Soft Weather

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

Part 0: two frames, two photographers, one register

Read from the design archive's look collection on 2026-08-25. Both frames were published by Colossal within days of each other and neither photographer has anything to do with the other:

• The Winston Taylor Photo Portrait Prize shortlist. A figure draped backwards over a round hay bale in a English field, arms hanging, a small white dog running into frame at the right edge.
• Jennifer Latour's botanical assemblages. Cut flowers wired onto a bare branch, standing in dune grass above a Irish sea inlet with a blue boat at anchor.

Different countries, different subjects, different intentions. What they share is exact enough to design against:

Two independent sources arriving at the same rendering is what separates this from a nice photograph. It is the house style of contemporary prize and gallery photography, and it was the largest coherent register in a 261-image photography and film window.

Part 1: the requirements, ranked by what ruins a frame

1. The sky. Under heavy cloud the brightest thing in frame is a featureless white two or three stops above everything else, and it is usually a third of the picture. Clipped, it stops being weather and becomes a hole.
2. Modelling. Flat light removes shape. A face with no cheekbone and a landscape with no form is the failure this register is always one step away from.
3. Muted, with survivors. The palette is low chroma and the picture depends on one or two saturated things standing inside it. Every operation that mutes a frame also mutes those.
4. Honest skin. In prize photography a cast reads as a filter, and a filter reads as a lesser photograph.
5. Soft without being soft-focus. Medium-format film resolves enormous detail and never reads crisp. Those are different properties and the recipe has to separate them.

Requirements 1 and 2 pull against each other. Protecting the sky means softening the top end, and softening a base under already-flat light is how a picture becomes dull.

Part 2: Fujifilm answered the base question before it was asked

The Pro Neg pair is the only place in the lineup where Fujifilm publishes a rule rather than a preference. Std is "optimized for use in situations with good lighting and moderate contrast, such as studio photography". Hi "boosts contrast, creating shadows even in flat light".

This bank has already recorded the consequence: the choice between them is a lighting decision rather than a taste one, which is why North Light sits on Std for controlled interiors and names Hi as its exterior variant.

Heavy overcast is the definition of flat light, so requirement 2 is answered by the manufacturer.

The distinction is easy to misread and worth stating carefully. Pro Neg. Hi is not adding global contrast, which is a tone-curve operation and one this recipe performs in the opposite direction. It is putting modelling back: separating a cheekbone from a cheek, a fold from a sleeve, a hedgerow from the field behind it, when the sky has become a single enormous softbox and taken all of that away. Global range and local shape are different problems and this recipe uses different tools on each.

The rejections

Sharing a base with Fashion Editorial

Fashion Editorial is also on Pro Neg. Hi and the two do not collide anywhere. It runs … with Shadow … for deep studio blacks. This runs … with Shadow … for an open low end. It sits at Color … for polish. This sits at Color … for restraint and pairs it with Color Chrome Effect …, which that recipe leaves at Weak.

Worth keeping both visible, because a reader comparing them learns more about what a simulation is than either recipe teaches alone.

Part 3: the pair that carries the colour

Requirement 3 is a contradiction stated as a requirement, and it needs two controls pointing opposite ways rather than one compromise.

• Color … takes the field down. Straw, sage and weathered timber are already low chroma. A notch of restraint stops the greens creeping toward postcard when the sun does briefly threaten.
• Color Chrome Effect … protects the survivors. Fujifilm's account of the control is that it detects gradation and creates tonality in colours that are already saturated, rather than raising saturation. In a frame that is overwhelmingly muted it has almost nothing to act on, so its whole effect lands on the red shirt, the hydrangea and the craspedia.

The pairing is the recipe. Color … with the Chrome effect … produces a dull picture with flat accents. … with it … produces a normal cheerful landscape. Only the pair produces a quiet field with something alive in it.

This is the same mechanism Austurbane uses for a refracted spectrum in a beige room, and it generalises: when a look is muted with exceptions, mute with Color and protect with Color Chrome Effect.

Color Chrome FX Blue sits at …. A white sky and a milky sea are the two largest blue-adjacent areas in this register and neither is deep. Off would leave a pale sky flat. Strong on X-Trans V, which already renders blue more deeply than X-Trans IV, would give the weather a drama it does not have.

Part 4: the curve, and the one setting that is not optional

Highlight … is the milky-highlight signature, and Pro Neg. Hi needs a real value there because it is a contrast-adding base rather than a soft one.

… rather than … is the decision that protects the base choice. Fujifilm's dynamic-range modes underexpose and then lift everything below the highlights, so two stops of headroom would open the low end and remove the modelling the simulation was picked to supply. One stop covers the sky. Choosing … here would undo a base decision with a range setting, which is the kind of quiet contradiction that makes a recipe feel wrong without anyone being able to say why.

Shadow … keeps the low end open, because this register has no true blacks in it.

Exposure compensation is the setting most likely to be skipped and the one that decides whether any of the rest works. A reflective meter reads a bright grey scene and pulls it toward middle grey, which produces the muddy version of this look that people make by accident and then blame on the weather. The working default is +2/3.

Part 5: honest limits

• No frames. Creative-design tier and nothing more.
• Two images is a small sample. Austurbane had a written entry with an era, a palette and a 24-image corpus behind it. This has two published frames and an inference. If a third and fourth frame from the same register contradict the reading, the reading moves.
• No era, no name, no attribution. The collection handed over pictures and provenance and deliberately no analysis. Everything characterising this register was read off the frames, which is a weaker starting position than a catalogued aesthetic and is why this section is longer than Austurbane's.
• The base decision rests on a phrase rather than a measurement. "Creating shadows even in flat light" is Fujifilm's description of Pro Neg. Hi, not a published curve. No characteristic curve exists for any simulation, so the whole argument in Part 2 is qualitative.
• The recipe is on the wrong side of its own rule the moment the sun appears. It is chosen by the weather, so it is the only look here with a stated expiry condition rather than a subject range.
• Medium format is not reproducible. The rendering being referenced comes partly from a larger negative and a longer lens at the same field of view, and no setting supplies either.

Last updated: 2026-08-25