Two numbers, two different questions
Integrated loudness (LUFS) is how loud the file sounds, over its whole length, weighted the way hearing is. It is what streaming platforms normalise playback against.
True peak (dBTP) is how high the signal actually goes once it is reconstructed into a waveform. It is what predicts whether the file clips.
A quiet file can clip. A loud file can have headroom. Neither number tells you the other, which is why a tool that prints one and not the other is half a check.
Your DAW's peak meter is lying to you
It shows the highest sample value. Playback does not play samples — it reconstructs a continuous waveform through them, and that waveform routinely goes higher than any individual sample.
So a file showing −0.1 dBFS can clip in a lossy encoder, in a converter, or on a phone. Nothing was wrong with the file as a set of numbers; the problem appears when it becomes a waveform again.
DropWave measures the real thing, as ITU-R BS.1770-4 specifies. Above 0 dBTP the row turns and says clips — the one unambiguous fact in the whole result. More →
Loudness, gated properly
Integrated LUFS is not an average of the file. The standard throws out silence and quiet passages before averaging, which is exactly why a proper reading and a naive average are different numbers — and why two tools can disagree if one of them skipped it.
DropWave measures to ITU-R BS.1770-4, gating included. More →
Reading the numbers
| You see | What it means |
|---|---|
| True peak above 0 dBTP | It will clip on playback. Lower the ceiling and re-export. |
| True peak −1 to 0 dBTP | No headroom for a lossy encoder. Frequently comes back over after MP3 or AAC. |
| True peak −1 dBTP or lower | The common recommended ceiling for anything headed to a lossy codec. |
| Loudness around −14 LUFS | The figure most often cited for streaming normalisation. |
| Loudness −8 to −6 LUFS | Typical of modern club and electronic masters. |
| Loudness −23 LUFS | EBU R 128 broadcast. |
None of these are rules. A loud master is not wrong — it will simply be turned down by anything that normalises, and the extra loudness bought you less dynamic range at the same playback level.
Where this fits
DropWave answers "what is this bounce" in one drop. That is the question you have when a file is finished, or when someone sends you one.
It is not a metering plugin for your master bus while you are mixing — momentary loudness, short-term and a live histogram are a different tool for a different moment.
Requirements
macOS 13.0 (Ventura) or later. Universal — Apple silicon and Intel. Measured locally; nothing is uploaded, because DropWave never connects to the internet. Closed beta — the Discord is the way in.