What EXPECTED TP tells you

A digital file is a list of numbers: the level at one instant, tens of thousands of times a second. A sample-peak meter reports the largest of those numbers, which is a fact you can verify by reading the file.

But nobody listens to a list of numbers. A converter has to draw a continuous waveform back through those points, and the curve it draws does not stay inside them. Between two samples that both sit at -1 dB, the reconstructed waveform can arc above both. That arc is real signal. It leaves the converter, and if it goes above 0 dB something downstream clips, even though every number in the file was below the line.

You cannot see between two samples by looking harder at them. You have to reconstruct the curve, and to reconstruct it you make more samples: upsample, look at the denser version, take the peak of that.

How much denser matters. ITU-R BS.1770 specifies a minimum of 4x oversampling for true-peak measurement. Four times catches most inter-sample peaks but still under-reads, because the true maximum can sit between two of the new samples too. Mastering Suite measures at 16x with a 512-tap-per-phase polyphase filter, which is a deliberately expensive choice for a measurement that only has to be right rather than fast.

EXPECTED TP is calculated from the ceiling setting with a fixed 0.01 dB allowance. It helps make that configured relationship visible in the interface.

It does not inspect the music passing through the plug-in. It cannot replace a meter on the actual output, and it does not predict or guarantee the measured true peak of a rendered, encoded, or exported file.

Mastering Suite shows an EXPECTED TP figure beside the ceiling. It is derived from the ceiling setting with a fixed 0.01 dB allowance; it is not a measurement of the audio output and it does not guarantee the true peak of a rendered, encoded, or exported file.

It is a readout, not a parameter: there is nothing to turn, and switching the display off produces a byte-identical file. It is not a correction or an output meter. Use a true-peak meter on the rendered file when a delivery specification requires a measured limit.

For streaming delivery the practical advice has not changed: leave real headroom. Platforms transcode to lossy formats, and lossy encoding can move peaks in ways that cannot be inferred from the source file. A ceiling of -1.0 dBTP is a common starting point.

EXPECTED TP is not a new target to chase. Treat it as a clear indication of the configured ceiling relationship, then use a true-peak meter on the actual output and exported file when compliance matters.