Exposure question - How do you explain this paradox?

[CamInit]

I'm not quite sure what you are asking. At first I thought this page might be of help...

http://www.luminous-landscape.com/tutorials/expose-right.shtml

after sometime I thought these two might be relevant...

http://www.sekonic.com/classroom/classroom_2.asp
http://www.sekonic.com/classroom/classroom_21.asp

now I honestly don't know... :dunno:

 

[Priscilia]

trying to understand why it is a paradox.

do a spot metering on the faces (unless you are telling me one is black the other is white).

No pokiemon, nothing to do with spot metering or not... Let's assume you have some solid sophisticated light meter that does incident metering to determine the correct exposure for you.

My question really is (There I go repeating again!!!), why is it that the camera can hold details in the darker colored clothes but not the white one?

Now let's judge the situation separately, one at a time: If you just look at the dark clothed friend on the camera's LCD, you'll say Good, I got the exposure spot on. Look at the histogram, Yup, no clipping on the highlights. Exposure is CORRECT. BUT.... If you look at the white shirt friend, you'll see it "burnt out". Look at the histogram and you'll see some over-exposure. So the exposure is WRONG.

That's the paradox I'm driving at. SAME lighting, one fixed exposure, but 2 different interpretations. So how can incident metering be the 100% justification for a correct exposure?

 

[Priscilia]

I'm not quite sure what you are asking. At first I thought this page might be of help...

http://www.luminous-landscape.com/tutorials/expose-right.shtml

after sometime I thought these two might be relevant...

http://www.sekonic.com/classroom/classroom_2.asp
http://www.sekonic.com/classroom/classroom_21.asp

now I honestly don't know... :dunno:

Don't worry... I know it's very academic.... and the points I'm bringing up have very intricate implications.

As a photographer, we all know how to handle it. That's why a few here tried to suggest to me the HOW part. One CS member even referred me to a book which is not what this post is about.

What I'm asking is the WHY. It's something I've taken for granted all this while....

 

[Priscilia]

Priscilla,

Academic or not, this is a very interesting topic. :think:

I'll wait to see what others say on the subject first. Got to dig rusty brains, and even then, dunno if brain will fall apart.

Ha... phew, can I assume you do get the point I'm trying to raise?

To put in another way.... if reflected light does have a part to play in determining exposure, then I feel it must be "incorporated" into the exposure "equation".

1. Common sense and pure physics tell us a white surface reflects more light than a black one. So if you use this explanation, then perhaps we can explain the situation I brought up -- the white shirt reflects more light into the camera's sensor than the darker one, so for a given exposure, the white shirt is over-exposed.

2. But this is not what we have commonly know about exposure in photography. Reflected light fools the camera. For that reason, we get underexposure in snow and overexposure in a black panther, for example. Use a light meter or grey card, as we are told! But then, that will not guarantee getting the right exposure.

So how do you reconcile the 2 different observations in 1 and 2?

Ok, its getting late now... Need to take off my philosophical cap now....

 

[sabee]

That's the paradox I'm driving at. SAME lighting, one fixed exposure, but 2 different interpretations. So how can incident metering be the 100% justification for a correct exposure?

I do not get this statement, where do you get the notion that incident metering is a 100% justification for correct exposure? This seems to be the cause of your 'paradox' or confusion. Some guidelines to using a lightmeter even specifically mention compensation of 1/2 - 1 stop depending on whether your subject is light or dark.

 

[Dream Merchant]

Ha... phew, can I assume you do get the point I'm trying to raise?

To put in another way.... if reflected light does have a part to play in determining exposure, then I feel it must be "incorporated" into the exposure "equation".

1. Common sense and pure physics tells us a white surface reflects more light than a black one. So if you we this explanation, then perhaps we can explain the situation I brought up -- the white shirt reflects more light into the camera's sensor than the darker one, so for a given exposure, the white shirt is over-exposed.

2. But this is not what we have commonly know about exposure in photography. Reflected light fools the camera. For that reason, we get underexposure in snow and overexposure in a black panther, for example. Use a light meter or grey card, as we are told! But then, that will not guarantee getting the right exposure.

So how do you reconcile the 2 different observations in 1 and 2?

Ok, its getting late now... Need to take off my philosophical cap now....

Perfectly. Your point that is.

It's not too difficult to fanthom the confusion.

I'm waay too beat right now, and would love to hear other views first.

 

[pokiemon]

No pokiemon, nothing to do with spot metering or not... Let's assume you have some solid sophisticated light meter that does incident metering to determine the correct exposure for you.

My question really is (There I go repeating again!!!), why is it that the camera can hold details in the darker colored clothes but not the white one?

Now let's judge the situation separately, one at a time: If you just look at the dark clothed friend on the camera's LCD, you'll say Good, I got the exposure spot on. Look at the histogram, Yup, no clipping on the highlights. Exposure is CORRECT. BUT.... If you look at the white shirt friend, you'll see it "burnt out". Look at the histogram and you'll see some over-exposure. So the exposure is WRONG.

That's the paradox I'm driving at. SAME lighting, one fixed exposure, but 2 different interpretations. So how can incident metering be the 100% justification for a correct exposure?

it is an inherent problem not so much a paradox. take our human eye looking at 2 cars - matt car metallic car under sun. metallic car is more glaring than matt car even though same light. that's cos the eye is using reflective reading and not incident light.

so incident light should be used as a benchmark adjusting for reflective light.

your example is pretty extreme. so unless the camera has 2 built-in metering system to set 2 different exposures for your example A and B scenario.

 

[Priscilia]

I do not get this statement, where do you get the notion that incident metering is a 100% justification for correct exposure? This seems to be the cause of your 'paradox' or confusion. Some guidelines to using a lightmeter even specifically mention compensation of 1/2 - 1 stop depending on whether your subject is light or dark.

Great, actually you do see my point, though indirectly.

Incident light alone will not determine the exposure. That's what most (well, as far as my memory goes, for me, ALL books) photography books will tell us.

 

[J-Chan]

Perfectly. Your point that is.

It's not too difficult to fanthom the confusion.

I'm waay too beat right now, and would love to hear other views first.

Yea, not enough brain power to ramage through this confusion.. Though this is prob one of the more thought-provoking threads so I've come across so far.. :thumbsup:

 

[sabee]

Great, actually you do see my point, though indirectly.

Incident light alone will not determine the exposure. That's what most (well, as far as my memory goes, for me, ALL books) photography books will tell us.

Perhaps then that their notion of the "correct" exposure refers to the technical aspect rather than the "correctness" of the end product. I prefer to worry only about the end product.

 

[Priscilia]

it is an inherent problem not so much a paradox. take our human eye looking at 2 cars - matt car metallic car under sun. metallic car is more glaring than matt car even though same light. that's cos the eye is using reflective reading and not incident light.

so incident light should be used as a benchmark adjusting for reflective light.

your example is pretty extreme. so unless the camera has 2 built-in metering system to set 2 different exposures for your example A and B scenario.

It's getting clearer here....

You mentioned this:

"so incident light should be used as a benchmark adjusting for reflective light."

I agree with the above statement. But you'll never find it in photography books. In those books, they'll tell you we get under or overexposure cos the camera uses reflective metering which fails if the scene is off from middle grey. Only incident light reading will tell you how to set exposure. Which I find to be an insufficient condition.

But I don't think the example I've given is extreme though. It's commonly encountered.

Now to go one step further: To what extent should we judge or at least estimate reasonably in what proportion should incident or reflected light determine a correct exposure?

For example, if you want to photograph a black shiny car. Do you underexpose cos it's black, or over expose cos it's shiny? Or perhaps, keep the exposure as what the camera tells you cos they sort of even out? (Assume you don't have the luxury of using your DSLR's LCD or histogram.) I suppose this one comes with experience.

(I'm putting the above in bold and color cos I find it to be an interesting example, on 2nd thoughts.)

 

[Priscilia]

Perhaps then that their notion of the "correct" exposure refers to the technical aspect rather than the "correctness" of the end product. I prefer to worry only about the end product.

I guess it's also difficult to put everything down in words.

 

[Priscilia]

Yea, not enough brain power to ramage through this confusion.. Though this is prob one of the more thought-provoking threads so I've come across so far.. :thumbsup:

Haha.... That's how a lazy Sunday afternoon caused me to behave, which is not quite me usually....

 

[chuanzuya]

worst comes to worst do a bracketing of +-2.0,3.0 or 4.0 stops and take 3,5 or 9 images, bring to pc do HDR proccessing and tonemapping xD

 

[cutecdo]

It's getting clearer here....

You mentioned this:

"so incident light should be used as a benchmark adjusting for reflective light."

I agree with the above statement. But you'll never find it in photography books. In those books, they'll tell you we get under or overexposure cos the camera uses reflective metering which fails if the scene is off from middle grey. Only incident light reading will tell you how to set exposure. Which I find to be an insufficient condition.

But I don't think the example I've given is extreme though. It's commonly encountered.

Now to go one step further: To what extent should we judge or at least estimate reasonably in what proportion should incident or reflected light determine a correct exposure?

For example, if you want to photograph a black shiny car. Do you underexpose cos it's black, or over expose cos it's shiny? Or perhaps, keep the exposure as what the camera tells you cos they sort of even out? (Assume you don't have the luxury of using your DSLR's LCD or histogram.) I suppose this one comes with experience.

(I'm putting the above in bold and color cos I find it to be an interesting example, on 2nd thoughts.)

I agree with this too. "so incident light should be used as a benchmark adjusting for reflective light."

But your black shiny car example doesn't hold up very well. Go shoot a white shiny car and a black shiny car to see why. The colours are the main factor that determine the amount of compensation you dial in. The 'Shiny' part just show more spectacular hight-lights, at least that's what i've encountered so far.

 

[Atarandas]

Perhaps then that their notion of the "correct" exposure refers to the technical aspect rather than the "correctness" of the end product. I prefer to worry only about the end product.

I agree with this. :thumbsup: Exposure isnt a science where we can define in way to say that , at this parameter , its the best exposure , and at that its not good.

Essentially , exposure is about the tones and details we can see . I would define correct exposure loosely. As long the exposure is such that it shows what the photographer wanted to show or create , its a good exposure. Even though ur camera tells you otherwise.

I dont think theres a paradox at all as in the first place , the exposure theory of incident light isnt the utimate law that defines exposure and explains behaviors of exposure for all conditions. Its just a guide for general illustration and understanding. Just like any other physics law , it holds true for certain preset conditions . Beyond those parameters , the properties will change .

But then its really a good thread to explore the science and arts of exposure heh

 

[Jed]

Hang on.

I've just come back from a job so I'm not totally clued in yet (black shirts, v white shirts, no less). But I'm a little confuzzled about what the exact nature of the problem, or the paradox is.

Firstly, an incident light reading basically measures the amount of light falling on a given scene. This means that using that setting under that lighting, middle grey should reproduce as middle grey.

Now with the dark shirt/light shirt example, is that dark shirt exactly the same number of stops darker than the light shirt is brighter? If the dark shirt is not as dark as the light shirt is light (eek for the expression) then it stands to figure that it's entirely probable that the light shirt will not reproduce as well. Have a further look at the zone system for a more thorough examination of this.

Add to that the fact that dynamic range of the sensor is very pertinent. Firstly if your tone curve is linear (which it isn't) then if the dark shirt is exactly the same number of stops darker than the light shirt is brighter, and you have accurately metered for middle grey, then they should reproduce similarly. But, especially when it comes to digital sensors, the tone curve tends to be S shaped, headroom in the shoulder area is poor, and dynamic range tends to be limited, with particularly poor performances in the highlight areas especially when any single channel comes close to being clipped.

Finally, you also have to bear in mind that most highlight clipping on the camera triggers if any of the RGB channels is clipped, rather than if all are clipped. It's entirely possible that a light shirt that might appear to be clipped is actually within tolerances.

Now also consider, if you reversed the lighting situation. So instead of the two shirts in sunlight, you have them in a dimly lit room. The dark shirt is likely to block up, while the light shirt would likely be properly exposed.

I'm still not entirely sure where the paradox comes in, but I suspect it's somewhere in the whole "correct exposure" concept. Incidence metering takes care of your mid point, after that a scene with lots of light tones will appear light, a scene with dark tones will appear dark, and if the dynamic range of the scene exceeds the sensor's capabilities then you will end up with clipped highlights, or clipped shadows. Because of sensor design at the moment, clipped highlights are more likely than clipped shadows generally speaking, with more latitude in the shadow areas than in the highlights.

Have a quick look at any dynamic range graph on DPReview for example, or any other reliable test site. As an example I had a quick look at the D3000 review on that site, and at ISO 200 the useable shadow range is 5.1EV, and the useable highlight range is 3.6EV (below and above mid grey respectively). That's an extra 1.5 stops of latitude in the shadow areas, and why your dark shirt is AOK while the light shirt gets clipped.

 

[Jed]

For example, if you want to photograph a black shiny car. Do you underexpose cos it's black, or over expose cos it's shiny? Or perhaps, keep the exposure as what the camera tells you cos they sort of even out? (Assume you don't have the luxury of using your DSLR's LCD or histogram.) I suppose this one comes with experience.

That totally depends on your camera and metering choice and what else is in the background. The great bane of matrix metering is what while it does a good to reasonable job the majority of the time, you as the photographer have absolutely no clue what it's doing. With a LOT of experience with the camera you'll have some idea what it's going to do, but even then you're not guaranteed.

So it might do a good job, it might be fooled by the black, you might have exceedingly bright reflections that totally fool it or it might be intelligent enough to get ignored. On the other hand if you spot meter or even centre weight, you will have a fairly good idea what's going to happen and you'll be able to adjust accordingly. This situation is totally a reflected light issue as well, theoretically and taking into account what I have said about dynamic range, tonal curves, and lack of highlight headroom in my previous post, incident lighting should correctly expose the scene.

But if your highlights are 10 stops brighter than mid grey then they're going to blow, if your shadows are 10 stops darker than mid grey then they're going to get blocky. If you have a D3000 and the highlights and shadows were as much as four stops over/under mid grey respectively, then only your highlights will blow.

 

[Atarandas]

Hang on.

I've just come back from a job so I'm not totally clued in yet (black shirts, v white shirts, no less). But I'm a little confuzzled about what the exact nature of the problem, or the paradox is.

Firstly, an incident light reading basically measures the amount of light falling on a given scene. This means that using that setting under that lighting, middle grey should reproduce as middle grey.

Now with the dark shirt/light shirt example, is that dark shirt exactly the same number of stops darker than the light shirt is brighter? If the dark shirt is not as dark as the light shirt is light (eek for the expression) then it stands to figure that it's entirely probable that the light shirt will not reproduce as well. Have a further look at the zone system for a more thorough examination of this.

Add to that the fact that dynamic range of the sensor is very pertinent. Firstly if your tone curve is linear (which it isn't) then if the dark shirt is exactly the same number of stops darker than the light shirt is brighter, and you have accurately metered for middle grey, then they should reproduce similarly. But, especially when it comes to digital sensors, the tone curve tends to be S shaped, headroom in the shoulder area is poor, and dynamic range tends to be limited, with particularly poor performances in the highlight areas especially when any single channel comes close to being clipped.

Finally, you also have to bear in mind that most highlight clipping on the camera triggers if any of the RGB channels is clipped, rather than if all are clipped. It's entirely possible that a light shirt that might appear to be clipped is actually within tolerances.

Now also consider, if you reversed the lighting situation. So instead of the two shirts in sunlight, you have them in a dimly lit room. The dark shirt is likely to block up, while the light shirt would likely be properly exposed.

I'm still not entirely sure where the paradox comes in, but I suspect it's somewhere in the whole "correct exposure" concept. Incidence metering takes care of your mid point, after that a scene with lots of light tones will appear light, a scene with dark tones will appear dark, and if the dynamic range of the scene exceeds the sensor's capabilities then you will end up with clipped highlights, or clipped shadows. Because of sensor design at the moment, clipped highlights are more likely than clipped shadows generally speaking, with more latitude in the shadow areas than in the highlights.

Have a quick look at any dynamic range graph on DPReview for example, or any other reliable test site. As an example I had a quick look at the D3000 review on that site, and at ISO 200 the useable shadow range is 5.1EV, and the useable highlight range is 3.6EV (below and above mid grey respectively). That's an extra 1.5 stops of latitude in the shadow areas, and why your dark shirt is AOK while the light shirt gets clipped.

:thumbsup::thumbsup: A very nice n sweet technical explaination

 

[cabbySHE]

I am not qualified to explain your doubt, however, I think this guy can....and a very detailed one.

His name : Christian Bloch, author of the book...the HDRI Handbook under rockynook.

Get a copy, you should have all the answer and in details too.