What Is Reciprocity Failure?

Film obeys a simple bargain called the reciprocity law: exposure is intensity × time. Halve the light, double the time, and the negative comes out identical. Open up a stop, halve the shutter — same density. Every exposure calculation you've ever made rests on this trade working.

Between about 1/10,000s and 1 second, it does. Outside that range, film starts to cheat you.

Why film fails in the dark

A silver halide crystal doesn't record light one photon at a time. It needs several photons to arrive close together to build a stable latent-image site — a tiny cluster of silver atoms that development can amplify. In bright light, photons arrive in a flood and every crystal that should fire, fires.

In dim light, photons trickle in. A crystal absorbs one, waits… and the single-atom seed decays before its partner arrives. The energy is simply lost. The film is receiving light and quietly discarding some of it — so the dimmer the light, the less efficient the film becomes. That's low-intensity reciprocity failure: at long exposure times, film effectively loses speed while you're exposing it.

What it does to your pictures

• Underexposure — the obvious one. Your meter says 10 seconds; the film needed 20.
• Contrast build-up — the sneaky one. Shadows are dimmer than highlights, so they fail harder. Long exposures don't just come out thin, they come out contrastier, with shadows falling off a cliff.
• Color crossover — the color-film special. The three emulsion layers fail at different rates, so long exposures drift green or magenta in ways no white-balance slider fully undoes. Some photographers filter for it; most of us call it atmosphere.

The correction is film-specific

There is no universal fix — every emulsion fails at its own rate, which is why the correction lives in datasheets, not folklore:

• Ilford/Harman films publish a power curve: corrected = metered^p. HP5 Plus runs p = 1.31 — a metered 10s becomes ~20s. Kentmere Pan 200 runs p = 1.26.
• Kodak's T-Grain films (T-Max, and now Ektapan) are gentle: p ≈ 1.10–1.15 barely moves until you're deep into minutes.
• Fuji Acros is the legend: almost no correction out to two minutes. Astrophotographers love it for a reason.
• Slide film punishes you: Velvia 50's table turns a metered 16 seconds into roughly a minute — and its window of trustworthy behavior is narrow.

In practice

1. Meter the scene as usual.
2. Apply your film's correction — not a rule of thumb from a forum post.
3. On very long exposures, bracket one frame with extra time; film's latitude forgives over-exposure far more than under.
4. Tripod, cable release, and patience.

Fixing it in Analogly

This is exactly what the Reciprocity tab is for:

1. Pick your film stock. Search or filter by manufacturer — stocks that carry published reciprocity data are marked, and the moment you select one the screen drops you straight to the time input.
2. Enter your metered time. Tap a quick preset (1 second to 4 minutes) or drag the fine slider out to 10 minutes.
3. Read the corrected exposure. The adjusted time and the compensation in stops sit side by side, with a bar that shows metered vs. corrected at a glance — open your shutter for the longer number.
4. Respect the honesty warning. Push past the film's tested range and Analogly tells you the result is an extrapolated estimate instead of pretending it's gospel.

Every correction uses the film's own published math — power curves for the Ilford, Harman, and Kentmere families, lookup tables for Kodak and Fuji — never a one-size-fits-all rule of thumb.

The dark is where film gets weird. Go shoot it anyway.