Few simple photographic tools are the subject of more mystery and misinformation than the 18 per cent grey card. It is widely touted as the exposure tool for experts, but nothing could be much further from the truth. It is a reasonable substitute for an incident light meter in colour photography, where exposure is keyed to the highlights, and it has some modest use in black and white, as described below.
Even its origins are shrouded in mystery. Kodak Grey Cards seem first to have appeared in the late 1940s or early 1950s; certainly, the 1941 Kodak Reference Handbook was still recommending the use of a (Kodak yellow) paper packet for determining exposures for Kodachrome, using a similar technique to the later grey card approach. To quote directly from the Handbook:
"An Eastman photographic paper envelope should be put in front of the subject, facing the camera. A reading is made by holding the meter close to the envelope without shading it. For average subjects the indicated exposure should be doubled."
The basic purpose of a grey card is to provide a standard target for an exposure meter, thereby allowing the photographer to take a meter reading without being influenced by the reflectivity or colour of the subject.
Portraiture was the classic use for a grey card but even there Kodak recommended them only for colour. Frances used an 8x10 De Vere with 5x7 inch reducing back for this shot on Ilford FP4, but exposure determination was with an incident light meter. The lens was a 21 inch (533mm) f/7.7 Ross, probably made before the Great War.1
Black velvet and white paper obviously have very different reflectivities, but equally obviously, if you put the same subject in front of either, you will need the same exposure in order to maintain the same image density for the subject. If you read from a grey card, this problem is removed.
Slightly less obviously, different colours influence meter cells to different degrees, according to the colour sensitivity of the meter. Pentax spot meters, for example, recommend 1-2/3 stops more exposure than the meter indicates if you read off yellow, or 1-1/2 to 1/2/3 stops less exposure if you read off indigo or purple. Values for other colours are in the Pentax instruction book or on page 92 of our book Perfect Exposure. If you read from a grey card, this problem is removed.
Seafood, Mertola, Portugal
The yellows and oranges here could fool some meters, but equally, you can aim off on the basis of experience or (easier still) take an incident light reading. After all, producing a grey card in a crowded market might draw odd looks. From memory Roger used his Leica M4-P and 90/2 Summicron to shoot this on Agfachrome 1000.
Of course, an incident light reading has exactly the same effect as a grey card reading, as it is also uninfluenced by subject reflectivity or colour. Our own belief is that grey cards originated at a time when incident light meters were rare and expensive. Usually, an incident light reading is much easier.
In effect, taking an incident light reading via a reflected light dome or Invercone is the same as taking a reading from a sheet of white paper. In fact, the old name for incident light metering was 'artificial highlight' metering. This will be covered at greater length in the paid module Exposure for Transparencies (and Digital).
The only significant difference between a grey card and a sheet of white paper is the reflectivity. White paper reflects about 90 per cent of the light falling on it, or 5x as much as an 18 per cent grey card. If you read a sheet of white paper, you therefore need to give 2-1/3 stops more exposure than if you read a grey card. Variations in whiteness are sufficiently small that you are probably only looking at +/- 1/6 stop for most white papers, or +/- 1/3 stop at most.
Although the difference between a grey card and a maximum diffuse highlight (under the same light) is pretty much constant at 2-1/3 stops, the difference between a grey card and the darkest shadow in a picture is much less constant. This is especially true if the subject is not uniformly lit. As noted in the free module on subject brightness ranges, it is quite possible for the the brightest highlight in a church interior to be 500x or even 1000x brighter than the darkest shadow -- a brightness range of 9 or 10 stops.
If you are shooting transparency film, or using a digital camera, your maximum exposure is limited by the fact that the highlights will 'blow' to a featureless white if you over-expose. If need be, you must be willing to sacrifice shadow detail. This is where a highlight reading or grey card reading can be useful.
With negative film (black and white or colour) there is no such problem: you just give enough exposure to be sure of shadow detail. You can recover the highlights via curtailed development (see the paid-for module on development technique) or a softer paper grade (another paid-for model, paper grades, not finished yet) or dodging and burning(another paid-for module -- sorry).
St. Martin's, Noize
Roger took his (spot meter, shadow index) exposure reading from the roof timbers because he wanted texture and detail in there. This entailed dodging to keep the detail -- but any other metering method would probably have given no detail at all in the roof. KowaSIX, 85/2.8 lens, Maco Cube 400c.
In addition to providing an exposure target, a grey card also provides a known subject tone (and colour) which can be reproduced as a known image tone (and colour). Of course a wide range of other comparison standards could be used but the grey card has the advantage of providing a visual mid tone in a neutral grey: if there are any colour shifts, they are comparatively easy to see.
Obviously colour is irrelevant when grey cards are used in black and white photography, and as a means of exposure determination for black and white they may also seem to be of limited use. After all, ISO film speeds for negative films are determined by the minimum exposure required to give a specified shadow density, as explained in the free module ISO speeds.
The brightness of an 18 per cent grey card clearly bears no fixed relation to the brightness of the darkest shadows, so from a strict ISO point of view, it is worthless in determining exposure. On the other hand it does provide a known density for printing and sensitometry and it is of some use with subjects having a very short brightness range (free module).
St. Martin's, Noize
In the original print there is (just) texture and detail in the highlights on the wall and
on the overhead beam, but of course the dark interior of the church is pitch black. Where would
you put a grey card for such a reading? Far easier, as in the shot above, to take a spot meter
reading of the darkest part of the roof timbers in which you want texture and detail. Technical
information as above. (Roger)
In the former case -- a standard density -- the importance of a grey card can be overestimated. After all, how often do you want to reproduce an 18 per cent grey exactly? A photograph is a matter of interpretation: what looks right, is right, and this often means slight increases or decreases in density until you get a print that you are happy with.
In the latter case -- subjects with a very short brightness range, such as a misty day -- it is possible that an exposure based on shadow detail or even on overall brightness may result in a negative that is thin and tonally unsatisfactory. Taking a reading from a grey card is likely to give a more satisfactory image. Alternatively, you can simply give a stop or even two stops extra exposure as compared with a spot reading of the darkest area, or a general broad-area meter reading. This will be just as satisfactory and is a lot easier.
Many Zone System users are completely wedded to grey cards both for initial calibration and for subsequent exposure determination. There are enough books on the Zone System to explain all this in great detail, and as we are not devotees of the Zone System, we do not propose to recite their arguments here. If anyone is puzzled at our lack of interest, we refer them to our short free module on the Zone System
Contrary to popular belief, the overall, average reflectivity of an 'average' scene is not 18 per cent: rather, it is 12 to 14 per cent. This is the assumption on which traditional broad-area reflected-light exposure meters were based. It is surprising that an average outdoor scene should so consistently reflect 12 to 14 per cent of the light falling on it, but it is the case. Of course there are many non-average scenes.
Unfortunately the 18 per cent myth has been so widely reported that many people get quite angry when you tell them that it is a myth. All we can say to them is, go back to the original Kodak research (Jones and Condit in the 1930s) and look it up for yourself.
The important thing about an 18 per cent grey, however, is that it is a visual mid-tone. In other words, if you make (say) 20 or 30 grey cards, ranging in reflectivity from about 5 per cent to about 90 per cent, most people will pick the 18 per cent card as a mid tone. This is related to the fact that our visual perception is much better able to distinguish subtle variations in light tones than in dark ones.
The best way to use a grey card depends on the meter you are using. If you are using a conventional broad-area reflected-light hand-held meter -- and remember, grey cards long antedate commercially available through-lens metering -- you should get the most precise exposure if you place the card at the subject position and angle it so that it splits the difference between the camera position and the key light. Read the card from as close as you can without shadowing the card with the meter.
You can see from the shadows that the lighting is only very slightly forward of pure side-lighting, so the correct way to angle a grey card would be pretty close to 45 degrees to the camera-subject axis. A much easier approach, and more reliable, is to use an incident light meter with dome or Invercone pointing straight at the camera, which in this case was a Linhof Technikardan with a 210/5.6 Schneider Symmar. Film was 4x5 inch Fuji RDP2.
The difference between 12 per cent and 18 per cent is near enough half a stop, and if you place the meter square-on in front of the subject, pointing towards the camera, and take a reading that way, you will get slight under-exposure. This should however be within the latitude even of colour slide films and digital sensors.
If you are using a spot meter, you may do better to place the card square-on in front of the subject and take a spot reading from the camera position. The middle index on many spot meters does seem to be calibrated to 18 per cent grey, not 12, 13 or 14 per cent.
If you already have a grey card and find it useful, do not let us discourage you from continuing to use it. If on the other hand you do not have a grey card, and wonder what all the fuss is about, you are not alone. We have grey cards, but we almost never use them: there is invariably a quicker, easier way of achieving the same results, or better.
For slide film, an incident light reading is better.
For black and white, a spot meter reading of the shadows is better.
For subjects with a very short brightness range, give one or two stops extra exposure or use an incident light meter.
Go to the unillustrated list of modules (in either alphabetical or date order)
or go to the illustrated list of modules
or go to the home page
last updated: 15/02/05
© 2005 Roger W. Hicks