incident light vs reflected light?

[antitrust]

i bought a photography book from borders (those brief overview but covers everything from equipment to even darkroom processing pretty well) recently and i've been reading abit.

and when it came to the light meter section, the book recommends an INCIDENT light meter over the camera's built in REFLECTED light meter as it is more accurate.

now the question is, why use INCIDENT light metering when after all, the light still needs to be REFLECTED off the subject into your sensor/film? :dunno:

 

[synapseman]

Okaayy, I'm no expert in this light metering thing. I just bought a lightmeter myself so did a bit of research and this is what I gathered...

Basically, reflected light is affected by the colour, texture and/or material of a subject, which in turn affects its reflectivity. If there's say a model wearing a black outfit, the reflected light hitting the camera sensor might indicate that it's a very dark scene, overcompensate and cause the blacks to be grey and most everything else to be blown out.

Conversely, shooting some shiny object may fool camera (reflected) meters into thinking there's too much light, cutting down on exposure and thus resulting in underexposure.

For incident light readings, the amount of light that falls onto a subject is the "correct" amount of light for the scene, and so exposure settings should be set to accomodate this amount of light.

(Er, experts out there: Is this right?)

 

[catchlights]

in a simple word.

let say a light has incident reading of f8 fall on a black object, you exposed it as f8 (according what incident meter suggested), it come out as black object.
if you use a reflected meter to take reading of the black object, it shows f16, so if you exposed it at f16, needless to say the black object will become gray tone.

so when you only have reflected light method to take reading, if you meter on a black object, you need to close down, when you meter on a white object, you need to open up.

Hope this help.

 

[roygoh]

in a simple word.

let say a light has incident reading of f8 fall on a black object, you exposed it as f8 (according what incident meter suggested), it come out as black object.
if you use a reflected meter to take reading of the black object, it shows f16, so if you exposed it at f16, needless to say the black object will become gray tone.

so when you only have reflected light method to take reading, if you meter on a black object, you need to close down, when you meter on a white object, you need to open up.

Hope this help.

Idea is right but I think you mixed up the f-numbers in your example. If an object shot with f/8 would turn out black, how can it turn out grey when shot with f/16 under the same lighting?

If incident light meter recommends f/8, then reflected light metering on a black object under the same lighting would probably recommend a larger aperture (f/4 maybe...), and reflected light metering on a white object would then recommend a smaller aperture (f/16).

 

[catchlights]

Idea is right but I think you mixed up the f-numbers in your example. If an object shot with f/8 would turn out black, how can it turn out grey when shot with f/16 under the same lighting?

If incident light meter recommends f/8, then reflected light metering on a black object under the same lighting would probably recommend a larger aperture (f/4 maybe...), and reflected light metering on a white object would then recommend a smaller aperture (f/16).

oppsss... you are right, I have it mixed up liao, thanks for pointing out the mistake.

 

[Splutter]

It's always easier to use incident metering to get more accurate results. However, sometimes it's just impossible to do so, especially in the case of taking photos of a mountain range a few thousand km away.

 

[creampuff]

I think you can get a more concise understanding of incident light metering here:
http://www.sekonic.com/classroom/classroom_21.asp

That said, incident light reading is often more accurate simply because it isn't affected by the intensity of light reflecting off the subject. This is particularly true in a studio context where more than one light source is used or if the subject or model varies.

Incident light reading can still be used to measure exposure for landscapes so long as the light intensity measured is the same as say the mountains in the background.

 

[student]

#1 INCIDENT LIGHT METERING IS NOT MORE ACCURATE THAN REFLECTED LIGHT METERING!

All metering methods are equally accurate for their intended purposes.

An incident light metering measures the amount of light falling on the subject. If the subject is black the amount of light reflected will be little and the sensor/film will record black. If the subject is white, there will be much reflection, and the sensor/film will record the subject as white. This has great advantage for situations such as studio lighting where one can bring the meter to the subject.

However as mentioned elsewhere, incident metering is not ideal for landscape photography, unless one is doing "micro-landscape photography" such as phptographying a small patch of grass/flowers right beside you.

For the usual landscape photography, a spot metering should be the preferred method. It is extremely rare that the light falling on the photographer would be exactly the same as the light falling on the mountains.

#2 HOWEVER, IT IS VERY DIFFICULT TO USE INCIDENT LIGHT METERING TO MEASURE THE RELATIONSHIP OF LIGHT VALUES WITHIN THE SAME SUBJECT IN THEIR CLOSE PROXIMITY.

IN SUCH SITUATIONS, INCIDENT METERING IS FAR INFERIOR TO SPOT METERING.

For example, if you want to know the difference in light values between the right side of the face and the left, the incident metering will not be extremely accurate, because the measurement is often from a light sensitive dome which receives light from all over.

This kind of accuracy may not be very important today, given the widespread use of digital process. But in analogue B&W photography, for accurate reading of light values for the purpose of deteriming how to develop the negative, spot metering is almost a must.

 

[Robbie J]

You need to have experience in using a separate hand-held light meter.
When I first got mine (some 16 years ago) I got pretty awful results(on slide film). It was only after using it regularly that I learned to judge the colour and reflectivity of the stuff I was photograhing... and make the necessary compensations. Of course that was with reference to shooting film. With digital things changed. Whereas with film one tended to overexpose slightly (negative film) with digital the game has changed somewhat. You can use your histogram to judge the correct exposure (after the fact) although this means an extra step or two in your workflow. The bottom line is good hand-held light meters are EXPENSIVE. If you can well afford it why not? If not, use the money to buy a better lens.

 

[lsisaxon]

It's always easier to use incident metering to get more accurate results. However, sometimes it's just impossible to do so, especially in the case of taking photos of a mountain range a few thousand km away.

Either make an assumption that the light falling in front of you is the same as the light falling on the mountain thousands of km away or use a spot meter.

 

[lsisaxon]

#1 INCIDENT LIGHT METERING IS NOT MORE ACCURATE THAN REFLECTED LIGHT METERING!

All metering methods are equally accurate for their intended purposes.

An incident light metering measures the amount of light falling on the subject. If the subject is black the amount of light reflected will be little and the sensor/film will record black. If the subject is white, there will be much reflection, and the sensor/film will record the subject as white. This has great advantage for situations such as studio lighting where one can bring the meter to the subject.

However as mentioned elsewhere, incident metering is not ideal for landscape photography, unless one is doing "micro-landscape photography" such as phptographying a small patch of grass/flowers right beside you.

For the usual landscape photography, a spot metering should be the preferred method. It is extremely rare that the light falling on the photographer would be exactly the same as the light falling on the mountains.

#2 HOWEVER, IT IS VERY DIFFICULT TO USE INCIDENT LIGHT METERING TO MEASURE THE RELATIONSHIP OF LIGHT VALUES WITHIN THE SAME SUBJECT IN THEIR CLOSE PROXIMITY.

IN SUCH SITUATIONS, INCIDENT METERING IS FAR INFERIOR TO SPOT METERING.

For example, if you want to know the difference in light values between the right side of the face and the left, the incident metering will not be extremely accurate, because the measurement is often from a light sensitive dome which receives light from all over.

This kind of accuracy may not be very important today, given the widespread use of digital process. But in analogue B&W photography, for accurate reading of light values for the purpose of deteriming how to develop the negative, spot metering is almost a must.

It is very likely that you can always change the dome to a disc for measuring the light from individual source, or on some meters, the dome can be retracted. Like what Splutter mentioned, it is easier to get a correct reading from an incident light meter than with a spot meter. But once mastered, the spot meter can really give what you hope to achieve.

 

[student]

It is very likely that you can always change the dome to a disc for measuring the light from individual source, or on some meters, the dome can be retracted. Like what Splutter mentioned, it is easier to get a correct reading from an incident light meter than with a spot meter. But once mastered, the spot meter can really give what you hope to achieve.

You are right that for most, it is easier to use an incident meter. Changing a dome to a disc does not matter much because it is still gathering light generally rather than specifically.

But incident metering is not inherently more accurate than reflected metering.

Spot metering gives a lot more information to any scene than any incident metering. Whether one wishes to use the information, or need to use the information is another matter.

 

[Splutter]

Doc, what I mean is that incident gives you the true measurement of light, while taking readings of reflected light gives you the percentage of how much light is reflected off. Afterall, the reflected reading will only be the same as the incident reading if you are measuring a 18% grey surface right?

 

[student]

Doc, what I mean is that incident gives you the true measurement of light, while taking readings of reflected light gives you the percentage of how much light is reflected off. Afterall, the reflected reading will only be the same as the incident reading if you are measuring a 18% grey surface right?

I am a little lost to try to understand what you are trying to say.

What is the "true measure of light?". I think scientists will have a different ways to measure light, but I do not think that is relevant to us here.

But let me put it in practical terms. Note that earlier I had written that the use of spot metering may not find a use for most digital users, given the powerful tools they have with PP.

To illustrate, I will use my own photography processes.

Let us say I am making a photograph of a flower next to a window. There is some light differences obviously on different parts of the petals. In my previsualisation, I wish to make the differences even more that what the amblient light was. In other words, I want to increase the contrast more than the ambient light. If I were photographing this flower digitally, I will just use curves to tweak the contrast. But since I am using B&W film, I will have to increase the development to increase contrast.

Using an incident meter will allow me to take a picture that will approximate the inherent contrast. But this meter cannot tell me how much to increase my development for what I want to achieve. The flower petals may only be something like 5 cm size. No incident metering can tell me the difference between the lighter and darker area of this petal. Only the spotmeter can do that.

 

[antitrust]

so after all that's been said, am i correct to draw the following conclusions:

1. Each type of metering has its own uses as discussed by student
2. INCIDENT metering is more useful when the lighting of the scene is more balanced (as discussed by catchlights for the black and white subject) in a sense unlike the example of the uneven light distribution on the flower petals that student discussed earlier on (in which case SPOT metering will be ideal)

but you see, either i'm missing out on a big big point here or that one of my questions still hasn't been answered. (pardon me. trying to digest all these info!)

so why is incident light metering more recommended? afterall, light still needs to be REFLECTED OFF THE SUBJECT into the film/sensor?

anyway, to all who has replied, thanks alot for sharing!

 

[student]

so why is incident light metering more recommended? afterall, light still needs to be REFLECTED OFF THE SUBJECT into the film/sensor?

Incident light is more recommended because

#1 It removes the problem of where and how to measure reflected light. A reflected light meter can take meterings from many parts of the subject/scene. And if you do not know much about light, wrong metering may ensue. Camera makers try to reduce this problem by such things like matrix metering, evaluative metering etc, where the light values are compared to a bank of other images and then give you an approximate. Most of the time the results are pretty good!

The incident meter takes away this problem, because how much light is reflected depends on the reflectivity of the subjects.

#2 It is a lot easier to learn to use the incident meter than the spot meter.

 

[creampuff]

All metering methods are equally accurate for their intended purposes.

An incident light metering measures the amount of light falling on the subject. If the subject is black the amount of light reflected will be little and the sensor/film will record black. If the subject is white, there will be much reflection, and the sensor/film will record the subject as white. This has great advantage for situations such as studio lighting where one can bring the meter to the subject.

The first part of your explanation is correct but your second sentance onwards (in blue) make no meaning as subject reflectance is never an issue when making incident readings.

Incident metering reads the intensity of light falling on the subject, it provides readings that will create accurate and consistent rendition of the subject’s tonality, color and contrasts regardless of reflectance, background color or brightness or subject texture.

However as mentioned elsewhere, incident metering is not ideal for landscape photography, unless one is doing "micro-landscape photography" such as phptographying a small patch of grass/flowers right beside you.

For the usual landscape photography, a spot metering should be the preferred method. It is extremely rare that the light falling on the photographer would be exactly the same as the light falling on the mountains.

Yes a spot meter will give a precise reflected light reading but if one doesn't have one in hand, one can get by with an incident reading, assuming the light levels are similar as mentioned by lsisaxon. I have used made incident reading from my old Sekonic and Minolta meters and have managed to get acceptable exposures using this method. Again if the light levels are similar for an average scene, even the sunny 16 rule can work in such situations.

#2 HOWEVER, IT IS VERY DIFFICULT TO USE INCIDENT LIGHT METERING TO MEASURE THE RELATIONSHIP OF LIGHT VALUES WITHIN THE SAME SUBJECT IN THEIR CLOSE PROXIMITY.

IN SUCH SITUATIONS, INCIDENT METERING IS FAR INFERIOR TO SPOT METERING.

For example, if you want to know the difference in light values between the right side of the face and the left, the incident metering will not be extremely accurate, because the measurement is often from a light sensitive dome which receives light from all over.

This kind of accuracy may not be very important today, given the widespread use of digital process. But in analogue B&W photography, for accurate reading of light values for the purpose of deteriming how to develop the negative, spot metering is almost a must.

Yes spot metering is ideal in situations where incident readings are impossible and normal reflected readings are unreliable, such as measuring distant subjects, backlight, extreme brightness range, highly reflective surfaces, or when the main subject may be moving.

However it is not entirely correct to say that an incident meter will not be extremely accurate because the light dome receives light from all over. I had a Minolta Spot Meter F and comparing to my Minolta Flash Meter IV or III, the readings are pretty much similar whether using ambient or flash readings (subject is in a studio). If a dome is inaccurate, then a flat diffuser could be used. In fact I dare say many photographers in a studio use incident readings to measure lighting ratios rather than a spot meter. Spot metering would be good if you're using the Zone System to pre-visualise your exposures.

The incident meter takes away this problem, because how much light is reflected depends on the reflectivity of the subjects.

Again you've got yourself confused here, light meters using incident readings measure light falling on the meter and is not affected by subject reflectance.

 

[creampuff]

but you see, either i'm missing out on a big big point here or that one of my questions still hasn't been answered. (pardon me. trying to digest all these info!)

so why is incident light metering more recommended? afterall, light still needs to be REFLECTED OFF THE SUBJECT into the film/sensor?

anyway, to all who has replied, thanks alot for sharing!

I think you still don't get it. Whether one uses the built in meter of the camera, use an external handheld exposure meter in reflected light mode or a spot meter you're taking a reflected light meter reading from the camera position. Your exposure reading will be affected by the light reflected off the subject.

On the other hand, for an incident meter reading, one measures light falling on the subject
and taken from the subject position. As it is unaffected by differences in subject color or reflectance, it will accurately record the amount of light falling onto the subject. In the majority of situations, an incident reading is extremely accurate and records tones, colors and values correctly.

 

[student]

The first part of your explanation is correct but your second sentance onwards (in blue) make no meaning as subject reflectance is never an issue when making incident readings.

I suggest you go and study a little more, or read carefully what I was actually saying.

Incident metering reads the intensity of light falling on the subject, it provides readings that will create accurate and consistent rendition of the subject’s tonality, color and contrasts regardless of reflectance, background color or brightness or subject texture.

Correct. Now please explain how and why.

Yes a spot meter will give a precise reflected light reading but if one doesn't have one in hand, one can get by with an incident reading, assuming the light levels are similar as mentioned by lsisaxon. I have used made incident reading from my old Sekonic and Minolta meters and have managed to get acceptable exposures using this method. Again if the light levels are similar for an average scene, even the sunny 16 rule can work in such situations.

I am sure that you can understand that the word "acceptable" is not the same as "accurate". Have you work with landscape photographers before? Real landscape photographers?

However it is not entirely correct to say that an incident meter will not be extremely accurate because the light dome receives light from all over. I had a Minolta Spot Meter F and comparing to my Minolta Flash Meter IV or III, the readings are pretty much similar whether using ambient or flash readings (subject is in a studio). If a dome is inaccurate, then a flat diffuser could be used. In fact I dare say many photographers in a studio use incident readings to measure lighting ratios rather than a spot meter. Spot metering would be good if you're using the Zone System to pre-visualise your exposures.

Again you failed to read properly. I did not say that incident meter is not accurate. I said that for the purpose of determining the relative light values such as between both sides of the face, or the relative light values on a 5 cm petal, the incident meter cannot do a good job, and therefore will be useless to me to determine how much development adjustment to make.

Please go an try to determine the lighting ratio on a petal 5 cm diameter with the incident meter.

I did not want to talk about the ZS. I was afraid it would overwhelm the mental capacity of most people here.

Again you've got yourself confused here, light meters using incident readings measure light falling on the meter and is not affected by subject reflectance.

Again you confused yourself. The meter of course is not affected by the reflectivity of the subject because the meter measures light falling on the subject. I was explaining why the incident meter will give a reading and render the relative values accordingly. That one can use the incident meter is precisely the result of reflective nature within the subject.

 

[creampuff]

I suggest you go and study a little more, or read carefully what I was actually saying.
Correct. Now please explain how and why.

No need to be so condescending. I have taken pictures for many, many years and part-owned a commercial studio before, so thanks for the tip that I need to study a little more. Maybe you forgot that incident metering reads the intensity of light falling on the subject so when you said,

An incident light metering measures the amount of light falling on the subject. If the subject is black the amount of light reflected will be little and the sensor/film will record black. If the subject is white, there will be much reflection, and the sensor/film will record the subject as white.

you are in fact describing how a reflected meter reading works.

I am sure that you can understand that the word "acceptable" is not the same as "accurate". Have you work with landscape photographers before? Real landscape photographers?

Point taken but I did agree that a spot meter would give a precise reading. I have also stated that if one doesn't have a spot meter on hand, an incident reading can be used assuming the light levels are somewhat similar. Certainly not as precise as a spot reading but it can give a workable estimation of the exposure value. Was it really necessary to ask if I have worked with landscape photographers? Short answer, yes I have.

Again you failed to read properly. I did not say that incident meter is not accurate. I said that for the purpose of determining the relative light values such as between both sides of the face, or the relative light values on a 5 cm petal, the incident meter cannot do a good job, and therefore will be useless to me to determine how much development adjustment to make.

Please go an try to determine the lighting ratio on a petal 5 cm diameter with the incident meter.

I did not want to talk about the ZS. I was afraid it would overwhelm the mental capacity of most people here.

Bro, your first example in post #8 was of lighting values of the right and left side of the face, in which you said,

the incident metering will not be extremely accurate, because the measurement is often from a light sensitive dome which receives light from all over

I respectfully disagree with you on this as I have used incident readings so many times this way and even comparing with a spot meter that can measure ambient and flash readings, the incident readings are pretty accurate.

Next you talk about determining the lighting ratio of a petal 5 cm with an incident meter!:nono:

No incident metering can tell me the difference between the lighter and darker area of this petal. Only the spotmeter can do that.

Please understand if you want to measure something so small, yes I could use a spot meter or or use a Booster II to my Minolta handheld meter (reflected light reading) or better still use a mini receptor (incident light reading). Do note there are macro and micro photographers who do use a mini receptor to take incident readings of small objects and inacessible areas where the meter will not fit between the lens and the subject.

Again you confused yourself. The meter of course is not affected by the reflectivity of the subject because the meter measures light falling on the subject. I was explaining why the incident meter will give a reading and render the relative values accordingly. That one can use the incident meter is precisely the result of reflective nature within the subject.

Please stop telling others that they are confused as your last sentence is clearly incorrect.

If my subject to photograph was a shirt; whether the shirt is black, grey or white in colour, the exposure should all be similar if you take an incident reading from the subject towards the camera. An incident reading here is gonna be accurate because it is not affected by the reflective nature of the subject. So nothing to do with "reflective nature within the subject."