A darkened room, one lit screen, a person seated well back

01 — Contrast

Contrast beats resolution

At normal seating distance most people cannot resolve the extra pixels, but everyone sees a black that is actually black. Dynamic range is the difference you notice; pixel count is the number on the box.

Fig. 01 · The only light in the room is the one being measured.

Photo: Tima Miroshnichenko / Pexels

Measured transfer What a lit room leaves Target

The shape of the problem

A display does not reproduce an image. It attempts a curve, and the room subtracts from the bottom of it.

Ambient light does not dim a picture evenly. It lifts the floor. Everything meant to be black arrives as the same dark grey, the steps between the lowest tones collapse into each other, and the curve the grade was made against no longer exists in the room you are sitting in.

Light first: It is free to remove and it moves the floor further than any setting does.
Then geometry: Angles and distances, measured from the seat rather than judged from behind the equipment.
Correction last: It trims what exists. It cannot restore what the room has already taken.

A hand uses a stylus to edit photo contrast settings on a tablet beside camera lenses

04 — Calibration

What calibration corrects

Not preference — the gap between what the signal specifies and what the panel emits.

Fig. 02 · Correction is arithmetic applied to a signal, not an improvement to a panel.

Photo: Jakub Zerdzicki / Pexels

Recording studio mixing console flanked by studio monitors and dual computer screens

07 — Treatment

First reflections

The earliest reflected arrivals smear the image. Find them with a mirror and a second person.

Fig. 03 · The earliest reflected arrivals are the ones that smear the image.

Photo: RDNE Stock project / Pexels

Field notes

Short entries · no equipment named

01 · CONTRAST

The room lifts the black

Any light in the room raises the floor and throws the ratio away.

Spend on blackout and dark surfaces.

03 · LIGHT IN THE ROOM

Wall colour

Every surface facing the screen bounces some of it back.

Spend a tin of dark matte paint on the screen wall.

05 · PLACEMENT

Height and the ear

Tweeter at ear height, seated, is the whole rule.

Spend on stands before electronics.

06 · BASS

Nulls and peaks

Move the seat a metre and the bass changes completely. Nothing is broken.

Spend on moving things, not on adding level.

07 · TREATMENT

What furniture already does

A carpet, curtains and a full bookshelf are treatment.

Spend nothing first: count what the room already has.

08 · THE CHAIN

Bit depth and banding

Visible steps in a gradient are a depth problem, not a panel fault.

Spend on the source and the link, not the panel.

In almost every finished room, the weakest component in the chain is the room.

Who keeps the lights off

Rooms and rigs we calibrate our recommendations against.

Eight sections · every entry reachable without a dropdown

The whole register →

01ContrastThe thing everybody can see.4 entries 02PanelsWhat each technology can and cannot do.4 entries 03Light in the RoomThe uncontrolled variable.3 entries 04CalibrationWhat is actually being corrected.4 entries 05PlacementGeometry, not preference.4 entries 06BassA standing wave, not a sound.3 entries 07TreatmentFirst reflections and what to do about them.3 entries 08The ChainSignal, scaling and where quality is lost.3 entries

Hands adjusting a speaker angle on a stand

05 — Placement

Geometry, not preference

Angle, distance and height are three numbers that decide whether a soundstage exists. They are set with a tape measure, from the seat, and they are almost never where the furniture wants them.

Fig. 04 · Toe-in is set from the seat with a tape, not judged from behind the cabinet.

Photo: Jens Mahnke / Pexels