---
title: "Sample Peak vs True Peak (dBFS vs dBTP) | Gawergy Audio"
description: "Learn how sample peaks differ from reconstructed true peaks, what dBFS and dBTP mean, why inter-sample peaks occur, and where to measure final delivery."
canonical_url: "https://gawergy.com/learn/sample-peak-vs-true-peak"
md_url: "https://gawergy.com/learn/sample-peak-vs-true-peak.md"
last_updated: "2026-08-27"
date_published: "2026-08-27"
---

# Sample Peak vs True Peak: What Producers Need to Know

A sample-peak meter checks the values stored at discrete sample instants. A true-peak meter estimates the maximum of the reconstructed waveform between those instants. That is why a signal can remain below full scale by sample value and still produce a higher dBTP reading after reconstruction or later processing.

## Key takeaways

- Sample peak measures stored samples; true peak estimates the continuous-time waveform those samples represent.
- Between-sample maxima can exceed the largest sample and may overload later conversion or processing stages.
- dBFS is used for digital sample level, while dBTP denotes a true-peak estimate relative to full scale.
- Measure the final output after clipping, gain, limiting, sample-rate conversion, and other delivery processing.

## What sample peak measures

Digital PCM audio stores amplitude values at regularly spaced times. A sample-peak meter finds the largest absolute stored value, often displaying it in decibels relative to digital full scale (dBFS). It is direct, efficient, and useful for detecting whether sample values reach a fixed numerical boundary.

The meter does not look between those stored instants. AES-R7 explains that peak-sample meters are common but can under-read the actual maximum of the reconstructed signal. The limitation is not that samples are inaccurate; it is that the highest point of the continuous waveform often occurs between sampling instants.

## What true peak estimates

Playback reconstruction converts the sample sequence into a band-limited continuous waveform. That curve does not connect sample points with straight lines. It can rise above adjacent sample values. These maxima are commonly called inter-sample peaks, though true peak is the broader standardized measurement term.

ITU-R BS.1770 defines a true-peak estimation method using an oversampled meter and specified filtering. The standard reports the result in dBTP. AES educational guidance summarizes the practical reason: estimating between-sample peaks helps avoid unexpected overload in later stages such as conversion or sample-rate processing.

> **True peak is an estimate with a defined method**
>
> A meter oversamples and filters the PCM signal to approximate the reconstructed maximum. Different non-compliant displays or quality settings should not be assumed equivalent to a BS.1770-conforming measurement.

## dBFS and dBTP are related but not interchangeable

*Peak-level terms*

| Term | Reference | Typical use |
| --- | --- | --- |
| dBFS | Digital full scale for sample values. | Sample level, headroom, and fixed digital boundaries. |
| Sample peak | Largest stored sample magnitude. | Fast digital peak metering. |
| True peak | Estimated reconstructed waveform maximum. | Delivery headroom and downstream overload risk. |
| dBTP | Decibels relative to the true-peak full-scale reference. | Reporting a standardized true-peak estimate. |

A peak shown below 0 dBFS does not guarantee the true peak is below the same numerical level in dBTP. Conversely, a true-peak reading does not describe perceived loudness. Loudness and peak level are separate measurements defined for different questions.

## Why sample and true-peak readings differ

The difference depends on waveform phase, frequency content, sample rate, filtering, and processing. High-frequency components near the sampling limit can produce reconstructed shapes whose maxima fall between samples. Limiting or clipping can create new high-frequency content and change the relationship further.

Sample-rate conversion uses filters and new sample instants, so peaks can change even when the audible program remains nominally the same. Lossy coding can also change waveform peaks. Output gain or EQ after a ceiling stage plainly creates a new level condition. This is why measuring only an intermediate plugin is not enough for final delivery.

## Why a clipper ceiling may not equal final true peak

A clipper ceiling usually defines its nonlinear transfer behavior in the sample domain. If a plugin includes dry/wet mixing, the dry path can return peaks above the wet ceiling. Downsampling filters, output gain, and later processors can alter sample and reconstructed maxima. Only documentation for a specific product can establish a true-peak guarantee.

Oversampling inside a clipper can reduce aliasing and may improve the representation of its nonlinear curve, but internal oversampling is not automatically a final true-peak limiter. JUCE and FabFilter documentation both describe oversampling as a process with filtering, CPU, latency, and implementation tradeoffs.

## Measure at the final delivery output

1. Complete the intended mix and mastering chain, including output gain and any final limiter.
2. Render at the required sample rate, bit depth, and format using the actual delivery path.
3. Measure the rendered output with a true-peak meter appropriate to the specification.
4. Check the entire program, not only a loud loop, because the maximum can occur anywhere.
5. If a client, broadcaster, or platform provides a requirement, follow that current source rather than a generic internet target.
6. Keep an archival master and create delivery variants when different destinations genuinely require them.

This is measurement guidance, not an exact final-ceiling workflow. A specific delivery can require additional standards, channel formats, codecs, or tolerances beyond the concepts described here.

## Common peak-measurement mistakes

- Assuming a sample ceiling guarantees the same true-peak value.
- Reading dBTP as a loudness measurement.
- Measuring before output gain, sample-rate conversion, or the final limiter.
- Treating oversampling in any nonlinear plugin as a delivery guarantee.
- Copying a platform target without checking its current official documentation and playback context.
- Ignoring sound quality because a meter passes a numerical threshold.

Meter placement is part of the measurement. A true-peak meter before an output trim describes the signal before that trim; a meter on a monitoring bus may include calibration or room processing that is not printed. Label the measurement point and confirm that it represents the file or stream you intend to deliver.

Stereo and multichannel programs also require the channel handling defined by the relevant standard or destination. Do not infer a complete delivery pass from one channel's peak or from a meter whose channel configuration is unknown.

Finally, preserve the distinction between compliance and aesthetics. A file can meet a true-peak limit and still sound distorted, weak, or overly dense. It can also sound excellent while requiring a different delivery version for a particular specification. Measurement verifies a boundary; listening evaluates the production.

## About G-Clipper Pro

G-Clipper Pro's Ceiling controls its clipping behavior and its views help show peak reshaping. As with a normal clipper, use a final standards-appropriate true-peak meter and delivery processor when a dBTP requirement applies.

## Sources & References

- [ITU-R BS.1770-5: Loudness and true-peak measurement algorithms](https://www.itu.int/rec/R-REC-BS.1770-5-202311-I/en)
- [EBU Tech 3341: Loudness metering and maximum true-peak level](https://tech.ebu.ch/docs/tech/tech3341.pdf)
- [AES-R7-2018: Considerations for accurate peak metering of digital audio signals](https://aes.org/standards/standards-development/project-status/)
- [Audio Engineering Society: True peaks and inter-sample peaks](https://aes.org/resources/audio-topics/loudness-project/learn-more/)
- [JUCE DSP documentation: Oversampling](https://docs.juce.com/develop/classjuce_1_1dsp_1_1Oversampling.html)

## Continue Reading

- [Audio Clipping Glossary](https://gawergy.com/resources/audio-clipping-glossary)
- [Soft Clipper on the Master](https://gawergy.com/learn/soft-clipper-on-master)
- [Why Clipping Can Sound Louder](https://gawergy.com/learn/why-clipping-makes-audio-louder)

## Sitemap

See the full [Gawergy.com sitemap](https://gawergy.com/sitemap.md).