Light reading when facing sun

[Godfrey]

Incident meters are designed to be read pointed at the light source for your main subject, meaning if the subject is in direct sun or shade, the meter's disc should be in the same light....

No. You point the incident meter at the camera from the position of the subject, not at the light, to get the correct reading. To wit:

Sekonic L-478 Owner's Manual p23:

4-2-1. Measuring with Incident Light Function ...

Use either extended or retracted (flat) lumisphere to measure incident light. Point the lumisphere at the camera (lens optical axis) from a position close to the subject and then measure. ...

Sekonic L-398A Owner's Manual p6:

Incident Light Measurement
...
c. From position where subject is to be measured, point Lumisphere q in direction of camera. ...

 

[Ko.Fe.]

If here is bright sun, I'm using S16. Shutter speed is film ISO. And then I just look at the amount of light of my subject. And adjust aperture.

 

[Chriscrawfordphoto]

No. You point the incident meter at the camera from the position of the subject, not at the light, to get the correct reading. To wit:

That's true in most cases, but incident meters do not accurately measure backlit scenes; a single reading with the dome pointed at the camera in such light will produce an overexposed image. Not a big deal with negative film, but the shot would be ruined on slide film or digital.

For backlit scenes, you have to use special techniques like the one I show in the tutorial I linked earlier in the thread.

 

[charjohncarter]

That's true in most cases, but incident meters do not accurately measure backlit scenes; a single reading with the dome pointed at the camera in such light will produce an overexposed image. Not a big deal with negative film, but the shot would be ruined on slide film or digital.


For backlit scenes, you have to use special techniques like the one I show in the tutorial I linked earlier in the thread.

Right: the tendency is to use incident meters with back lighting. And as Chris' tutorial simply explains that you will get an incorrect reading with just a one time incident reading. It isn't tricky to do one of two methods Chris suggests. The hard part is recognizing when real back lighting exists. And Chris helps with this too.

Here is one that ISN'T back lighting that I used one domed incident reading. It looks a little back light but really the foreground and the background are the same exposure. I just wanted the faces to be properly exposed, and I almost did it:

Kodak EktaChrome expired 2003 by John Carter, on Flickr

 

[jarski]

Direct light while reading something not too demanding? Solution: sunglasses 😎

 

[Godfrey]

That's true in most cases, but incident meters do not accurately measure backlit scenes; a single reading with the dome pointed at the camera in such light will produce an overexposed image. Not a big deal with negative film, but the shot would be ruined on slide film or digital.

For backlit scenes, you have to use special techniques like the one I show in the tutorial I linked earlier in the thread.

I wasn't saying no to that, just to the general notion of how to use an incident meter ... 🙂

I'll have to read your workflow. To be sure, I've been doing this so long I no longer remember how I figured out how to meter the scene since I just set the aperture and shutter speed from experience and get what I like. 😀

G

 

[olifaunt]

It sounds like it would be easier to just get a spot meter and point it at whatever part of the scene you want to be a midtone.

 

[MCTuomey]

No. You point the incident meter at the camera from the position of the subject, not at the light, to get the correct reading. To wit:

disagree - won't work consistently.

what I tried to say, poorly, is that the incident reading should be taken in the same light as the most important element of the subject, assuming you want to expose that element at "18% gray." For example, if you're making a picture of a person wearing a hat standing backlit, and you take an incident reading with the meter tight to the person's face under the hat with the disc facing shadow, you will get a reasonable exposure.

Same condition, if the camera was in direct light, and you pointed the meter at the camera standing in the vicinity of the subject, you would underexpose the subject's face.

that's all I was trying to say. Bob said it much more simply above.

 

[MCTuomey]

It sounds like it would be easier to just get a spot meter and point it at whatever part of the scene you want to be a midtone.

yep, which is why spotmeters are so nice to have. but if all you have or want is an incident meter, you have to learn how to use the thing properly.

 

[Godfrey]

disagree - won't work consistently.
...that's all I was trying to say. Bob said it much more simply above.

Rob-F described how to meter for backlit and partially backlit scenes well.

But disagreeing with the fundamental way that you use an incident meter by the manufacturer's instruction isn't worth commenting on further. Good luck with your experiments.

G

 

[charjohncarter]

Rob-F described how to meter for backlit and partially backlit scenes well.

Mostly I use and have forever Rob-F's method. But I have about 15 rolls of 120 E-6 film and I just ordered a roll of 35mm Ektachrome plus a roll of Elitechrome so I will to continue to use Chris' method for these low latitude films. So far it has worked well.

 

[Solinar]

Ultimately, how much of an adjustment you can make depends on the latitude of the film.

I myself will take both an incident and a reflected reading, and then decide which one to favor. If the incident says 1/125 at 5.6, and the reflected says f/16, I might try 1/125 at 8 and a half. And bracket!

I use negative film and that's good advice.

Years ago, I would use a fill flash to bring the shadows within two stops of a brightly lit background. Due to the age of my three favorite 35mm cameras, I rarely bring a flash with me.

 

[MCTuomey]

Rob-F described how to meter for backlit and partially backlit scenes well.

But disagreeing with the fundamental way that you use an incident meter by the manufacturer's instruction isn't worth commenting on further. Good luck with your experiments.

G

I think you just did comment further ... thanks for the tart encouragement. May your tolerance for differing opinions improve.

 

[retinax]

This discussion about whether to point the meter to the the light source or the camera is getting to what I think I'd the achilles' heel of incident meters. I think the manuals say point it to the camera because that will give something reasonable without additional interpretation, but it's not correct per se. Assume you're photographing, for the sake of simplicity, a tree trunk. As you move around and meter from the different angles from which you have part sun, part shade on the tree trunk, pointing the meter at the camera will give different readings. But you're photographing the same subject, just different proportions of it sunlit and in shade! Would you vary the exposure from different angles if you were using a spot meter? Probably not. You might want to bias it toward the sunlit or the shaded part, but that depends on your idea of the final photograph, not the angle to the sun and thus how much of the tree trunk in your picture is sunlit and how much in shade, as long as you have parts of both.
Pointing the meter at the light source will give an exposure for the highlights. If you know what you're doing, better do that, as well as a shadow reading, and make an informed decision rather than rely on the meter to find a compromise for you, IMHO.
But we're digressing from what the OP asked.

 

[retinax]

Finally we have the critical point: your intention. Yes, the light falling on the scene is the same and so if your intention is to render "the subject" the same, then the exposure would be the same. There's no need, in your example, to meter the scene again: the original metering of the sunlight would still apply since your intent (to render the sunlit portion normally) is the same. The light [ETA: falling on the subject] hasn't changed and neither has your intended rendering, so the exposure would be the same too.

Here's a concise answer to your question, OP, to which I might add that you need to remember making sure that you're not accidentally metering in the shade of your own body when metering for something sunlit.

 

[Chriscrawfordphoto]

This discussion about whether to point the meter to the the light source or the camera is getting to what I think I'd the achilles' heel of incident meters. I think the manuals say point it to the camera because that will give something reasonable without additional interpretation, but it's not correct per se. Assume you're photographing, for the sake of simplicity, a tree trunk. As you move around and meter from the different angles from which you have part sun, part shade on the tree trunk, pointing the meter at the camera will give different readings. But you're photographing the same subject, just different proportions of it sunlit and in shade! Would you vary the exposure from different angles if you were using a spot meter? Probably not. You might want to bias it toward the sunlit or the shaded part, but that depends on your idea of the final photograph, not the angle to the sun and thus how much of the tree trunk in your picture is sunlit and how much in shade, as long as you have parts of both.
Pointing the meter at the light source will give an exposure for the highlights. If you know what you're doing, better do that, as well as a shadow reading, and make an informed decision rather than rely on the meter to find a compromise for you, IMHO.
But we're digressing from what the OP asked.

The white dome on an incident meter simulates the three-dimensionality of most real-world subjects, and allows the meter to take into account light coming from the sides with a single camera-direction reading. This works great most of the time, but it fails in backlit or severely sidelit scenes.

 

[Godfrey]

I think you just did comment further ... thanks for the tart encouragement. May your tolerance for differing opinions improve.

Not an opinion. Quotes from the manufacturers' instructions.

I've been using those instructions to meter with incident meters since 1967. It's given me perfect reference exposures for all average subjects (and of course backlighting is not an "average subject" situation).

The purpose of the hemispherical integrating dome and pointing to the camera position without shading the meter is so that the meter sees the light incident to the subject from the perspective of the camera lens. That's why they recommend this metering technique.

I just quote from credible sources what I know works, that is, from the people who designed the meters and how to use them. But you'll argue as if you know better anyway. Enjoy your experimentations.

G

 

[Rob-F]

This discussion about whether to point the meter to the the light source or the camera is getting to what I think I'd the achilles' heel of incident meters. I think the manuals say point it to the camera because that will give something reasonable without additional interpretation, but it's not correct per se. Assume you're photographing, for the sake of simplicity, a tree trunk. As you move around and meter from the different angles from which you have part sun, part shade on the tree trunk, pointing the meter at the camera will give different readings. But you're photographing the same subject, just different proportions of it sunlit and in shade! Would you vary the exposure from different angles if you were using a spot meter? Probably not. You might want to bias it toward the sunlit or the shaded part, but that depends on your idea of the final photograph, not the angle to the sun and thus how much of the tree trunk in your picture is sunlit and how much in shade, as long as you have parts of both.
Pointing the meter at the light source will give an exposure for the highlights. If you know what you're doing, better do that, as well as a shadow reading, and make an informed decision rather than rely on the meter to find a compromise for you, IMHO.
But we're digressing from what the OP asked.

This is an important point Retinax is making. Pointing the meter at the sun will give the shortest exposure, since it maximizes the amount of light striking the meter. So it will result in all areas of the photo, from highlight to shadow, receiving the minimum exposure, and being rendered at a lower value. But we don't want highlights rendered at a low value. We want them to look bright. That's why we call them highlights.

The hemispherical dome integrates the light from all directions over a 180 degree solid angle. When we point the meter at the camera, it gives an honest reading of the light striking the subject, as seen from the camera's position.

The incident meter will generally give the same exposure as a reflected reading taken from an 18 percent gray card. Either method will give an exposure that is at the midpoint (is it OK if I call that zone 5), allowing the number of stops from highlight (zone 9) to midpoint to equal the number of stops from midpoint to shadow (zone 1). The middle tone in the subject will then be rendered as a middle tone, while all others will fall where they belong in relation to it. The highlights look bright, and the shadows look dark. That's the theory behind the incident meter.

Sometimes the subject is too light overall, and we want to bring it down to a more medium value. And sometimes it is too dark overall, and we want to bring it up to a medium value. Then we use a reflected light meter. And sometimes we take both readings and average them. The Spectra Combi 2 has provision for an incident reading; a reflected reading; or an average of the two, with a single button press!

When to do what? Judgment will always be required.

 

[Richard G]

When to do what? Judgment will always be required.

The most important line in the thread.

 

[Godfrey]

The most important line in the thread.

100% Agreed!