File facts
Measurements computable from the decoded file: for example peaks, loudness, clipping indicators, stereo correlation, bandwidth or codec-grid traces.
/// White paper · English edition
A second read,
before import.
This white paper explains what DigWash can honestly compute from one distributed audio file — and names, just as plainly, what depends on the master, the room, the playback level and the DJ's own taste instead.
§ Read the labels first
Three different kinds of statement.
File risks
Calibrated, advisory interpretations of those facts. They are prompts to inspect or listen, not automatic rejection.
Venue or taste outcomes
Judgments that depend on the room, the system and the crowd on the night. The file doesn't contain any of those, so DigWash leaves this call to you.
§ Three diagnostic layers
What a file is, does, and becomes.
Each layer carries its own evidence grade, from standards-backed measurement to hypothesis-grounded design — so a given warning always comes with a sense of how much weight to put on it.
- 01 Provenance and integrityEvidence chapter · strongestIs the file what it claims to be? Fake-lossless files, transcodes and codec traces that survive a lossless rewrap — the strongest, most directly implementable evidence in the corpus.Read the chapter ↗
- 02 Rig and dynamics safetyEvidence chapters · standards-backedWhat will this file do to a system? Standards-backed true peak, loudness, clipping, headroom and dynamics facts that protect a system from clipping and over-limiting, without predicting a venue outcome.Read the chapter ↗
- 03 Spatial translation and spectral fatigueMethod-grounded · partly ongoing researchWhat will this file become once it's summed, played loud, or heard off-axis? Mono-sum loss, stereo/phase translation and psychoacoustic outliers, framed as listen-check prompts, plus the explanation layer that ranks which risk deserves attention first — never a single quality score.Read the chapter ↗
§ Continue