Exposure control in a raw developer: misnomer?

Started Jan 30, 2013 | Discussions
dosdan
Regular MemberPosts: 373Gear list
Like?
Exposure control in a raw developer: misnomer?
Jan 30, 2013

As I understand it, the "Exposure Triangle" is composed of Scene Luminance, Shutter Speed, Aperture (or should that be "F-number"?). These 3 alone determine the number of photons captured by the sensor.

ISO Sensitivity plays a role in the rendered brightness of the outputted image. (As well as helping to mitigate the contribution to the total read noise of the ADC noise floor in most cameras.)

So, I take the phrase "exposing a shot" to refer to the capturing process, not the rendering process. With in-camera JPEG, these two processes are combined. When shooting raw, the development process is separated and postponed.

But most raw development programs have an "Exposure" adjustment control. But isn't this adjusting the brightness of the rendered output? If so, it has nothing to do with exposure and is a misnomer.

So what should this control be labelled? "Rendered Brightness"?

What about a suitable term for the combined capturing & rendering process, that doesn't confuse it with exposing?

Dan.

 dosdan's gear list:dosdan's gear list
Pentax K20D Pentax K-5 Pentax Q Pentax K-01 Pentax K-3
Draek
Senior MemberPosts: 2,028Gear list
Like?
Re: Exposure control in a raw developer: misnomer?
In reply to dosdan, Jan 30, 2013

The technically-accurate term would be "Exposure Compensation", but of course there's already something else under that name, and no less accurate. Perhaps we could switch the latter to "Metering Compensation" and "Exposure" to "Exposure Compensation", and we'd have two distinct, accurate names for the two actions.

Or perhaps we could keep it as "Exposure", since nobody seems to be confused by it in any case.

 Draek's gear list:Draek's gear list
Fujifilm FinePix F70EXR Samsung TL500 Canon PowerShot A1200 Sony Alpha DSLR-A390
Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 15,000
Like?
Re: Exposure control in a raw developer: misnomer?
In reply to dosdan, Jan 30, 2013

dosdan wrote:

As I understand it, the "Exposure Triangle" is composed of Scene Luminance, Shutter Speed, Aperture (or should that be "F-number"?). These 3 alone determine the number of photons captured by the sensor.

ISO Sensitivity plays a role in the rendered brightness of the outputted image. (As well as helping to mitigate the contribution to the total read noise of the ADC noise floor in most cameras.)

So, I take the phrase "exposing a shot" to refer to the capturing process, not the rendering process. With in-camera JPEG, these two processes are combined. When shooting raw, the development process is separated and postponed.

But most raw development programs have an "Exposure" adjustment control. But isn't this adjusting the brightness of the rendered output? If so, it has nothing to do with exposure and is a misnomer.

So what should this control be labelled? "Rendered Brightness"?

What about a suitable term for the combined capturing & rendering process, that doesn't confuse it with exposing?

But Dan, that would make sense...

It seems that (on these forums, anyway) it is by far more the exception than the rule when the distinction between Exposure:

http://en.wikipedia.org/wiki/Exposure_%28photography%29#Photometric_and_radiometric_exposure

http://en.wikipedia.org/wiki/Exposure_value#Formal_definition

... and Image Brightness is understood. Further, there seems to exist a common tremendous intertial farce that (despite the definition of Exposure being cited) opposes letting go of the very common misnomer (which seeks to include an ISO rating, and relates only to camera settings themselves, as opposed to including Scene Luminance).

That (technically incorrect) usage of the phrase Exposure Value (EV) with reference to camera settings only is so widespread that camera companies play along, as well. For an example of that ...

... see this post: http://forums.dpreview.com/forums/post/50618520

Reply   Reply with quote   Complain
jcharding
Senior MemberPosts: 2,004Gear list
Like?
I'd argue that scene luminescence is the misnomer
In reply to dosdan, Jan 30, 2013

It assumes that the photographer seeks some nirvana of perfectly exposed scene where luminescence actually matters mathematically.

I disagree with Detail Man, as usual, and argue that ISO remains third point of the triangle.  All three of aperture, shutter speed and ISO are photographer decisions. Made to move towards whatever that person wants to photograph. Scene luminescence, or scene brightness, isnt  a photographic decision. It simply is. It affects the shot, but isn't something the photographer can generally change short of turning on a bunch of lights.

-- hide signature --
 jcharding's gear list:jcharding's gear list
Fujifilm FinePix X100 Olympus PEN E-P2 Olympus PEN E-P3 Fujifilm X-Pro1 Olympus OM-D E-M5 +10 more
Reply   Reply with quote   Complain
jcharding
Senior MemberPosts: 2,004Gear list
Like?
Examples
In reply to jcharding, Jan 30, 2013

The scene luminescence in this shot in Antelope Canyon was actually much brighter. I don't care. I don't want a perfectly exposed scene.

This scene was also brighter. I also didn't care.

This next scene was darker. Still don't care.

Scene luminescence only matters in that how it affects your ISO, shutter speed and aperture values. Do you care about noise. Do you care about subject matter motion. Do you care about depth of field. A lot of the time the goal is not to perfectly and mathematically accomplish something.

I'd also point out that all of ISO, aperture and shutter speed affect rendered brightness.

-- hide signature --
 jcharding's gear list:jcharding's gear list
Fujifilm FinePix X100 Olympus PEN E-P2 Olympus PEN E-P3 Fujifilm X-Pro1 Olympus OM-D E-M5 +10 more
Reply   Reply with quote   Complain
Hugowolf
Forum ProPosts: 11,269
Like?
Re: I'd argue that scene luminescence is the misnomer
In reply to jcharding, Jan 30, 2013

Luminescence is its name, it hasn't been missnamed, it isn't a misnomer. Whether it is part of a triangle is another matter.

And adding light isn't part of photography? Perhaps you would like to put that to the Studio and Lighting Forum?

Brian A

Reply   Reply with quote   Complain
sherwoodpete
Veteran MemberPosts: 7,761
Like?
Re: Exposure control in a raw developer: misnomer?
In reply to Draek, Jan 30, 2013

Draek wrote:

The technically-accurate term would be "Exposure Compensation", but of course there's already something else under that name, and no less accurate.

Hold on. If exposure compensation is used, it biases the exposure meter in the camera, which in turn leads to a greater or smaller quantity of photons being captured.

Whatever you do after the event during raw development isn't going to change the number of photons that were captured.

If you want a technically accurate term it would be "push-processing" or "pull-processing", which is how it is referred to when a film is processed differently in order to give a different effective ISO rating.

Perhaps we could switch the latter to "Metering Compensation" and "Exposure" to "Exposure Compensation", and we'd have two distinct, accurate names for the two actions.

Or perhaps we could keep it as "Exposure", since nobody seems to be confused by it in any case.

It seems that isn't so.

Regards,
Peter

Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 15,000
Like?
Re: I'd argue that scene luminescence is the misnomer
In reply to jcharding, Jan 30, 2013

jcharding wrote:

It assumes that the photographer seeks some nirvana of perfectly exposed scene where luminescence actually matters mathematically.

Here's what the word that you used actually means:

Luminescence is emission of light by a substance not resulting from heat; it is thus a form of cold body radiation.

http://en.wikipedia.org/wiki/Luminescence

I disagree with Detail Man, as usual, ...

Then I must be a reliable indicator of some sort, then. Glad to know that I provide some direction.

... and argue that ISO remains third point of the triangle.

Could you possibly define the specific origins of "the triangle", and speak to it's scholarly accuracy ?

All three of aperture, shutter speed and ISO are photographer decisions. Made to move towards whatever that person wants to photograph.

They are camera settings only. You demonstrate yourself in the images that you posted that your eye seeks differing amounts of rendered output Image Brightness in different images. Thus, you are (whether you like it or not) implicitly subject to Scene Luminance as a factor that influences your choice of camera settings (such as F-Number, Shutter Speed) and/or processor/editor settings.

Scene luminescence, or scene brightness, isnt a photographic decision. It simply is. It affects the shot, but isn't something the photographer can generally change short of turning on a bunch of lights.

A light bulb just "went off" somewhere ...

... Scene luminescence only matters in that how it affects tyour ISO, shutter speed, and aperture values.

ISO setting to reach some Image Brightness only matters in how it is affected by Scene Luminance.

Do you care about noise.

Image Signal/Noise Ratio is precisely why (some) photographers care about (the real) Exposure. You have made my point (about differences between light and ISO setting being ignored) beautifully.

Do you care about subject matter motion. Do you care about depth of field. A lot of the time the goal is not to perfectly and mathematically accomplish something.

What are you arguing for ? Being blithely unaware of all of the above parameters ? Is that better ?

Reply   Reply with quote   Complain
Great Bustard
Forum ProPosts: 23,228
Like?
Absolutely correct.
In reply to dosdan, Jan 30, 2013

dosdan wrote:

As I understand it, the "Exposure Triangle" is composed of Scene Luminance, Shutter Speed, Aperture (or should that be "F-number"?). These 3 alone determine the number of photons captured by the sensor.

Yes.

ISO Sensitivity plays a role in the rendered brightness of the outputted image. (As well as helping to mitigate the contribution to the total read noise of the ADC noise floor in most cameras.)

Yes.

So, I take the phrase "exposing a shot" to refer to the capturing process, not the rendering process. With in-camera JPEG, these two processes are combined. When shooting raw, the development process is separated and postponed.

Yes.

But most raw development programs have an "Exposure" adjustment control. But isn't this adjusting the brightness of the rendered output? If so, it has nothing to do with exposure and is a misnomer.

So what should this control be labelled? "Rendered Brightness"?

Yes.

What about a suitable term for the combined capturing & rendering process, that doesn't confuse it with exposing?

Or we could just use the existing terms as they are defined.  It's funny, 'cause if you ask someone who considers ISO part of exposure, and you ask them if f/2.8 1/100 ISO 100 and f/5.6 1/100 ISO 400 result in the same exposure for a given scene, it's exceedingly rare that the answer to this simple yes or no question will take the form of "yes" or "no".

Reply   Reply with quote   Complain
dosdan
Regular MemberPosts: 373Gear list
Like?
Re: Absolutely correct.
In reply to Great Bustard, Jan 30, 2013

Great Bustard wrote:

It's funny, 'cause if you ask someone who considers ISO part of exposure, and you ask them if f/2.8 1/100 ISO 100 and f/5.6 1/100 ISO 400 result in the same exposure for a given scene, it's exceedingly rare that the answer to this simple yes or no question will take the form of "yes" or "no".

GB, ask them this: if they consider these 2 exposures to be equivalent, what about the 2x worse shot noise SNR in the ISO400 shot due to the 1/4x smaller total number of photons collected during the exposure period?

Dan.

 dosdan's gear list:dosdan's gear list
Pentax K20D Pentax K-5 Pentax Q Pentax K-01 Pentax K-3
Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 15,000
Like?
Re: Absolutely correct.
In reply to dosdan, Jan 30, 2013

dosdan wrote:

Great Bustard wrote:

It's funny, 'cause if you ask someone who considers ISO part of exposure, and you ask them if f/2.8 1/100 ISO 100 and f/5.6 1/100 ISO 400 result in the same exposure for a given scene, it's exceedingly rare that the answer to this simple yes or no question will take the form of "yes" or "no".

GB, ask them this: if they consider these 2 exposures to be equivalent, what about the 2x worse shot noise SNR in the ISO400 shot due to the 1/4x smaller total number of photons collected during the exposure period?

Perhaps if the (RAW processor) control in question was renamed to "Image Noise Amplification Factor" - as more positive valued settings of such a digital-amplification control-slider indeed do just that (in cases where a higher Radiant Exposure of the image-sensor would have helped to minimize relative Photon Shot Noise while increasing Brightness, and a correspondingly lower ISO setting would have helped to minimize relative Read Noise) - it would very likely get just a little bit more attention ?

Good luck convincing Adobe's marketing executives on that one, however !

It does not have to be "mathematically perfect" in order to be "demonstrably better" to the eyes.

DM...

Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 15,000
Like?
Re: Exposure control ?
In reply to dosdan, Jan 30, 2013

Perhaps for some photogs, rather than endlessly obsess about the "correct" or the "proper" exposure, one may at times simply satisfy themselves with any sort of exposure whatsoever ? ...

Reply   Reply with quote   Complain
dosdan
Regular MemberPosts: 373Gear list
Like?
Re: Exposure control in a raw developer: misnomer?
In reply to dosdan, Jan 30, 2013

dosdan wrote:

As I understand it, the "Exposure Triangle" is composed of Scene Luminance, Shutter Speed, Aperture (or should that be "F-number"?). These 3 alone determine the number of photons captured by the sensor.

Yes, it should f-number.

Consider: a 100/F2 lens fully open has a much bigger aperture (diameter; area) than a 50/F2 lens. What determines the amount of light reaching the sensor is the f-number (the ratio of FL to effective aperture dia.) So, when fully open, a 100/F2 and 50/F2 lens both admit the same amount of light. Same with a 300/F4 and a 28/F2.8 lens, both closed down to ƒ/8. So swapping between two lenses (assuming the same transmission loss), set to the same f-number, should result in the same exposure.

("T-stop", instead of f-number, is used when the transmission loss is factored in.)

(f-number/f-ratio/f-stop is a ratio of ƒ (focal length) / entrance pupil dia. Just as ½ is bigger than ¼, so ƒ/2 is a bigger ratio, and for the same FL, a bigger area, than ƒ/4.)

So it should be:

The Exposure Triangle is: Scene Luminance, Shutter Speed, f-number (or T-stop, if you're pedantic).

Dan.

 dosdan's gear list:dosdan's gear list
Pentax K20D Pentax K-5 Pentax Q Pentax K-01 Pentax K-3
Reply   Reply with quote   Complain
Mike_PEAT
Forum ProPosts: 10,191Gear list
Like?
There's lots of misnomers, crop factor is another one!
In reply to dosdan, Jan 30, 2013

Calling ANY sensor that's smaller than 135/35mm format (the so-called "full frame") as having a "crop factor" is anther one.

If you use a DX lens on a DX camera there is no cropping taking place, therefore no crop factor. There is a MULTIPLICATION FACTOR IF you want to compare it to a reference format, i.e. 135/35mm format.

Don't know how many times I've had a Canikony user ask me what the crop factor of my FourThirds camera is, when the lens attached is designed for the Four Thirds mount, so no cropping is taking place. Bridge & p&s cameras don't have removable lenses, yet they are said to have a crop factor even though you can't mount SLR lenses to them.

And what happens when you compare a larger format camera, do you then have a NEGATIVE crop factor?

Unfortunately Wiki has it wrong promoting the wrong use of the term "crop factor"...the ONLY time you have cropping taking place is when you have a lens on your camera designed for a larger format. For instance if I take a medium format (4x5) lens and put it on a 135/35mm ("full frame") camera, you'd have a crop factor on the full frame camera!

 Mike_PEAT's gear list:Mike_PEAT's gear list
Lytro Light Field 16GB
Reply   Reply with quote   Complain
AlbertInFrance
Senior MemberPosts: 2,190Gear list
Like?
Re: Exposure control in a raw developer: misnomer?
In reply to dosdan, Jan 30, 2013

For me, the exposure at the image plane is defined by the subject reflectivity, the incident illumination, the aperture and the shutter speed. The image exposure is further affected by the photosensitivity of the sensor (ccd, film ...).

With a camera where you can physically change the sensitivity of the sensor (increasing the gain or pushing film development) then the image exposure is affected by that change. There may be other effects (noise, grain, contrast ...) but the image density (for film) or the recorded numerical values (digital) for a given image plane exposure will be modified. This will allow you to modify the image plane exposure.

If you have a camera where the photosensitivity of the sensor is fixed then any post-processing changes merely select different ranges of values from those that were recorded when the photo was taken. Thinking of a classic H&D curve you can use different parts of the curve (print only the highlight end or just the shadow end). Eventually you will move into the shoulder or the toe and lose tonal range.

Similarly, if you got the image exposure 'wrong' (non-optimal) you can make adjustments in post processing, but ultimately you are restricted by the range of tones that were recorded by the camera. For a digital system the amount of leeway you have depends on the relative H&D curves of the RAW file and the JPEG, print or whatever that is used for display.

-- hide signature --

Albert
Every photograph is an abstraction from reality.

 AlbertInFrance's gear list:AlbertInFrance's gear list
Canon PowerShot G15 Fujifilm X-E1 Sigma 70-300mm F4-5.6 APO DG Macro Fujifilm XF 18mm F2 R Fujifilm XF 35mm F1.4 R +2 more
Reply   Reply with quote   Complain
Jack Hogan
Senior MemberPosts: 3,717Gear list
Like?
Linear vs Non-linear Brightness
In reply to dosdan, Jan 30, 2013

dosdan wrote:

As I understand it, the "Exposure Triangle" is composed of Scene Luminance, Shutter Speed, Aperture (or should that be "F-number"?). These 3 alone determine the number of photons captured by the sensor.

Correct, but if we want to be precise we need to know what the light source is and speak of the average luminance at the sensor (L), exposure time (t) and f-number (N). For instance, with the spectral power distribution of a 5000K blackbody illuminant, the number of photons incident on the sensor is approximately:

13000 * Lt / N^2 photons per micrometer squared

ISO Sensitivity plays a role in the rendered brightness of the outputted image. (As well as helping to mitigate the contribution to the total read noise of the ADC noise floor in most cameras.)

In addition to OOC rendered brightness, depending on what definition of ISO your camera manufacturer chose, the ISO indicator of Sensitivity also interacts with your camera's metering and controls an intermediate step/choice, that of scaling (analogically and/or digitally) the photon count and storing it in the raw data file.

So, I take the phrase "exposing a shot" to refer to the capturing process, not the rendering process. With in-camera JPEG, these two processes are combined. When shooting raw, the development process is separated and postponed.

Correct, I would complement your statement by adding the scaling/storing component:

1) Expose (objectively determine the number of incoming photons)
2) Scale and Store (objectively scale the incoming photon count and store it in the raw data)
3) Render (present the raw data in a subjectively pleasing manner on the output medium of choice)

But most raw development programs have an "Exposure" adjustment control. But isn't this adjusting the brightness of the rendered output? If so, it has nothing to do with exposure and is a misnomer.

So what should this control be labelled? "Rendered Brightness"?

How about linear brightness (as opposed to non-linear, 'regular' brightness)? I think most people would be confused, and that's why Adobe combined the two in a single slider in LR4/ACR7 - misnaming it 'Exposure'. EC is really not that bad, as long as one uses the Saturation definition of ISO: within limits (and discounting noise and other effects), such a slider has in fact the same effect as dialing in an equivalent amount of in-camera Exposure Correction on the mean photon count as stored in the raw file.

Jack

Reply   Reply with quote   Complain
quadrox
Senior MemberPosts: 1,018Gear list
Like?
Film terminology perhaps?
In reply to dosdan, Jan 30, 2013

dosdan wrote:

As I understand it, the "Exposure Triangle" is composed of Scene Luminance, Shutter Speed, Aperture (or should that be "F-number"?). These 3 alone determine the number of photons captured by the sensor.

ISO Sensitivity plays a role in the rendered brightness of the outputted image. (As well as helping to mitigate the contribution to the total read noise of the ADC noise floor in most cameras.)

So, I take the phrase "exposing a shot" to refer to the capturing process, not the rendering process. With in-camera JPEG, these two processes are combined. When shooting raw, the development process is separated and postponed.

But most raw development programs have an "Exposure" adjustment control. But isn't this adjusting the brightness of the rendered output? If so, it has nothing to do with exposure and is a misnomer.

So what should this control be labelled? "Rendered Brightness"?

What about a suitable term for the combined capturing & rendering process, that doesn't confuse it with exposing?

Dan.

When shooting film, you would typically get a negative image on the roll of film itself. To develop the negative into a proper (positive) image, you must project the negative onto the film again. Of course, during this process you have the possibility to adjust exposure again by adjusting the length of the exposure of the negative onto the positive medium. This way the positive could be developed darker or brighter than the original negative was recorded.

Calling this adjustment exposure seems quite correct to me, and by extension calling the digital equivalent in RAW processing exposure seems fair as well.

Disclaimer: I know only the basics of shooting and developing film and know not if the above is actually correct, but it seems to me that it should be.

 quadrox's gear list:quadrox's gear list
Canon EOS 5D Mark II Canon EOS 7D Canon EF 85mm f/1.8 USM Canon EF-S 17-55mm f/2.8 IS USM Canon EF 55-200mm f/4.5-5.6 II USM +2 more
Reply   Reply with quote   Complain
sherwoodpete
Veteran MemberPosts: 7,761
Like?
Re: There's lots of misnomers, crop factor is another one!
In reply to Mike_PEAT, Jan 30, 2013

Mike_PEAT wrote:

Don't know how many times I've had a Canikony user ask me what the crop factor of my FourThirds camera is, when the lens attached is designed for the Four Thirds mount, so no cropping is taking place.

I think I've seen at least one Olympus four-thirds lens actually marked with both the actual focal length, and also having in large letters the 35mm equivalent focal length displayed on the side too.

When manufacturers do things wrongly, its difficult to expect people to not be follow their lead.

Regards,
Peter

Reply   Reply with quote   Complain
sherwoodpete
Veteran MemberPosts: 7,761
Like?
Re: Film terminology perhaps?
In reply to quadrox, Jan 30, 2013

quadrox wrote:

dosdan wrote:

As I understand it, the "Exposure Triangle" is composed of Scene Luminance, Shutter Speed, Aperture (or should that be "F-number"?). These 3 alone determine the number of photons captured by the sensor.

ISO Sensitivity plays a role in the rendered brightness of the outputted image. (As well as helping to mitigate the contribution to the total read noise of the ADC noise floor in most cameras.)

So, I take the phrase "exposing a shot" to refer to the capturing process, not the rendering process. With in-camera JPEG, these two processes are combined. When shooting raw, the development process is separated and postponed.

But most raw development programs have an "Exposure" adjustment control. But isn't this adjusting the brightness of the rendered output? If so, it has nothing to do with exposure and is a misnomer.

So what should this control be labelled? "Rendered Brightness"?

What about a suitable term for the combined capturing & rendering process, that doesn't confuse it with exposing?

Dan.

When shooting film, you would typically get a negative image on the roll of film itself. To develop the negative into a proper (positive) image, you must project the negative onto the film again. Of course, during this process you have the possibility to adjust exposure again by adjusting the length of the exposure of the negative onto the positive medium. This way the positive could be developed darker or brighter than the original negative was recorded.

Calling this adjustment exposure seems quite correct to me, and by extension calling the digital equivalent in RAW processing exposure seems fair as well.

Disclaimer: I know only the basics of shooting and developing film and know not if the above is actually correct, but it seems to me that it should be.

Although what you say makes sense, that's possibly more like adjusting the brightness. It doesn't give the same results as altering the processing of the film itself.

Regards,
Peter

Reply   Reply with quote   Complain
John Sheehy
Forum ProPosts: 16,026
Like?
Re: Examples
In reply to jcharding, Jan 30, 2013

jcharding wrote:

I'd also point out that all of ISO, aperture and shutter speed affect rendered brightness.

Only if you shoot JPEGs or accept converter defaults.  There is nothing in a RAW file that dictates what brightness is possible from the file.  The only issues are noise in the shadows, and clipping of the highlights.

ISO exposure index doesn't even have to be established until the image is displayed, and a displayed image doesn't even need to have a single ISO; different parts of the image or different tonal ranges can have different exposure indices.

In fact, if you had a super-DR image and a super-DR monitor in a dark room with charcoal black walls, and walked in from the sunlight, the ISO would change from low to high as your eyes adjusted to the dark areas.  Existing notions of ISO exposure indices are dependent on temporary technology issues, where DR of display devices is limited.

Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads