exposing negative films

With negative films -- black and white or colour -- there is no such thing as a single correct exposure. Grievous under-exposure will give poor tonality and empty shadows, and grievous over-exposure will give a very dense negative with poor sharpness and (if the over-exposure is very great indeed) 'blown' highlights with no detail at all.

In between these two extremes there is a wide range of usable exposures, and quite a wide range of good or even excellent exposures. The original research (at Kodak in the late 1930s and early 1940s) used the concept of the 'first excellent print'. Each of a range of 'typical' subjects was photographed across a wide range of exposures, and the best possible print was made from each negative.


Waterfall, Julian Alps, Slovenia

This picture has a long tonal range, and there is important detail in both the highlights (the waterfall and the wet rock) and the shadows (the cave and the foliage).

Contrary to popular belief, there is next to no danger of 'blowing' the highlights through over-exposure of the negative. As long as there is adequate detail in the dark areas, the lighter areas can usually be brought in by use of a softer grade of paper or by dodging and burning (paid-for module).

Exposure determination was simply by spot-metering the darkest area where Roger wanted texture and detail: he checked both the cave mouth and the foliage, to see which was darker. Tripod-mounted Nikon F, 200/3 Vivitar Series 1 with 2.8x Soviet-era orange filter, Ilford HP5 Plus at 500 in Ilford DD-X, Ilford Multigrade Warmtone.

choosing the first excellent print

A wide range of people were asked to choose the 'first excellent print' from each series of 'typical' subjects. With the early prints in a series, each increase in negative exposure made for an obviously better print. Then, fairly suddenly, the quality would stabilize and then remain good for quite a while until it began to deteriorate again. The point at which the quality stabilized was taken as the 'first excellent print', corresponding to the minimum desirable exposure. This in turn formed the basis for Kodak film speeds, which later became first ASA and then ISO(free module).

The trouble is, the original 'first excellent print' research was done with 5x7 inch contact prints, uncoated lenses and an audience that did not consist entirely of keen photographers with a full awareness of the maximum possibilities of the photographic process: Kodak was quite understandably after the mass market. The result of this was that the 'first excellent print' may not always have been quite as excellent as the one made from the next negative that had received more exposure again. To this day, some photographers prefer to exposure their negatives a little more than others do: usually, anything from a third of a stop to a stop more.

keeping exposure to the minimum

Keeping exposure to the minimum is desirable for several reasons. It allows a faster shutter speed (better action freezing, less camera shake) or a smaller aperture (more depth of field) or both. It makes for a less dense negative, which prints faster: obviously, a dense negative requires a longer exposure in the enlarger and some scanners may be unable to penetrate dense negatives. It gives finer grain with conventional black and white films, though colour negative films and chromogenic films actually give finer grain with over-exposure. Finally, the negative is sharper: a lens-film combination that resolves 90 line pairs per millimetre (lp/mm) at the minimum desirable exposure may drop to 75 lp/mm at a stop more and 60 lp/mm at two stops more. This is true even with colour negative and chromogenics.

Exposure is therefore a compromise which depends on personal preference and priorities, and this is one of the reasons why manufacturers always urge their customers to establish personal film speeds. Other reasons include variations in equipment (meters and shutter speeds), and variations in metering technique and development technique (including choice of developer, which can affect the true ISO speed of the film). It is quite normal for different photographers to rate the same film at anything from one-half or even one-quarter of its ISO speed to twice its ISO speed or more. This would mean that an ISO 400 film might be rated at anything from 100 to 800.

Xaghra Mill, Gozo

Even when you are photographing something like this old windmill, the exterior of which can be seen in the gallery, you may need to keep exposure to a minimum. Sure, the camera is on a tripod -- but you may only have a few seconds before you are in someone's way, and besides, the less time the shutter is open, the less time there is for someone to kick the tripod leg. Voigtländer Bessa-R, 12/5.6 Ultra-Wide-Heliar, Fuji Acros at 64 in Ilford DD-X, Ilford Multigrade Warmtone. This picture also appears in Rangefinder.

personalizing film speeds

The only way to get the film speed that best suits you, for any given film, is by personal test with your equipment, your subject matter, and your preferences. ISO film speeds are only a starting point, but they are the best starting point we have. If you really understand sensitometry and know how developers work, you may be able to devise a better starting point, but even then you will need to refine the actual speed you need via personal testing. There are three ways to do this.

sensitometric testing

The first, and usually the least satisfactory, is by shooting uniform targets -- a large sheet of white paper will do -- at varying exposures at known illumination levels; reading the negative densities with a densitometer; and plotting a graph of negative density against the logarithm of the exposure (D/log E curve, H&D curve, characteristic curve). This will tell you about minimum usable densities and contrast.

the zone system

Closely related to this is the second approach, the Zone System, which is a simplified version of the same thing with much of the testing reduced to rote. There are plenty of books on the Zone System, so there is little point in going into it here. Suffice it to say that there are at least as many good photographers who do not use the Zone System as do, so it cannot be essential. There is more about why we do not use or recommend it in a free module.

hurter and driffield

The basic work on sensitometry was done by Ferdinand Hurter (1844-1898) and Vero Charles Driffield (1848-1915). Their ground-breaking paper introduced the characteristic or d/log E curve, also known as the H&D curve after its inventors. This was published in 1890. Apart from the naming of Zones (which is a work of genius -- see the free module), the Zone System added nothing to Hurter and Driffield's work. It certainly does not begin to compare with the First Excellent Print research carried out by Kodak in the 1930s.


Back roads, Gozo.

A perennial problem with the Web is the conflict between speed of download and detail. In the original print, though sadly not in this web version, there is tone and texture in the white lighthouse on the upper third and in the dark foliage around the white flowers in the foreground. In Zonespeak these are Zone VII (lightest area with texture and detail) and Zone III (darkest area with texture and detail). This is using the old, symmetrical 9-Zone system rather than the later asymmetrical 10-Zone or the ultimate 11-Zone which is normally only possible with contact printing. Voigtländer Bessa-R, 35/1.7 Ultron, 2x yellow B+W filter, Ilford HP5 Plus at 500 in Ilford DD-X on Ilford Multigrade.

real-world iterative testing

After full sensitometric testing and the Zone System, the third approach is simple iterative testing with real subjects. We very strongly recommend this approach as both easiest and best.

Ideally, it is done in two stages. The first stage -- as little as one roll should be enough -- consists of shooting mundane subjects that are more or less the sort of thing you normally shoot, but without trying to get Fine Pictures. This gives you your basic exposure and development time, and if the pictures are too contrasty or too flat or badly exposed or tonally nasty, you haven't lost any good images.

For the first roll, set your meter to the ISO speed; use your meter to give you a 'best guess' at the xposure (see below, Metering Technique'); and make three exposures of each subject, at the best guess and at one stop more and less. Develop as recommended by the film or developer manufacturer, or at your best guess; look at it to see which exposures give adequate shadow detail; print the ones that look best.

If you are happy with the first roll, you can go straight to the second stage, which is taking real pictures. Otherwise, shoot another roll, with a revised ISO speed as necessary, and change the development time to get more or less contrast (there is a paid module on developing black and white films). If there is plenty of shadow detail at the ISO rating, but only the faintest traces at the one-stop-under, by all means split the difference. Let's say it's ISO 400: you could try EI 500 or 650. Or if you prefer the tonality at a stop over, you could use EI 250 as your starting point. Bracket again, but this time by 1/2 stop or 2/3 stop (1/3 stop isn't worth bothering with when you are shooting negatives). Again, develop and print.

By now you should have a pretty good idea of the optimum exposure index (EI) and development time, but if you haven't, repeat the process until you are happy. You might also want to have a look at the paid module on choosing films: we firmly believe that there is no point in persisting with a film or developer that is giving you difficulty.

Then you can start shooting real pictures, using the EI and development time you have established. Do not regard either as sacred. If over the course of the next few films you find that you could do with a bit more exposure or a bit less contrast, change the EI or development time or both. This goes against the strict scientific method, where you change only one thing at a time, but to be honest, photography ain't that scientific, and you can aim off on the basis of experience: a little more development may add as much as 1/3 or even 1/2 stop, but a little less development will lose speed a lot faster.

Pumpkins, Slovenia.

On this trip we tried shooting HP5 at both 200 (in Ilford Perceptol, for finer grain) and 500 (in Ilford DD-X, for more speed). The simple truth was that it was more trouble than it was worth. Tonality was excellent in both cases, and we don't mind the extra grain from the DD-X. You may feel differently, of course. This one ended up on 200 (which it should have) but the waterfall at the beginning of this module ended up on 500. It is however worth knowing that a slow film in a speed-increasing developer will almost invariably be finer grained than a fast film in a fine-grain developer, and the speeds may not be all that different: Ilford FP4 in Ilford DD-X is about ISO 200, and HP5 in Perceptol is about 250. Nikon F, 200/3 Vivitar Series 1, Soviet-era 2.8x orange filter, Ilford Multigrade Warmtone toned with Fotospeed selenium.

exposure and format

Some photographers use different EIs and development times for different formats. For that matter, some manufacturers recommend this. They do this less often than they used to because they grew tired of explaining why, though the reasons are simple enough.

First, increased exposure means bigger grain and less sharpness. With 35mm, this is far more significant than with roll-film or cut film. Keeping exposure to the bare minimum consistent with good tonality is much more important. With larger formats, there is much less of a penalty for half a stop or so of extra exposure, so you can afford a bigger safety margin against under-exposure. Also, many feel it gives better tonality.

Second, it was historically more usual to print 35mm with condenser or condenser-diffuser enlargers, and larger formats with condenser-diffuser or pure diffuser enlargers. The optimum contrast for the three types differs, with condenser enlargers requiring the lowest contrast and pure diffuser enlargers requiring the most. It therefore made sense to recommend a longer development time (= higher contrast) with the larger formats. The same is true of contact prints, especially with 'alternative' processes which often require high contrast negatives.

Third -- though this is rare -- there are a few films where the actual emulsion formulation differs from one film to another, even with the same name.

Minnis Bay Stores

On 8x10 inch (here, a Gandolfi Precision with 110/5.6 Schneider Symmar XL) doubling the exposure has no real disadvantages: you just get more and more shadow detail. The limit to the exposure was set by the time the owner could keep still. From memory the exposure on Ilford HP5 Plus was two minutes at f/16.

tonality versus minimum exposure

At the risk of repetition, exposure is a matter of priorities: minimum exposure (with all its desirable characteristics) versus good tonality.

Fine grain is the easiest of all film characteristics to see and compare, and for many unsophisticated users it is the be-all and end-all. Some manufacturers shamelessly cash in on this, making films that are very fine grained but tonally awful.

Sharpness and resolution are easy to compare too, though they are not the same thing. Surprisingly, nor are fine grain and sharpness. Chromogenic films, for example, are much finer grained than conventional films of the same speed, but they are also less sharp.

As noted above, with conventional black and white negative films, the less the exposure, the finer the grain, though with chromogenics and colour films, increasing the exposure will give finer grain. With any film, the less the exposure, the higher the sharpness and the higher the resolution. This holds true even into the realms where tonality -- the hardest of all film qualities to quantify -- suffers grievously.

This is why many photographers, including ourselves, habitually give slightly more exposure than is sufficient to give the amount of shadow detail required by ISO standards: we prefer the tonality. How much extra exposure we give depends on the metering technique, the subject and the lighting.


Sunshade, South of France.


Frances used her Alpa 12 S/WA with a 35/5.6 Rodenstock Apo-Grandagon on 6x9cm -- the equivalent of a 15mm shift lens on 35mm. The film was Ilford HP5 Plus, which has a true ISO speed of around 500 to 650 in the developer we used, Ilford DD-X, but she set a film speed of only 400, ensuring over-exposure by 1/3 to 2/3 stop.

The meter reading, with a Pentax Digital spot meter, was taken from the shadowed area on the lower left. This is not the darkest area in the picture, because it has been held back (dodged) in printing, but it was the darkest area in the subject.

She made a correction for the known tendency of this lens and shutter to give us negatives that are somewhat less exposed than we like, and added 1 stop to the reading. We don't know why we have to do this with this lens (alone of all our Alpa lenses) but we do -- an important illustration of matching film speeds to the equipment.

The print is on Ilford Multigrade Warmtone, selenium toned. This picture is the frontispiece of Rangefinder.

meter types

There are essentially four types of meter: spot meters, broad area reflected light meters (integrating meters), incident light meters (artificial highlight meters), through-lens meters. Each is dealt with in turn below, but before we do so, it is worth remarking on an extraordinary fact which we also note elsewhere.

Compare any two meters and you will quite likely find discrepancies of 1/3 stop or more in the readings they give. Now give the same meter to two different photographers in turn and ask them to read a series of subjects. Once again, you will quite likely find discrepancies of 1/3 stop or more even when they meter exactly the same subject. Part of this is down to where they point the meter, and part (with many meters) depends on how they interpret the scales: one may round up while the other rounds down.

And yet, most people get good exposures with most meters. Latitude helps, no doubt, but there is something subtler at work, a series of what must be subconscious adjustments. We are indebted to Garry Coward-Williams, editor of Amateur Photographer, for this intriguing thought.

spot meters

Anything other than a spot meter reading is a compromise. This is because the most important criterion for negative exposure, other than tonality, is adequate shadow detail. The easiest way to ensure adequate shadow detail is to measure the brightness of the darkest area in the subject in which you want detail and texture, and base your exposure on that. Hence, logically, a spot meter is the very best way to meter negatives. And what you need to read is the darkest area in which you want detail and texture.

Left, a Pentax Digital spotmeter -- the one that Frances uses all the time. Right, an SEI Photometer, the earliest professional-quality spot meter and still (in the eyes of some) the best. The Pentax reads one degree; the SEI, half a degree or less because of its comparison spot design. The SEI is much slower to use, even though this is the updated LED version, but it can read a brightness range of 1,000,000:1 and focuses down to a hand-span.

angle of view

A spot meter, with its very narrow angle of view (usually just 1 degree), allows you to read even small areas accurately and had two further advantages. First, you can take your readings at a distance: useful if you cannot get closer or do not want to get closer (think about tigers). Second, it corrects automatically for the haziness and loss of contrast that comes with distance: you will not get the same reading if you walk up to a tree and read the brightness of its trunk as you will get if you measure it from a hundred yards away.

shadow detail

Because your exposure is based on the same criteria as the ISO standards for film speed -- shadow detail -- a spot meter reading often allows use of the full ISO speed with absolute confidence that there will be no under-exposure, and it requires the minimum of interpretation.

expense and inconvenience

On the other hand, spot meters are expensive to buy and inconvenient to use, and it is perfectly possible to get perfectly adequate readings without one. If you do not use a spot meter, you just have to apply more rules of thumb and take more care to err on the side of over-exposure.

metering technique with spot meters

Taking a spot meter reading of a grey card is not the answer, because the only way to discover how much darker the deepest shadow is than the grey card is to measure the shadow directly. In a flatly lit scene, the shadows might only be a stop or two darker than the grey card; in a very contrasty scene, they might be five stops or more. In the former case, reading the grey card will recommend overexposure, which won't matter too much, but in the latter case it will recommend quite significant underexposure, which may matter a lot.

There may be a little to be said for reading a grey card as well, because a shadow reading of a scene with a very short brightness range (such as a landscape on a misty day) may result in unpleasant tonality, but quite honestly you will get around that if you simply to give a stop more than the shadow reading indicates.

Chateau, near Loches.

For 'open' shadows on a sunny day, exposure must be generous: it would have been very easy to have pools of impenetrable blackness on the shadowed side of these walls, while still retaining entirely credible tonality in the sky, buildings and foliage. A spot meter reading of the dark foliage beside the chateau is best but a through-lens reading of a shaded area, such as the dark side of the wall on the right, will be entirely adequate. Frances used her Voigtländer Bessa-T with a 21/4 Color-Skopar and 2x yellow B+W filter. Ilford XP2 printed on Ilford Multigrade Warmtone.

reading the darkest and lightest areas

For maximum control you should read both the darkest area in which you want texture and detail and the lightest area in which you want texture and detail.

You then base the exposure on the shadow reading and the development time on the highlight reading: for a short brightness range you give increased development to increase the contrast, and for a long brightness range you give reduced development to compress the contrast as described in the free module on brightness ranges or (in more detail) in the paid-for module on negative development. But unless you are developing each exposure separately, or all the exposures on a single roll have the same brightness range, this is not feasible: you have to resort to 'average' or 'standard' or 'normal' development.

Dog and barn, Slovenia.

You cannot really see it in this web image, but in the original print there is texture and detail in the wall behind the dog, inside the barn. Exposure determination was by a spot meter reading of that wall, the darkest area in which Roger wanted texture and detail. Nikon F, 200/3 Vivitar Series 1 with Soviet-era 2.8x orange filter, Ilford HP5 Plus at 500 in Ilford DD-X.

At this point, the less well-informed devotees of the Zone System will say, "Aha! This is the Zone System after all!" The truly ignorant will add, "They are using the Zone System and don't know it." Well, actually, we are using the same basic sensitometry of which the Zone System is a sub-set. In our view the Zone System both over-simplifies basic sensitometry and over-complicates it. It does both in much the same way, over-complicating it by omitting a good deal of the science on which it is based and over-simplifying it by reducing much of the rest to a series of more or less rote tests. This is not to say that the Zone System won't work. If you like the Zone System, don't let us discourage you. Just don't imagine it is the only path, or the best, or the original.

spot meter indices

The first commercially successful spot meter, the SEI Photometer, still commands enormous respect today, and it had only two indices: shadow and highlight. The shadow index was for use with negative films, where exposures are keyed to the shadows, and the highlight index was for use with transparency films, where exposures are keyed to the highlights. There is more about 'keying' exposures in the free module on ISO speeds.

Modern spot meters normally add a third 'mid tone' index, keyed to an 18 per cent grey card. This has the advantage under controlled lighting that you can reproduce a known tone at a known value, while adjusting the lighting so that both the highlights and the shadows fall within the brightness range that the film or digital sensor can handle.

the i.r.e. scale

Many modern spot meters have an I.R.E. (Institute of Radio Engineers) scale from 1 to 10. I.R.E. 1 is the shadow index, 2-1/3 stops below the grey card index, and I.R.E. 10 is the highlight index, 2-2/3 stops above the grey card index. The total brightness range is therefore 5 stops, 64:1, which can be accommodated in the dynamic range of many reproduction media.

Unless you are in full control of the lighting, the shadow index (I.R.E. 1) is the most important. It is the only one that guarantees adequate shadow detail. As we have already said, you never know how much darker than a grey card your shadows will be.

Some other meters omit the full I.R.E. scale and provide just shadow, mid-tone and highlight indices, though in all the ones we have seen, shadow and highlight correspond to I.R.E. 1 and 10 respectively.

In all cases, the metering technique is identical. We know we have said this before but we will say it again. You read the darkest area in which you want detail and texture, and read off your exposure accordingly against the shadow index. You do NOT use the mid-tone index, because if you did, the meter would recommend 2-2/3 stops of overexposure. Nor do you try to find a mid-tone and read that. Judging a mid-tone is extremely difficult, and reading it is usually pointless anyway.


The I.R.E. (Institute of Radio Engineers) scale can be clearly seen on this Pentax spot meter. It is the lowest of the scales: from the top, ASA (ISO), shutter speed, aperture, EV (exposure value, in very dark red) . The meter scale, visible through the eyepiece, reads in EVs.


I.R.E. 1, on the left, is the shadow index; I.R.E. 10, on the the right, is the highlight index. The big red triangle is the mid-grey index -- substantially useless to anyone who understands what a spot meter is for. Here EV 8+2/3 is alongside I.R.E. 1.

if you don't have a shadow index

If you do not have a shadow index, you can do one of two things. You can give 2-2/3 stops less exposure than indicated by the mid-tone index; you can re-set the film speed 2-2/3 stops faster (100 becomes 650, 400 becomes 2500); or (if the meter allows it) you can dial in 2-2/3 stops compensation.

In practice, if you wish, you can use some other value than 2-2/3 stops. With just 2 stops you will always have plenty of shadow detail; with 3 stops, with most meters, you will have the maximum safe speed of the film. Or you can use 2-1/2 stops and still have a good safety margin.


colour negatives

The rules for colour negatives are exactly the same as the rules for black and white negatives. The big difference is that a colour print looks strange much sooner than a black and white print if you do too much dodging or burning.


Tree Roots

With a 6x7cm negative -- Frances used a Linhof Super Technika IV with a 105/3.5 Schneider Xenar -- you don't have to worry very much about sharpness and grain. In fact, if you expose colour negative (and chromogenic films) generously, the grain will actually be smaller, though there will still be a loss of sharpness. The ideal thing to meter here would be the shadows between the tree roots. We have forgotten which film it was except that it was Kodak.

spot meters and film speeds

As already noted, you can often use the ISO film speed without worrying about under-exposure. There are two main exceptions to this rule.

The first is if you habitually under-develop your film as compared with the ISO standard contrast. If you normally develop to a gamma of 0.56 instead of 0.62, you may well find that you need to rate the film at 1/2 stop or more below the ISO speed.

The second is if you prefer the tonality you get with a little extra exposure. We do. We often use an EI 1/3 or even 2/3 stop less than the actual ISO. With any other metering technique, 1/3 stop is within experimental error; with spot metering, it is a meaningful number.

In practice, though, the developers we normally use give about 1/3 to 2/3 stop of true ISO speed increase with the films we normally use. For example, Ilford HP5 Plus in Ilford DD-X is ISO 650. We can therefore set EI 500 with confidence, and we often use ISO 400. We cannot over-stress though that other developers (especially fine grain developers and Rodinal) may well lose speed. If we develop the HP5 Plus in Perceptol, the true ISO is at most 320, so we set 200 on the meter.

When you first buy a spot meter, or when you are trying a new film, it is a good idea to shoot your first test roll with brackets at +/- 2/3 stop: few meters allow 1/2 stop, and 1/3 stop is hardly worth bothering with. This is more accuracy than we recommend with other meters, but then, you are metering more accurately.

interpreting spot meter readings

As already noted, spot meter readings don't require interpretation in the same way as readings taken with other meters. Once you have established your working film speed, you will always get adequate shadow detail. With subjects having a very short tonal range, however, we would recommend on the basis of experience giving a stop more exposure in the interests of tonality.


Barn, La Chatre, France.

There is no problem of interpretation with spot metering. You simply look for the darkest area in which you want texture and detail, and meter it. In this case, it was a shadow area -- not the deepest -- on the far wall. A trick worth remembering, though, is that the darkest parts of most pictures are towards the bottom of the frame. You may be deceived by contrasts which make you think that something nearer the top is dark, but usually, if you meter it, you will find that the darkest part is nearer the bottom of the picture than the top. Frances used a Voigtländer Bessa-T with 50/1.5 Nokton with 2x yellow B+W filter for this picture, shooting on Ilford XP2.

broad area reflected light meters

Broad area reflected light meters measure the light reflected from a broad area and integrate it. Classic examples include the old Weston Master or the Gossen Lunasix. The measuring angle of such meters is typically around 30 degrees, though it can be as wide as 60 degrees or as narrow as 15 degrees.

Such meters are a great deal better than no meter at all, but they do require a certain degree of interpretation and finesse because they do not measure the shadows directly in the same way as a spot meter. It is quite possible to 'favour' the shadows, pointing the meter towards a dark part of the subject rather than a bright part, but it is still an approximation.

After this, it is a matter of common sense and experience. For example, most users of such meters angle the cell slightly towards the ground when taking a reading, so that the sky does not have too much influence. Too much sky would of course introduce a tendency to under-exposure. Because different photographers have different ideas about exactly how much to angle the meter, they are also quite likely to have different ideas about the optimum film speed.

Likewise, an unusually light subject (a snow scene, for example) will normally be under-exposed if the user does not interpret the results somewhat, while an unusually dark scene (a black cat in a coal cellar) will often be over-exposed for similar reasons: the meter tries to interpret both as 'average' scenes, reflecting about 12 to 14 per cent of the light falling on them (and not 16 per cent as most people believe).

interpreting broad area readings

With experience, a broad area reflected light meter can be as good as any other, but until you have enough experience to use it with absolute confidence, it is as well always to err on the side of over-exposure: the penalties for this are much less than the penalties for under-exposure. An easy way to ensure that you err on the side of over-exposure is to set the film speed somewhat lower than the ISO speed, as described in the table below. Even then, you will do well to add a couple of stops for unusually bright subjects, such as snow scenes; for backlit subjects; and for subjects with a long brightness range, such as very sunny days where there are deep shadows in which you want texture and detail.

When using a broad area reflected-light meter, it is a good idea to angle it down slightly -- maybe 10 to 15 degrees -- so that it is not unduly influenced by reading too much sky.

Very sunny days: set film speed 1 stop lower e.g. EI 200 for ISO 400 or 50 for 100

Hazy sun: set film speed 2/3 stops lower e.g. EI 250 for 400 or 64 for 100

Cloudy bright (clear shadows): set film stop 1/3 stop lower, e.g. EI 320 for 400, 80 for 100

Dull days: use the full film speed and consider increasing development time

If all this seems too precise, just give 1/2 stop to 1 stop more than the meter indicates (or set the meter 2 stop lower) and make the adjustments by guesswork according to your assessment of the conditions.

Snow, Pyrenees

Frances used a Kodak chromogenic film for this shot, with a red filter. She used a spot meter but if you were using a broad-area reflected-light meter (including through-lens meters) you would need to give between two and three stops more than indicated by a reading off the snow. Alternatively you could take a reading straight off the blue sky, which is a remarkably reliable mid-tone -- though even then, an extra half stop would probably do no harm. Voigtländer Bessa-T, 28/1.9 Voigtländer Ultron, Kodak chromogenic film, B+W 8x red filter, on Ilford Multigrade Warmtone with a touch of selenium toner.

incident light meters

An incident light meter measures the light falling on the subject and therefore removes at a stroke the need to compensate for subjects of unusually high or unusually low reflectivity. The old name for incident light meters gives the game away: they used to be called 'artificial highlight' meters, the highlight in question being the white sheet or dome that covers the meter cell when the meter is used in incident mode.

The big problem with incident light meters is exactly the same as the problem with reading grey cards: you are not measuring the shadows directly, and you do not know how dark they are. If they are 5 stops or less darker than the artificial highlight, no problem. If they are up to 7 stops darker, you should still get printable shadow detail, but the tonality may not be as good. And if they are more than about 7 stops darker, you will have no shadow detail at all: your negative will be under-exposed. Incident light meters are perfect for transparencies and digital sensors, where exposures are keyed to the highlights, but they are less than ideal for negatives.



An incident light reading is best taken from the subject position, with the incident light dome pointed towards the camera. If you cannot stand next to the subject -- think of politicians, or tigers, or other unapproachable creatures, or the middle of Niagara Falls -- you can often take a reading under equivalent lighting, that is, where the light falling on the meter is the same as the light falling on the subject. It is usually all right to angle the meter slightly upwards, like this, but it is better to hold it parallel with the ground.

interpreting incident-light readings

The main corrections you need to make are for holding texture in small areas of unusually low reflectivity, and for very long brightness ranges. If there is dark fur in the picture, for example, you may be well advised to give anything up to a stop of extra exposure (fur 'eats' light) and on a bright sunny day with small areas of deep shadow in which you want detail you should give about the same increase.

incident light readings at night

At night, brightness ranges are often very long indeed, and you usually have to sacrifice some shadow detail. An incident light reading is often the best you can do, but it is by no means unusual for the meter to lack the sensitivity that is needed. A useful trick in this case is to take an incident light reading from the subject position, but pointing the bare meter cell (no incident light diffuser or Invercone) at the most important light source illuminating the subject. Give 5x the indicated exposure -- in other words, add two stops, and round to the nearest value above that. This will normally ensure adequate exposure.



Bicycles, Paris.

Roger finds bicycles hard to resist as a photographic subject, even at night: "I'll tell the truth: I didn't meter this exposure. Instead, I guessed. A very wide range of exposures is normally acceptable in night scenes, as long as you don't under-expose, so a good bet is to work at full aperture (or close to it) and at the longest exposure you reckon you can safely hand hold. I've forgotten the exposure but it was probably about 1/30 second at f/1.7." But the trick with the bare meter cell, described above, would have been a good starting point. Bessa-R, 35/1.7 Ultron, Ilford HP5 Plus at 500 in Ilford DD-X, printed on Ilford Multigrade Warmtone.

through-lens meters

In a sense, these do not deserve a heading to themselves as all are merely variations on broad-area reflected-light meters: the so-called 'spot' option is normally far wider than the 1 degree of a true spot meter. Where they differ from other broad-area meters is that they sample the metering area and accord it different weights, whether this is as simple as the white spot on the blind of a Leica MP or as complex as the multi-point meters for which Nikon and Canon are so well known.

With all through-lens meters, however, it is important to realize three things. The first is that while each advance does indeed make it likelier that they will achieve a very good exposure for slide film or a digital sensor, 'a very good exposure' is not the same thing as 'the best possible exposure'. The second is that even the cleverest meter can be fooled: all that advances do is reduce the number of occasions on which the photographer actually has to think. And the third is that they are optimized for slides and digital sensors, which is to say, the exposure is keyed to the highlights. With the multi-point meters would be very easy to install a program which gave optimum exposure for negative films, using exactly the same data, and it would probably be even more reliable than the usual programs, but demand must be tiny, and the incentive to provide such a program is more than cancelled out by the likelihood that people would choose the wrong program and then blame the camera.


A subject like this can be surprisingly hard to meter. Overall, the total amount of light reflected is likely to be greater than the 12 to 14 per cent that the meter assumes -- and because there are relatively few dark areas, mostly well towards the edge, even the cleverest metering system could easily be fooled into recommending over-exposure. The yellow of the stone might also influence some meters more than others. Our 'best guess' on the basis of experience would be to give at least 1/2 stop more than recommended by a through-lens meter and preferably 1 stop, though 2/3 stop should suffice. Frances used a Voigtländer Bessa T with a 50/1.5 Nokton, shooting on Ilford XP2.

interpreting through-lens readings

Because through-lens meters are essentially broad-area reflected light meters, they are subject to the same constraints and must be interpreted in the same way. The cleverer they are, the harder it is to know exactly which corrections are called for, and when, and how big. Out inclination, once again, is to set an EI that is about 2/3 stop slower than the ISO speed, and to whack in substantially the same corrections as for any other broad-area reflected-light meter. In this way, we can be reasonably confident that any errors we make will be in the direction of overexposure, which is the way we like it.

guesswork and deliberate over-exposure

The vast majority of subjects can be well exposed via simple reliance on almost any meter; or by reference to an exposure table of the sort that used to adorn the inside of film boxes; or indeed by guesswork. We quite often find ourselves setting an exposure, then checking the meter almost as an afterthought. If it agrees, all well and good; if it does not, we ask ourselves whether we were mistaken, and if so why, or whether our guess is not likely to be closer to the optimum exposure than the meter's offering. From the above, you should be able to see where meters can go wrong, and why you might want to use settings other than those recommended by the meter.

Cat's eye view.

Frances saw this as she imagined a cat would see it. She didn't have a spot meter with her so she simply guessed the exposure, using the 'sunny 11' rule (below) and then opening up another stop to be absolutely sure of adequate detail on the shaded side of the tree. A through-lens reading, favouring the shadows, confirmed the exposure. Voigtländer Bessa-T, 28/1.9 Voigtländer Ultron, Ilford XP2 printed on Ilford Multigrade Warmtone.

the sunny 16 rule

An oft-quoted and simple rule of thumb is that on a sunny day you can set the aperture to f/16 and set the shutter to the reciprocal of the IS0 speed. This will work very well with slide film with the sun behind your shoulder, but with negative film you will often do better to use a 'sunny 11' rule instead in order to get good shadow detail.

For hazy sun, open up one stop (f/8); for 'cloudy bright' (things still cast shadows) open up two stops (f/5.6); and for dull, overcast days open up three to five stops (f/4, f/2.8, f/2). Or, of course, take the hit on shutter speed: instead of 1/125 at f/5.6 on a cloudy bright day, go for 1/60 at f/8.

For side lighting, open up a stop in full or hazy sun, and half a stop on a 'cloudy bright' day. On a dull day, with no clear shadows, there is no such thing as side lighting...

For back lighting, open up two stops in full or hazy sun and one stop on a 'cloudy bright' day.


don't panic

It is possible to worry much more about lighting than you need too: after all, it is only when subjects get really tricky that serious, heavy-duty metering is required. On the other hand, it is worth remembering that Ansel Adams allegedly reckoned that when he acquired a spot meter, his exposures increased on average by a full stop. If any proof were needed of the desirability of erring always on the side of over-exposure, this must surely be it.



His Holiness Dalai Lama, Lhasa Uprising Day (March 10th), Dharamasala.

This was taken in the early 80s and the exposure is lousy. In the original print you can just about differentiate His Holiness's shoulders from the background, but after JPEG compression and downsizing for the Web you have to look hard. Another stop of exposure would have been ideal -- and easy with a spot meter. Nikon F, 200/3 Vivitar Series 1, Ilford HP5, further details long forgotten.


the bottom line: a rule of thumb

Can it all be reduced to a simple rule of thumb for when you don't have a sophisticated meter, or you do have one but you don't have time to use it?

Yes. The rule of thumb is easy enough, and it is this:

Make your best guess at the optimum exposure and then make two exposures, one at the best guess and one at two stops over. At worst, you may lose a bit of sharpness and incur some extra grain in the shot that is two stops over, but you should have a printable negative.



Painter, Honfleur

Frances shot this on Ilford XP2 with a 'baby' 35mm Rollei which did not have a functioning meter -- so she guessed, erring on the side of generosity. Ilford's chromogenic films (XP1, XP2, XP2 Super) are all extremely forgiving of over-exposure -- but then, so is Ilford HP5, Roger's favourite ISO 400 film.




If you have found this module informative, you may wish to look at the following books:

Black and White Handbook

Lighting for Photographers

Perfect Exposure

Quality in Photography

Go to the list of modules

or go to the home page

or support the site with a small donation.

© 2006 Roger W. Hicks