A contrast filter passes some wavelengths and absorbs others, so it always reduces the total light reaching the film. Left uncorrected, that loss underexposes the negative. The filter factor is the figure that quantifies the loss and tells you how much exposure to add back.

What the Factor Represents

A filter factor is a multiplier applied to the unfiltered exposure. A factor of 2 means the filtered scene needs twice the exposure to place the same density on the negative; a factor of 8 means eight times. The number accounts only for light lost to absorption, not for the contrast effect the filter produces.

The figure is not arbitrary. As Ansel Adams sets out in The Negative, a factor is measured by finding the extra exposure needed to hold the same negative density on a Zone V subject — an 18% grey card — under daylight of roughly 5500K, the mixture of direct sunlight and skylight that the standard tables assume. Move away from that illuminant and the published number stops being correct. This grey-card-at-5500K derivation is why every factor is fundamentally a property of the emulsion’s spectral response, not of the glass alone: it depends entirely on what the film sees through the filter. For that reason the figure is published per emulsion, and a manufacturer’s value for your actual stock should always be preferred over a generic one.

The cleanest proof of that emulsion dependence is not a filter at all but a film. Shoot Ilford Ortho Plus 80, an orthochromatic emulsion sensitive only to blue and green, and a red subject renders dark while a blue sky renders light with no filter fitted — the inverse of what a red filter does on panchromatic stock. The contrast effect, and the factor that goes with it, live in the film.

A Working Factor Table

The generic Wratten contrast filters under daylight (5500K) carry these factors, with the stop equivalent alongside:

Filter	Daylight factor	Tungsten factor	Stops (daylight)
No. 8 yellow	2	—	1
No. 15 deep yellow	2.5	—	1 1/3
No. 11 yellow-green	4	—	2
No. 21 orange	3	—	~1 2/3
No. 25 red	8	5	3
No. 29 deep red	16–20	—	4 to ~4 1/3

Treat these as starting points. Ilford’s own published figures for FP4 Plus list noticeably different numbers for the same glass: No. 8 yellow 1.5 in daylight, No. 15 deep yellow 2.0, No. 11 yellow-green 3.0, No. 21 orange 2.3, No. 25 tricolour red 6.0, No. 58 tricolour green 6.0. Set Kodak’s generic No. 25 of 8 against FP4 Plus’s published 6.0 (4.0 in tungsten) and the thesis is settled in a single line: identical filter, two manufacturers’ numbers, because the spectral responses differ. The primary references for all of this are the Kodak Photographic Filters Handbook (Publication B-3) and the Ilford FP4 Plus and HP5 Plus technical datasheets — not a generic web table.

Why It Shifts With the Source

A factor is valid only for the spectral content of the light used to measure it. Tungsten runs at about 3200K against daylight’s 5500K: it is far richer in red and far poorer in blue. A red or orange filter therefore discards proportionally less of a tungsten source and its factor falls, while a blue filter discards more and its factor rises. The Wratten No. 25 shows it plainly — a factor of 8 in daylight drops to 5 in tungsten on Kodak’s generic figures, and FP4 Plus mirrors the shift at 6.0 daylight to 4.0 tungsten. Same glass, same film, a different correction, purely because the source spectrum changed.

Converting to Stops

A stop is a doubling of exposure, so the conversion is logarithmic: stops equal the base-2 logarithm of the factor. log2(2) = 1, log2(4) = 2, log2(8) = 3. If your calculator has no base-2 key, use the change-of-base form:

stops = log10(factor) / log10(2)

For the intermediate values this matters. A factor of 5 gives log10(5)/log10(2) = 2.32, a shade over 2 1/3 stops; a factor of 3 (the orange No. 21) gives log10(3)/log10(2) = 1.58, just under 1 2/3 stops. A No. 8 yellow at factor 2 asks for one stop, a No. 25 red at 8 for three.

How you spend those stops is a free choice between the two exposure controls, and the arithmetic differs for each. Shutter speed scales linearly with the factor; aperture moves by the stop count. Say the meter reads 1/250 at f/11 and you fit a No. 25 red — factor 8, three stops. You can drop the shutter to 1/30 at f/11 (1/250 divided by the factor of 8), or open up to f/4 at 1/250 (three stops wider), or split the difference. The negative density is the same either way; only the depth of field and motion rendering change.

Stacking, and the Metering Trap

When two filters are combined the factors multiply and the stops add. A No. 8 yellow (factor 2, one stop) over a No. 11 green (factor 4, two stops) gives factor 8 and three stops in total. Treat the result as approximate: overlapping absorption bands and the extra glass surfaces push the real combined loss slightly higher than the simple product.

The last pitfall is metering. Through-the-lens cells read the filtered light directly, which sounds ideal, but silicon and CdS photodiodes are disproportionately sensitive to red relative to most panchromatic emulsions. Meter through a deep red — No. 25 or No. 29 — and the cell sees more red transmission than the film will record, so the camera tends to under-expose, often by around a stop on many bodies. For deep reds, take an unfiltered hand-held reading and apply the published factor by hand. Yellow and orange generally meter acceptably through the lens.

The canonical demonstration of all of this is Monolith, the Face of Half Dome, made by Ansel Adams from the Diving Board in Yosemite on 17 April 1927. He first made an exposure through a K2 yellow (Wratten No. 8), then swapped to a deep-red Wratten No. 29 for a second exposure that rendered the blue sky near-black — choosing the filter for its tonal effect, paying its factor, and previsualising the print he wanted. It is the image that prompted him to coin visualisation, and it is the whole subject of this post in one frame: a filter chosen for what it does to tone, and the exposure adjusted to pay for it.

Take any of these numbers as a first approximation, then settle it on your own materials: bracket a Zone V grey card with and without the filter on your actual film and developer, read the densities, and trust the negative over the table.