There are already several modules on metering, but this one is a bit more basic. It is for people who are wondering about buying a hand-held meter, and how to use it if they do. It is also a warning against buying a spot meter unless you understand a good deal of the theory behind metering. Spot meters allow unparalleled control if you really know how to use them, but if you don't, they are expensive, time consuming and less use than an old Weston Master.
Chateau de Bourg-Archambault
The easiest way to get a good exposure is to choose an easy subject under easy lighting. Quite honestly, you could have got a good exposure of this with a single-use camera, instead of the Leica M8 that Roger used. Any metering system, including the very simple through-lens meter of the M8, should give good results with an 'average' subject and the sun over your shoulder, even if you follow it blindly..
Before you read any further, it is important to realize that just about any meter or metering technique can be made to work, as long as you know what you are doing, and as long as the meter isn't broken or defective. Ignore anyone who tells you that their metering technique is the only one that works: this is a subject that attracts a lot of gurus, monomaniacs and armchair theoreticians. The main difference between the different meters and systems is the type and extent of the corrections you need to make with out-of-the-ordinary subjects.
ANY metering technique will work fine with 'average' subjects if the meter is used properly. The further you get from 'average' subjects, and from simple meters, the more you will need to know about how to use the meter and how to interpret the meter reading, but equally, the more you know about how to interpret meter readings, the likelier you are to get a good exposure of a 'difficult' subject.
Trees and rocks
As well as realizing that just about any metering technique can be made to work, it is important to realize that just about any materials can be made to work too, though inevitably, some materials will suit some photographers and some subjects better than others. This is from a 4x5 inch Polaroid Type 55 PN negative (no longer available) but it could equally well have been taken with at least half of the films on the market today. Camera: Toho 45. Lens: 120/6.8 Schneider Angulon with heavy red filter (8x - possibly too heavy). Metering: incident + 1 stop (for extra shadow detail).
A lot of understanding metering is to do with subject brightness ranges (SBRs), the range from the darkest part of the subject to the brightest. This is a sufficiently important concept that there is a whole module devoted to it. Short SBRs (e.g. an overcast day) are easiest to meter; long ones (e.g. interiors with patches of sun and deep shadows) are hardest.
Most of this module is concerned with metering 'average' subjects, with varying SBRs. At the end, though, we look at unusually light subjects (e.g. snow scenes) and unusually dark ones (such as the proverbial black cat in a coal cellar), and then at subjects with very long brightness ranges such as night shots and traditional wedding parties.
Neil and Leslie's wedding day
An 'average' subject (photographically, not emotionally) where skin tones are of primary importance but you don't want to lose detail in either Neil's dark suit or Leslie's wedding dress - or the white flowers in the bouquet either. In practice, a tiny bit can be sacrificed in both the suit and the flowers, because this will look better than a lower-contrast, 'flat' image, but you really want to capture as much detail in both the shadows and the highlights as you possibly can. Never let technical considerations (maximum tonal range) be subjugated to aesthetics.
Frances was shooting Ilford XP2 in a Nikon F with 75/2.5 Voigtländer Color-Skopar, and took a spot meter reading (with her Pentax Digital) off Neil's suit. The same exposure could have been arrived at by taking a reflected light reading and giving 1/2 stop to 1 stop extra to get more detail in the suit, or taking an incident reading and giving 1/2 stop to 1 stop less to keep detail in the dress.
XP2 can capture a very long tonal range, so the only important tricks were first, getting enough detail in the shadows and second, choosing the right grade of printing paper. Of course you could also adjust contrast in a scanned image.
First, you need to understand that there are two basic types of meter: reflected and incident.
Reflected meters measure the light reflected from the subject, and incident meters measure the light falling on the subject. Several meters are illustrated in this module, but quite a few more are illustrated in a module of their own, Choosing Exposure Meters.
There are several kinds of reflected-light meters. The most basic kind is variously known as broad-area or integrated. It measures the average of all the light reflected back from the subject: it 'integrates' this reading. The 'angle of view' (more accurately, angle of measurement) is typically 30° for most modern meters, though with old selenium-cell meters (the sort that don't need batteries) it can be as wide as 60°. Most hand-held meters are broad-area reflected light meters, though most can also be used as incident meters (see below). Many older camera meters are broad-area reflected light meters: so are all non-through-lens (non-TTL) camera meters, and some TTL.
Broad-area meters rely on two things. One is that average outdoor subjects reflect a surprisingly constant 12-14% of the light falling on them (and not 18% as many mistakenly believe). The other is the latitude of the photographic process. Film negatives can stand quite a lot of over-exposure, but not much under-exposure, and digital 'negatives' (raw files, DNG, etc.) can stand quite a bit of under-exposure, but very little over-exposure.
Ancient and Modern
The Gossen Sixtomat on the left is much more sensitive; can quickly and easily be switched from reflected (30° angle of acceptance) to incident; is lighter; has a big digital readout; and takes an AA battery. The Weston Master III on the right has a 60° angle of acceptance; needs an accessory two-part Invercone for incident light metering; and relies on a moving needle and lots of tiny numbers - but it was discontinued decades ago, needs no battery and was still working fine in 2010.
Some older through-the-lens (TTL) meters in cameras, and even a few new ones, are basically broad-area, with the addition of 'centre weighting'. In other words, they attach more importance to the light reflected back from the centre of the image. This relies on the principal subject being in the middle of the picture, or at least, on the photographer's metering that way, and then recomposing. Centre weighted meters are regarded as being more 'idiot proof' than broad-area meters, but they can still be fooled by unusually bright or unusually dark subjects, or subjects with unusual brightness distribution.
Many modern TTL meters are 'multi-sector' or 'multi-segment' or 'matrix'. They measure several points in the field of view; compare these (in effect) with a library of pictures that were deemed to be 'successfully' exposed; and calculate the exposure accordingly. These are probably the most 'idiot-proof' reflected-light meters of all, but they are far from infallible. Also, because they rely on reading the light at numerous points within the image area, they are really only feasible with SLR cameras, where the meter cells can be placed in the diverted light path. Stop and think about trying to put (say) five metering sectors in an interchangeable-lens rangefinder camera, where there's nothing between the lens and the film or sensor except the shutter, and you'll see that it's not actually feasible.
Finally, there are 'spot' and 'semi-spot' meters. As their name suggests, these have a much smaller measuring angle than others. Most true spot meters are 1°, like the Pentax Digital on the right, though the SEI Photometer is 0.5°. In-camera ttl 'spot' meters are a good deal less precise, and of course their angle of view depends on the lens fitted. A much more accurate description would be 'semi-spot'. The same is true of the Tele attachment for hand-held Gossen meters. Spot and semi-spot meters are the least idiot-proof of all, but if you know what you are doing, they offer the maximum control over exposure.
The old name for these explains a lot. They used to be known as 'artificial highlight' meters. The meter cell is covered with a translucent white plate, dome or cup of some kind. This is the 'artificial highlight'. You could get the same effect by measuring a sheet of white paper with a reflected-light meter, and making the appropriate compensation for the fact that you are reading the brightest part of the subject instead an overall average.
Most hand-held meters have sliding or clip-on domes to allow them to be used for incident light metering. Incident light metering is all but idiot-proof for most subjects, provided you can take a reading from the subject position or from a position that is identically lit. It is however an invitation to failure if you are shooting negatives (colour or black and white) and the subject brightness range (SBR) is unusually long. This brings us on to the second thing you need to know: what (and how) to meter. Before you do this, you have to think about the differences in how films and sensors respond, and what can be represented in the final image.
Reflected and incident meters
Above, the Gossen Sixtomat Digital has its (captive) incident light dome pushed to one side to allow reflected readings: the cell (behind its little window) is uncovered, and the meter reads straight ahead. Below, the dome is slid into place over the cell for incident light readings, where it covers near enough a 180° hemisphere, reading light from all sides.
Imagine a bright, sunny beach with deep caves in the cliff behind it. If you walked into those caves, your eyes would adapt, and you could see at very low light levels. But if you are on the beach, and looking at them from some distance away, they are just black holes.
Now consider taking a picture. On a bright, sunny day with ISO 100 film (or the digital sensor set to ISO 100), the sunny beach might call for 1/125 second at f/11. Walk a few paces into the cave, and you might need 1 second at f/2.8. This is a brightness range of about 2000:1 or 11 stops.
Perhaps surprisingly, this is inside the subject brightness range that the right film can handle, with the right processing, but even 1000:1 (10 stops) is pushing your luck with most films and sensors, and plenty will run out of steam at 500:1 (9 stops).
All right, it isn't a beach - but exactly the same arguments apply. In colour, you can afford to give less exposure than in black and white because colour never allows you to 'see into' the shadows to the same extent as black and white. Ideal metering technique for slide or digital: incident, cut 1/2 stop to 'pop' (saturate) the colours.
In any case, the maximum brightness range of a print is about 200:1 (a bit over 7 stops), and that's from pure black to pure white. You will be lucky to see texture and detail across a range of 100:1 (a bit under 7 stops). Good colour will only be reproduced across a range of less than 50:1, and quite possibly as little as 32:1 (5 stops). You therefore have two choices. One is to ignore part of the shadow detail, or part of the highlight detail, or both. The other is to compress the information captured on the film or sensor in some way, so that you can get more of it on the print.
The classic example of this is a colour slide, where most people don't even try to capture the shadows at the taking stage. Sticking with our sunny beach, we dare not over-expose the image, or the highlights will 'blow' and it will look very nasty indeed. We therefore 'key' the exposure to the highlights, and let the shadows go dark. Actually, there is more detail in the dark areas of an underexposed slide than most people realize, and it is possible to dig it out via scanning. But as a projected slide, the shadows will appear very dark.
Lantern, Bastille Day
In a picture like this, you don't want a lot of detail in the background, so you want it as dark as possible, and you aren't worried about the fact that the hole in the top of the candle-lantern has blown to a clear white. The easiest way to meter is to take a series of incident light readings before you start shooting, to discover the exposures in the brightest and darkest areas, and base your exposures on a combination of these and guesswork.
Actually, a machine print from a negative ignores a good deal of the brightness range in much the same way as a slide does, except that it's done at the printing stage. Let's say the negative has captured a brightness range of 500:1. This is entirely possible. The machine then prints a 200:1 section out of that 500:1 range. This explains the generous latitude of most negative films. It's overexposed? No matter: the 200:1 section is taken out of the denser part of the negative. It's underexposed? No matter: the 200:1 section is taken out of the thinner part. This sort of latitude explains how single-use cameras, with a fixed shutter speed and aperture, can work at all.
An easy subject to meter, because the important part of the SBR is very short indeed. As long as you have enough exposure to put some texture in the darkest part of the door, or in the upright of the door on the right, the rest just isn't a problem. A prime candidate for an incident reading.
The example of a single-use camera also illustrates very well that you can be extremely sloppy with metering - even, as in this case, to the extent of omitting it altogether - and still get an image. You can now see why we said, early in the module, that almost any metering technique can be made to work.
Suppose, though, that you want the maximum possible useful information on your negative or digital file, with a view to compressing rather more of it onto a print later. Before we look at this, we need to look at how you can compress the information.
Courtyard, Gran Canaria
Including the gate in the foreground suggests why this courtyard is devoid of people: no-one is allowed in. You need the white of the lightest wall to be only fractionally darker than paper-base white, but dark enough the the darker areas show up. As with the earlier black and white image in this module, this is Ilford XP2: there is no need to worry about arcane films or fancy developers if you get your exposure right. Frances took a spot reading of the darkest part of the door, shooting with a 35/2.8 PC-Nikkor on a Nikon F.
And again as with the earlier picture, she could have achieved a very similar result - certainly, one you would be unable to tell apart in the final print - by taking an incident light reading and cutting the exposure by a stop, or taking a broad-area reflected-light reading and increasing it by a stop.
There are three ways of doing this.
The first is simply by reducing the contrast. With traditional black and white in the wet darkroom, this can be done either by reducing film development time or by using a softer grade of printing paper. This often works very well but can lead to a somewhat 'flat' or 'muddy' image.
The second option, with digital imaging, is HDR (High Dynamic Range). You take at least two exposures, and combine them electronically. Again, it often works very well, but it can look unnatural, more like a painting than a photograph. Of course, you may be as happy with a 'painterly' look as with a 'photographic' look.
The third choice is via dodging and burning, local increase or decrease of exposure: a sort of primitive HDR, going back to the earliest days of photographic printing. Most software programs allow an analogous process when dealing with digital files. It requires quite a lot of skill. Horrible 'haloes' around badly-dodged heads are commonplace.
It's a truly rotten print, but then, it was a truly rotten negative: Roger shot it in the mid-1970s, on 'pushed' Ilford HP5 rated at EI 1600, so the girl's jumper was clear film. In an attempt to differentiate her face better from the background (it would otherwise have been far too dark) he used a mask a few centimetres from the paper to hold back the exposure locally for a part of the exposure. Result: halo. Once you've made a few bad halos yourself, you'll become sensitized to them in the work of others.
Under the heading of 'ignoring the extremes' we have already looked at colour slides, where we 'key' the exposure to the highlights because 'blown' (overexposed) highlights are irrecoverable. Digital sensors work the same way - and, as with scanned slides, there is often a lot more that can be dug out of the shadows.
Negative films are different, though. Here, the brightness range that the negative can capture is far more than most people realize, so the risk of blowing the highlights is very small. This means you can afford to 'key' your exposure to the shadows. If you key your exposure to the highlights, it is quite likely that the shadows will block to a solid black.
A further question arises here, though. Many scanners simply cannot 'see through' high negative densities, so a negative that would print perfectly well in an optical enlarger, or with a professional scanner, may be too dense in the highlights (the darkest part of a negative, remember) to scan. The only way to find out if your scanner works with dense negatives is to try it. This is one reason why we like Ilford XP2 for scanning. Another is that 'Digital ICE' scratch and spot removal works with XP2 and not with conventional black and white films. Another is that we just plain like the tonality, especially when wet printed.
This brings us back yet again to the truth that any metering system can be made to work. Biasing a reading simply means using the meter in the way that is most suitable for the subject matter, equipment and medium in use.
Back street, Rhodes Old Town
This is a good example of needing to suit equipment, metering and materials to the subject in hand, and a bad example of doing so. Roger used Polaroid Type 55 PN for this shot, when conventional 4x5 inch film with increased exposure and reduced development would have been more suitable in order to reduce the contrast of the negative. He gauged exposure by reading the darkest area (the foliage on the right) from close-up with a broad-area reflected light meter (Gossen Variosix). Then, when Frances printed it, the glaring whites didn't look right so the whole image is printed slightly darker. The camera was a Toho FC45; the lens, a 120/6.8 Schneider Angulon.
Let's start with the simplest meters of all, broad-area reflected. When shooting landscapes on negative film, the experienced photographer will angle the meter slightly downwards, instead of pointing it straight at the subject with the cell at 90 degrees to the ground. Or indoors, he or she will 'favour' the darker areas of the subject, rather than the lighter ones. 'Favouring' is simply an old-fashioned term which means that if one area is darker, that's the general direction in which to point the meter.
When shooting the same subject on slide film, or even with digital, this sort of bias might lead to over-exposure. The easy thing to do here is to use an incident light meter: remember the old name, the 'artificial highlight'. Hold the meter so that it measures the light falling on the subject (in the brightest area, if you are indoors) and use that as your reading. Alternatively, 'favour' the lighter areas. Some meters have subsidiary indices to help you do this.
On the left, Frances is taking an incident reading, with the meter pointed slightly upwards. The meter is at the subject position, pointing towards the camera. In strong side-lighting, you may want to point the meter along an imaginary line bisecting the subject-camera axis and the subject-light axis.
On the left, she's taking a reflected reading, with the meter angled slightly downwards, pointing at the subject from the camera position. The meter is a Gossen Sixtomat Digital in both cases.
Using an incident meter for metering negatives will almost certainly lead to a loss of shadow detail, unless the SBR is very short. The solution? Easy. Give an extra stop - or set the film speed on the meter at half its real value.
This is why many experienced photographers prefer very simple meters to very complicated ones. With a minimum of experience, you can bias the meter reading to get the result you want. With a multi-sector meter, you are always trying to second-guess whatever you are trying to override.
Biasing is all very well, but if you want the absolute maximum of useful information on your negative or file, you need to go a step further. Actually, with slides and digital, you normally don't. An incident light reading will do all that you need, almost all the time: the 'artificial highlight' again. The only time you need to worry is when you are physically unable to get to the subject position and you can't take an equivalent reading at the camera position. At that point, a spot meter reading of the brightest area in which you want texture and detail is the only way you can, with absolute reliability, key the exposure to the highlight.
Electricity sub-station, Daroca, Spain
Frances shot this with the then-new Zeiss Ikon and (as far as we recall) the 35/2 Biogon, though it might have been the 50/2 Planar. Either way there was an orange filter on it, and the camera was loaded with Ilford HP5 Plus which we planned to develop in Ilford DD-X for a true ISO of around 650. The camera meter was set to EI 400, for 2/3 stop extra exposure in the interests of better tonality
She checked the camera meter reading against a 1° spot reading of the shadows under the insulators. They were too dark to read easily: 3 stops down from a clear blue sky, and 3 stops down from the the camera reading. This is what she wanted: she did not want too much detail in the shadows. Of course she had to make a 2-stop (4x) allowance for the filter when taking the spot reading without a filter, and these adjustments are always approximations: there is no guarantee that two different meters will have the same spectral response.
With negatives, the only 100% reliable way to key the exposure to the shadows is to take a direct reading of the shadows, with a spot or semi-spot meter. Choose the darkest area in which you want texture and detail, and meter that. 'The darkest area in which you want texture and detail' is not necessarily the darkest area in the whole scene, of course. You can always decide, on aesthetic grounds or because you want a shorter shutter speed or smaller aperture, to let some areas go black.
But wait. We have just recommended spot readings of the highlights in one case, and the shadows in the other. In the former case, if we use a mid-tone index on the meter, then the highlight will be reproduced as a mid-tone and everything else will be darker: the whole thing will be under-exposed. In the latter case, if we use a mid-tone index on the meter, then the shadow will be reproduced as a mid-tone and everything else will be lighter: the whole thing will be over-exposed.
Well, yes. This is why the mid-tone index on a spot meter is a snare and a delusion, and why it is almost never used by anyone who understands how spot meters and speed indices work. Indeed, it is why there was no mid-tone index on the first commercially successful spot meter, the SEI Photometer. On a properly designed spot meter there is a shadow index, typically 2 to 3 stops below the mid-tone (on the Institute of Radio Engineers scale used in television, I.R.E. 1 is 2-2/3 stops down) and a highlight index, typically 2-3 stops above the mid-tone (I.R.E. 10 is 2-1/3 stops above). You may already have noticed that the I.R.E. range spans 5 stops, the range which we have already pointed out as being realistic for reproducing colour.
The white scale either side of the big red index triangle is the Institute of Radio Engineers scale, where 1 is the shadow index and 10 is the highlight index. They are 5 stops apart, a brightness range of 32:1.
Because the Pentax spotmeters are very old designs, they are still labelled ASA instead of ISO. But they still work very well.
The mere fact that spot meters are supplied with a mid-tone index at all, presumably in response to public demand, is why we firmly believe that the vast majority of spot meters are bought and used by people who do not really know what they are doing. Anyone who tries to read a mid-tone is wasting his or her time. A spot meter guarantees good shadow detail if you read the shadows. If you read the brightest highlight in which you want texture and detail, it also tells you the subject brightness range. This can be very valuable if you are shooting sheet film and process each sheet individually to control contrast. But if you are just using it to measure mid-tones, then quite honestly, you are being saved by the latitude of the neg/pos system: you'd be better off with another meter, and a better understanding of what you are doing.
Early in this module, we pointed out that broad-area reflected light meters rely on the rather surprising fact that average outdoor subjects reflect about 12-14% of the light falling on them. What happens, though, if the subject in front of your camera (or more importantly, in front of your meter) reflects more or less light?
Most of us have had the experience, early in our photographic careers, of metering a snow scene with a broad-are reflected light meter; following the meter reading; and ending up with an underexposed slide or negative. This is because a snow scene can reflect as much as 90% of the light falling on it, six or seven times as much as the 'average' scene for which the meter is designed. The meter treats it like an average subject, and recommends two to three stops less exposure than is needed. Even the cleverest multi-segment meters can be fooled by this.
When you are faced with an unusually light subject, the answer is however quite simple. Give at least one stop more exposure than the meter indicates, and quite possibly two stops. This will still result in slight under-exposure, but that is normally what you want in order to get a bit of texture in the snow, instead of its blowing to a featureless white. This is counter-intuitive: there is so much light around that there is a temptation to cut exposure still further from the meter reading, thereby underexposing still more. But if you think it through, you realize it's right.
We talk about a beach, then provide a picture of a ski resort. We talk about snow, and then provide a picture of a Venetian blind. Believe it or not, there is a good reason for this.
It is all too easy to read about a snow scene, and then to think that the advice given applies only to snow scenes, whereas in fact, it applies to all very light or white subjects.
We were staying with Frances's nephew Dane in California when we woke up and saw this. Roger took the picture on ISO 100 slide film, Kodak Elite Chrome 100, one of the least forgiving of all media. He gave 2 stops more than a reflected-light reading recommended: enough to hold texture in the darker part of the blinds, while blowing the brightest part to a featureless white which was still slightly darker than clear film (between the slats and the window-frame).
The camera was a meterless Leica M4P, the lens a 90/2 Summicron. It would have been a lot more difficult to guess what compensation to apply if he had been using an SLR with a super-clever multi-point metering system..
Alternatively, use an incident light meter. This will not be influenced by the brightness of the subject, but in most cases you will still need to cut the exposure slightly (typically 2/3 stop, though 1/2 stop or 1 stop will do) in order to get the texture in the snow. The same applies with any bright white object: a white cat, a damask tablecloth, a bride's dress. Or a white Venetian blind.
The proverbial black cat in a coal cellar is a good example here. The overall scene may reflect as little as 2-3% of the light falling on it, again, a 6x difference as compared with 'average'.
Once again, the remedy is simple. Give anything from one to two stops more than the meter indicates. As with the light subject, this is less compensation than you might think, but you want a modest amount of over-exposure in order to get some detail in the cat's fur. Black fur 'eats' light, so in this specific case, one stop less than the meter indicates is normally ideal.
With an incident light meter, you will however need to increase the exposure somewhat in order to get detail in the fur. About a stop extra is the likeliest bet.
This broken fire-jack hung above our main barbecue in the courtyard. Both it and the wall behind it are blackened with years of smoke. A hasty test revealed that the wall reflects about 5 stops less light than a piece of white expanded polystyrene packing material. In other words, it would be just possible to record texture and colour in both, with very careful exposure and no further manipulation.
Not only does careful exposure save time in post-production (scanning, electronic image manipulation or printing): it also results in a more natural-looking picture. .
It might seem that we have covered this already, with our example of the sunny beach and the deep caves. So we have, but not fully. Two further examples are worth considering.
First, there are subjects with a very long tonal range. A good example is a wedding party, with the bride in white (symbolizing purity and innocence) and the groom in black.
Shooting on negative, ideally a spot reading of the groom's trousers and the bride's dress will be inside the 5-stop range of the I.R.E. scale. If they are not, you may be able to afford to cut exposure by about a stop without the groom's trousers looking like a featureless black hole, but don't sacrifice the detail on the bride's dress! Shooting in the shade, or on an overcast day, is much easier than shooting in bright sun. The happy couple's eyes won't be screwed up against the light, either. We've already seen this earlier in the module, with Neil and Leslie's wedding.
If you are rash enough to shoot slide film at a wedding, take an incident light reading and cut the exposure by 1/3 to 1/2 stop to keep detail in the dress. With digital, check the histogram and the picture on the back of the camera.
The tonal range here is immense, from the scene outside the door, via the smith's shirt, to the interior of the smithy. Obviously, the smith's face is the important tone, but if we exposed for a normal light flesh tone, his shirt would be even more mercilessly 'blown' than it is here. More to the point, how do you determine the exposure anyway?
Broad-area metering will almost certainly result in too light a background, with the smith himself far too light: you would need at least a stop less than the meter indicated. This would not matter too much with a negative if you were printing it yourself- generous exposure is often a useful buffer - but with slides, digital or automatic printing it could easily be a disaster.
An incident light reading from the smith's position would probably be the best bet, or (if you had time) a spot reading off his shirt (I.R.E. 1) and his cheek (either a mid-tone or I.R.E. 4, a dark mid-tone). Another possibility is to guess that his face is a stop or two darker than it would be outside, and base your exposure on that. Or - let's be honest - you could just bracket +/-2/3 stop around your best guess. As far as he recalls, Roger used Kodak Elite Chrome 100 with a 35/1.4 Summilux on a Leica M-series..
Second, a night shot. The remarkable thing here is that a very wide range of exposures is often acceptable. The lightest frames may have patches that are blown to a featureless white, but as long as they are not too big, this is surprisingly often acceptable. The darkest frames may consist of only tiny pools of light in vast areas of blackness, but once again, this is surprisingly often acceptable. The two exposures may be as much as four stops apart, or occasionally even more, and normally, inside this range, it can be quite hard to decide which of three images, each separated by half a stop, is the most successful. It's all a question of mood.
And as we've already said, you can always bracket.
Cyclist, Times Square
It's a very old picture, probably from around 1990, taken on slide film with a Leica M4P and a 35/1.4 Summilux. Obviously some of the brightest areas are blown completely, but half a stop either way wouldn't make much difference and a stop either way would almost certainly be acceptable. Of course it's important if you're shooting negative not to underexpose, and if you're shooting transparency or digital raw, there's a limit how far you can overexpose.
As with the lantern shot above, it's a good idea to take a series of meter readings before you start shooting, to establish the limits of exposure required. A useful trick if your meter isn't sensitive enough is to point the bare cell (no incident dome) straight at a light source, from the subject position or from somewhere that is similarly lit, and then give 1/5 of the exposure indicated.
Some of the information given above may seem a bit daunting: all that stuff about biasing and keying, and using different media (negative, slide, digital). But the truth is that as we have said, any meter or metering system can be made to work. All you need is a bit of common sense, a little basic information of the kind given above, and a modicum of experience. Unfortunately, many of the most passionate self-appointed experts on metering lack at least one of these desirable qualities
One last piece of advice is that when you are in the studio, it makes sense to keep the lighting ratios (SBR) strictly under control. It is all too easy to set up a lighting plot that looks dramatic. only to find that then you take the picture, the highlights are too bright and the shadows too dark. Use a spot meter if you have it, or alternatively, use the incident dome and point it (from the subject position) at each light in turn. The difference should only rarely exceed 4 stops (16:1), and then, only with a subject of reasonably even reflectivity. Nikon F; Sigma APO zoom; Kodak slide film; incident light metering.
Go to the list of modules
or go to the home page
or support the site with a small donation.
© 2010 Roger W. Hicks