Resolve & Colour line · stop 07 of 14 · 28 min · members
How a delivery-first grade is structured when someone else has to open your project and finish it.
Free with an account
Membership is free: an account opens all 86 script pages. The Lab, Studio Canvas and the paid guides need the $99 pass, paid once. Already signed in on this browser? The page opens by itself.
The difference
Which is a different design goal from a grade built to look good.
On a personal project the grade only has to satisfy you, and its structure can be whatever got you there. On a broadcast or agency job, someone else will open the project — to make a change, to produce a different version, to finish it when you are unavailable.
That changes the priorities. Legibility, predictable node positions and separation of concerns matter more than elegance, because the cost of a confusing tree is paid by someone under time pressure who cannot ask you.
The fixed tree is the answer: the same nodes, in the same order, with the same names, on every clip of every project.
The layers
Two levels with a clean division of responsibility.
Timeline level carries anything that applies to the whole programme: the input transform, the output transform, any broadcast-legal limiter, and a version-specific adjustment if one is needed.
Clip level carries the image work: exposure, balance, secondaries, look.
The value of this split is that producing an alternative deliverable — a different gamma, a different colour space, a version for a second platform — is a timeline change. Nobody has to touch eighty clips.
The clip chain
An empty node is free and removes a decision.
Every clip gets the same serial nodes in the same order:
Name them. An inherited project where node three is always secondaries is navigable; one where it varies has to be read node by node.
Additional work attaches as a layer or parallel branch to the node it belongs to, never as an extra serial node on the end. The chain stays four long on every clip in the project.
Legality
Not per clip, and not a substitute for grading within range.
Broadcast specifications define legal limits for luminance and saturation. A single limiter on the timeline, after the output transform, catches excursions.
It is a safety net, not a workflow. Grading beyond legal and relying on the limiter to pull it back produces flattened highlights and a look you did not choose. Grade within range and let the limiter handle the occasional overshoot.
Check on scopes rather than by eye, and check the whole programme rather than the shots you remember being bright. Excursions usually appear on a shot nobody thought about.
Handover
Assume the person opening it has no context and no time.
Before delivering the project file:
Ten minutes of tidying. It is also, in practice, the thing that gets you recommended — colourists who receive a legible project remember who built it.
The real argument
Automation, matching and review all require predictability.
Beyond handover, a consistent tree enables things a variable one cannot. A script can adjust exposure on every clip when the node is always in position one. Matching shots is faster when you know where each correction lives. Reviewing your own work three months later takes minutes rather than an hour of orientation.
The discipline feels pedantic on the first clip and pays on the fortieth, on the handover, and on every project after this one.
It is also the difference between having a method and having a habit of getting there somehow.
1:1 · two hours