The active curve
The Active curve in the inspector’s Curve section is the .acv that was
applied in Photoshop when the negative for this print was made. Load it with
Open… (or ⌘L, or by dropping the file on the window); the recents menu
keeps the last five, which matters because a printing session usually
re-uses the same curve all day.
It has one job: the correction the app exports is composed with it, so what you get back is a replacement for that curve — swap it in, don’t stack it on top. With no active curve loaded the export corrects from linear, which is the right answer for a negative printed without one.
Clear removes it and reverts to linear.
Aim: what “correct” should mean
By default the target is a literal reproduction: every source tone should
print at the same value it has in the source image. The Aim slider bends
that target into an S-curve, and the readout tells you the midtone slope it
produces — linear, ×1.20, ×1.35, and so on. ×1.35 means midtones
render 1.35 times steeper than a literal reproduction, with shadow and
highlight separation compressing in exchange.
This is a look, not a fix. Set it once for the rendering you want from the process, check it in the Preview view, and leave it: the exported curve carries it.
Print range: literal or printable
The Print range button toggles between two targets:
- Literal — aim at the nominal tone scale. Source tones darker or lighter than the print can actually make collapse onto a single value in the negative.
- Aim at print range — aim at the tone range the print can really produce. Nothing collapses, but the whole scale compresses slightly, so midtones no longer land on their nominal values.
Leave it off when you are proofing or measuring against a reference; turn it on for a pictorial negative, especially one with large deep-shadow areas — that is exactly the case where the literal target throws away separation.
What gets exported
⌘E, the Curve section’s Export Adjusted Curve…, or Export →
Adjusted Curve… (.acv) writes a Photoshop Version 4 curve. Three
properties are worth stating plainly:
- It replaces the active curve.
- Tones the image never showed are carried over from the active curve unchanged. Where there is no measurement there is no correction; the app does not extrapolate one.
- It carries the Aim and Print-range settings that were in force when you exported.
The filename is built from the scan and your process metadata, e.g.
raw-2026-08-04-0005-Platinum-Palladium-Bergger-COT320-adjusted.acv.
ICC profiles for soft-proofing
Export → ICC Profile… asks for a folder and writes two grayscale profiles:
| File | What it describes |
|---|---|
<name>-as-printed.icc |
What this print does to tone — the measured response, with no correction applied |
<name>-next-print.icc |
What the next print should look like once the exported .acv is in use |
Install them in ~/Library/ColorSync/Profiles, then use View → Proof
Setup → Custom → Device to Simulate in Photoshop. The profile description
carries the process, paper, coverage and endpoints, so the two stay
distinguishable in that list.
Two practical notes the app also reminds you of when it exports:
- Photoshop caches profiles by filename. If you overwrite a profile you have already used, quit and reopen Photoshop or it keeps the stale copy — and Simulate Paper Color stays greyed out.
- Soft-proofing is display-only. It never alters your document, its profile, or the digital negative.
Proof, in the app
The toolbar’s Proof toggle applies the same profile to the source image on the canvas, in the Both view’s source pane, so you can see the tone and the black lift your process imposes without leaving the app.

It becomes available once an analysis has produced a profile. The tonal compression and black lift it shows are measured from your print; the warm media white is the D50 reference, not a measurement of your paper’s tint — that would need per-patch colorimetry.
Preview
The Preview view is the other half of the same question: the current print as scanned, next to a simulation of what the next print will look like once the correction is applied.

If the right-hand pane does not look like an improvement, do not export the curve — check registration, coverage and the detected endpoints first.