native sample rate | 24-bit sources | do not upsample |
| Working files and mixdowns | the project rate | 24-bit or 32-bit float | float files do not clip inside the file |
A 16-bit file has a theoretical dynamic range of about 96 dB and a 24-bit file about 144 dB. The details, and when dither is needed, are in sample rate and bit depth for mastering.
The mastering chain in one line
Corrective EQ โ compression (if needed) โ tonal EQ or saturation (if needed) โ stereo adjustments (carefully) โ true-peak limiter โ dither (only when reducing bit depth).
There is no single correct order, but the limiter stays last because it sets the final peak level. Each stage is explained in the mastering chain, step by step, and the difference between the two dynamics tools in compressor vs limiter.
Checklist before you release
โ Compared with a reference track at matched loudness, not louder.
โ Integrated loudness where you intended it, and true peak at or below your ceiling (for Spotify: below -1 dBTP).
โ Checked in mono: the vocal and the low end are still there.
โ Listened start to finish at a normal volume, including the first and last seconds.
โ Exported at the native sample rate, 24-bit, unless the destination asks for something else.
โ Checked on headphones and on a phone speaker.
Using it in Quantara
The studio shows momentary loudness and peak levels while you play, and the master preview reports integrated loudness, true peak and loudness range, so the numbers on this page can be checked on your own track instead of guessed. A free account includes one free export per account (MP3 320, with a watermark), no card required. Open the studio.
Frequently asked questions
There is no single number. Spotify normalizes to -14 LUFS integrated, so a louder master is turned down on its default setting. Master for the song, keep true peak below -1 dBTP, or below -2 dBTP if the master is louder than -14 LUFS (Spotify's recommendation).
Yes. EBU R 128 states that LUFS is equivalent to LKFS, the unit used in ITU-R BS.1770.
Keep the project's native sample rate (44.1 kHz or higher) and export 24-bit. Spotify asks for 24-bit when that is the native master and says not to downsample a master made above 44.1 kHz. Use 16-bit with dither only for CD.
Last, after EQ and compression, because it sets the final peak level. Only dither comes after it, and only when the bit depth is reduced, for example to 16-bit.
Sources
Loudness normalization on Spotify โ Spotify for Artists
Audio file formats for Spotify โ Spotify for Artists
Apple Digital Masters (April 2021) โ Apple
Audio requirements โ Apple Podcasts for Creators
EBU R 128 v5.0: Loudness normalisation and permitted maximum level of audio signals (2023) โ European Broadcasting Union
EBU Tech 3341 v4: Loudness Metering, 'EBU Mode' metering to supplement EBU R 128 (2023) โ European Broadcasting Union
Control video volume on your device โ YouTube Help
Compact Disc Digital Audio โ Wikipedia
Audio bit depth โ Wikipedia