subject brightness range

Some parts of a scene are brighter than others. This is no surprise. But to get the best possible pictures, we need to know two things. First, how much brighter are the brightest parts than the darkest parts? Second, how do we use this information to get the best possible picture? What follows is concerned principally with black and white photography: colour is covered fairly briefly at the end.

the range of ranges

On a misty day, the difference in brightness between the lightest part of a scene and the darkest could be 4:1 or less. In other words, the lightest areas might only be four times as bright as the darkest. It is difficult to compose a picture with a brightness range that is much lower.

At the other extreme, on a brilliantly sunny day with deep shadows -- caves on a beach, for example -- and the sun in shot, the brightness range could easily be 1,000,000:1.

In practice, an 'average' outdoor brightness range is taken to be something around 128:1 (log brightness range 2.1). This often disregards tiny areas of very dark shadow, which will 'block up' to a featureless black, and equally tiny areas of very bright highlight, which will 'blow' to a featureless white in the final image.

This 'average' is based on research done around Rochester, New York in the late 1930s and early 1940s and has never been bettered, but it needs to be regarded with some suspicion.

 

 

Tree and sky

This is a pretty 'average' sort of subject, except that a snapshot might have been expected to have people in it. Frances shot it on Ilford XP1 (yes, it's that old) with an 8x red filter. In fact she slightly under-exposed, but fortunately there's no important shadow detail and the tree merely looks stark. Nikkormat and 90/2.5 Vivitar Series 1 Macro.

the importance of geography

The further south you go, the greater the average brightness range. In the California desert or the Greek islands, or Southern Japan, for that matter, the 'average' range could easily be twice that around Rochester, at 256:1. Go north, on the other hand, and the average range gets smaller. In London, which is as far north as Edmonton, Alberta, the 'average' range across the year might be half that of Rochester, at 64:1.

This means that if you apply the same exposure criteria in all three places, and develop your films in the same way, the pictures taken in London are quite likely to look fairly flat and dull, while the ones taken in Greece may well look overly contrasty.

Railway viaduct over canal, Manchester

Manchester is a long way north of London, and London is pretty far north in its own right. Even on a sunny day, therefore, you are likely to need to develop your films for a bit longer than if you live and shoot in California. It might even make sense, other things being equal, to develop your films for a little longer in winter than in summer -- though few if any photographers will go that far.

the importance of subject

Imagine a bride and groom on a sunny day. The bride, a fair-skinned blonde, is a traditionalist, wearing white to symbolize innocence and purity. The groom, being both a traditionalist and a realist, is wearing dark clothes, including black trousers. The difference in reflectivity between a white satin dress and black wool trousers can reach 7 stops -- 128:1 -- even if the fabrics are evenly illuminated: if the trousers are in shadow, it can exceed 8 stops, 256:1.

Next, imagine that same bride on her honeymoon, on the beach. She is temporarily wearing a light-coloured swimsuit. The maximum brightness range, from skin highlights to the shadow she casts on the sand, is unlikely to exceed 6 stops or 64:1. Without the shadow, it may well be as little as 4 stops or 16:1, or even less.

Finally, overcome by the heat and aware of the dangers of too much sunbathing, she goes into the shady hotel room, strips off and sprawls on the bed. Assuming she has not caught the sun too badly, the total brightness range might be around 2 stops or 4:1.

 

Louise

 

One scenario we omitted from the above story is the bride dressing -- a scene which will, if she is in her bedroom lit by a window like this, be pretty contrasty. Frances used a fast lens (50/1.5 Nokton) on a Voigtlander Bessa R2 and shot on Ilford XP2 Super, famous for its ability to capture a long tonal range. Then she printed on grade 2 paper.

'double' lighting

Now imagine the church where they were married. She is also a keen photographer, and wants to photograph it when they get back. She's lucky: it's another sunny day. Or is she lucky? The sun is streaming through the windows, but where it strikes the floor the flagstones are a full 7 stops brighter than they are even a pace or two away from the sunlit spot, and a bit over 9 stops brighter than in the shadowed parts of the church: that's 128:1 and over 500:1. The darkest areas of the hammer-beam roof are no less than 11 stops darker than the sunlit floor, a brightness range of 2000:1 or thereabouts. This is 'double' lighting, where the brightest parts of the shaded areas may well be darker than the darkest parts of the well-lit areas. These readings, incidentally, were verified in the 11th century church in the village in which we live.

Chapel, Sumeg Castle, Hungary

This is almost as nasty a piece of double lighting as you could hope not to see: it could only be worse in brilliant sunlight. There is important detail in areas that are window-lit, and in areas that are all but pitch dark. To make matters worse, the dark areas were so dark that they could not even be measured with a spot meter. Roger guessed at the exposure after taking a reading in the darkest area where his Gossen Spotmaster would give a reading, but still didn't give enough. There is a great deal of dodging and burning (paid module) in this picture -- but it was still, ultimately, under-exposed. Alpa 12 WA, 38/4.5 Zeiss Biogon, 66x44mm format, Ilford HP5 Plus.

'average' reflectivity

Another important point concerns 'average' reflectivity. The same Kodak researchers who discovered 'average' brightness ranges also discovered that an 'average' scene reflects 12 to 14 per cent of the light falling on it (not 18 per cent as is widely believed -- see the short free module The 18 per cent grey card). This is remarkably useful and consistent for outdoor photography, but even outdoors there are plenty of scenes that do not conform: snow, for example, can reflect 90 per cent, a beach scene may average 50 per cent or more, and woodland may be 5 per cent or less. Indoors, especially under 'double' lighting as described above, any 'average' may well be completely useless.

Aosta

Stone (and stucco) can be mightily deceptive. Light stone can reflect far more light than 'average', but honey-coloured stone that looks quite light to the eye can come very close to a mid-tone. Also, a great deal depends on where the main interest is. Here, it is clearly in the buildings: dark clothes and dark hair are almost lost against the dark arches (though they read better in the original print) but this actually helps preserve the bright, sunny, contrasty mood. Frances shot this on XP2 with a Nikon F and a 70-210/2.8 Sigma Apo.

print brightness range

The darkest black achievable with a top-flight black and white paper, optimally processed, reflects about 1/200 as much light as the pure, paper-base white: a brightness range of 200:1. Few prints will achieve this: a more realistic brightness range is between 100:1 and 125:1. What is more, there is a difference between the total range -- the purest white and deepest black -- and the dynamic range, within which there is texture and detail. This is very unlikely to exceed 100:1, and may well be more like 64:1.

From this it is fairly clear that it is impossible to represent even an 'average' 128:1 subject with absolute accuracy in a print, with the print brightness matching the subject brightness one-for-one. There are therefore four possibilities. You can sacrifice the highlights or shadows or both; you can use additional light to reduce the brightness range; you can dodge or burn during printing; or you can compress the tonal range in some way, so that the brightness range in the print represents, but does not reproduce, the brightness range of the subject. It is most convenient to look at these in reverse order.

compressing the tonal range

Conveniently, the camera starts this process for us. In any camera-lens system, there is some degree of flare. Flare is non-image-forming light that bounces around inside the lens and the camera body. Some is reflected back out of the front of the lens; some is absorbed by the blacking and baffles; and some ends up on the film. Inevitably, it has a greater effect in the shadows (the darkest areas of the image) than in the highlights (the brighter areas of the image), so the overall brightness range of the subject is compressed even at the image-forming stage, before the image is captured on the film.

the flare factor

The 'flare factor' is the degree to which the subject brightness range is compressed at this stage. A perfect camera and lens would have a flare factor of 1, so that for a subject brightness range of 128:1 the image brightness range would be 1x 1/128 = 128:1. A good, modern multi-coated lens on a large format camera with a deep lens hood and a well-blacked bellows can come surprisingly close to this.

A flare factor of 2, quite common in the days when Kodak did their original research, would imply an image brightness range of 2x 1/128 or 64:1. This would not be good by modern standards, but it would not be impossible, especially with an old or cheap zoom. And a flare factor of 4, quite likely with an old box camera, would reduce the image brightness to 4x 1/128 or 32:1.

After allowing for the flare factor, the other ways to compress the total tonal range are by cutting the development time and printing on softer paper. These options are (or will be) covered in separate chargeable modules. There is more about flare in the free module on density, d/log e curves and related topics.

 

Yosemite

 

Distance flattens contrast, which is one reason why long-focus and telephoto lenses are sometimes unfairly described as 'low contrast'. Roger shot this on Ilford XP1 (a low contrast film) using an old 400/5 Telyt on a Visoflex III and a Leica M2 -- but Frances regained contrast by making a lith print on Paterson Warmtone. In the opposite direction, you can lose contrast by printing on a soft grade of paper.

dodging and burning

Because a negative can capture a very much greater brightness range than can be represented 1:1 on a print, it is entirely possible to 'dodge' (give reduced exposure locally) or 'burn' (give additional exposure locally) to lighten or darken specific areas of the print. Done well, this looks much more natural than you might expect, because it echoes the way our eyes normally work. We scan a scene, and our irises automatically contract when we look at a bright area and expand when we look at a dark area, so we can look at bright areas without being dazzled and into shadows without finding them too dark. There is a chargeable module on dodging and burning, or you could buy Darkroom Basics which has step-by-step illustrations of how to do it.

.

using additional light

In the studio, or for that matter in the open if you have the resources of Hollywood at your disposal, this is easy. You just lighten the shadows until the brightness range is exactly where you want it. This works even in colour (there is more about colour later). Trying to lighten the shadows with on-camera flash is likely to be less successful, especially if the too-bright areas are nearer the camera and the too-dark areas further away, in which case 'fill flash' will actually make things worse.

 

Toe shoes

You don't even need a great deal of light in the studio, and the studio needn't be any more than a corner of the garage. Frances shot this on 6x7cm Fuji Astia using a Linhof Technikardan with a rollfilm back and 210/5.6 Schneider Symmar.

discarding detail

 

This is not as silly an option as it sounds. In some pictures you can use blocked up shadows as a part of the composition, though large areas of featureless 'blown' highlight are rarely successful, and in many pictures the brightest and darkest parts of the image are so small that you do not really miss them.

 

Window, St. Clement's, New Romney

There is no detail in the leading on the window; precious little on the mullions; and effectively none on the inside wall to the right of the window. How much does any of this matter? And yet the contrast cannot really be made higher, or you would lose the texture and gentleness of the shadows on the whitewash. As far as we recall Roger shot this with a Voigtlander Bessa-R and a 35/1.7 Voigtlander Ultron. Film was definitely Ilford XP2 Super.

brightness ranges in colour photography

Colour images -- slides or prints -- can represent colours only across a very limited range of brightness. Too dark, and the colours become muddy and desaturated; too light, and they 'wash out'.

All of the options available in black and white are also available in colour. There is a remarkable range of filters from Tiffen that in effect introduce controlled flare, and these are extremely useful, especially with digital cameras. They are somewhat misleadingly called 'Ultra Contrast', which suggests that they increase contrast instead of reducing it as they do. They require some practice to master, but as there are no film costs involved in digital photography this is not too serious.

Apiary, Xemxija

This apiary is Phoenician or possibly Roman; the bees built their nests in the 'windows' which go through to the cave beyond, whence the honey was extracted. This is a 4x5 inch shot on Fuji RDP2 using a 110/5.6 Schneider Symmar-XL on a Toho FC45X. Roger took an incident light reading and didn't even bracket (free module), but the trees on the right and the shaded side of the staircase were lightly dodged in Adobe Photoshop. The rock-cut cave tomb and the foliage on the upper centre left, below the workmen, was left to go dark. Xemxija is (as you might guess from the spelling) in Malta; it is pronounced 'Shem-shee-yah'.

In conventional colour photography, it is possible to increase contrast slightly by increased development or reduce it by decreased development, but this is generally regarded as an undesirable side-effect and most manufacturers have reduced contrast changes to almost negligible proportions. Although different contrast grades of paper are available in colour, they offer nothing like the range of the different contrast grades in black and white. Digital image processing, whether from digital capture or scanned film, offers far more scope here.

Dodging and burning can be used, but become more obvious, faster, when they are overdone in colour. Fill flash is widely used, and a few cameras do it automatically and very well indeed. Then again, digital cameras more or less have to, because of the very limited tonal range they can capture.

the bottom line

Subject brightness ranges are something that you eventually become more or less instinctively aware of, but until this happens, you need to make a conscious effort to judge brightness ranges (and even to measure them, if you use a spot meter), just as you have to train yourself to check backgrounds and foregrounds (chargeable module) when you look through the viewfinder.

Go to the list of modules

or go to the home page

or support the site with a small donation.

© 2005 Roger W. Hicks