Does one sometimes also use exposure compensation on incident metering?

[BobD]

Yes, but that was not the question. The question was getting the exposure before adjusting for filters.

The question is:
"Does one sometimes also use exposure compensation on incident metering?"

There is no mention of filters or before or after anything in the question.

Why not just admit you made a mistake? You won't die. I promise.

 

[RalphLambrecht]

I have noticed that they always talk about 50% grey for reflective metering, but it seems unclear if this is what incident metering works towards as well, so it is unclear if I should make the same EC decisions here?

no compensation required; leave that to whoever calibrated your meter(it's manufacturer).

 

[MattKing]

This is an example of a scene where I believe that I used incident metering, and I adjusted the exposure from what was recommended:

The light was low, of low contrast, and the subject was dark.
So I made a decision to add additional exposure, in order to have a negative with more to work with.
I wouldn't have done that if the scene had a wider Subject Luminance Range. The narrow SLR (aka SBR) allows you to move the exposure up the curve, without damaging the highlights.
This, by the way, is a scan of a print.

 

[Bill Burk]

This is an example of a scene where I believe that I used incident metering, and I adjusted the exposure from what was recommended:
View attachment 275922

The light was low, of low contrast, and the subject was dark.
So I made a decision to add additional exposure, in order to have a negative with more to work with.
I wouldn't have done that if the scene had a wider Subject Luminance Range. The narrow SLR (aka SBR) allows you to move the exposure up the curve, without damaging the highlights.
This, by the way, is a scan of a print.

This is a perfect example of when you would compensate because you know you can and you know it can help.

 

[Pieter12]

This is an example of a scene where I believe that I used incident metering, and I adjusted the exposure from what was recommended:
View attachment 275922

The light was low, of low contrast, and the subject was dark.
So I made a decision to add additional exposure, in order to have a negative with more to work with.
I wouldn't have done that if the scene had a wider Subject Luminance Range. The narrow SLR (aka SBR) allows you to move the exposure up the curve, without damaging the highlights.
This, by the way, is a scan of a print.

Curious as to where the meter was positioned?

 

[Alan Edward Klein]

The only time I rigorously used incident metering was for commercial color product shots, studio under hot lights, in the field natural light. Otherwise I find reflective meter to work better for me, I carry a small 18% gray card or use my the palm of my hand if I'm in trickily and bracket. If you shoot sheet film you use Phil Davis Beyond The Zone System, he advocated using a incident meter to determine scene brightness range, and then the magic or was it the wonder wheel or software to calculate shutter and F stop for a given film.

What did you mean, Paul?

 

[Mr Bill]

The short answer is "No, never."

Agreed.

Love the short answers. Too often technical types start answering questions you didn't even ask, and you get even more confused

Unfortunately the short answer given is wrong. It would have been better to to have answered, "No, not usually."

Speaking as the so-called technical type I don't wanna confuse you, so you might wanna stop reading here (I'm gonna give a couple examples). First is the example of detail shots of an old-time time steam locomotive in a flat-black finish. If you shoot this with an unadjusted incident meter reading, you'll may end up thinking, "One day I may have another opportunity to try this." A second example is in doing studio portrait work. If you have subjects with VERY dark complexions, shot on something like Portra 160 film and printed on a professional color paper, this will be marginally in the tolerance of the film. If the subject is wearing any dark, near black clothing, those parts may get very grainy on a moderate sized enlargement. I say this about portrait work on the basis of extensive real experience.

Running off on a tangent I am an admirer of the (long deceased) physicist Richard Feynman. In one of his lectures he made the point that he always likes to teach things "correctly" from the start, meaning that the student will never have to go back and "unlearn" anything. But there was at least one place where he stated that some specific thing was too difficult to learn immediately. So he was gonna teach some wrong things in order for the students to get a grasp on the subject, after which they would then have to come back and unlearn the wrong things.

In this situation with the incident meter there is no need to teach the wrong way - it is easy enough to say "most of the time" rather than "never." Then, as you gain experience by screwing up, you can modify your understanding of weaknesses of metering techniques. There are actually a handful of fine points related to incident meters that can significantly affect the reading.

This loco is one where straight up incident metering would have failed...

 

[MattKing]

Curious as to where the meter was positioned?

In the cleft itself, but not in the shadow of the stump, pointing back at the camera.
I used the light integrating dome, because that is the only choice on a Digiflash or Profisix F (can't remember which).
I probably took an incident reading in the fully shadowed part as well, to satisfy curiosity.

 

[Sirius Glass]

Unfortunately the short answer given is wrong. It would have been better to to have answered, "No, not usually."

Speaking as the so-called technical type I don't wanna confuse you, so you might wanna stop reading here (I'm gonna give a couple examples). First is the example of detail shots of an old-time time steam locomotive in a flat-black finish. If you shoot this with an unadjusted incident meter reading, you'll may end up thinking, "One day I may have another opportunity to try this." A second example is in doing studio portrait work. If you have subjects with VERY dark complexions, shot on something like Portra 160 film and printed on a professional color paper, this will be marginally in the tolerance of the film. If the subject is wearing any dark, near black clothing, those parts may get very grainy on a moderate sized enlargement. I say this about portrait work on the basis of extensive real experience.

Running off on a tangent I am an admirer of the (long deceased) physicist Richard Feynman. In one of his lectures he made the point that he always likes to teach things "correctly" from the start, meaning that the student will never have to go back and "unlearn" anything. But there was at least one place where he stated that some specific thing was too difficult to learn immediately. So he was gonna teach some wrong things in order for the students to get a grasp on the subject, after which they would then have to come back and unlearn the wrong things.

In this situation with the incident meter there is no need to teach the wrong way - it is easy enough to say "most of the time" rather than "never." Then, as you gain experience by screwing up, you can modify your understanding of weaknesses of metering techniques. There are actually a handful of fine points related to incident meters that can significantly affect the reading.

This loco is one where straight up incident metering would have failed...

Sometime giving a full answer overwhelms the person who asked the question. I rather is a usable answer followed by a conversation that develops answers for the near future. It the inquisitive one wants more details, I take the time to explain them. We are in a running conversation on Photrio, so people are free to come by and delve deeper. In the past there was one knowledgeable poster who would provided extremely detailed answers to Hasselblad questions that would befuddle people who were knowledgeable about Hasselblads. Ask me a question and I will, if able to answer it. When you want more detail I will provide it. When one has a one time opportunity to take a photograph, then if needed take a backup with a variation. However I still do not understand the added adjustment you made to the photograph of the engine.

 

[Bill Burk]

Unfortunately the short answer given is wrong. It would have been better to to have answered, "No, not usually."

Sure, I agree with you too. But what would be the right answers on a true false quiz?

If I were going to make a list of advantages and disadvantage for different types of meters it would look like this:

 

[BrianShaw]

Touche

Checkmate

 

[Mr Bill]

Sometime giving a full answer overwhelms the person who asked the question. I rather is a usable answer...

Well, great! But in this case, instead of your answer, "No, never," wouldn't you agree that "No, most of the time," would have been a better answer?

Here's my beef with your answer: you don't know what impact your answer will have on the OP. If they see you as an ultimate authority they may lock into their mind, "never." So when they take that as an absolute, and IF they eventually run into severe exposure errors, their brain is initially crippled in attempts to troubleshoot. They "KNOW" that their metering is not at fault cuz you "NEVER" compensate for an incident meter reading. They chase the wrong things - did I get bad film, was it stored wrong, was my processing bad, etc.? Or they may blow a once-in-a-lifetime opportunity cuz they knew definitely that there was no need to bracket exposure.

I may be especially sensitive to these things because I personally had a number of such biases in my younger days. And they were difficult to overcome until I eventually accepted that my original sources were, shall we say, not that good. In 40+ years of full-time work in photography, largely tech work, I've worked with A LOT of people with similar mental blocks in certain specific areas. My experience is that It is much more difficult to overcome these misconceptions than it was for them to originally be implanted.

Anyway, I just don't understand how you can continue to defend your use of "never" rather than "most of the time"

 

[BrianShaw]

The topic is exposure compensation with incident light metering. Even though two extreme examples have been offered it isn’t clear to me how much EC was given or if it even substantially improved the exposure. I suspect not.

For me the best (and perhaps only needed) “correction” has been that regarding the use of filters.

Lets face it... when incident reading is taken in same or equivalent light as the subject, it’s darn near impossible to get an exposure that is unworkable.

 

[Bill Burk]

Well, great! But in this case, instead of your answer, "No, never," wouldn't you agree that "No, most of the time," would have been a better answer?

Here's my beef with your answer: you don't know what impact your answer will have on the OP. If they see you as an ultimate authority they may lock into their mind, "never." So when they take that as an absolute, and IF they eventually run into severe exposure errors, their brain is initially crippled in attempts to troubleshoot. They "KNOW" that their metering is not at fault cuz you "NEVER" compensate for an incident meter reading. They chase the wrong things - did I get bad film, was it stored wrong, was my processing bad, etc.? Or they may blow a once-in-a-lifetime opportunity cuz they knew definitely that there was no need to bracket exposure.

I may be especially sensitive to these things because I personally had a number of such biases in my younger days. And they were difficult to overcome until I eventually accepted that my original sources were, shall we say, not that good. In 40+ years of full-time work in photography, largely tech work, I've worked with A LOT of people with similar mental blocks in certain specific areas. My experience is that It is much more difficult to overcome these misconceptions than it was for them to originally be implanted.

Anyway, I just don't understand how you can continue to defend your use of "never" rather than "most of the time"

I have no problem with your longer, more correct answer. You are right. And my response wasn’t meant to throw back at you. Yes I need to be more responsible with answers.

When it comes to basically understanding how a meter sees in terms of average gray, the fundamental difference between the two is with a reflected light meter you must adjust if the scene is not average, otherwise for example your snow will be gray.

With an incident meter, the meter isn’t fooled by subjects whose reflectance distribution works out to something that’s not average gray. So you don’t compensate, the incident meter already gives a result that does not depend on subject reflectance.

The longer answer you gave applies as well. You’re a good source of knowledge here.

The Kodak Gray Card is an example of what we are talking about. The short answer to using the gray card is to hold it up and meter it and use what the meter says. The Kodak Gray Card instruction sheet goes into more detail about what angle to hold it and how you still compensate in cases where the subject has unusual distribution of reflectance.

 

[DMJ]

The problem are not the answers here. This kind of question would've been bashed in a place like Stackoverflow.

If you need to compensate for an incident reading, your metering technique is wrong. The point of using a meter is that you get shutter and aperture values that you can dial-in in your camera. Incident metering can be a very creative (and precise) tool. I would only dial-in exposure compensation in the meter (not in the camera) to deal with specific films and or/camera issues. My light meter has a "Correction Values" settings that can be used when using light-reducing filters. Incident metering is not just point at the camera and meter. It can be used with the Zone System as explained in the book "Beyond the Zone System" (already mentioned). Basically, it gives you an average of light falling at the dome of the meter from 180 degrees. Knowing that fact, your experience and creativity will help you make decisions on where to meter. Should be place in front of the camera? Maybe. Is this how I meter ? No. Now, if you are shooting a landscape form inside a cave, your are using the wrong tool.

Short answer now: You achieve "exposure compensation" by placing the meter at the right spot.

 

[wiltw]

If one starts with the fundamental assumption that reflected light metering assumes that either...

• the scene content averages to 18% grey, or
• the target is assumed to be a mid-tone (18% grey) target

then when deviations occur, such as the classic bride-in-white-snow scene or the tuxedo-in-black-coal-mine examples, then one can understand that 'Exposure Compensation' was provided as a means to adjust the reflected light reading whenever neither of the above assumptions is True.

And if that is assumed to be the rationale for the creation of 'Exposure Compensation', then EC is a superfluous concept for an Incident meter, since the incident meter has no knowledge of the scene, it sees only the light falling on the scene.​

A completely separate concept is that a particular scene is 'best exposed' at a combination of settings that neither the reflected light meter nor incident light meter suggests, and because the photographer feels that this recording is 'better' than one that a 'normal subject' might need, it is an 'adjusted exposure' to suit what the photographer's mind desires, rather than what either instrument suggests.
The photos in posts 29 and 33 are both clearly an example of that second concept, where the details in the shadow areas is the key element of the shot, which would have been lost if exposed only by the 'exposure based upon the light falling on the scene' concept fundamental to the incident meter. (The reflected light meter would have suggested an exposure 'too bright', in which the black and near-black comes out too 'midtone'...neither reflected nor incident would have been 'best'.)

 

[darkroommike]

"Generic incident light meter" (or, I suppose, "generic reflected light meter") is a unicorn. Meters all have their idiosyncrasies, battery meters will do much better in low light that one that is self powered, and old tired selenium cells can be weak, too. It's important to test what ever meter you use and yes, you may need to use exposure compensation under low light conditions if it has been a few decades since your meter has been serviced.

 

[Mr Bill]

But what would be the right answers on a true false quiz?

Bill, I don't think a true/false quiz is really relevant here. Something I've observed about such quizzes in the photographic realm is that the "correct" answers are more a reflection of what level the instructor intends it to be. At a sort of novice level one answer is "correct." ( your chart, for example.) At a higher level the answers are "wrong."

At any rate such quizzes are mainly a means for an instructor to evaluate what a student has "learned." They don't really have much bearing on real-world professional photography.

Maybe I should give some of my background. I spent probably about 15 or 20 years heavily involved with professional level photo lab work with a large portrait chain outfit in the US. A lot of this was as QC manager, staff of 5 or 6 people overseeing mainly process control, chem mix results as well as chemical analysis for regeneration of certain chemicals. Unusual color problems, they would ultimately end up in my department. In later years heavily involved with studio design and lighting systems.

The company had several dozen field techs for studio installation/maintenance. (Mainly they did mechanical installation of lighting, motorized background systems, shoot anchors into the floor, run network systems, etc. I'm thinking about 40 incident meters in their tool kits (of course these gradually "disappeared" on a regular basis). Probably about 80% of the techs could get mostly reliable incident meter readings, but the problems were so bad that we WOULD NOT ALLOW a new studio to open until we had received and processed test film. Until you deal with this sort of thing it's hard to believe it can be so difficult to explain to a reasonably competent tech person how to get reasonably accurate incident meter readings. I think perhaps the people here who think it's easy (and reliable) are primarily photographers who actually have a much deeper knowledge base than they realize - they are already tuned in to many of the pitfalls so they don't make the obvious mistakes. If something is way off the experience photographer tends to notice, and think, "that doesn't seem right". So it may seem simple. But apparently it's not.

Case in point... a couple hours drive from our headquarters (a good-sized Midwest USA city) was a university with a pretty strong photography program. When I was the QC guy I used to get 3rd year photo students as summer interns (a couple times longer, maybe, at the request of the school, if the student needed to make a little more money). The first few years I was a little surprised at how useless (to us) the students were. Only one, as I recall (out of 10 or 12?) was worthwhile to the company cost-wise (he had run their school processing lab). The rule was that i had to give them some experience pertinent to their studies, but could use them for SOME drudge work. So if we had some studio test projects on the back burner maybe we'd have them work on that. As I recall they generally needed something like a 20 or 30 minute hands-on "orientation" session with an incident meter in one of our test studios in order to even get their first reading, much less know if it was legit. Third-year university photography students! (I have yet to have my first formal photography class, so please forgive my ignorance about what is taught.)

 

[BrianShaw]

A completely separate concept is that a particular scene is 'best exposed' at a combination of settings that neither the reflected light meter nor incident light meter suggests, and because the photographer feels that this recording is 'better' than one that a 'normal subject' might need, it is an 'adjusted exposure' to suit what the photographer's mind desires, rather than what either instrument suggests.

At first I thought this to be wordsmithing but upon reflection is very important thought. A good distinction to remember even though both are exposure adjustments.

 

[Pieter12]

if it has been a few decades since your meter has been serviced.

Pretty hard to get a meter serviced nowadays.

 

[Mr Bill]

If one starts with the fundamental assumption that reflected light metering assumes that either...

• the scene content averages to 18% grey, or
• the target is assumed to be a mid-tone (18% grey) target

then when deviations occur, such as the classic bride-in-white-snow scene or the tuxedo-in-black-coal-mine examples, then one can understand that 'Exposure Compensation' was provided as a means to adjust the reflected light reading whenever neither of the above assumptions is True.

And if that is assumed to be the rationale for the creation of 'Exposure Compensation', then EC is a superfluous concept for an Incident meter, since the incident meter has no knowledge of the scene, it sees only the light falling on the scene.​

A completely separate concept is that a particular scene is 'best exposed' at a combination of settings that neither the reflected light meter nor incident light meter suggests, and because the photographer feels that this recording is 'better' than one that a 'normal subject' might need, it is an 'adjusted exposure' to suit what the photographer's mind desires, rather than what either instrument suggests.
The photos in posts 29 and 33 are both clearly an example of that second concept, where the details in the shadow areas is the key element of the shot, which would have been lost if exposed only by the 'exposure based upon the light falling on the scene' concept fundamental to the incident meter. (The reflected light meter would have suggested an exposure 'too bright', in which the black and near-black comes out too 'midtone'...neither reflected nor incident would have been 'best'.)

I agree 100% with what you said, but I would add an extra point. The lighting must be primarily frontal, meaning from the general direction of the camera.

There's one more obvious pitfall, really beyond the scope of this thread, but significant for color photography. Briefly, one cannot use a color film in an unknown lighting condition and expect an incident meter to give a usable reading.

I'm sure you (wilt) already know this (your posts are always on the money), but... Consider the case of a daylight balanced color film. If one takes this into a tungsten lighting situation, unknowingly, the light is gonna be deficient in bluish light, compared to reddish light, by roughly 2 f-stops. (The film data sheets will describe how to deal with this.) But the novice photographer, trusting the incident meter, will end up with the blue-sensitive layer underexposed by about 2 f-stops. The best emergency workaround (short of adding flash), for a color-neg film, is to increase exposure by a stop or two. This will at least get the yellow dye layer (blue sensitive) up off the toe, and into a usable range - more or less correctable at printing.

 

[Luckless]

Any light metering method gives you a starting value.

A reflective meter reading requires more careful interpretation to settle on a meaningful starting value, while an incident meter requires less.

In simple cases with basic lighting you can directly use a reliable incident meter's reading without needing much additional consideration of it, and you will get what most photographers would regard as a 'usable exposure'. This is more than enough for a wide range of photographic efforts. Even taking one step back and using Sunny 16 rules to determine your exposure settings will result in very useable and perfectly acceptable exposures for your photographs.

But your meter is still a tool to be used in the photographic arts, and you can always explore ways to push your tools further and refine your photographic expression to achieve images that may be more to your liking and goals.

This all comes down to understand what the tool is actually doing, and what your other tools [the camera's lens, shutter, and film selection] can do, and then learning what adjustments are needed for how the light is in order to achieve the images that you are after.


You don't have to make use of any of it, but you should at least eventually study and experiment enough to be aware of it before dismissing it entirely.

 

[BrianShaw]

Gosh... and the there’s this:

“The exposure increase in daylight may vary with the angle of the sun and the time of day. In the late afternoon or the winter months, when daylight contains more red light, green and blue filters may need slightly more exposure than usual.”

 

[Bill Burk]

Mr Bill,

I'm humbled by your experience. You are right, it is more correct to say, "No, Not usually". And your detailed stories clearly illustrate why, keep 'em coming.

I still think "No, Never" is funny. Of course it's wrong, but it emphasizes the chief difference between an incident meter and a reflected light meter.

I really want to get the point across, that an incident meter doesn't read significantly differently if you have a black cat in a coal mine or a polar bear in the snow, because it's looking at the light. But a reflected light meter will need significant correction to avoid overexposing the black cat or underexposing the polar bear.

You and I have never been cross with each other. I'm sorry that my post seems to have rubbed you the wrong way, and I was a bit grumpy this morning too. My son's upset with me for tearing a page from his sketchbook, so I got comeuppance for my evil thoughts.

I'll leave this post with a positive thought. I came across an old email exchange while searching for another email. I had been doing some research about 12 years ago and was given an email address, so I had a brief correspondence with Dr RM Callender who studied Hurter and Driffield. In one of his emails he wrote:

* Ross's Law says that "exposure depends on the result you want". As an astronomer, Ross gave a long exposure to record faint stars, but shortened his exposures when photographing bright stars. It all makes sense.

 

[Mr Bill]

And your detailed stories clearly illustrate why, keep 'em coming.

Thanks, there's too many, though. It would bore 99% of the members here to death. Most of em are better told over beer in the local bar. Plus, I only tell the ones that make me look good. Fwiw I was somewhat lucky (?) to have worked at a, shall we say, frugal outfit. They never wanted to spend the money to hire real expertise - it was always, "find a way to do it with what you have." We used enough film and paper that we generally had access to some of the best technical experts from whatever vendors we were using. So I always saw myself as working for less than I was worth, but considered it as tuition for the education I was getting - deep inside access that would not have been possible hardly anywhere else. But everything was confidential so I would never post anything online unless I knew of a place where it was previously published (or it could be easily discovered independently). I think that a side effect of this is that some people probably think that I'm "internet educated," with little real-world experience. But the reality is that I learned just about everything from the ground up (except never had the "opportunity" to clean toilets though).

. I'm sorry that my post seems to have rubbed you the wrong way...

No, I'm always glad to discuss and argue points, and i think you're great for this. I know that you like to experiment with things, and reason your way through, and I applaud this sort of thing. I'd just say, keep learning and keep having fun with it.