Divided Light

An original design for the Neo-Impressionist daylight look: the same lifted, no-black tone scale as Open Air with the shadows deliberately cooled, violet carrying nearly half the palette, and a white point close to paper. Derivation in the notes.

Painterly reference: Paul Signac and the Divisionists, measured from ten public-domain reproductions against a twelve-painter control group. Named for the principle rather than the painter: Divisionism is the practice of dividing a colour into its components and placing them separately, and the measurable consequence is a picture whose lights and darks sit at opposite ends of the temperature scale.

Provenance. Built 2026-10-05 from ten Signac oils, public domain, cropped to the canvas and verified by eye. Measured against the same twelve-painter control group as the three earlier painting recipes.

What the corpus measures

Two findings carry the recipe, and the first is the one that makes it a separate recipe rather than a variant.

The shadows are actively cool. Signac's shadow b* measures −8.55. Every other painter in the twelve-painter group sits at or above zero: Monet +3.21, Rembrandt +1.12, Caravaggio +0.49, Sargent +4.32. Signac is the only one whose darks are blue, rather than merely less yellow than his lights. Against highlights at +3.13, that is a 11.7-unit swing across the tone scale in the opposite direction from the Impressionist case.

Violet is nearly half the palette. At 42.8% of coloured pixels it is the largest single hue share of any painter measured, ahead of Monet's 26.9% and far ahead of Rembrandt and Caravaggio at 0.0%. Yellow is second at 22.0%. A picture built from violet and yellow in those proportions is the complementary-contrast theory the Divisionists wrote down, arriving as a measurement.

Why Astia, again, and why the recipe is still separate

This is the second recipe in the bank on Astia from the painting seam, and the base argument is the same as Open Air's: soft at both ends for a black of 30.0 under a white of 88.3, saturation maintained at a moderate 14.0, and no per-hue agenda to rearrange a palette that is 42.8% violet.

What separates this from Open Air is three notches of white balance and the reason for them. The two recipes share a base, a tone curve and a dynamic-range setting, because both corpora are high-key with no black in them. They differ where the measurement differs: Monet's shadows at b* +3.21 take a warm notch, Signac's at −8.55 take two cool ones. One painter lets the shadow stay the colour of the light, the other paints it the complement. That is the whole of Impressionism against Neo-Impressionism, and it survives as a three-notch swing on one axis.

If you want a single recipe for both, use Open Air. This one is for when the shadow is genuinely blue, which in practice means open shade under a clear sky.

The rest

• White Balance 0 Red and +2 Blue, which is the only recipe in this folder shifting toward Blue at all. It is aimed at a shadow b* of −8.55 and it works because the highlights carry so little warmth to lose, at b* +3.13 against Monet's +5.45. Red stays at 0 because shadow a* is +1.77 and highlight a* +0.84, which is near-neutral at both ends.
• Color Chrome FX Blue Off, and at 42.8% violet this is the single most load-bearing setting in the table. FX Blue deepens blues toward navy. Nearly half this palette is violet, and it is the light part of the picture rather than the dark part. Deepening it inverts the look.
• Color +1 rather than Open Air's 0. Measured chroma is 14.0 against Monet's 11.9, and Astia is gentle enough that one notch lands rather than overshoots.
• DR400 with Shadow −2, the same pairing as Open Air and for the same reason: a black point of 30.0 is reached by lifting, and the highlight headroom is a bonus. ISO floor 500, which is this sensor's conversion-gain step.
• Exposure +2/3 to +1, higher than Open Air. The median measures 67.4 against Monet's 59.7 and the white point 88.3 against 77.8, so this is the brighter of the two looks and the meter will fight it harder.
• Clarity 0. Pointillism is paint applied in separate touches, which is a surface property, and a local-contrast control does not reproduce it. The same admission Open Air makes about brushwork applies here with more force, because the technique is the whole point of the movement. The recipe delivers the colour structure and nothing of the touch.

Variants

Movie mode

Ports well. Movie mode drops Grain, Color Chrome Effect and Clarity, and Clarity is already 0 while Color Chrome Effect is only Weak. Astia, the white balance, DR400 and the tone curve all carry, which is where the look lives. See Video / Movie Mode, settings, differences, and photo→video translation.

Validation tier: Creative design, field-test pending. Grounded in Fujifilm's published design statements for Astia and each base rejected above, in this bank's Dynamic Range & Tone Curve, the contrast controls tone-curve table, and in a CIELAB measurement of ten verified reproductions against a twelve-painter control group, with two estimators.

No frames have been shot with it.

Test priority: in order

1. A sunlit subject with open shade beside it, metered at +1. The whole recipe in one frame. Measure the shadow's b*: the target is about −8.5, which is the one number that separates this from Open Air. If it comes back positive, you are shooting Open Air's condition and should use that recipe.
2. Anything violet or blue-violet in the frame. At 42.8% of the reference palette this is the colour the look is made of, and Color Chrome FX Blue Off is what protects it. It should read light and luminous rather than navy.
3. The white point. The target is L* 88, which is brighter than any other painting recipe here. If the frame reads grey, the exposure is too low before anything else is wrong.
4. Shoot the same subject with this and with Open Air back to back. The pair is the deliverable: one warms the shadow, one cools it, and everything else is identical. If you cannot tell them apart, the white-balance shift is not reaching far enough.
5. Skin, if you must. The variants table backs the blue off for a reason, and knowing where the line sits is worth one frame.

What would count as the recipe being wrong rather than needing a tweak: if a +2 Blue shift cools the highlights as much as the shadows, then the mechanism this recipe depends on, that a global shift reads differently where chroma is low, is too weak here, and the cool shadow has to come from shooting open shade under a blue sky rather than from the camera.

Pro tip: One number separates this from Open Air. Monet's shadows measure b* +3.21 and Signac's −8.55. Everything else about the two recipes is the same, because both painters abandoned black. Choose by looking at the shadow: if the sky is filling it and it looks blue, this one. If the shadow looks like the light, Open Air.

Pro tip: Violet is 42.8% of this palette, the largest hue share measured anywhere in this seam. FX Blue Off is not a default here, it is the setting protecting nearly half the frame from being pulled toward navy.

Pro tip: Meter higher than feels right, then higher again. A median of L* 67.4 and a white point of 88.3 make this the brightest painting recipe in the bank, and every automatic system in the camera is built to pull it down.

Pro tip: The technique is not in the recipe and cannot be. Divisionism is separate touches of unmixed colour sitting side by side on canvas. Clarity 0 is the recipe declining to pretend a local-contrast control reproduces that. What transfers is the colour theory, which is the half a camera can actually do.

HEIF-safe. Clarity sits at 0, so nothing is silently dropped. See Grain & Detail. Grain Effect, Sharpness, Clarity, High ISO NR.

Creative design (field-test pending). See the recipe and the Knowledge base.

Part 0: method, and a fourth corpus failure

Fourth recipe from the painting seam, measured with the instrument and control group described in the notes Open Air's Part 0: twelve painters, one instrument, only differences between them treated as findings.

That seam has now produced four distinct ways a painting corpus can be silently wrong. Three were found in September: gilt frames, works on paper, and a name collision. This recipe found the fourth.

Google Art Project stores gigapixel captures as tiles. A first Seurat harvest returned sixteen files, of which thirteen were pieces of a single painting named -x0-y0, -x1-y1 and so on. Measured as a corpus that is one picture weighted thirteen times, with the tiles of its sky and its lawn counted as separate works. The fix is a filename rule, and the general lesson is the same as the other three: the fault was invisible in the numbers and obvious in a contact sheet.

Part 1: the corpus, and why it is Signac

Ten Signac oils, public domain, cropped to the canvas, verified by eye. Nine Seurat oils were gathered the same way after the tile purge and are used for comparison rather than for the recipe.

The two painters were measured separately and then together, and the separate figures are the reason the recipe is built on one of them:

Combining them produces a corpus whose two halves disagree by a factor of three on the headline quantity, and whose spread is wider than either painter alone. That is a blended average rather than a characterisation, and a recipe built on it would describe neither man.

Signac is the corpus because his signature is both stronger and tighter: the largest violet share measured anywhere in this seam, the only genuinely negative shadow b* in twelve painters, and the lowest per-image spread of the two. The recipe carries a variant row for the Seurat reading, which is the same look at lower amplitude.

The images were fetched into a scratch workspace and are not in this collection.

Part 2: the measurement

Finding 1: the only painter with cool shadows

Shadow b* measures −8.55. Every other painter in the group is at or above zero, from Caravaggio's +0.49 to Sargent's +4.32. Signac is the only one whose darks are blue rather than merely less yellow than his lights.

This is a different claim from the one Open Air rests on. That recipe's finding is that Monet drains less colour temperature from his shadows than anyone else, at −2.24 against a group running to −24.34. This one is that Signac puts the opposite temperature there. Monet's shadows are warm and his lights are warmer. Signac's shadows are cold and his lights are warm, a 11.7-unit swing across the tone scale.

That is complementary contrast, which the Divisionists wrote down as theory and which turns out to leave a measurable trace.

Finding 2: violet is nearly half the palette

The largest single hue share recorded in this seam, ahead of Monet's 26.9% violet and Sargent's 64.5% yellow is the only larger single family anywhere in the group. Violet against yellow at roughly two to one is the complementary pair the movement built its method on.

Finding 3: the brightest corpus measured

A median of 67.4 under a white point of 88.3 makes this the brightest painting corpus in the seam, above Monet at 59.7 and 77.8. Combined with a black point of 30.0 it is a picture living almost entirely in the upper half of the tone scale.

Part 3: the base

Four requirements, and the answer is the same simulation as Open Air.

1. Soft at the shadow end, for a black of 30.0.
2. Soft at the highlight end, for a white of 88.3 that must stop short of paper.
3. Saturation maintained at 14.0, which is moderate.
4. No per-hue agenda, with 42.8% of the frame riding on violet staying violet.

Astia meets all four, and requirement four is where every rival falls.

Why Eterna lost, harder than last time

Eterna is soft at both ends with real latitude, and on tone it is arguably the better base for a corpus this high-key. It loses on the same number as in Open Air, with more force: its blues are cyan-shifted by design, and violet here is 42.8% rather than 26.9%. Pushing nearly half the frame toward cyan would not tint the look, it would replace it.

Part 4: the separation from Open Air, stated plainly

The two recipes share a base, a dynamic-range setting, a tone curve, Sharpness and Clarity. They differ by three notches of white balance and one of Color. That is a narrow gap, and this bank has rejected looks for being narrower, so the case has to be made rather than assumed.

It rests on one measured quantity. Monet's shadow b* is +3.21. Signac's is −8.55. Those are opposite signs, not a difference of degree, and they correspond to two documented and genuinely opposed positions about what a shadow is. One is light that has less of the sun in it. The other is light of the complementary colour.

The settings express that honestly: a warm notch against two cool ones, on an otherwise identical curve, because the curve is what the two movements agree about. Both abandoned black, so both recipes lift to 30 and above.

The practical test in the recipe is to shoot both back to back. If the pair is indistinguishable in the field, the shift is not reaching and the two should collapse into one recipe. That is stated as a falsification condition rather than left for a reader to wonder about.

Part 5: honest limits

• No frames have been shot with it.
• Ten paintings. Workable because the per-image spread on the headline figure is the tightest in the seam at 7.5, and not enough for a fine distinction.
• Seurat and Signac disagree substantially, which is recorded in Part 1 rather than averaged away. This is a recipe for Signac's reading of the movement, and the variants table says what to change for Seurat's.
• These are reproductions, so the endpoints belong to a photograph of a painting rather than to pigment.
• Divisionism is separate touches of unmixed colour on canvas, a surface property no camera setting reproduces. Clarity 0 is the recipe declining to pretend otherwise. What transfers is the colour theory.
• A reference tells you what a look is and never whether a recipe reproduces it.

Part 6: where it sits in the bank

Seventh Astia original and the second from the painting seam. The pair with Open Air is the point of it: same tone curve, opposite white balance, built from two painters working the same ground a generation apart.

Across the four painting recipes the shadow-temperature axis now runs end to end: Signac at b* −8.55, Monet at +3.21, Sargent at +4.32, Caravaggio at +0.49 with highlights at +24.83. Raking Light and this recipe are the two extremes of the seam, and they are as far apart as the bank's creative folder gets.

Last updated: 2026-10-05