Keep level, dynamics and speaker volume separate
For a simple conceptual example, turning an entire recording down by the same amount moves quiet and loud passages together. A process that selectively reduces louder passages changes their relationship. These are different tasks even if both can reduce the loudest moment heard in a room.
EBU’s R128 publication specifies an average programme target of -23 LUFS and uses Loudness Range and Maximum True Peak Level descriptors. Those measures answer different questions. A single peak reading cannot stand in for the programme’s average loudness or its changing dynamics.
Use the appropriate context for a target
R128 is a named broadcast recommendation, not evidence that every streaming service, Canadian programme or local player uses one universal target. EBU’s streaming supplement explicitly discusses services using higher targets and different reproduction environments.
Record the actual delivery specification when preparing material. Do not normalise a local file to a value copied from an unrelated platform and claim it is now correct for all distribution. Keep production measurements, any supplied loudness metadata and the receiving player’s behaviour separately identifiable.
Distinguish a measured adjustment from a listening preference
EBU’s production guidance discusses normalising the signal versus using metadata, loudness processors and loudness range. The workflow selected matters: an instruction carried as metadata and a permanently processed audio waveform are not the same resulting object.
For a home viewing problem, note whether the goal is matching levels between programmes or making quiet dialogue easier to hear without very loud effects. A device’s named dynamic-range or night-listening feature should be checked in its own manual. Do not treat a preference for that feature as proof that the source violated R128.
Make a useful comparison record
Keep the same source passage, output route and playback setting while comparing one documented control. Record what changed: overall level, intelligibility of quiet speech or difference between quiet and loud passages. A receiver volume number alone is not a calibrated sound-pressure measurement.
No loudness meter, programme or television was tested for this article. It explains the standards distinction and a listening task without prescribing a universal mastering value, certifying compliance or claiming normalisation recreates missing audio information.
Sources and scope
Scope notes distinguish published documentation, supplied information and editorial methods. They do not report a hands-on service test.
- EBU: R128 recommendation — Primary publication summary: -23LUFS programme target and separate loudness-range/true-peak descriptors.
- EBU: production guidelines2023 — Primary technical guide: signal versus metadata, loudness processing and range concepts; no programme measurement performed.
- EBU: loudness in streaming2023 — Primary supplement: differing platform targets and reproduction contexts, not a universal streaming-master instruction.
