SL-103 / Colour as Effect

The Two-World Grade

Intercut locations the audience can tell apart by grade alone. No captions.

The brief

Two intercut locations distinguishable by grade alone, so the audience always knows where they are without a single caption.

Why it's hard

The model averages both halves toward one pleasant look, so the cut reads as a lighting change rather than a different world.

The craft key

Never harmonise the two looks — the separation is the information. Repeat a prop and the lens so grade is the only variable.

Skill Tests

03 / 28

Animal anatomy Human anatomy Biomechanics Physics & mass Water simulation Crowd dynamics Multi-shot continuity Face consistency Camera movement Unbroken take Motivated lighting Cutting rhythm Hands & fingers Hair & fabric Reflection & refraction Scale & depth Impossible geometry Period accuracy Music sync Aging & time Natural asynchrony Eye-lines & gaze Particle behaviour Motivated light source Consequence & persistence Time manipulation Graphic-layer discipline Colour as storytelling

Every challenge is scored against the same vocabulary, so you can compare them. Lit cells are what this one tests.

Starter prompt

Built with the CEMI MediaMax system for Google Veo 3.1. Prompts stay in English — that is the language these models take instructions in. Treat it as a starting point and start varying it.

the-two-world-grade

One clip, one hard cut at the midpoint, two intercut locations coded by grade alone in the style of Steven Soderbergh's storyline-graded work. WORLD A: a coastal net loft — bleach-bypass amber, blown window highlights, crushed silver blacks, low saturation with warm skin; handheld 28mm at f/4, operator swaying, dirty foreground of hanging monofilament and a salt-fogged pane. WORLD B: an inland records basement — cold green-cyan, dense shadow, fluorescent flicker, no warm tone anywhere; same 28mm, same operator sway, dirty foreground of a dusty file spine rail across the lower frame. The SAME brass key sits on a bench in both halves so the audience reads location, not lighting accident. No captions, no titles. Materials: tarred twine, oxidised net needles, damp cardboard boxes, steel shelving with paint chipped to primer.

Derived from the CEMI MediaMax production canon — cemi.media

Style reference: Steven Soderbergh's storyline colour-coding

Why it's built this way

A colour code works only when the difference is categorical and consistent — two locations, two fixed grades, no blending between them. Repeating the same lens, operator behaviour and prop across the cut isolates grade as the single variable the audience is reading. Never harmonise the two looks; the separation is the information.

Watch for

The model averaging both halves toward one pleasant teal-and-orange look, so the cut reads as a lighting change rather than a different world.

Ran a challenge? Show us what you got.

Share your attempt — the triumphs and the glorious failures. The most instructive entries get featured on the board.