Film Speed and Exposure Index

The number on the box is not a property of the film in the way that its thickness is. It is the answer to a question — how much light does this material need to reach an agreed point on its curve, when processed in an agreed way? — and changing the agreement changes the answer. That is not a scandal. It is what the manufacturers themselves say, in the small print, on every datasheet in this course’s corpus.

This page defines the criterion the course uses, works a speed from a published curve with it, and then spends the rest of its length on the four things that move the answer: the developer, the light, the criterion and the exposure time.

Speed is a position on the curve

Contrast is the slope of the characteristic curve. Speed is its position along the exposure axis. A fast film reaches a given density with less light, so its curve sits further to the left.

To turn “further to the left” into a number you need two agreements: a criterion point on the curve, and a rule fixing the development, because otherwise more development would let anyone claim any speed they liked. Every speed system in history is a pair of answers to those two questions, and the systems disagree because the answers do.

The course’s speed criterion

This is the second of the two conventions this part fixes for the whole course. Later parts cite it rather than restating it, exactly as they cite the contrast-index convention.

!Why the course states its own criterion

The relevant standard is sold, not published, and buying it would grant no right to reproduce it. So the course does not buy it. It states the criterion in full, cites ISO 6 by number as the standard the criterion is modelled on, and marks every figure derived under it as the course’s own measurement under the course’s own criterion. The construction below follows the two-step method printed in full in Kodak’s publication H-740, a freely available manufacturer document the course may describe at length.

A page in this course may write “EI 320, measured under the criterion defined in Part XIII, which is modelled on ISO 6”. It may never write “the ISO speed is 320”.

One further honesty. The edition the course names is the second, ISO 6:1993, and the course has not verified whether that edition is still current or has since been revised or withdrawn. Since nothing here is quoted from it and nothing here claims conformance to it, that uncertainty changes no figure on the page — but a reader who needs the standard itself should check its status rather than take this citation as current.

The definition

Pure Silver speed criterion. Plot diffuse density against log exposure in lux-seconds. Determine base-plus-fog D₀ from an unexposed strip processed alongside. Locate point m where the curve reaches density D₀ + 0.10. Locate point n at 1.30 log exposure units to the right of m. The development is at the criterion when the density at n exceeds the density at m by 0.80, within ±0.05. When that condition holds, H_m is the exposure at m, and the speed is

S = 0.80 ÷ H_m, with H_m in lux-seconds

Speed from the criterion exposure

S = 800 ÷ H_m, with H_m in millilux-seconds

The same thing in millilux-seconds

Three things are worth pausing on.

The criterion is a density above the base, not an absolute density. That is why base-plus-fog has to be measured on a control strip every time: if D₀ drifts, m drifts with it.

The 0.80-over-1.30 condition is a development rule, not a speed rule. It exists so that the speed is quoted for a comparable degree of development. Notice what it is: 0.80 ÷ 1.30 = 0.615, an average gradient of about 0.62, and the ±0.05 tolerance makes that a window of 0.58 to 0.65. The speed criterion and the contrast conventions on the previous page are the same arithmetic seen from two directions.

The formula is a definition, not a physical law. The 0.80 in the numerator is a chosen constant that makes the resulting numbers land on the familiar series. Change the constant and every speed in the world changes by the same factor, and nothing about any film would be different.

The link to cite is /part-13-sensitometry/film-speed-and-exposure-index/#the-courses-speed-criterion. A later page that reports a speed or an exposure index points at that anchor rather than restating the criterion, for the same reason the contrast convention is stated in exactly one place.

Locating m and n in practice

The speed criterion on a published curve

• Measured densities, Kodak H-740 sample film
• The 1.30 by 0.80 criterion chord, m to n

Show the numbers behind this plot

Eleven measured density points rise from 0.18 at log exposure 0.30 to 1.70 at 3.30. Base plus fog is 0.18, so the criterion point m sits at density 0.28, which the curve reaches at log exposure 0.90. Point n is 1.30 further along at log exposure 2.20, where the curve reads 1.08. The density difference between them is 0.80 exactly, so the development satisfies the criterion, and the speed follows from the exposure at m.

Series	Log exposure (millilux-seconds)	Density
Measured densities, Kodak H-740 sample film	0.30	0.18
Measured densities, Kodak H-740 sample film	0.60	0.20
Measured densities, Kodak H-740 sample film	0.90	0.28
Measured densities, Kodak H-740 sample film	1.20	0.45
Measured densities, Kodak H-740 sample film	1.50	0.64
Measured densities, Kodak H-740 sample film	1.80	0.82
Measured densities, Kodak H-740 sample film	2.10	1.01
Measured densities, Kodak H-740 sample film	2.40	1.20
Measured densities, Kodak H-740 sample film	2.70	1.39
Measured densities, Kodak H-740 sample film	3.00	1.58
Measured densities, Kodak H-740 sample film	3.30	1.70
The 1.30 by 0.80 criterion chord, m to n	0.90	0.28
The 1.30 by 0.80 criterion chord, m to n	2.20	1.08

The eleven points are Kodak's published densities for its sample film; the chord is the criterion construction applied to them. The density difference comes out at exactly 0.80 over 1.30 of log exposure, so this development qualifies.

Work it through on those numbers. D₀ = 0.18, so m is at density 0.28, which the curve reaches at log H 0.90. Point n is at log H 0.90 + 1.30 = 2.20, where the curve reads 1.08. The difference is 1.08 − 0.28 = 0.80 exactly, so the development qualifies. Then

S = 800 ÷ 10^0.90 = 800 ÷ 7.94 = 100.7

The worked speed

which rounds to a speed of 100. In lux-seconds the same point is log H = 0.90 − 3.00 = −2.10, and S = 0.80 ÷ 10^−2.10 = 0.80 ÷ 0.00794 = 100.7. Two units, one answer, as they must be.

The two scales, and why ISO 400/27° is one speed written twice

Every speed appears in two forms. The arithmetic form doubles when the film gets twice as fast: 50, 100, 200, 400. The logarithmic form, written in degrees, adds a constant instead: 18°, 21°, 24°, 27°. They are the same rating.

The relation between them can be derived rather than looked up, which is the honest route for a course that does not hold the standard.

S° = 1 + 10 log₁₀ S, rounded to the nearest whole degree

The degree scale

Check it against published ratings in this course’s corpus, which is how you know a derivation is right:

Published rating	S	10 log₁₀ S	+ 1	Rounded	Printed degrees
ILFORD Ortho Plus, tungsten	40	16.02	17.02	17	17°
ILFORD Pan F Plus	50	16.99	17.99	18	18°
ILFORD Ortho Plus, daylight	80	19.03	20.03	20	20°
Fomapan 100 Classic	100	20.00	21.00	21	21°
ILFORD FP4 Plus	125	20.97	21.97	22	22°
ILFORD SFX 200	200	23.01	24.01	24	24°
ILFORD HP5 Plus	400	26.02	27.02	27	27°

Seven published ratings, seven agreements. That is what a derivation is worth: the course did not have to be told the relation, and you did not have to take it on trust, because the manufacturers’ own sheets check it. Notice too that the +1 offset never does anything except shift the whole scale — the useful content is the 10 log₁₀, which turns a doubling into 3.01 degrees and so makes one stop almost exactly 3° and one third of a stop almost exactly 1°.

That table’s structure is worth knowing. Kodak’s workbook prints the standard series in rows of three — 32, 40, 50 · 64, 80, 100 · 125, 160, 200 · 250, 320, 400 · 500, 650, 800 — where going down a column doubles the speed and going across a row is a third of a stop. The reason a third of a stop is the step is arithmetic: log 2 = 0.30, a third of that is 0.10, and 0.1 in log H is a tidy interval to work in. Kodak also flags the two members of the series that do not follow the pattern, 650 and 1250.

ILFORD names the ancestry directly on the Pan F Plus sheet, which gives the rating as ISO 50/18°, 50 ASA, 18 DIN. ASA is the arithmetic scale and DIN is the degree scale, and the modern notation simply prints both.

Why speed depends on the process, the light and the criterion

On the process. ISO 6 — cited by number — is a standard for film and process systems rather than for films alone, and the manufacturers write accordingly. Kodak’s workbook states plainly that different developers yield different film speeds and that the developer specified by the standard method is similar to D-76. ILFORD goes further and prints the process on the sheet: FP4 Plus is rated ISO 125/22 to daylight, and its rating was measured using ID-11 at 20 °C with intermittent agitation in a spiral tank. Change any of that and the number the same criterion returns will move.

On the light. The single clearest demonstration in the corpus is ILFORD’s Ortho Plus, one film with two published ratings: ISO 80/20° to daylight sources and ISO 40/17° to tungsten sources, both in ID-11 developed to normal contrast. That is a whole stop, and nothing about the emulsion changed — only the spectral distribution of the light that fell on it. An orthochromatic film is blue and green sensitive and blind to red, and tungsten light is poor in blue and rich in red, so a lux of tungsten does less photographic work than a lux of daylight. ILFORD even notes that the 135 cassettes are DX coded for 80, so a camera reading the code will be a stop out under tungsten unless you override it. This is why the standards specify an illuminant as well as a criterion.

On the criterion. Move the criterion point and every film’s speed moves, not necessarily by the same amount, because films differ in toe shape. A criterion placed higher up the curve flatters short-toe films; one placed lower flatters long-toe ones. That is the deep reason speed systems cannot simply be converted into one another by a constant: they rank films differently, not just number them differently.

Practical ratings versus foot speed

Manufacturers are explicit that the number on the box is a recommendation, and they say what kind.

ILFORD prints, on HP5 Plus, FP4 Plus and Pan F Plus alike, that the recommended exposure index range is based on a practical evaluation of film speed and is not based on foot speed, as is the ISO standard. HP5 Plus is rated ISO 400/27°, and the same sheet claims good image quality anywhere from EI 400 to EI 3200 given extended development in named developers. Those are two different kinds of claim on one page, and ILFORD distinguishes them.

Kodak writes of T-Max 100 that its nominal speed is EI 100, determined in a manner published in ISO standards, and then advises using the rated speed or making tests to determine a rating that meets your needs.

Foma gives Fomapan 100 Classic a nominal speed rating of ISO 100/21° and adds that its wide exposure latitude gives good results overexposed by 1 EV, as ISO 50/18°, or underexposed by 2 EV, as ISO 400/27°, without any change in processing. Read carefully, that is a claim about latitude and printability, not a claim that the film has three speeds.

!Foot speed, and why the phrase matters

“Foot speed” means a speed criterion taken from the foot of the curve — a point low on the toe, close to where shadow detail begins. It is a demanding criterion, because it asks the film to have recorded something in the deepest tones. A “practical evaluation”, by contrast, asks whether the negatives print well, which tolerates thinner shadows if the mid-tones and highlights are good.

The two questions have different answers, and a manufacturer that says which one it answered is being more honest than one that prints a bare number. When ILFORD says its EI range is not foot speed, it is telling you that at EI 3200 the shadows will not be there — and that the pictures may still be worth making.

Where the number came from before

Hurter and Driffield’s system is fully verifiable and worth knowing because its logic is the opposite of the modern one. They defined the inertia i as the exposure at which the produced straight line cuts the exposure axis — the least exposure marking the beginning of what they called the period of correct representation — and took speed as its inverse. Their published actinograph speed was

S = 34 ÷ i, with i the inertia in candle-metre-seconds

The Hurter and Driffield speed number

and they worked it for the plates of the day: Ilford “ordinary” came out at 17, “rapid” at 24 and “special rapid” at 60. The virtue of the construction was that the inertia point barely moves with development time, so the number was a property of the plate. Its vice is that it depends on a straight line existing to produce backwards, which is exactly the weakness Sheppard and Mees identified in 1907 and the reason the modern criterion abandoned it for a point on the toe plus a development condition.

Scheiner, DIN, Weston and BS are where this page stops. Their names are everywhere in the older literature; this course has not verified their dates, their defining criteria or their conversion factors against any source meeting its standard, and it will not print a conversion table it cannot source. What it can say is narrow and true: ILFORD equates ASA with the arithmetic scale and DIN with the degree scale on its own datasheets, and Kodak’s workbook records that ANSI was formerly the ASA and that ISO adopted the ANSI method, so ASA and ISO arithmetic numbers are the same numbers. Anything finer than that awaits a source.

!Why there is no conversion table on this page

Tables converting Scheiner to DIN to Weston to ASA circulate widely, in books and online. This course prints none of them, and the reason is not caution for its own sake.

A speed number means nothing without the criterion behind it, and these systems used different criteria — which is to say they did not merely number films differently, they could rank them differently. Several of them were also revised more than once, so a conversion is only correct for a stated revision, and a table that does not name its revision cannot be checked. This course holds none of the defining documents, so it has no way to name the criterion or the revision for any of them, and under the course’s provenance rule a figure it cannot verify does not go on the page.

If you own an old meter with one of these scales, the route that does not depend on anybody’s table is the one the rest of this part teaches: expose a step wedge, apply the criterion above, and measure the offset between what your meter says and what your film does.

Exposure index: a documented decision

An exposure index is the number you actually set on the meter. It is not a measurement of the film. It is a decision about how to expose, and it is worth nothing unless the process it was chosen for is written down beside it.

That is the whole distinction the course keeps between film speed and effective film speed. Speed is a rating under a criterion. Effective speed is what you get from your film in your developer at your time and temperature, judged by your own criterion, and a personal exposure index is your documented choice in the light of it.

A personal EI that means something reads like this: EI 320 for HP5 Plus in ID-11 1+1 at 20 °C, 13 minutes, agitation ten seconds each minute, chosen so that a subject shadow placed two and a half stops below the meter reading records 0.10 above base-plus-fog on my step wedge. Every clause is doing work. A personal EI that means nothing reads: I rate HP5 at 320.

📓What to record when you set a personal EI

Film and batch; developer, dilution, temperature, time and agitation; the criterion you used and the density it asks for; the measured base-plus-fog of the control strip; the wedge and whether it was calibrated; how the densities were read and with what uncertainty; the date. Any figure quoted without those is a preference, not a measurement, and the course asks you to say which you are offering.

Speed loss and gain: what really moves

Four things are commonly said to change film speed. They do not all change the same thing.

Developer choice genuinely moves the criterion speed, and the manufacturers say so. A solvent fine-grain developer dissolves some halide before it can be developed and grows more compact grains with lower covering power, so the same exposure yields less density, including at the criterion point — a real speed loss. Part VIII’s solvent lesson is the mechanism. A speed-increasing developer works the other way, developing more of the marginal grains.

Dilution and agitation move the time to a given contrast more than they move the criterion speed, because the criterion holds development constant by construction: satisfy the 0.80-over-1.30 condition and you have already normalised for it.

Push processing is the one to be careful about, and this course has not found a source in its corpus that settles it. The argument against calling it a speed increase is geometric: extended development rotates the curve about a near-fixed point low down, so it raises the mid-tones and highlights a great deal and the deep shadows very little. If the shadow density was not recorded, no development creates it. The argument that something real happens is that the criterion is not a shadow-detail criterion in the ordinary sense, and extended development does move where the curve crosses it — and ILFORD, which knows its own films, publishes extended-development times for HP5 Plus up to EI 3200 while telling you in the same document that this is a practical evaluation and not foot speed. The course’s position is that pushing buys printable mid-tone contrast and buys very little shadow detail, and that ILFORD’s own wording is the best short statement of it.

Underexposing and calling it a push changes nothing at all. It is worth saying once.

Reciprocity failure as a speed effect

At long exposures the reciprocity law stops holding, and the sensitometric symptom is a loss of effective speed — the same total exposure delivered slowly produces less density. ILFORD describes it exactly that way: reduced efficiency in forming stable development centres at low light levels. Part IV owns the mechanism.

ILFORD publishes a per-film correction. Its equation is

T_c = T_m^p

ILFORD reciprocity correction

where T_m is the metered time in seconds, T_c the corrected time, and p a factor ILFORD gives for each film: HP5 Plus 1.31, FP4 Plus 1.26, Pan F Plus 1.33, Delta 100 1.26, Delta 400 1.41, Delta 3200 1.33, SFX 1.43, XP2 1.31, Ortho Plus 1.25, Kentmere 100 1.26, Kentmere 400 1.30. ILFORD’s own worked example: a metered 10 seconds on HP5 Plus becomes 10^1.31 = 20.4, rounded to 20 seconds. And the sheet states that exposures of one second or less need no compensation at all, which is the practical boundary for the step-wedge lab later in this part.

Two further points from the same document, both of which matter here rather than in Part IV. It notes that contrast is increased with long exposures, because different parts of the image sit at different light levels and so suffer different amounts of reciprocity failure, and that pulling development may be needed as a result. And it warns that at very low light levels the accuracy of the light measurement itself becomes a variable, so some trial and error is required.

Foma states the same phenomenon in table form for Fomapan 100 Classic, under its older name the Schwarzschild effect: no correction from one five-hundredth to half a second, ×2 at 1 second, ×8 at 10 seconds and ×16 at 100 seconds. Comparing that with the ILFORD exponent for a similar film is a worthwhile exercise in reading two makers’ conventions against each other.

The name comes from Karl Schwarzschild, whose 1900 paper states the reciprocity law as equal blackening for equal products of intensity and time and then measures its failure on gelatine plates, reporting an exponent of 0.86 for the material he tested. That single number, in a paper about astronomical photography, is the ancestor of every reciprocity table since.

Why a home sensitometer cannot give you an ISO speed

It can give you a great deal. It cannot give you that.

The light is wrong. The criterion assumes a specified sensitometric illuminant. Your enlarger lamp is not one, and as Ortho Plus demonstrates, a whole stop can hang on the spectral distribution alone.

The densitometry is wrong. The criterion assumes a specified diffuse geometry. A lux meter under an enlarger, a spot meter on a light box or a flatbed scanner each measure something related but not identical, in an undeclared spectral band.

The exposure axis is nominal. An uncalibrated wedge’s steps are nominal, as Part IX established, and the absolute exposure at the clear edge is only as good as your lux meter’s calibration.

The process is your own, which is the one item on this list that is a feature rather than a fault.

What survives all four is comparison, and comparison is what changes decisions. Relative speed between two developers, measured on the same lamp with the same wedge on the same afternoon, is a sound measurement: the systematic errors are common to both and largely cancel. A personal exposure index, documented as above, is a sound decision. Neither is an ISO speed and neither needs to be.

Do the maths

🧮From a strip of numbers to a speed and an EI

You expose a 21-step wedge on FP4 Plus, contact, under a lamp giving a measured 32 lux-seconds at the clear edge, and develop in ID-11 1+1. The control strip reads base-plus-fog 0.11. Reading the strip gives density 0.21 at wedge step 14 and density 1.01 at wedge step 5.

1. Turn wedge steps into log exposure. The clear edge is log 32 = 1.51. A 21-step wedge is nominally 0.15 per step, and step 1 is the clearest at density 0.05, so step k has density 0.05 + 0.15(k − 1). Step 14: density 2.00, so log H = 1.51 − 2.00 = −0.49. Step 5: density 0.65, so log H = 1.51 − 0.65 = 0.86.

2. Check the development against the criterion. m is the point at D₀ + 0.10 = 0.11 + 0.10 = 0.21, which is step 14, at log H −0.49. Point n is 1.30 further along, at log H −0.49 + 1.30 = 0.81. Your reading at step 5 is at log H 0.86, close but not on it; interpolating between neighbouring steps gives a density at 0.81 of about 0.98. The rise from m to n is 0.98 − 0.21 = 0.77, which is inside the 0.80 ± 0.05 window, so the development qualifies — just.

3. Compute the speed under the criterion. H_m = 10^−0.49 = 0.324 lux-seconds. S = 0.80 ÷ 0.324 = 2.47.

4. Notice that the answer is absurd, and find out why. FP4 Plus is rated 125, and a criterion speed of 125 puts H_m at 0.80 ÷ 125 = 0.0064 lux-seconds, or log H −2.19. Your strip put it at −0.49. The gap is 1.70 in log H, a factor of about fifty, and 2.5 is not a speed but a diagnosis: the exposure axis is wrong. Thirty-two lux-seconds is roughly what a printing paper wants, not a camera film. Either the lux meter was read with the lens wide open and the wedge out of the beam, or the neutral-density filtration a sensitometer needs to bring a camera film into range was left out. Sanity-check every home speed determination against the box speed like this, because a large error in the exposure axis is easy to make and completely invisible in the shape of the curve.

5. What you can trust anyway. The shape is unaffected: gamma, contrast index and the 0.77 rise from m to n are all differences, and a constant error in the exposure axis cancels out of every one of them. So the strip still tells you your development is at the criterion window, and a second strip in a different developer would still give you a sound relative speed. That is the honest division of the spoils: shape and comparison are yours, absolute position is not.

6. Set an exposure index. Now suppose you run two strips side by side, one in each of two developers, and the comparison shows that developer B reaches the criterion at two thirds of a stop less exposure than developer A. Two thirds of a stop is 0.20 in log H, a factor of 1.58. If you were rating the film at EI 125 in developer A and you are satisfied with those negatives, then the same shadow placement in developer B calls for 125 × 1.58 ≈ 200, and you set EI 200 — recording, with it, every clause the notebook callout above asks for. Note what has and has not been claimed: the factor of 1.58 is a measurement, the choice of 125 as the starting point is a judgement about your own prints, and the resulting 200 inherits both.

Recap

• Speed is the exposure needed to reach a criterion point on the curve, under a stated development. Position along the exposure axis, not slope.
• The course’s criterion is fixed on this page: m at 0.10 above measured base-plus-fog, n at 1.30 log units further, development set so the rise is 0.80 ± 0.05, and S = 0.80 ÷ H_m in lux-seconds. It is modelled on ISO 6, cited by number and quoted nowhere.
• The development condition is an average gradient of about 0.62, which is the same arithmetic as the contrast conventions of the previous lesson.
• The two scales are one rating: arithmetic doubles per stop, degrees add about 3 per stop, and ILFORD prints ASA for the first and DIN for the second on its own sheets.
• Speed depends on the process and the light. ILFORD publishes the developer, temperature and agitation it measured in, and gives Ortho Plus two ratings a stop apart for daylight and tungsten.
• Practical ratings are not foot speed, and ILFORD says so in as many words; an exposure index is a documented decision, not a property of the film.
• Long exposures cost speed and add contrast; ILFORD’s correction is T_c = T_m to the power p, with no correction needed at one second or less.
• A home chain gives relative speed and a personal EI, not an ISO speed, because the illuminant, the densitometry and the exposure axis are all outside the criterion’s assumptions.

Check your understanding

Question 1. A strip has base-plus-fog 0.14. Where is the criterion point m, and what has to be true 1.30 log H further along for the speed to be quotable under this course criterion?

• m is at density 0.10; the density 1.30 further along must be 0.80
• m is at density 0.24; the density 1.30 further along must be 1.04, within 0.05
• m is at density 0.24; the density 1.30 further along must be 0.80, within 0.05
• m is at the toe; there is no condition on the second point

Show the answer and why

Answer: m is at density 0.24; the density 1.30 further along must be 1.04, within 0.05

The criterion is a density above the measured base, so m sits at 0.14 plus 0.10 = 0.24. The condition on point n is on the density difference, not on its absolute density: n must be 0.80 above m, so 0.24 plus 0.80 = 1.04, with the tolerance of plus or minus 0.05 giving a window of 0.99 to 1.09. Getting this wrong in the third option is the commonest slip, and it matters: reading the condition as an absolute density would make the qualifying development depend on how foggy the film was, which is the opposite of what the condition is for.

Question 2. A curve satisfies the development condition with the criterion point at log H = −2.40 lux-seconds. What speed does the course criterion give, and what would you write on the page?

• S = 200; write "the ISO speed is 200"
• S = 200; write "speed 200, measured under the criterion defined in Part XIII, which is modelled on ISO 6"
• S = 0.0032; write "the film is very slow"
• S = 320; write "ISO 320/26 degrees"

Show the answer and why

Answer: S = 200; write "speed 200, measured under the criterion defined in Part XIII, which is modelled on ISO 6"

H at m is 10 to the power minus 2.40, which is 0.00398 lux-seconds, and 0.80 divided by 0.00398 is 201, rounding to the standard value 200. The wording matters as much as the arithmetic. The course does not purchase the standards and quotes no part of them, so it cannot claim that a figure is an ISO speed; what it can do is state the criterion in full, name the standard the criterion is modelled on, and label the figure as its own measurement. A reader can then reproduce it, which is more than an appeal to a document they cannot read would give them.

Question 3. ILFORD publishes Ortho Plus at ISO 80/20 degrees to daylight and ISO 40/17 degrees to tungsten, both in the same developer. What does the one-stop difference measure?

• A manufacturing variation between batches
• The spectral sensitivity of the film against the spectral output of the source: an orthochromatic emulsion is blind to red, and tungsten light is rich in red and poor in blue
• A difference in development between the two tests
• An error in one of the two figures

Show the answer and why

Answer: The spectral sensitivity of the film against the spectral output of the source: an orthochromatic emulsion is blind to red, and tungsten light is rich in red and poor in blue

Nothing about the emulsion changes between the two figures. What changes is how much of the incident light the emulsion can use. Lux is a photometric unit weighted by the response of the human eye, and the eye is most sensitive in the green while an orthochromatic film is most sensitive in the blue, so a lux of tungsten light, which is heavily weighted to the red end, does less photographic work than a lux of daylight. This is exactly why the standards specify an illuminant alongside the criterion, and it is why ILFORD warns that a DX-coded cassette reading 80 will be a stop out under tungsten.

Question 4. A metered exposure of 30 seconds is needed on FP4 Plus, whose ILFORD reciprocity factor is 1.26. What corrected time should you give, and what else should you consider changing?

• 38 seconds; nothing else
• 30 times 1.26 = 38 seconds; reduce development slightly
• 30 to the power 1.26 = 73 seconds; consider reducing development, because long exposures raise contrast
• 30 divided by 1.26 = 24 seconds; increase development

Show the answer and why

Answer: 30 to the power 1.26 = 73 seconds; consider reducing development, because long exposures raise contrast

The factor is an exponent, not a multiplier: the corrected time is the metered time raised to the power p, so 30 to the power 1.26 is about 73 seconds, not 38. That distinction grows with the exposure, which is why multiplying works passably at a few seconds and fails badly at a minute. ILFORD adds the second half of the answer in the same document: contrast increases with long exposures, because the shadows sit at a lower light level than the highlights and therefore lose proportionally more to reciprocity failure, so pulling the development may be required. A long night exposure is a contrast problem as well as an exposure problem.

Question 5. Two developers are tested on the same film, the same afternoon, with the same lamp and the same uncalibrated wedge. Developer A reaches the criterion at log H −1.80 on the strip axis and developer B at −1.50. What can be claimed?

• Developer A gives an ISO speed of 50 and B of 25
• Developer A gives twice the effective speed of B, a sound relative result, even though neither absolute figure can be quoted as a speed
• Nothing, because the wedge is uncalibrated
• Developer B is faster, because its number is larger

Show the answer and why

Answer: Developer A gives twice the effective speed of B, a sound relative result, even though neither absolute figure can be quoted as a speed

The difference between the two is 0.30 in log H, which is exactly one stop, so A reaches the criterion with half the exposure B needs: twice the effective speed. That comparison is sound because every systematic error in the chain, the lamp spectrum, the meter calibration, the wedge step error and the reading geometry, is common to both strips and cancels in the difference. What does not survive is the absolute position of the axis, so neither minus 1.80 nor minus 1.50 may be turned into a speed number and published. This is the central practical lesson of home sensitometry: differences are trustworthy, positions are not.

Question 6. Why does the course criterion impose a condition on the development at all, rather than simply reading the exposure at a fixed density above base?

• To make the arithmetic harder
• Because without it, more development would raise the curve and appear to give any speed you liked, so speeds could not be compared
• Because the standard requires it and no reason is given
• Because base-plus-fog would otherwise be impossible to measure

Show the answer and why

Answer: Because without it, more development would raise the curve and appear to give any speed you liked, so speeds could not be compared

Development rotates the curve, and although it moves the deep toe least, it does move it. Without a condition tying the degree of development down, a manufacturer or a student could develop further, watch the criterion point slide left, and report a higher speed for the same film. The condition, a rise of 0.80 over 1.30 of log exposure, pins the contrast at an average gradient near 0.62, so every speed quoted under the criterion refers to a comparable degree of development. It is also why the standard is written for film and process together rather than for films alone, and why a datasheet that quotes a speed without naming the developer is telling you less than ILFORD does.

Check my answersClear