Broadcast target
job_d091979d9a4b4896EBU R128, the European broadcast standard.
ffmpeg -i https://cdn.rendobar.com/assets/examples/sample.mp4 -af "loudnorm=I=-23:TP=-2:LRA=7" -c:v copy -t 5 out.mp4audio filter
The FFmpeg loudnorm filter
Normalises perceived loudness to a target LUFS, following EBU R128.
3 of 3 variants measured Measured 2026-08-20
The part that catches people
Single-pass loudnorm is a live estimate and will not hit the target exactly. Two-pass is the accurate form: run once with print_format=json, read the measured values back, pass them into a second run. Most people run one pass and wonder why the result is off by a decibel.
Measured, not described
Every row below is a real job. The same 1280x720 five-second H.264 source went through each variant, so the numbers are comparable to each other and to every other filter page on this site. Encode time is the FFmpeg step on its own, separated from the download and the upload.
| Variant | Encode | Output | Cost |
|---|---|---|---|
| Broadcast target | 411 ms | 363.0 KB · 1280x720 | $0.0029 |
| Streaming target | 447 ms | 363.0 KB · 1280x720 | $0.0019 |
| Podcast target | 424 ms | 363.0 KB · 1280x720 | $0.0019 |
The fastest variant here is not the smallest. Broadcast target finished the encode in 411 ms, while Streaming target produced the smallest file at 363.0 KB. Which one matters depends on whether you are paying for compute or for storage.
The exact commands
Streaming target
job_eef8ac206a734c73Where most streaming platforms normalise to.
ffmpeg -i https://cdn.rendobar.com/assets/examples/sample.mp4 -af "loudnorm=I=-14:TP=-1:LRA=11" -c:v copy -t 5 out.mp4Podcast target
job_980262922ca04f6aThe common spoken-word compromise.
ffmpeg -i https://cdn.rendobar.com/assets/examples/sample.mp4 -af "loudnorm=I=-16:TP=-1.5:LRA=9" -c:v copy -t 5 out.mp4How these numbers were produced
Each variant was submitted to the Rendobar API as an ffmpeg job against the same public source file, and the response was recorded verbatim. Encode time is the API's own per-step timing for the FFmpeg step. Output size is the byte count of the file the job produced. Cost is what the job was actually billed.
Nothing here was estimated, and the whole set is re-runnable, so if a figure looks wrong it can be checked rather than argued about. The 72 variants behind this filter reference cost $0.193082 in total to measure.
Frequently asked questions
What does the FFmpeg loudnorm filter do?
Normalises perceived loudness to a target LUFS, following EBU R128.
What is the most common mistake with loudnorm?
Single-pass loudnorm is a live estimate and will not hit the target exactly. Two-pass is the accurate form: run once with print_format=json, read the measured values back, pass them into a second run. Most people run one pass and wonder why the result is off by a decibel.
How long does loudnorm take to run?
On a 1280x720 five-second H.264 source, the fastest variant on this page (Broadcast target) spent 411 ms inside FFmpeg. That figure is the encode step alone, measured separately from the download and the upload, so it is the filter's own cost rather than the round trip.
Do I need FFmpeg installed to use loudnorm?
Not with Rendobar. Every command on this page ran through the Rendobar API against a URL, with no local FFmpeg install and no server. The job id next to each row is the real job that produced the numbers.
Other filters, measured the same way
atempo Changes audio playback speed without changing pitch. volume Scales audio gain by a ratio or in decibels. afade Fades audio in or out over a duration, with selectable curve. aresample Converts audio sample rate, and can compensate for timestamp drift. silenceremove Strips silent passages from audio, by threshold and minimum duration.
Run loudnorm without installing FFmpeg
Every command on this page ran through the API against a URL. No binary, no server.