How trichrome
scanning works

What a colour negative actually is

A colour negative carries three dye layers — cyan, magenta and yellow — sitting on an orange-tinted base. The dyes record how much red, green and blue light hit the film. The orange is a mask: a deliberate correction built into the film so that a darkroom printer could compensate for the dyes' own impurities.

To read that negative correctly you need to know how much of each dye is present. That is a density measurement, and it needs two things: a known light, and a way of measuring one dye without also measuring its neighbours.

Why white light makes it harder

Photograph a negative on a white light table and your sensor measures it through its own red, green and blue filters. Those filters are wide and they overlap. A camera's red channel sees some of what the magenta dye is doing; the green channel sees some of the cyan. The measurement of each layer is contaminated by the other two.

On top of that, the orange mask sits under everything, so a large and uneven amount of the signal is the base rather than the picture. Software then has to separate all of this apart after the fact — which is why single-capture conversions tend to need a nudge on every frame, and why two frames from the same roll can land in different places.

Trichroma Lab does take a white-light capture anyway, because most people have a lightbox and almost nobody has an RGB lamp. The un-mixing is twelve numbers — three rows of four — fitted at one station on frames photographed both ways, so it is a measurement of that camera and that lamp rather than a guess about yours. What the single shot still cannot do is measure the film base the pipeline grades against, since it only ever sees that base through the camera's filters; so you name the emulsion, and the roll's own base tunes that entry to your development batch.

It costs something, and the number is worth knowing: against three bands on a full 36-frame roll, about five counts a channel. That is the price of the convenience, not a different pipeline.

What three narrow bands change

Light the negative with a single narrow band instead — say 665 nm, deep red — and expose one frame. Now almost the only thing modulating that light is the cyan dye. Do the same at 525 nm for magenta and 455 nm for yellow, and you have three clean, nearly independent measurements of the three layers.

The channels no longer contaminate each other, so nothing has to be un-mixed afterwards. This is what a minilab scanner does, and it is why a minilab's colour is repeatable in a way a one-shot capture struggles to be.

A lamp like that is not something you can buy off a shelf, which is the honest obstacle to all of this. The one every frame on this site was shot under is Scanlight, an open-source narrowband light source — hardware and firmware both published, so you can build one or have one built. Trichroma Lab has no connection to that project beyond being a grateful user of it.

The denominator: film base and a measured light

Density is a ratio. D = −log10(frame / light). Without knowing what the bare light measures, the numbers have no absolute scale at all — so three shots of the empty gate, one under each band, are as much a part of the scan as the frames.

The other half of the denominator is the film base itself: the clear rebate around the picture. That is unexposed film, so whatever it measures is the base of that roll on that day, developed in that tank. Read it off the rebate and the orange mask stops being something to guess at.

Registration

Three separate exposures mean three separate captures, and nothing on a copy stand is perfectly still. Sub-pixel alignment before recombination is what keeps colour fringing off the edges — the shift is measured in fractions of a pixel and resampled, not rounded to the nearest whole one. Trichroma Lab reports the median and worst drift for every roll, so you can see when the stand moved.

And then the colour

All of the above gets you clean densities. Turning densities into a photograph — the curves, the cross-channel matrix, the tone, the look — is a pipeline, and which pipeline you pick is what makes a scan look like a lab scan or like something else.

Trichroma Lab is an independent implementation of a pipeline of the kind a Fuji Frontier runs — pinned stage by stage by generated parity vectors on every build, and set beside real Frontier scans rather than merely described.