Stupid question

[cyrus]

OK so I admit that as a newbie non-professional photographer, there's one thing that's been bugging me & I can't quite get my head around it. Unfortunately, I'm the sort of fellow who has to know why things are a certain way; I'm not satisfied with rote learning.

OK so we all know that if you use a reflection lightmeter settings to photograph a white cat in snow, or a black dog in coal, you'll end up with gray cats in gray snow and gray dogs in gray coal. That's because of the old 18% dull gray standard (and lets not get into the debate of 16% vs 18%)

In other words, the lightmeter is just a photoelectric cell, whose readings have been calibtrated so that it will convert the amount of light present into aperture/shutter speed settings that will result in 18% dull gray images.

This applies to reflection lightmeters - light meters that read the amount of light that is reflected off of the subject. And that's why the Zone System adjustments are necessary etc etc.

This does not apply to incident light meters - light meters that measure the amount of light falling on the subject.

So here's my question: why the difference between incident v. reflected light meters? Whether the light they're reading is reflected or incident, all lightmeters still have to have to be calibrated, and the calibration of both types is set to result in 18% gray images. If you use a reflected lightmeter to take readings off of a white cat in white snow, that should be pretty much the same as measuring the light falling on it, right?

SOmeone take me through the mental steps!

[Walter Calahan]

Nope

An incident meter photoelectric cell is measuring light filtered through the plastic dome that covers it as the light falls to the subejct. The cell is calibrated to read that light at 18% gray, so it can only see light as 18% gray regards of the tonal range of the subject.

With a reflective meter you need to use an 18% gray card when photographing the white cat (is it truly white?) against the snow (is IT truly white?).

Now I have no idea if what I said is a set of mental steps? Some simply think I'm mental. HA!

Hope that helps. It at least helps me figure things out because of my non-engineering mind. I'm an artist. Grin.

http://www.walterpcalahan.com

[Mark Sampson]

The incident meter measures the light falling on the subject, and integrates it to middle gray. This removes from the meter reading the variable reflectance of the subject, be it white, black, or inbetween. An all-black subject absorbs more (and reflects less) of the light falling on it than the 18% gray; an all-white subject absorbs less light (and reflects more). Thus the variability of reflected meter readings. Not a stupid question at all.

[Dan Fromm]

When you meter incident, you don't meter the light reflected from the subject, you meter the light the subject receives. So when you meter incident an 18% grey subject will be recorded on film, if you follow the meter's advice, as 18% grey. Blackest black will be recorded as blackest black. And so on.

When you meter reflected, whatever you meter (the average of the whole scene, a selected part of it, ... ) will be recorded as 18% grey.

Got it now?

[Marko]

Well, I'm the newbie (to LF) myself, but the way I see it, it's pretty simple:

The light reflected off a subject is only a fraction of the light shone on the subject. That is the light that will fall on the film surface. The reflected light meter meters that light without interpretation.

The incident light, on the other hand, is the full-strength light, it will get absorbed and reflected by the subject in variable ratios, depending on properties of the surface it falls on. Incident light meter measures that light and then interprets it as if it was bounced off a grey card.

So, if you apply the reflected light measurement as you took it, without correction, you will record the surface you measured as middle-grey in tone and everything else will be relative to that.

If you apply the incident light measurement in the same way, you will get the gray card recorded as middle gray with everything else relative to that.

Hope this makes sense.

[Emmanuel BIGLER]

If you use a reflected lightmeter to take readings off of a white cat in white snow, that should be pretty much the same as measuring the light falling on it, right?

Well, pretty much but not quite for several reasons.
In reflected mode the sensor reads what falls inside a certain limited angle of view. The good ol' Lunasix®/Luna-pro® reads inside a 30° cone, a spotmeter reads inside a 1 to 5° cone. (in the past there have been several sub-degree spotmeters... but they are hard to use)
In incident light reading you use a diffusor, either a white sphere or a planar diffusor (if to want to measure like a lux-meter).
So the metering conditions are very different and according to the scattering properties of your cat & snow, you'll get different reflected readings where the incident reading by definition will yield the same value.

So, all white cats do not reflect/scatter light the same way. Some well-polished cats may have a glazing aspect, they will fool your readings ;)

----

As far as snow is concerned, one of my best puzzling experience is trying to meter a snow-covered landscape, with dark trees and tree-covered slopes in the distance (with snow on ground, a superb view), in winter by overcast sky.
Sure, you can object that snow landscapes demand a pure blue sky ;-)
By overcast sky, some weird metering things do happen to the snow-covered landscape and neither the reflected nor the incident metering with a broad angle (e.g. 30° and above) will give you a direct usable value, you have to correct by yourself.

If you look at a snow-covered roof against an overcast sky, close to the horizon, you'll find that the roof is whiter than the grey sky, altough all light reflected by snow comes from the sky ; if you agree that there is no heat source inside a regular woolen blanket, you'll agree that there is no light source inside snow : so why does the grey sky appears to be less white than snow ! An overcast sky is far from being a uniform gray card, whether you aim at the horizon or vertically, the readings can differ by 1 - 2 f/stops, the horizon being darker.

A reference about this strange phenomenon i.e. snow being whiter than the overcast sky can be found in the famous textebook by Prof. Minnaert :
http://en.wikipedia.org/wiki/Marcel_Minnaert
The Nature of Light and Color in the Open Air, Dover, 1954, ISBN 0-486-20196-1

More recently reprinted than the Dover paperback, with many new color photographs !
Light and Color in the Outdoors, Springer, 1993, ISBN 0-387-97935-2, and 3-540-97935-2

[Alan Rabe]

All that is said above is true. But there is still another issue that bears discussion that has a great deal to do with which type of light meter you use.
If you use a spot meter and you meter a white object in light and a black object in the same light you will find at best 4 stops difference. Likewise if you meter a white object in shade and a black object in shade you will again find 4 stops difference. The difference between the white in light and the black in shade is 6.5 stops. This is an at best situation with a clear bright sky. Any haze or overcast will reduce the differences. So the classic example is a lump of coal in a snow bank. Whether you meter for the snow or coal if you straight print the snow as white the coal will be medium grey, likewise if you print the coal as black the snow will be medium grey.
This is why I like spot meters better. It will show the narrow difference and you can filter accordingly to get the maximum split on the negative to make life in the darkroom easier. An incident meter won't tell you anything about the scene you are photographing so you won't see the short range until you try to print it.
And if you use a Zone VI modified meter you can meter thru the filter and see exactly what effect it will have on the scene.

[Greg Lockrey]

If you measure the white or black scene with a reflective meter, it "thinks" it is measuring 18% grey. So the white will become zone V and the black will become zone V as well. That's why you need to use the 18% card which is zone V. Hope this clears the mud up for you.

[Alan Davenport]

An incident meter tells you how much light there is.

A reflected meter tells you how much of it is reflected by the subject.

[C. D. Keth]

An incident meter tells you how much light there is.

A reflected meter tells you how much of it is reflected by the subject.

Cyrus, this is a good simple explanation. The incident meter lets the colors and reflective properties of objects in the scene determine the brightness on film.

[photographs42]

You people are making this way too hard. A light meter reads the amount of light within its field of view, averages it, and reports the settings required to render it neutral gray. It doesn’t know if the light is reflected or not. How would it know?

So the key to using a light meter is knowing what you are measuring.

Jerome

[Greg Lockrey]

For reflected light setting, Zone V or 18% reflectance you can use green grass for your standard if you don't have your card handy. No grass, then the palm of your hand is about 36% reflectance or zone VI. It'll al least get you close.