top of page

What Is True Peak, and Why Does It Matter for Streaming?

1 day ago
4 min read

If you've ever exported a master that never touched 0 on your DAW's meter, uploaded it, and then heard a crunchy little edge on the streaming version, true peak is probably part of the story. It's one of those terms that shows up on every mastering spec sheet and in every limiter's settings, but almost nobody explains what it is in plain language. So here's how I think about it.

What is true peak?

Digital audio is stored as samples, individual snapshots of the waveform taken thousands of times a second. A normal peak meter (a sample peak meter) just looks at those snapshots and reports the highest one.

The problem is that the real, continuous waveform doesn't stop at the samples. When the file is converted back to analog, or resampled and encoded for streaming, the signal gets rebuilt between those snapshots. On loud, heavily limited material, the rebuilt waveform can swing higher than any individual sample. Those are called inter-sample peaks.

True peak is a measurement that estimates those in-between peaks. A true peak meter oversamples the signal, looking at it at a much finer resolution, so it can catch overs a sample peak meter would miss. The method is defined in the ITU-R BS.1770 standard, the same family of standards behind LUFS loudness measurement. You'll see it written as dBTP (decibels true peak).

In plain terms: sample peak tells you what's in the file. True peak tells you what the file is likely to do once it's played back or converted.

Why it matters for streaming

Here's where streaming comes in. Services don't usually play your WAV exactly as you delivered it. Spotify, for example, says it processes delivered files internally to prepare them for both compressed and lossless streaming. Lossy encoding (Ogg/Vorbis and AAC on Spotify) changes the waveform slightly, and those changes can push peaks higher than they were in your master.

If your master is already sitting right at the ceiling, that extra overshoot has nowhere to go. The result can be clipping or added distortion in the encoded version, the kind of harsh, gritty edge on snares, cymbals, and loud vocals that wasn't in your mix.

Spotify spells this out in its loudness guidance for artists. Its mastering tips say to:

  • Keep your master below -1 dB TP (true peak) max, which Spotify says is best for lossy formats and makes sure no extra distortion is introduced in transcoding.

  • If your master is louder than -14 dB integrated LUFS, keep true peak below -2 dB TP, because louder tracks are more susceptible to extra distortion when encoded for streaming.

Spotify also says those guidelines apply to tracks delivered for lossless playback, so this isn't only a "lossy listeners" issue.

True peak also affects how loud you get played

There's a second, less obvious reason to care. Spotify normalizes playback toward -14 dB LUFS. Quieter masters can be turned up toward that target, but Spotify says it considers the track's headroom and leaves 1 dB of headroom for lossy encodings when it does.

Spotify's own example: a track measured at -20 dB LUFS with a true peak maximum of -5 dBFS only gets lifted to -16 dB LUFS, not all the way to -14. So true peak isn't just a distortion safeguard. On a quiet, dynamic master, it can also limit how much the platform will turn you up.

I covered the loudness side in more detail in How Loud Should My Master Be for Spotify?. True peak is the other half of that conversation.

Sample peak vs true peak in practice

A few habits that keep me out of trouble:

  • Meter true peak, not just sample peak. Most modern limiters and loudness meters have a true peak mode or a dBTP readout. Use it on the final master.

  • Set your limiter's ceiling with true peak in mind. If your limiter has a true peak option, turn it on for streaming masters and set the ceiling to the headroom you're aiming for. If it only limits sample peaks, leave more room and check the result on a true peak meter.

  • Check the whole song, not a loop. The highest true peak often lands on a big chorus hit or a final crash, not the part you've been looping all afternoon.

  • Check after any processing that follows the limiter. Dither, sample rate conversion, or a final gain change can all move the peaks. Measure the exact file you're delivering.

Does every master need to hit -1 dBTP exactly?

No. The numbers above are ceilings, not targets. A dynamic acoustic record might peak well below that and sound great. What I care about is that the master never relies on the last fraction of a dB to sound good.

It's also worth remembering that chasing a hotter master to "win" usually backfires on a normalized platform. Spotify turns louder masters down toward -14 LUFS, so the extra limiting doesn't buy you level. It just buys you distortion risk, and per Spotify, a tighter true peak ceiling on top of it.

A quick true peak checklist for your next release

Before you upload a master for streaming:

  • Measure integrated LUFS on the full song.

  • Measure true peak on the full song, in dBTP.

  • For Spotify, stay under -1 dBTP, or under -2 dBTP if the master is louder than -14 LUFS.

  • Listen to the loudest sections at a normal volume and on a second system, like earbuds or a car, for any harshness.

  • Measure the final export, not the session.

If you want a second set of ears on a mix or master that's headed for streaming, take a look at my online mixing and mastering. I'm always happy to talk through where your levels and peaks are sitting and what they'll do once the platform gets hold of them.

— Jim

If you'd like one-on-one help with your own mix, vocals or audiobook, you can book an online session with me.

Comments


© 2026 East Coast Recording Company  

215-287-0717  | 780 Falcon Circle Suite 300, Warminster, PA 18974

  • Grey Facebook Icon
bottom of page