|
Two Myths of Metering: A Film User's Guide
By Ctein
Myth #1: Photographers Can Measure Film Speeds Accurately
The ways in which photographers obsess about their exposures are frequently unhealthy, often
unproductive, and occasionally downright incorrect. Precise exposure is your goal: Work toward
being able to keep that slop to a minimum so that your results are predictable.
Precision, though, is not the same as absolute accuracy. Whenever I hear photographers comparing
their exposures on the 1/3-stop level, I wonder if they're assuming (incorrectly) that everything
operates so accurately that they can talk meaningfully about such small exposure differences.
Have you ever thought about how many places errors can creep into your exposures?
 |
|
Accurate metering is the key to photographing scenes with high luminance ranges, like
this photograph of moonlight reflecting off the ocean. But accurate metering is much harder
to achieve than most photographers know. © Copyright by Ctein
|
|
I don't mean human error--I mean the kinds of errors that occur even if you are perfection
embodied: errors in shutter speeds, errors in lens apertures, errors caused by not taking
lens transmission into account, errors in film speed (ISO), speed variations introduced in
processing, and all the possible errors in metering.
These kinds of randomly distributed errors compound by what's known as the "root-mean-square"
rule: the square of the total error is the sum of the squares of all the individual errors.
Of course, that's the average result. Sometimes everything cancels perfectly; at other times
they all add up against you.
In the real world it's not at all unusual for camera shutter speeds to be in error by 1/4 stop.
Lens apertures may also deviate markedly from what the inscription on the barrel says.
Manufacturers frequently are a bit optimistic about a lens' maximum aperture, in particular
for complex zoom lenses and telephotos. The smallest apertures can be in error because of
simple mechanical tolerances, most commonly with short focal-length lenses. I've seen errors
of a half-stop at both extremes, and errors of 1/6 to 1/4 stop aren't uncommon. A well designed
optic may lose less than 1/4 stop, but some are worse.
Film manufacturers don't control the inherent emulsion speed to better than 1/6 stop. The
actual film speed depends upon processing, and the tightest process specs don't guarantee
better than 1/6-1/5 stop. In combination, you can expect to have random errors of 1/4 stop
in film speed.
Put it all together and you're not likely to be able to determine a film's true ISO to better
than 1/2 stop, even with perfect metering technique. That's why I never give absolute ISO ratings
in film reviews; I always rate film speeds in comparison to other films, with all the film
exposed on the same equipment with the same settings and processed at the same time.
Myth #2: Light Meters Are Calibrated For 18% Gray
This is the one that's going to get me the hate mail.
A light meter doesn't have any way of knowing how reflective the subject really is; it sees
only how much light the subject reflects its way. To make an exposure recommendation, the meter
has to make an assumption about how reflective the average subject is to provide the greatest
number of successful exposures. The measured average effective reflectance for outdoor scenes
in the middle latitudes during midyear is 12%. If light meters had been invented by folks
living in Antarctica, they would be calibrated to about 30%.
Yes, I know 18% is what everyone says, and people will swear on a stack of Ansel Adams'
diaries that it's true. This subject has caused no end of confusion and misinformation imparted
to young and eager photographers, including myself. It wasn't until the late 1980s that Dick
Dickerson at Kodak managed to drill the truth into my thick skull, and he had to send me a copy
of the International Standards Organization (formerly the American National Standards Institute)
document to do it.
 |
|
When the primary subject is the illumination, normal metering methods go out the window.
To photograph the sinuous curls of flame from a campfire, opening up three stops from the meter
reading gave me good results, but it's really about guesswork and bracketing.
© Copyright by Ctein
|
|
Manufacturers created ANSI PH3.49-1971 to provide an industry standard for handheld
meters--it's advisory, not mandatory, but it is what they agreed on. It specifies that meters
should be calibrated to about 12% gray, with an allowable error of plus or minus 2%.
Calibration reflectance doesn't appear explicitly in the ANSI specs; it's built into the
correction constants. Photo Topics and Techniques, a Kodak publication (ZA-S1)
published in 1980 by Amphoto, has a fine chapter ("Exposure") written by Jerry O'Neill that
lays out quite nicely the logical chain that goes from illuminance to film exposure,
explaining how the 12% reflectance is incorporated into the standards. It's out of print,
but worth looking for.
Why then even have 18% gray cards? Probably because 18% gray is visually middle-gray on the
Munsell scale, so it's a good visual reference point. But I don't have any smoking-gun
documents to prove that, like I do for meter calibration.
Ever wondered why Kodak recommended holding a gray card at 45 degrees to the sun when
making a reading? Holding the card at 45 degrees to the light source reduces the apparent
brightness of the card by 0.3; 18% times 0.7 is just about 12%.
What happens if you take light readings off a fully illuminated standard gray card, with a
meter that is calibrated for the ISO standard of 12%? An 18% card reflects about a half-stop
more light than a 12%-reflectance surface would. The meter "thinks" the illuminance is a
half-stop brighter than it really is and recommends an exposure setting that is a half-stop darker.
When you're exposing slide film, this will result in slightly dark slides, which many
photographers prefer for their richer colors. But, negative film users will find their films
a bit on the thin side. From this, it is easy to understand why so many photographers feel that
film manufacturers overrate film speeds and come up with personal exposure indices (EIs) that
are lower than the film's real ISO.
As with many ointments, there is a fly in this one. Along with most photographers, most repair
technicians don't know about the ANSI standard, and many camera manufacturers seem to ignore
the standard they wrote. There's a fair chance your light meter has been adjusted to 18%. If
so, it will read gray cards just fine. It will also overexpose the average outdoor scene by
a half stop. I've also run across automatic cameras calibrated as low as 8% or 9%. They tend
to underexpose average scenes. How do you find out what your meter is doing? Run some exposure
tests (but don't forget about Myth #1).
Note that some meters weight different parts of the scene according to their brightness to
increase or decrease the exposure. Center-weighted, segmented and matrix meters may give you
a reading equivalent to 12% when viewing a uniformly illuminated field, but they can't always
function as a 12% meter and do their job.
|
