3
\$\begingroup\$

I am struggling to understand what is going on inside the logic of an incident light meter in order to determine the exposure as opposed to a reflective light meter.

I understand that a reflective light meter assumes the measured spot to represent a middle gray tone, and accordingly displays exposure settings that will render the measured portion of the frame as a middle gray tone.

Now, when using an incident light meter, the meter should assume that the measured light is the same as that reflected by the subject. In order to provide a reading that will also render an 18% reflective object a middle gray tone in the photography, is it going to calculate that the measured incident light will be reflected by 18% by the subject, thus mulitplying the measured light intensity with 0.18 before treating it like a reflective light measurement?

In other words, assuming equal lighting conditions and an 18% gray card as a subject, in order for an incident and a reflective light metering to yield equal exposure settings, am I correct in assuming that the only difference of both modes is that in incident mode, a light meter needs to account for an assumed reflection of 18%, but otherwise works the same way as in reflective mode?

In this case, a light meter that is suitable for both modes would need to know what kind of measurement it is doing in order to yield correct results. For instance, according to the manual of Gossen Sixtomat F2, the only thing that a user needs to do to change between modes is to move the white dome accordingly. I assume that internally, there are switches to determine the position of the dome for the meter's electronics to distinguish between modes.

I wasn't able to find any related information in web searches. Maybe my understanding is fundamentally wrong, in which case I ask you to help me out...

Thank you. Leo

\$\endgroup\$
3
  • 1
    \$\begingroup\$ "the meter should assume that the measured light is the same as that reflected by the subject" . Maybe not and that would be the big difference. If you shoot at a white wall with a tiny card of 18% gray (too small to be taken in account by the metering) taped on it, with reflected light the exposure will make your wall look like 18% gray and the 18% gray card will be too dark. With incident light it will set the exposure set so that 18% gray looks like 18% gray, so you white wall remains white and the 18% gray card is exposed correctly. \$\endgroup\$
    – xenoid
    May 27, 2023 at 14:53
  • \$\begingroup\$ If everything you shoot is 18% grey it will make no difference. For the rest of us that shoot subjects which are both brighter and dimmer than 18%, and more colorful than a single intensity of grey, incident meters will give more accurate results quicker. \$\endgroup\$
    – Michael C
    May 28, 2023 at 3:04
  • \$\begingroup\$ When the world isn't anything like 18% gray you will see a huge difference. I've had to dial in more than 2 stops of correction when dealing with snow. Likewise, when your objective is far from 18%--say, the totally black cat I had growing up. \$\endgroup\$ May 29, 2023 at 2:48

4 Answers 4

3
\$\begingroup\$

The idea is quite simple... Add the right amount of plastic.

Here is a simple diagram of a reflective measure.

You have a light source (L) and some rays hit your gray target (C). This absorbs some rays and reflects 18% of the light to the lightmeter which has a hole (B) and a sensor (A)

enter image description here

When you move the plastic dome, position the lightmeter in the place of the gray card, and point it toward the light, the plastic dome reflects some light, disperses some other light which does not enter the hole (B) and, as if it is well designed, will spill the equivalent of 18% of the light it received inside the hole into the sensor.

You can make the dome bigger, smaller add more plastic, or remove the thickness of the dome so the right amount of light enters and gives you a consistent reading, either incident or reflected.

It does not need a switch, only more or less plastic.

enter image description here

\$\endgroup\$
4
  • 1
    \$\begingroup\$ Coincidentally, I was wondering about this just the other day... how much plastic is enough plastic? How does Gossen or Sekonic or whoever measure the light passing through the plastic? I guess using some kind of sophisticated industrial photometer. \$\endgroup\$
    – osullic
    May 28, 2023 at 17:05
  • \$\begingroup\$ I am sure because they are top manufacturers, but you could do that at home. Let's say you have a reflective-only light meter. You prepare a test scene with a gray card, and now you put your light meter pointing at the light. You know what reading you need, so you can start adding layers of vegetal paper until you get the measure you need. Of course, you have a rudimentary incident lightmeter, with no fancy dome. But it could be useful. Of course, designing a dome needs more measures, but you already have the photometer. \$\endgroup\$
    – Rafael
    May 28, 2023 at 18:06
  • \$\begingroup\$ @Rafael This is exactly what white-balance filters or lens caps do. Essentially turns your camera into an incident light meter, that also allows you to set the right white balance. \$\endgroup\$
    – scottbb
    May 29, 2023 at 15:14
  • \$\begingroup\$ Oh. Interesting photometer! I never tried that! \$\endgroup\$
    – Rafael
    May 30, 2023 at 1:54
2
\$\begingroup\$

You used the word "assume" quite a bit. Unfortunately, your assumptions about the assumptions made by the light meter are incorrect.

You are correct that the light meter is calibrated to 18% gray. But no, there is no "multiplication by .18" going on. Don't get hung up on the number 18% or 0.18. It's just a thing, a standard. Might as well call it "nominal gray", "middle gray", or even "arbitrary gray" for purposes of using the light meter.

A reflective light meter assumes a nominal reflectivity of the subject. But different objects and surfaces reflect light differently, including differing colors, which affect the light intensity entering the meter. Subjects darker than middle gray reflect less light, and thus will be metered (exposed) higher. Conversely, subjects brighter than middle gray will be metered darker. Smooth reflective surfaces, even dark ones, with specular light reflection will show up very bright. The meter doesn't know this.

Measuring incident light makes no assumptions about the subject, because it's not measuring the subject. It's measuring the environmental light falling on the subject.

In other words, assuming equal lighting conditions and an 18% gray card as a subject, in order for an incident and a reflective light metering to yield equal exposure settings, am I correct in assuming that the only difference of both modes is that in incident mode, a light meter needs to account for an assumed reflection of 18%, but otherwise works the same way as in reflective mode?

Well, of course you're assuming equal lighting conditions, because that's the scene and light you're measuring. But no, you're not correct in assuming. By spot metering the gray card, your exposure settings will render the gray card as middle gray. That's it.

The goal isn't to get an incident light meter and reflected meter to yield equal exposure settings. They are not interchangeable or "convertible" techniques.

  • Incident meters measure the light coming into the scene, from the perspective of the subject.
  • Reflective meters measure the light coming into the camera from the subject, from the perspective of the camera.

An incident light meter essentially tells you "you're in Sunny 16 conditions", or "you're in indoor lighting conditions". It tells you the intensity of light, in EV.

A reflected metering tells you how much you need to compensate your exposure based on the specific subject you're shooting, because it may be backlit, or have a dark background, or have other specific conditions that make the scene lighting particular, unique, and well, not boring gray.

See also:

\$\endgroup\$
2
  • \$\begingroup\$ The exposure settings should be the same from an incidence reading and a reflectance reading of mid grey. They should generate equal/interchangeable results. \$\endgroup\$ May 27, 2023 at 20:08
  • \$\begingroup\$ Thus the advantage of incident light meters is when the subjects aren't 18% or even grey. If you meter a black object with a reflectance meter and follow its recommendation you'll overexpose. If you meter a white dress with a reflectance meter and follow its recommendation you'll underexpose. If you meter the incident light with an incident meter, your black subject will be very dark and your white subject will be very bright. Both incident and reflective meters are tools which are each optimal for different use cases. \$\endgroup\$
    – Michael C
    May 28, 2023 at 3:01
1
\$\begingroup\$

The diffusion dome itself reduces the incident light transmission to equal ~18% grey reflectance. The meter sensor inside simply reacts to the intensity of the light that falls on it, reflected or incident. I.e. there are no "switches" required and the sensor doesn't need to know what the operating state is. Use the meter wrong and it will give you bad information... it doesn't know or care.

You can get adaptor disks that will do the same thing for your camera making it suitable for making incidence readings. This is just one version.

\$\endgroup\$
1
\$\begingroup\$

The first practical light meter was a thermometer. Researchers in England circa 1900’s, took temperature readings in sunlight and in shadow, a chart was consulted which returned the camera settings. Carts and tables gave hit or miss settings. The electric light meter arrived on the scene in the early 1930’s. These light meters averaged the scene’s brightness. Kodak laboratories discovered that one could take a close-up reading of a yellow film box. Seems this box, when properly exposed was rendered as a middle-gray tone on the finished film/print.

This discovery led to the use of a target placard with an 18% reflective surface. We call this a “gray card”. The 18% gray card became a useful tool for exposure determination. An alternative to carrying a gray card was a translucent cover that mounted over the entrance of the light meter. This device furnished a light meter reading that matched a reflective 18% gray card reading.

In this method, the meter is pointed, not at the subject but in a direction that reads the light that is about to play on the subject. The method is called an “incident” light reading. Incident is old French for about to happen.

The theory of these methods, if you can figure out the exposure setting that will render one tone of the scene correct, the rest fall in line by law. Thus, rendering the 18% “middle gray” correct, provides the best basis for exposure.

Let me add, Cine photographers (movie camera operators) relied almost exclusively on the incident method. Considering the cost expended on a Hollowood movie shoot, they were under pressure to get the exposure right the first time.

Today, our cameras use sophisticated metering technology coupled with computer chip logic, all this and the experience of the photographer, get the job done.

\$\endgroup\$
5
  • \$\begingroup\$ Good answer, but small quibble... this is the third time I've read Alan's aside claim that 'incident' is old French for "about to happen". That's not quite right, and a misunderstanding of the etymology of the word. \$\endgroup\$
    – osullic
    May 28, 2023 at 16:59
  • \$\begingroup\$ late Middle English: via Old French from Latin incident- ‘falling upon, happening to’, from the verb incidere, from in- ‘upon’ + cadere ‘to fall’. \$\endgroup\$ May 30, 2023 at 16:16
  • \$\begingroup\$ Yep, exactly. Not the same at all as "about to happen" which is a temporal concept. The actual etymology relates to an action/effect. \$\endgroup\$
    – osullic
    May 30, 2023 at 18:12
  • \$\begingroup\$ Definitions from Oxford Languages in·ci·dent1. likely to happen because of; resulting from. "the changes incident to economic development" 2. (especially of light or other radiation) falling on or striking something. "when an ion beam is incident on a surface" noun an event or occurrence. \$\endgroup\$ May 31, 2023 at 15:20
  • \$\begingroup\$ Yes, correct. In all of these definitions, incident refers to something happening or causing an action/effect, whether that "something" is light rays or economic development or an ion beam. The idea that 'incident' means or meant simply "about to happen" or "about to occur" is what's not correct. \$\endgroup\$
    – osullic
    May 31, 2023 at 15:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.