a new exposure triangle

I posted this before in "general photo techniques", but maybe this is a better place?

I've found various versions of "exposure triangles" in the web, and also here on dpreview, but if I see it well, they all just show a qualitative relation between ISO, shutter speed and aperture to the related "secondary effects" noise, motion blur and defocus blur/diffraction. But the most important parameter LIGHT simply is disregarded. Of course it is true, 4 parameters are too much for a triangle, but here is my try:

Imagine you print this first page on a thick cardboard:





Then you print these 2, preferrably on transparent film, and cut out all the triangles (they are just examples):









Now imagine you measured the light of the scene and you found e.g. sunshine with EV15. Place the transparent EV15 triangle on the cardboard diagram and align it with the edges for ISO, shutter speed and aperture. Wherever you place it (as long as camera and lens allow such settings) you will get the same exposure. So this is actually a quantitative exposure triangle, that can tell you the exact relation betwenn the 3 (or 4) parameters of exposure. But with the little gradients it also shows the qualitative differences in noise, motion blur and defocus blur/diffraction, just like all the other exposure triangles do.

What do you think? It is not intended to work with every day, but rather to teach beginners. Can this be helpful?

Best regards, Matthias

Those three elements are:

1. Scene luminance
2. t-stop (approximated by f-ratio)
3. shutter speed

The exposure is the density of the light that falls on the sensor, and ISO has nothing to do with exposure, except inasmuch as the ISO setting on the camera indirectly affects the f-ratio, shutter speed, or flash power.

masi1157 said:

What do you think? It is not intended to work with every day, but rather to teach beginners. Can this be helpful?

Click to expand...

My own view is that it is best not to have beginners start out with outmoded film-era ideas of exposure control. The three components of exposure are scene luminance, f-number (more accurately, as GB points out, T number) and shutter speed. Those alone determine the illuminance at the focal plane, which is what exposure is. The ISO control sets a target exposure, but since you have already introduced an absolute measure of exposure, EV (which should correctly be EV100 or LV, EV describes the combined setting of f-number and shutter - it is exposure without the scene luminance component) it is redundant in your triangle. Noise is controlled directly by exposure, not ISO, diffraction by f-number and motion blur by shutter speed.
--
Bob

Great Bustard said:

The exposure is the density of the light that falls on the sensor, and ISO has nothing to do with exposure, except inasmuch as the ISO setting on the camera indirectly affects the f-ratio, shutter speed, or flash power.

Click to expand...

And that makes it a fourth parameter in exposure. The first 3 control the amount of light falling on the sensor, the 4. (ISO) controls, what amount is needed for a correct exposure. That gives ISO a different quality, but it has the same relevance as the other three. Or would you know how to calculate exposure parameters without knowing the ISO of your sensor or film?

Regards, Matthias

bobn2 said:

My own view is that it is best not to have beginners start out with outmoded film-era ideas of exposure control. The three components of exposure are scene luminance, f-number (more accurately, as GB points out, T number) and shutter speed.

Click to expand...

Where do you see "film era ideas" in this model?

bobn2 said:

Those alone determine the illuminance at the focal plane, which is what exposure is. The ISO control sets a target exposure, but since you have already introduced an absolute measure of exposure, EV (which should correctly be EV100 or LV, EV describes the combined setting of f-number and shutter - it is exposure without the scene luminance component) it is redundant in your triangle.

Click to expand...

My original german version said "Lichtwert", which is LV, so it is simply a bad translation. I will replace EV with LV in the next version.

bobn2 said:

Noise is controlled directly by exposure, not ISO, diffraction by f-number and motion blur by shutter speed.

Click to expand...

I am not comparing different sensor sizes or anything, and it is meant as an aid for beginners. And within that model I find it clear enough to say, that noise is controlled by ISO. If you trade 2 ISO stops for 2 shutter speed stops, you get a rather similar photo, but with less motion blur and more noise.

Regards, Matthias

Great Bustard said:

Those three elements are:

1. Scene luminance
2. t-stop (approximated by f-ratio)
3. shutter speed

The exposure is the density of the light that falls on the sensor, and ISO has nothing to do with exposure, except inasmuch as the ISO setting on the camera indirectly affects the f-ratio, shutter speed, or flash power.

Click to expand...

Are you really trying to pretend that ISO is not part of the exposure? I have seen your work and I know you are not that ignorant, so you must just be fishing for an argument. If I change ISO, I must change one of the other exposure settings to compensate, just like if I changed the size of aperture or the speed of the shutter.

masi, I think your exposure triangle thing is pretty cool and would be pretty handy to see one in real life.

masi1157 said:

Great Bustard said:

The exposure is the density of the light that falls on the sensor, and ISO has nothing to do with exposure, except inasmuch as the ISO setting on the camera indirectly affects the f-ratio, shutter speed, or flash power.

Click to expand...

And that makes it a fourth parameter in exposure. The first 3 control the amount of light falling on the sensor, the 4. (ISO) controls, what amount is needed for a correct exposure.

Click to expand...

No it doesn't. It controls the exposure you choose to give. There is no such thing as a 'correct' exposure in digital photography.

masi1157 said:

That gives ISO a different quality, but it has the same relevance as the other three. Or would you know how to calculate exposure parameters without knowing the ISO of your sensor or film?

Click to expand...

You would simply work with the exposure that you select.

--
Bob

tkbslc said:

masi, I think your exposure triangle thing is pretty cool and would be pretty handy to see one in real life.

Click to expand...

Thanks. I have one here in front of me and it is really fun to play around with it. You can easily simulate what happens in P, S, A mode, when you turn the dial. You can show, in what direction the transparent triangle can grow in those modes, when the scene gets brighter. And much more.

Regards, Matthias

masi1157 said:

bobn2 said:

My own view is that it is best not to have beginners start out with outmoded film-era ideas of exposure control. The three components of exposure are scene luminance, f-number (more accurately, as GB points out, T number) and shutter speed.

Click to expand...

Where do you see "film era ideas" in this model?

Click to expand...

It is based on a notion that there is a 'correct' exposure that needs to be achieved for good results and that that 'correct' exposure is controlled by ISO. This is not true of digital. It is only true of film, and the the film emulation user interface commonly applied to digital cameras.

masi1157 said:

bobn2 said:

Those alone determine the illuminance at the focal plane, which is what exposure is. The ISO control sets a target exposure, but since you have already introduced an absolute measure of exposure, EV (which should correctly be EV100 or LV, EV describes the combined setting of f-number and shutter - it is exposure without the scene luminance component) it is redundant in your triangle.

Click to expand...

My original german version said "Lichtwert", which is LV, so it is simply a bad translation. I will replace EV with LV in the next version.

bobn2 said:

Noise is controlled directly by exposure, not ISO, diffraction by f-number and motion blur by shutter speed.

Click to expand...

I am not comparing different sensor sizes or anything, and it is meant as an aid for beginners.

Click to expand...

Nor was I, had I been doing so I would have said that noise is controlled by total light, not exposure.

masi1157 said:

And within that model I find it clear enough to say, that noise is controlled by ISO. If you trade 2 ISO stops for 2 shutter speed stops, you get a rather similar photo, but with less motion blur and more noise.

Click to expand...

ISO is unnecessary, you are simply saying that noise is controlled by exposure. ISO has indirectly become a proxy for exposure.
--
Bob

tkbslc said:

Great Bustard said:

Those three elements are:

1. Scene luminance
2. t-stop (approximated by f-ratio)
3. shutter speed

The exposure is the density of the light that falls on the sensor, and ISO has nothing to do with exposure, except inasmuch as the ISO setting on the camera indirectly affects the f-ratio, shutter speed, or flash power.

Click to expand...

Are you really trying to pretend that ISO is not part of the exposure?

Click to expand...

ISO is not a part of exposure:
http://en.wikipedia.org/wiki/Exposure_ (photography)
--
Bob

masi1157 said:

Great Bustard said:

The exposure is the density of the light that falls on the sensor, and ISO has nothing to do with exposure, except inasmuch as the ISO setting on the camera indirectly affects the f-ratio, shutter speed, or flash power.

Click to expand...

And that makes it a fourth parameter in exposure. The first 3 control the amount of light falling on the sensor, the 4. (ISO) controls, what amount is needed for a correct exposure.

Click to expand...

There is no such thing as a "correct exposure" -- if there were, the camera wouldn't have different metering modes. Or is the metering mode now another element of the "exposure pentagon"?

masi1157 said:

That gives ISO a different quality, but it has the same relevance as the other three. Or would you know how to calculate exposure parameters without knowing the ISO of your sensor or film?

Click to expand...

The ISO is simply a mapping from the captured image to the desired display brightness. In many cameras, it's also linked to an analog gain, which helps serve to reduce read noise.

In other words, the camera can choose the ISO more quickly and intelligently than the photogapher based on the chosen metering mode, just as a good automatic transmission can shift faster and more intelligently than a stick.

ISO control is an anachronistic kludge back to the days of film where the sensitivity of the recording medium was a function of the film speed as opposed to a digital sensor which has a fixed "sensitivity".

bobn2 said:

It is based on a notion that there is a 'correct' exposure that needs to be achieved for good results and that that 'correct' exposure is controlled by ISO. This is not true of digital. It is only true of film, and the the film emulation user interface commonly applied to digital cameras.

Click to expand...

And does any beginner use a DSLR, which does not emulate this behavior? So why shouldn't I teach him, what his camera does?

bobn2 said:

Nor was I, had I been doing so I would have said that noise is controlled by total light, not exposure.

Click to expand...

bobn2 said:

ISO is unnecessary, you are simply saying that noise is controlled by exposure. ISO has indirectly become a proxy for exposure.

Click to expand...

Then please tell me a way how to visualize all that in a way, that a beginner understands. Within the model, that I am using here, ISO has exactly the same relevance as shutter speed and aperture. That might not exactly represent the physics behind it, but it represents the behaior of a usual DSLR.

Regards, Matthias

bobn2 said:

ISO is not a part of exposure:
http://en.wikipedia.org/wiki/Exposure_ (photography)

Click to expand...

Did you try google with "exposure triangle"? You will find loads of triangles with shutter speed, aperture and -yes- ISO on the 3 ends, but none of them give a quantitative relation between them and the luminance of the scene. And that is, where my triangle might be an enhancement.

Regards, Matthias

tkbslc said:

Great Bustard said:

Those three elements are:

1. Scene luminance
2. t-stop (approximated by f-ratio)
3. shutter speed

The exposure is the density of the light that falls on the sensor, and ISO has nothing to do with exposure, except inasmuch as the ISO setting on the camera indirectly affects the f-ratio, shutter speed, or flash power.

Click to expand...

Are you really trying to pretend that ISO is not part of the exposure?

Click to expand...

It's not pretend -- it's a fact. The only role ISO plays in exposure is, as I said, how it indirectly affects an element of exposure in an AE (auto exposure) mode. For example, if you are in P mode, raising the ISO lowers the f-ratio and/or increases the shutter speed. The ISO setting doesn't change the exposure, the change in f-ratio and shutter speed does.

tkbslc said:

I have seen your work and I know you are not that ignorant, so you must just be fishing for an argument. If I change ISO, I must change one of the other exposure settings to compensate, just like if I changed the size of aperture or the speed of the shutter.

Click to expand...

It is the other settings that change (f-ratio, shutter speed, and/or flash power) that changes the exposure, not the ISO.

So, no, I'm not "fishing for an argument", I'm stating a fact. Changing the ISO no more affects exposure than changing the metering mode. Or are you suggesting that the metering mode is a fifth node of the "exposure pentagon" due to how it indirectly affects the three elements of exposure, like the ISO setting?

Great Bustard said:

It's not pretend -- it's a fact.

Click to expand...

Is it only the word "exposure", that you dislike here? Then please give me a better one. I think you won't deny, that my triangle reflects the relationship between shutter speed, aperture and ISO the the scene luminance, exactly as it is emulated in virtually all DSLRs.

Regards, Matthias

masi1157 said:

Then please tell me a way how to visualize all that in a way, that a beginner understands.

Click to expand...

Make an exposure triangle with the three elements of exposure:

• Scene Luminance
• t-stop (approximated with f-ratio)
• Shutter Speed

and explain how the metering mode, ISO setting, and flash power affect the three elements of exposure. Just not that hard. And if that's too confusing, well, hell -- just tell them to put the camera in Auto Mode, set Auto ISO, and be done with it.

Great Bustard said:

ISO control is an anachronistic kludge back to the days of film where the sensitivity of the recording medium was a function of the film speed as opposed to a digital sensor which has a fixed "sensitivity".

Click to expand...

May be. But then please tell me just one beginner's DSLR, that does not work this way. Don't forget, this is meant to teach a beginner how to use his camera.

Regards, Matthias

masi1157 said:

Great Bustard said:

It's not pretend -- it's a fact.

Click to expand...

Is it only the word "exposure", that you dislike here? Then please give me a better one.

Click to expand...

I like the word "exposure" just fine, and use the word as it is defined. Perhaps you are looking for a "better" term to call the brightness of the display photo. I call it "apparent exposure":

h ttp: www.josephjamesphotography.com/equivalence/index.htm#exposure

DPR calls it "light value":

http://www.dpreview.com/learn/?/Glossary/Exposure/Exposure_01.htm

masi1157 said:

I think you won't deny, that my triangle reflects the relationship between shutter speed, aperture and ISO the the scene luminance, exactly as it is emulated in virtually all DSLRs.

Click to expand...

If you called it a "Brightness Triangle", I wouldn't object. But just as the term "relative aperture" (f-ratio) was shortened to "aperture" (keeping in mind that there's also "physical aperture" -- iris -- and "virtual aperture" -- entrance pupil), I was kind of hoping that the same bastardization of the term "exposure" might be prevented.

Lost cause -- I know. But there was a reason exposure was defined the way it was in the first place. f/2.8 1/200 ISO 100 is not the same exposure as f/2.8 1/400 ISO 200, but the pics will have the same brightness (apparent exposure, if you will).

An exposure setting is and was always a combination of four things :

• Shutter speed
• Aperture ( as expressed in f-number )
• ISO

and, to relate that to the scene you need to know one more thing :

• Available light

These define how much light you are going to capture.

An abstract system of units ( "f-stops" ) has been used in conjunction with this in order to avoid the mess we'd all have in dealing with physical units, like area.

This system has been in use for many decades and not only works, but is the language used to allow photographers to reliably move between equipment and engineers to create a consistent and straightforward "interface" for users.

Any exposure setting omitting ISO is incomplete. It is not incidental or secondary, it is a fundamental part of the exposure setting. You cannot evaluate the exposure without it. It declares the sensitivity of the underlying medium used to capture the light. It cannot be omitted, inferred or taken for granted. It must be an explicit part of any exposure setting.

As astonishing as it will doubtless seem to you, the combined wisdom of several generations of photographers and engineers is quite capable of having hit upon the correct answer despite not having the benefit of your "insight".

--
StephenG

sjgcit said:

An exposure setting is and was always a combination of four things :

• Shutter speed
• Aperture ( as expressed in f-number )
• ISO

and, to relate that to the scene you need to know one more thing :

• Available light

These define how much light you are going to capture.

An abstract system of units ( "f-stops" ) has been used in conjunction with this in order to avoid the mess we'd all have in dealing with physical units, like area.

This system has been in use for many decades and not only works, but is the language used to allow photographers to reliably move between equipment and engineers to create a consistent and straightforward "interface" for users.

Any exposure setting omitting ISO is incomplete. It is not incidental or secondary, it is a fundamental part of the exposure setting. You cannot evaluate the exposure without it. It declares the sensitivity of the underlying medium used to capture the light. It cannot be omitted, inferred or taken for granted. It must be an explicit part of any exposure setting.

As astonishing as it will doubtless seem to you, the combined wisdom of several generations of photographers and engineers is quite capable of having hit upon the correct answer despite not having the benefit of your "insight".

Click to expand...

Welly welly welly well! That was quite a rant! OK, now to the facts:

http://en.wikipedia.org/wiki/Exposure_ (photography)

I was going to quote from the link, but hell, I would end up quoting the whole thing!

So, as I said before:

http://forums.dpreview.com/forums/read.asp?forum=1018&message=39349826

The only role ISO plays in exposure is, as I said, how it indirectly affects an element of exposure in an AE (auto exposure) mode. For example, if you are in P mode, raising the ISO lowers the f-ratio and/or increases the shutter speed. The ISO setting doesn't change the exposure, the change in f-ratio and shutter speed does.

It is the other settings that change (f-ratio, shutter speed, and/or flash power) that changes the exposure, not the ISO.

So, no, I'm not "fishing for an argument", I'm stating a fact. Changing the ISO no more affects exposure than changing the metering mode. Or are you suggesting that the metering mode is a fifth node of the "exposure pentagon" due to how it indirectly affects the three elements of exposure, like the ISO setting?

Or, more to the point:

http://forums.dpreview.com/forums/read.asp?forum=1018&message=39349885

Make an exposure triangle with the three elements of exposure:

• Scene Luminance
• t-stop (approximated with f-ratio)
• Shutter Speed

and explain how the metering mode, ISO setting, and flash power affect the three elements of exposure. Just not that hard. And if that's too confusing, well, hell -- just tell them to put the camera in Auto Mode, set Auto ISO, and be done with it.

So, here's a quick question for you: do f/2.8 1/200 ISO 100 and f/2.8 1/400 ISO 200 have the same exposure? A simple yes or no will do.

I'm off to sleep -- plenty of time for you to answer such a simple question.