Lifted blacks, and why flat looks like film.
Put a developed but completely unexposed strip of colour negative on a light table and it is not clear. It has a visible orange-brown cast and it blocks a measurable amount of light. That floor is called base-plus-fog, and it is the reason a photograph shot on film cannot contain a true black — there is no path through the material that transmits nothing.
Almost everything people mean by "filmic tone" follows from that one fact.
Density, not brightness
Film response is described by a characteristic curve: log exposure on the horizontal axis, density on the vertical. It is not a straight line. At the bottom there is the toe, where the curve is nearly flat because the emulsion needs a minimum quantity of light before any grains develop at all. In the middle there is a long, roughly straight section where density rises predictably with exposure. At the top there is the shoulder, where the crystals begin to run out and additional light produces less and less additional density.
A digital sensor has no toe and no shoulder. It is linear until it is full, and then it stops — abruptly, with no gradation. That single difference produces most of what people identify as the digital look: shadows that plunge to nothing, and highlights that clip flat instead of rolling off. The sensor is more accurate. It is also less forgiving, because forgiveness is precisely what a toe and a shoulder provide.
The one line that does most of the work
In the Halation pipeline the fade control is a single affine transform applied across the whole image: the signal is compressed slightly and then offset upward. At full strength it multiplies the image by 0.74 and adds 0.115. Black stops at roughly 11.5% instead of 0, and white pulls down to about 85.5%.
That is deliberately not a curve. It is tempting to model the toe with something elaborate, but the perceptual signal that reads as film is mostly the raised floor, not the shape of the approach to it. Once the floor is up, the eye stops finding the pure-black anchor it uses to judge contrast, and the image starts to read as a print rather than as a display.
Compressing the top at the same time matters more than it looks. Lifting the blacks alone produces a washed-out image, because the range still runs all the way to display white and the contrast now looks unbalanced rather than intentional. Pulling both ends in keeps the ratio plausible. It is the difference between a photograph that looks faded and one that looks flat — and flat is a compliment.
The threshold between filmic and muddy
Fade is the parameter most likely to be overdone, so it is worth being concrete about how far the presets actually push it, on a scale of 0 to 100:
- Golden and Silver — 6. Barely there. Enough to take the hard edge off the shadows without anyone naming it.
- Nocturne — 7. Night scenes need their blacks to stay convincing; more than this and the darkness starts to look like haze.
- Cinematic — 9. The classic small lift that makes teal shadows read as a grade rather than as a colour cast.
- Bloom — 13. A portrait stock reference, where softness is the point.
- Komorebi — 16. The highest in the set, and paired with negative contrast and reduced saturation. It is a high-key look built around the absence of a deep black.
Note that the ceiling across six quite different looks is 16 out of 100. Fade is almost never the loud parameter. Values above roughly 25 stop reading as a stock characteristic and start reading as an image that has been left in the sun.
Why order of operations matters
Fade is applied after the tone controls and before split toning. That sequence is not arbitrary. Toning is applied through a soft-light blend that is driven by luminance masks, and lifting the blacks first means the shadow tint lands on shadows that actually have some signal in them. Toning first and lifting afterwards would flatten the tint along with everything else, and the shadow colour would come out weak and desaturated.
This is the general rule the pipeline follows: anything physical happens where it is physically meaningful, and anything aesthetic happens where it behaves the way a photographer expects. Exposure and white balance run in linear light, because doubling the light is a real operation on real photons. Tone shaping, fade and toning run in a gamma-encoded working space, because that is the space in which sliders feel like the ones in Lightroom.
A test worth running
Take any image with a genuinely black region — a shadow under a car, a doorway, a dark jacket. Set every other control to zero and move fade alone from 0 to 20 and back. Watch the black patch rather than the subject.
Somewhere in that range there is a point where the image stops looking like a photograph of a dark thing and starts looking like a photograph that was printed. That point is different for every image, and finding it by eye once teaches more than any preset can.