What would Hollywood do?

David Lyga · Dec 18, 2014

Incident vs reflected light readings are not always the same. Which is correct? Understand the following scenario:

You have a group of twenty dark-skinned men from Africa sitting for a group portrait in the open shade, lit by open, cloudless sky, at midday. They are clothed in rather dark attire and the background is neutral. The film being used is T Max 100. The incident meter reading is EV 11, the reflectance meter reading is EV 9.

You have a group of twenty light-skinned women from Scandinavia sitting for a group portrait in the open shade, lit by open, cloudless sky, at midday. They are clothed in rather light attire and the background is neutral. The film being used is T Max 100. The incident meter reading is EV 11, the reflectance meter reading is EV 13.

Which meter reading is correct? My own answer to this question is that both are wrong and a compromise must be reached for best exposure. Thus, I would opt for a reading of EV 10 for the African men and a reading of EV 12 for the Scandinavian women, a decision which both respects and refutes the two rather disparate metering systems.

One is tempted to say that since both scenes are lit identically, the exposure for both would be likewise identical. But this just might not be true since one must, in such similar, but also dissimilar, situations, bias the exposure slightly, not completely, towards the predominance of tones in the particular scene. I cannot formulate a specific theory backing this up but, intuitively, it does seem to be the correct thing to do.

What would Hollywood do? - David Lyga

bdial · Dec 18, 2014

Hollywood would use an incident reading and modify the light if need be.

Unless you compensate, your reflective readings are only good if you're looking to render those tones at 18%. Once you place those zones where they should be you're going to be very close to the incident reading, at least, that's been my experience.

David Lyga · Dec 18, 2014

Thus, the compromise I proposed coalesces with what you infer. - David Lyga

DREW WILEY · Dec 18, 2014

Just ask the folks who repair & calibrate the light meters in Hollywood, or talk to a few actual cameramen. Pentax digital spotmeters are very popular, along with a few similar models. These make that kind of problem easy. But the exact exposure placement obviously is affected by the creative intent, the specific film, and whatever lighting range in the scene must be accommodated. Movie films tend to have even wider latitudes than color neg still film, but I've never had an issue with ethnic skintones. A classic dilemma might be a situations like a outdoor lit wedding, with not only a mix of races, but of dominant bright white garments versus dark suits etc. When in doubt, test more than one kind of film under analogous situations first. But here there's a more specific question: TMax black and white. Well, it's got a ton of wiggle room because it has a steep toe and long scale. But you do have to meter for the shadows because they drop
hard when they do, and watch out not to blow out the highlights. I've done lots of portrait work with this particular film in various formats,
35mm clear up to 8x10. Basic Zone System theory makes it easy enough. But since I successfully use spotmeters myself, that what I'll
recommend.

DREW WILEY · Dec 18, 2014

Oh, I do tend to use a pale light yellow-green filter for portraiture with TMax100. That's because it has somewhat more red sensitivity than
most pan film and can come out a bit pasty with pale complexions. (Hopefully, nobody will be wearing a peculiar shade of lipstick or makeup;
but you can always look thru the filter first to see the effect). If you have your ASA correct for your particular developer (I use the box speed of 100 for pyro, and a one stop compensation for this particular filter), then you can use a simple gray card reading, and everything will come out pretty much like the old "wing it" formula: pale Caucasian skintones will come out about a stop above middle gray (at Zone VI), very dark skin around a stop below (Zone IV), and most others somewhere in between. That leaves plenty of extra range for everything else potentially in the scene.

benjiboy · Dec 18, 2014

Are you shooting negative or reversal film?

DREW WILEY · Dec 18, 2014

He already stated TMax100 black and white.

Sirius Glass · Dec 18, 2014

bdial said:
Hollywood would use an incident reading and modify the light if need be.

Unless you compensate, your reflective readings are only good if you're looking to render those tones at 18%. Once you place those zones where they should be you're going to be very close to the incident reading, at least, that's been my experience.

Wot he done did gone and said.

Old-N-Feeble · Dec 18, 2014

Right... like charcoal on snow. Do you meter off of the charcoal or the snow? Both!! If you're shooting B&W negative film then you'll meter off the charcoal and set exposure to minus 2-4 EV, depending on how dark you want it in the print. Then you decide how much textural detail you want in the snow and where you want the high values placed so you can determine development time to adjust contrast on the film. If adjusted development time is necessary then tweak the exposure one way or the other to compensate for the speed change of the film as affected by development time.

DREW WILEY · Dec 19, 2014

One flaw in the general concept is that skintones are actually fairly reflective, so that even darker complexions have facial features significantly modeled by the incidence of the light, and can also have quite a bit of microtonality for the same reason. So they're never really that dark. It's not like reading a patch of flat paint. Therefore you're still going to be somewhere near the middle of the film range when it comes to exposure setting. In the studio this can obviously be fine-tuned by how the lighting is used, but even outdoors in strong lighting TMax can easily handle the situation if you meter sensibly.

BrianShaw · Dec 19, 2014

cowanw · Dec 19, 2014

David I thought 4 stops between Black skin and Scandinavian skin is a pretty big difference. Did you find this in real life?

Old-N-Feeble · Dec 19, 2014

I don't know about David but I never set exposure of even the darkest-skinned person to below EV-1 and never even the very lightest-skinned person to beyond EV+1. That's only 2 EV difference. When there was a mix of extreme skin tones then I erred toward overexposure unless shooting chromes.

DREW WILEY · Dec 19, 2014

ditto

Vilk · Dec 19, 2014

David Lyga said:
Thus, the compromise I proposed coalesces with what you infer.

:confused:

cliveh · Dec 19, 2014

bdial said:
Hollywood would use an incident reading and modify the light if need be.

+1

pbromaghin · Dec 19, 2014

In open shade, why use anything but an incident meter? Dark skin looks dark. The light skin next to it looks light. Just like it should. What's the big deal?

David Lyga · Dec 19, 2014

Well, my original answer remains the same: EV 10 for the Africans and EV 12 for the Scandinavians. True, black skin can 'shine' but even so, with the dark attire as well, I think that they deserve four times (NOT sixteen times) as much exposure as the Northern Ladies with their light blouses.

cowanw: Yes, four stops for such a difference in tonality is a 'real life' truth. Really, there is quite a difference here. I think that the attributes of both meters must be taken into account here. That is why I posited compromise and I think that Hollywood would agree, given the values contributed by both types of metering. - David Lyga

David Lyga · Dec 19, 2014

pbromaghin said:
In open shade, why use anything but an incident meter? Dark skin looks dark. The light skin next to it looks light. Just like it should. What's the big deal?

The 'big deal', I think, is that doing such is wrong. The dark skin needs some 'help' (which 'incident' does not offer), and the bright skin needs some 'cover' (which, likewise, 'incident' does not offer). Incident readings take no account of, provide no factor for, tonality and assume that everying is a mid tone. Reflectance takes no account of the AMOUNT of light hitting the subject and throws all its weight solely onto the part coming back to the camera. In effect, by compromising we are providing a needed 'bias' which would serve to slighly exaggerate the best attibutes of the varied skin tones.

Using purely incident readings would ignore this need. Using purely reflectance readings would serve to destroy the needed 'nuance' by (falsely) exaggerating the difference. Am I correct? - David Lyga

Old-N-Feeble · Dec 19, 2014

pbromaghin said:
In open shade, why use anything but an incident meter? Dark skin looks dark. The light skin next to it looks light. Just like it should. What's the big deal?

This is oversimplification. Incident metering does not resolve all problems... nor does reflected/spot metering. Only lots of experience teaches us how to cope with "meter readings" and how they transpose onto film and how this affects the final print.

David Lyga · Dec 19, 2014

Thus, maybe 'Old-N-Feeble' is neither too old nor too feeble. - David Lyga

Old-N-Feeble · Dec 19, 2014

David... I learned the hard way when I was a teen. I have forgotten much but not everything... perhaps because I worked in the field for three decades so certain challenges recurred from time-to-time. It's true I'm not that old but I'm definitely feeble with failing memory, decreasing cognitive plasticity and very limited mobility. I just hope to make one or two last images of some minor visual value before I'm gone.

My point is, I agree with your statements regarding human skin and exposure.

David Lyga · Dec 19, 2014

When you are gone so I will be gone. For the time being, keep being so very coherent, as you usually are so adept at being.

You said it right through inference: neither meter is a cure-all and what each type says has merit, but such merit needs to be modified. - David Lyga

pbromaghin · Dec 19, 2014

Old-N-Feeble said:
This is oversimplification. Incident metering does not resolve all problems... nor does reflected/spot metering. Only lots of experience teaches us how to cope with "meter readings" and how they transpose onto film and how this affects the final print.

Yes, it's oversimplication. But how complicated is open shade?

Old-N-Feeble · Dec 19, 2014

What does "open shade lighting" have to do with it other than decreasing overall contrast and micro-contrast by reduction or elimination of specular highlights? If one underexposes a persons skin tones they're forever lost on film and if one overexposes (blows out) skin tones on film they're forever lost. This thread began with a description of a scenario of lost skin tones. If it's morphing into something else then please clarify.

What would Hollywood do?

Privacy & Transparency

Privacy & Transparency