Assignment: Still Life and Controlled Light

The intention

Make one still life you meant to make, on a paper negative, and prove that you meant it by arriving at it through three logged iterations in which exactly one thing changed each time.

A still life is the first assignment because it removes every variable this part has not taught you to control yet. Nothing in the frame moves, so exposure duration costs you only patience. You own every light in the room, so the subject brightness range is yours to set rather than the weather’s. And the exposures are long — minutes to an hour — which matters more than it sounds: at f/200 on ISO 3 paper an error of one stop is an error of several minutes, and a mistake that big is one you can see, diagnose and fix in the same afternoon. The arithmetic of pinhole exposure is invisible outdoors in bright sun, where everything lands within a factor of two of everything else. Indoors it is the whole problem.

Learning objectives

By the end of this session you should be able to:

• Compute a paper-negative exposure indoors from an illuminance reading, and say which of your inputs is the least trustworthy.
• State, and then demonstrate, why moving the camera closer to a still life does not change the exposure, and what does.
• Measure the subject brightness range of a set you built and reconcile it against the paper’s published exposure scale.
• Place objects for near-far scale at a short focal distance and predict the magnification of each before exposing.
• Run three single-variable iterations and say which one mattered, with a reason drawn from light and tone rather than from taste.

Prerequisites

Seeing Like a Pinhole for framing and the intention field. Part VI’s exposure and reciprocity lesson for every number on this page, and its geometry lesson for the blur circle and the magnification. Part IV’s tray processing routine for the three baths, and its spectral sensitivity lesson for why the colour of your lamp changes the exposure.

Safety classification

Level A. The chemistry is the paper processing Part IV established and nothing else; what is new is a mains lamp left running for the length of a long exposure.

• Substances. Working-strength paper developer, citric acid stop at 1+19, ammonium thiosulfate rapid fixer at 1+4, in volumes of a litre or less.
• Energies. A mains lamp, and its heat, for tens of minutes at a time.
• Procedures. Trays and tongs indoors; a camera on a table or a tripod.
• Waste. Used fixer is silver-bearing and is collected.

What is not a hazard here, and why. None of the three baths generates a vapour, a gas or a dust: they are cold, dilute and aqueous, and the stop is citric rather than acetic. The silver stays locked in hardened gelatin until the fixer dissolves it, which is why the fixer is the one bath collected separately. The long exposure is not itself a hazard — a camera with a hole in it emits nothing — and the darkness you work in for loading is a trip hazard rather than a chemical one. What genuinely changes on this page is that a lamp which would be on for thirty seconds in ordinary use is on for forty minutes, close to card, paper and cloth that you have arranged around it on purpose.

Hazards

Paper developer. Hydroquinone’s aggregated GHS classification includes skin sensitisation and serious eye damage, and the working solution is alkaline. Tongs and gloves, as ILFORD instruct.

Stop and fixer. Mild irritants at working strength, eye irritants in particular.

Hot lamps and mains flexes. An incandescent or halogen lamp reaches a surface temperature that will scorch paper and card in contact with it, and this assignment invites you to build a set out of paper and card and then leave the lamp on for the whole exposure. Keep a hand’s width of air around the lamp, put nothing on top of it, use a lamp with an intact guard or shade, and run the flex where it cannot be walked into in the dark. Handle mains fittings dry, and switch off at the socket before moving a lamp rather than after. If you are using an LED lamp instead, this hazard is much reduced but not absent, because the driver electronics still warm and the mains flex is still a mains flex.

Cross-contamination. A trace of fixer in the developer gives blank or mottled sheets. The whole method here is comparison between iterations, so a drifting bath does not spoil one negative, it spoils the comparison.

Required PPE

• Nitrile gloves for wet work. HSE’s COSHH essentials sheet for manual development takes single-use nitrile as splash protection where the safety data sheet gives nothing more specific.
• Eye protection from before the first bottle is opened.
• Print tongs, one pair per tray, marked and never swapped.
• For the lamp, the protective measure is distance and a guard rather than anything worn.

Ventilation

An openable window or an extractor in the processing area, and not a sealed cupboard, which is ILFORD’s general instruction for darkroom spaces. Extraction is not among the controls here because none of the three baths produces a vapour that requires it; air movement is present for the ordinary reason that a small room with three trays and a hot lamp becomes stuffy, and this session depends on somebody who is still counting accurately at sheet twelve.

Materials

Item	Quantity	Notes
Variable-contrast RC paper	12 sheets	Nine for the plan, three spare. Cut to your back’s format in the dark.
The still-life objects	5 to 8	See the preparation. At least one that is genuinely dark and one that is genuinely light.
Background card or cloth	1 to 3	In two or three different tones, so background tone can be an iteration.
White card for a reflector	1 sheet, A3 or larger	Petrie’s “reflectors of white paper or card, or actual mirror”.
Tracing paper or greaseproof paper for a diffuser	1 sheet	Held clear of the lamp, never against it.
Black tape and black card	—	Flare control and masking.
A printed grey scale on your own paper	1	Made as in the commissioning session, so densities are judged against something.
Prediction sheet, folded	3	One per iteration, predictions on the left and observations on the right.

Chemicals

Chemical	Quantity	Form
Paper developer concentrate	100 mL	Diluted 1+9 to make 1 L. ILFORD use MULTIGRADE developer at 1+9 for 1 minute at 20 °C for RC paper.
Stop bath concentrate	50 mL	Diluted 1+19 to make 1 L; about 10 seconds. ILFOSTOP is a citric acid stop.
Rapid fixer concentrate	200 mL	Diluted 1+4 to make 1 L; 30 seconds for RC paper. An ammonium thiosulfate fixer.
Water	about 12 L	Dilution and washing.

Those are ILFORD’s published figures for their own materials. Use your own maker’s sheet if you are using something else, and record in the log which sheet the numbers came from.

Equipment

The camera with its flat back, at least two extension frames and the pinhole register; a table; a tripod or a stable box; a timer that will run to an hour; a lux meter or a phone app; a table lamp; three trays and a wash tray; a 1 L graduate; a thermometer; three pairs of tongs; a tested safelight; a loupe; and a light box or a bright window for reading the negatives.

Estimated cost

Cost band £. Twelve sheets of paper, some card, and chemistry you already have. Dated prices live in the laboratory planner.

Waste streams

Used fixer is silver-bearing and is collected, together with the first change of wash water, into the labelled container Part II established. Used developer and stop carry no silver. Fixed and washed paper is ordinary household waste. Local regulations govern, and they differ between authorities even within the United Kingdom.

Preparation

1. Build the set, and predict the magnification of every object in it

The set, from above, with the magnification of each object worked out before anything is exposed

1. 1Camera, f = 50 mm, sheet 4 × 5 in — frame covers 2.54 × u across the long side; at u = 150 mm that is 381 mm
2. 2Near object, u = 150 mm — m = 1/3; blur = 1.33 d; the deliberate foreground
3. 3Middle object, u = 400 mm — m = 1/8; blur = 1.13 d
4. 4Far object and background, u = 900 mm — m = 1/18; blur = 1.06 d — six times smaller than the near object at equal real size
5. 5Window, and a diffuser hung clear of the glass — the main light; measure the lux at the set, not at the window
6. 6White card reflector, about 45° — raises the shadow side; the cheapest control you have over the brightness range

Every distance on this plan is written into the log before the exposure, because m = f/u turns each of them into a predicted image size you can measure on the negative afterwards.

Choose five to eight objects, and choose them for tone before you choose them for charm. You need at least one that is genuinely dark — matt black card, a dark bottle — and one that is genuinely light, because the three-stop problem in the next section is invisible without both. Remember what the material will do to colour: the paper is blue-sensitive and gives out near 500 nm, so a red object will go nearly black and a pale blue one will go nearly white, regardless of how they look to you. That is a compositional tool, not a nuisance. Use it deliberately at least once and record what you expected.

Then measure. For each object, write u in the log and compute m = f/u and the image height m × h. This takes four minutes and it is the difference between a still life and a heap.

The extreme foreground is the pinhole’s native device. At a 50 mm focal distance an object 150 mm from the hole is rendered at a third life size, and its blur circle is d(1 + 50/150) = 1.33d — a third worse than a distant subject and still perfectly usable.

Why the near object dominates, and by how much

m = f/u = 1/3, so an 80 mm object images 27 mm tall — a fifth of the sheet's long side. The blur at this distance is 1.33 d, and the exposure is unchanged.

2. Settle the light, and measure it at the set

Petrie’s instruction from 1904 is still the right one: “The lighting is the most important element in photographing. No other requirement is so essential, for with bad lighting nothing can be done.” His method for finding a direction is worth copying exactly — hold the object with a moderately oblique light on its face, revolve it slowly in the plane of its face, and watch which features appear and disappear. Then set it down in the position that showed the most.

You have three candidate sources, and they are not equivalent to this material.

Source	What the paper sees	What to expect
North-facing window, daylight	The most blue and near-ultraviolet of the three, so the paper is at its fastest and closest to its rated speed. Glass attenuates the ultraviolet: ICNIRP record that glass strongly attenuates UVB and that window glass transmits some radiation down to about 310 nm, so the shortest wavelengths the paper could use are cut.	The shortest exposures, and the flattest, most even light. The starting point.
Tungsten or halogen lamp	Very little of what the paper can use. HARMAN’s guide time for their direct positive paper in a lit interior is about an hour, against 1 to 2 minutes in bright summer sun — roughly a factor of forty.	Much longer exposures, and a hard-edged directional light that is easy to shape.
Warm-white LED lamp	Unknown to this course, but not zero. ILFORD list “LED exposing heads designed for variable contrast papers” among the suitable exposing light sources for MULTIGRADE RC, so LED sources can certainly expose these papers.	Somewhere between the other two, and the assignment’s job is to find out where for your lamp.

📓What this course does not know, and what you are going to measure

No manufacturer publishes an effective speed for photographic paper used as a camera negative under a named illuminant. ILFORD publish an ISO paper speed determined the way paper speeds are determined, and the note that MULTIGRADE RC is roughly an equivalent film ISO of 3 to 6 — a range spanning a whole stop. Nobody publishes the correction for tungsten, and nobody publishes one for a domestic LED lamp. Part VI said so and refused to invent a number, and this page does the same.

What you can do is measure a ratio, which is exactly what the third iteration is for. Light the same set to the same illuminance at the paper’s position with two different sources, bracket both, and the difference in the sheets that match is your own effective-speed ratio between those two lamps. It is a real measurement of a real quantity, it is specific to your lamp, and it is one of the few numbers in this part that nobody could have handed you.

Measure the illuminance at the set, not at the window. Lay the meter or the phone face-up where the brightest important object stands, with its sensor pointing back at the camera. Part VI’s warning applies with full force: a phone’s ambient-light sensor was engineered to dim a screen, and it is far better at ratios than at absolutes. Since almost everything you do here is a ratio — this lamp against that one, this iteration against the last — that limitation costs you very little, provided you write the raw reading down and do not dress it up as a traceable measurement.

3. Compute the exposure, and choose the hole to land in a workable band

t = N² C / (E S)

Time from illuminance, from Part VI

t is the exposure in seconds, N the effective f-number, E the illuminance in lux at the subject, S the effective ISO of the material, and C the incident constant, which Part VI derived as C ≈ 310 from Kodak’s daylight table and a cited figure of 100,000 lux for midday summer sun. Take S = 3 for MULTIGRADE RC until your own log says otherwise.

Illuminance at the set	f/100	f/150	f/180	f/250
500 lux (a dim room)	34 min	78 min	112 min	215 min
1,000 lux	17 min	39 min	56 min	108 min
2,000 lux (bright interior)	8.6 min	19 min	28 min	54 min
5,000 lux (close to a bright window)	3.4 min	7.8 min	11 min	22 min
10,000 lux (sunlit interior)	103 s	3.9 min	5.6 min	11 min
20,000 lux (direct sun through glass)	52 s	116 s	2.8 min	5.4 min

Computed from the relation above at S = 3 and C = 310; nothing in the table is measured, and no reciprocity correction has been applied to any of it.

Read that table as a design tool rather than as an answer. Three sessions of processing and three iterations do not fit into 180 minutes if each exposure is an hour. Your job in preparation is to pick a combination — light level and pinhole plate — that puts the exposure between about two and ten minutes. That usually means working close to a bright window, or accepting a larger hole than the optimum and a correspondingly softer picture. Both of those are choices, and choosing them knowingly is the point of the exercise.

🧮DO THE MATHS: the still life that takes an hour, and why moving closer does not help

The set-up. Camera at f = 50 mm with the 0.278 mm plate, so N = 50/0.278 = f/180. The nearest object stands 150 mm from the pinhole. Paper at ISO 3. The meter, laid at the set, reads 800 lux.

The exposure.

t = 180² × 310 / (800 × 3) = 32,400 × 310 / 2,400 = 4,185 s = 70 minutes

Now the question people get wrong. The subject is 150 mm away, three focal distances. A lens photographer would apply a bellows factor here: at m = 1/3 that is (1 + 1/3)² = 1.78, three-quarters of a stop, and they would go to about two hours. Do not. Part VI proves that a pinhole has no separate bellows factor, because the f in N = f/d already is the image distance — there is nowhere else to measure it from. The illuminance at the film goes as (d/f)² with the subject distance cancelling out. Seventy minutes is seventy minutes whether the object is at 150 mm or at 15 m.

Subject at	m	Blur, as a multiple of d	Lens bellows correction	Pinhole correction
2 f = 100 mm	0.50	1.50	1.17 stops	0.00 stops
4 f = 200 mm	0.25	1.25	0.64 stops	0.00 stops
10 f = 500 mm	0.10	1.10	0.28 stops	0.00 stops

What does change the exposure is stacking a frame. Go from 50 mm to 120 mm with the same plate and N runs from f/180 to f/432: a factor of (120/50)² = 5.76, 2.53 stops, and seventy minutes becomes six and a half hours. That is the bellows factor arriving through the front door, and it arrives because N changed, not because the subject came closer.

A sanity check on the seventy minutes. HARMAN publish, for their direct positive paper in a pinhole camera, a guide time of “about 1 hour” for a lit interior. Their paper is slower than MULTIGRADE RC, their f-number is unstated, and their numbers cannot be transferred arithmetically — but a computed answer of seventy minutes landing beside a published anchor of about an hour is the kind of agreement that says the arithmetic is not wildly wrong. If you had computed six seconds or six hours, that is your signal to look for the error.

Computed exposure against effective f-number, on ISO 3 paper, at three indoor light levels

• 20,000 lux — direct sun through glass
• 10,000 lux — sunlit interior
• 5,000 lux — close to a bright window

Show the numbers behind this plot

Three rising curves of computed exposure time against effective f-number, for a material taken at ISO three, using the relation time equals N squared times three hundred and ten divided by the product of illuminance and speed. Each curve is a parabola, because exposure goes as the square of the f-number, so doubling the f-number quadruples the time. The lowest curve, for twenty thousand lux, which indoors means direct sun through glass, runs from fifty-two seconds at f slash one hundred through two hundred and seven seconds at f slash two hundred to four hundred and sixty-five seconds at f slash three hundred. The middle curve, for ten thousand lux, a sunlit interior, runs from one hundred and three seconds at f slash one hundred through four hundred and thirteen at f slash two hundred to nine hundred and thirty at f slash three hundred. The upper curve, for five thousand lux, which is a position close to a bright north window, runs from two hundred and seven seconds at f slash one hundred through eight hundred and twenty-seven at f slash two hundred to one thousand eight hundred and sixty at f slash three hundred, which is thirty-one minutes. A shaded horizontal band from one hundred and twenty to six hundred seconds marks the two-to-ten-minute working range this assignment aims for, and the practical use of the plot is reading off which pinhole plate keeps you inside that band at the light level you actually measured. At five thousand lux only apertures wider than about f slash one hundred and seventy stay inside it; at twenty thousand lux everything up to f slash three hundred does.

Series	Effective f-number, N	Computed exposure, seconds (before any reciprocity correction)
20,000 lux — direct sun through glass	100.00	52.00
20,000 lux — direct sun through glass	125.00	81.00
20,000 lux — direct sun through glass	150.00	116.00
20,000 lux — direct sun through glass	175.00	158.00
20,000 lux — direct sun through glass	200.00	207.00
20,000 lux — direct sun through glass	225.00	262.00
20,000 lux — direct sun through glass	250.00	323.00
20,000 lux — direct sun through glass	275.00	391.00
20,000 lux — direct sun through glass	300.00	465.00
10,000 lux — sunlit interior	100.00	103.00
10,000 lux — sunlit interior	125.00	161.00
10,000 lux — sunlit interior	150.00	233.00
10,000 lux — sunlit interior	175.00	316.00
10,000 lux — sunlit interior	200.00	413.00
10,000 lux — sunlit interior	225.00	523.00
10,000 lux — sunlit interior	250.00	646.00
10,000 lux — sunlit interior	275.00	781.00
10,000 lux — sunlit interior	300.00	930.00
5,000 lux — close to a bright window	100.00	207.00
5,000 lux — close to a bright window	125.00	323.00
5,000 lux — close to a bright window	150.00	465.00
5,000 lux — close to a bright window	175.00	633.00
5,000 lux — close to a bright window	200.00	827.00
5,000 lux — close to a bright window	225.00	1047.00
5,000 lux — close to a bright window	250.00	1292.00
5,000 lux — close to a bright window	275.00	1563.00
5,000 lux — close to a bright window	300.00	1860.00

Computed from t = N²C/(ES) with C = 310 and S = 3; not measured, and carrying no reciprocity correction, because nobody publishes one for paper. Every curve is a parabola: exposure goes as N², so one stop of aperture is a factor of √2 in f-number and a factor of two in time. Read it to choose the plate before you start, not to predict the answer. The curve is drawn to show the shape, not measured from a real material. Your own materials will differ, and measuring them is what the sensitometry part of the course is for.

4. Write the intention and the predictions, then fold the sheet

Before the first exposure, on the prediction side of the folded sheet: the intention in one or two sentences; the predicted image height of every object; the predicted exposure with its arithmetic shown; the two bracket times; and one sentence saying which tone in the set you expect to block up and which you expect to empty. Fold it over. You do not get to see it again until the negatives are dry.

The making

Stage 1 — The first exposure and its bracket (about 40 minutes)

Load two sheets. Expose the first at the computed time and the second at twice it, and if you have any doubt about the light, a third at half. Two stops of bracket rather than one is Part VI’s instruction for paper, and the reason is worth repeating: you are carrying an unquantified reciprocity correction, because no manufacturer publishes one for paper used as a camera negative, and at exposures of minutes it is certainly not zero.

Time the exposure with a timer you can hear, and do not stand in front of the camera. On a seventy-minute exposure you are part of the set.

Stage 2 — Process all of the first sheets together (20 minutes)

Developer 1 minute at 20 °C, stop about 10 seconds, fixer 30 seconds, wash. Process the bracket as a batch so that the only difference between the sheets is the one you introduced. Any drift in the developer between sheet one and sheet three destroys the comparison, and the comparison is the assignment.

Stage 3 — Read the negatives against the grey scale, then iterate (three cycles, 90 minutes)

Read each negative on a light box or against a window with your printed grey scale beside it, and answer three questions in writing before touching anything: which tone has gone so dense that no detail survives; which has stayed so thin that the paper base shows through; and did the objects come out at the sizes you predicted.

Then change one thing, and only one.

The three iterations, and the one variable each is allowed to move

1. 1Iteration 1 — placementMove one object, or the camera. Everything about the light and the exposure stays fixed. What you are testing is your prediction of m = f/u and of what the near-far stretch actually looks like on paper.
2. 2Iteration 2 — lightChange the direction, the source or the modifier: add the white card reflector, hang the diffuser, or swap the window for the lamp. Placement and computed exposure method stay fixed; the exposure time changes only because the measured illuminance changed.
3. 3Iteration 3 — exposureSame set, same light, a deliberate change of one stop in one direction, chosen from what iterations 1 and 2 showed. This is the one that establishes your working number for this material in this light.
4. 4After each — write whyOne sentence before the change saying what you expect it to do, and one after saying what it did. An iteration with no written expectation is not an iteration, it is another try.

The order matters. Placement first because it costs nothing to change and everything else depends on it; light second because it changes the exposure; exposure last because by then you know what you are exposing for.

Expected observations

• The exposure will be longer than feels reasonable. Minutes to an hour is normal indoors at these apertures. If your computed time comes out in seconds, check whether you used the paper’s ISO or a film’s.
• The bracket will not be symmetric. On paper, the sheet that is one stop over will usually look much closer to right than the one that is one stop under, because the material’s toe is long and its shoulder is not far above it. That asymmetry is a hint about where your effective speed really sits.
• The near object will be bigger than you expected, almost every time. A third life size on a 127 mm sheet is a substantial object.
• Reds will be dark and pale blues bright, and the set will not look the way it looked to you.
• The background will separate less than you hoped if you lit it with the same light as the subject. Backgrounds usually need to be brought forward or pushed back with their own light.

Contrast on a three-stop material

This is the section that turns a still life from an arrangement into a controlled experiment, and it rests on a published number.

ILFORD publish an ISO Range (R) figure for MULTIGRADE RC papers, to ISO 6846:1992, as a guide to matching a paper to a negative’s density range. For MULTIGRADE RC DELUXE the unfiltered figure is 90; for MULTIGRADE IV RC DELUXE it is 110. An ISO range of 90 means 0.90 log exposure units, and since one stop is 0.301 log units:

stops = R / 100 / log₁₀2 = 0.90 / 0.301 = 2.99 stops

Paper's exposure scale in stops

Three stops. That is the whole subject brightness range a paper negative can hold with detail at both ends. A film would give you seven or more.

📓How far this number can be pushed, and where it stops

What ILFORD measured is the paper’s exposure scale as a printing material, under their own processing recommendations, to a published standard. That is a real number about a real property.

What this course infers is that the same exposure scale governs the paper when it is exposed in a camera instead of under an enlarger, because it is the same physical quantity — the range of log exposures between the threshold of density and the shoulder. The inference is sound but it is an inference, and it comes with two cautions. First, the unfiltered column is the right one for a camera negative, since your subject is lit by whatever it is lit by rather than by a filtered enlarger; note that this is also the column with the highest speed, ISO P 500 against 240 with filters, which is roughly a stop. Second, at exposures of minutes the material is in reciprocity failure, and ILFORD themselves note for films that “the contrast is increased with long exposures”. Whether the paper’s scale narrows further at these durations is a thing this course does not know and cannot look up.

So use three stops as the design target and verify it on your own negatives, by putting a step-wedge or your printed grey scale in the corner of one exposure and counting how many steps survive. That measurement is worth more to you than the published figure.

What to do about three stops. Measure the range you have built: read the lux at the brightest important surface and at the darkest important surface, and take the ratio. A ratio of 8 is three stops and you are exactly at the limit. A ratio of 30 is five stops and two of them will be lost.

Three controls, in the order of how much they cost you:

1. The white card reflector, at about 45° on the shadow side. It raises the darkest important surface without touching the brightest, which is precisely what you need, and it is Petrie’s own remedy: “Any dark shadows should be lighted with reflectors of white paper or card, or actual mirror.” Expect it to buy one to two stops off the range. It costs nothing in exposure.
2. The diffuser, hung clear of the window or well clear of the lamp. It softens the transition and usually takes a stop or more off the range, at the price of a longer exposure, because it is throwing light away.
3. Changing the objects. Swapping the matt black bottle for a mid-grey one is not cheating. It is the same decision a painter makes.

What blocks and what empties. On a negative, the brightest parts of the subject go densest. “Blocking up” is a highlight so dense that no detail comes through it; “emptying” is a shadow so thin that only the paper base shows. Because the material’s whole scale is three stops, you generally have to choose which end to protect. Protect the end the intention needs, and say in the log which one you sacrificed.

Alternative route

The picture is the point, not the apparatus, and every route below produces a submittable negative.

• No mains lamp, or no wish to run one. Use the window and only the window. This assignment declares requiresMains because one of its two lighting options is a lamp on a stand, and the other is daylight through glass: the exposure arithmetic in the preparation stage is written for both, the measured illuminance at the set is what the computation actually consumes, and a north window on an overcast day is the steadier of the two sources. What you lose is control of the direction between sittings, which is why the log records the time of day and the sky.
• A window-sill camera. Set the camera on a sill or a table and build the still life on the same surface. Nothing here requires a tripod, floor work or lifting; the exposures are long enough that a camera resting on a book is as steady as anything.
• A card camera. If the modular body is not available to you, a stiff box with a measured hole and a measured depth does everything this assignment asks. Measure and log the focal distance; that is the only non-negotiable.
• A scanned negative instead of a contact print. Photographing the dry negative against a window with a phone and inverting it is an accepted working positive for every assignment in this part.
• A shorter session. The three iterations may be spread over three sittings. Log the light separately for each, because it will have changed.

The record to keep

Every field of Part VI’s exposure log, plus the three this part adds, for every sheet — including the ones you throw away, which are usually the informative ones.

Specific to this assignment, and easy to forget:

• The illuminance at the set for each source used, with what you measured it with and where the meter was lying.
• The subject-to-pinhole distance of every object, and the predicted image height of each.
• The measured brightness range as a ratio and in stops.
• The lamp, by type and wattage if it has one — this is the row that eventually gives you your own tungsten and LED corrections.
• Which prediction was wrong, in the words you wrote before the exposure rather than in a summary written afterwards.

The critique

One paragraph, written before you show the picture to anybody, following the critique sequence and naming causes rather than preferences:

1. Intention. What did you say you wanted, and which clauses of it happened?
2. Optics. Where is the softness, and is it the size of your hole? Where are the corners, and do they match the cos⁴ figure for your focal distance?
3. Exposure. Which bracket sheet did you keep, and what does that say about your effective speed in this light?
4. Tone. What blocked, what emptied, and was that the end you meant to sacrifice?
5. Composition. Did the near-far stretch do what you predicted from m = f/u?
6. Execution. Marks, leaks, flare, uneven development.
7. The one change. Stated as an action with a number in it, not as an aspiration.

Troubleshooting

What you see	Likely cause	What to do
Everything thin, almost blank	Exposure far too short — usually a film ISO used instead of the paper’s, or a meter reading taken at the window rather than at the set	Recompute at S = 3; re-read the lux where the objects stand
Everything dense, no separation anywhere	Exposure far too long, or a light leak over a long exposure	Check the leak test from Part VI before blaming the time; a leak that is invisible in 30 seconds is obvious in 40 minutes
Highlights blocked and shadows empty at the same time	Subject brightness range well over three stops	Reflector first, diffuser second, then change an object
Near object much softer than the rest	Correct and expected: blur = d(1 + f/u)	If it is much worse than 1.33 d, suspect a slot rather than a round hole, or camera movement
A pale band or fog on one edge	Flare from a light source just outside the frame	Black card flag between lamp and camera; check the interior blacking
One iteration lighter overall for no reason you changed	Exhausted or cold developer between batches	Process the batch together; log developer temperature every time
A soft, doubled edge on one object only	The object moved — a flower opening, a leaf drying, a lid settling	At forty minutes, some things you thought were still are not

Clean-up, storage and disposal considerations

Trays emptied in the order developer, stop, fixer, with the fixer into the labelled silver-bearing container. Rinse trays and tongs, and dry them; a tong that dries with fixer on it contaminates the next developer. Negatives dried face-up on a clean rack and stored flat in the sleeving Part VI specified, labelled to the log row rather than to a description.

Disposal. Used developer and stop carry no silver; used fixer does, and silver-bearing solution is collected rather than discharged, because the silver-thiosulfate complexes it contains are the thing the receiving works cares about. Local regulations govern, and they differ between authorities even within the United Kingdom.

What to submit

• One selected negative and its working positive.
• The three iteration logs, with the folded prediction sheets, unedited.
• One critique paragraph.
• One number: your best current estimate of the effective ISO of your paper in the light you used, and the range of exposures over which you believe it.

Further work

• The tungsten-against-daylight ratio, done properly. Light the same set to the same measured illuminance with each source, bracket both across five stops, and read off the difference. That gives you a number nobody publishes.
• The three-stop question, measured. Put a step wedge in the corner of an exposure and count the steps that survive at each of three exposure times. That is a crude characteristic curve, and Part XIII does it properly.
• A still life at 120 mm. The same objects at the long focal distance, from further back, at 2.5 stops more exposure. It is a different picture, and the log will tell you exactly why.