Exposure Compensation

[xpatUSA]

However there is something strange in Capture, I can't use the exposure-slider when loaded a JPG. I'm still wondering why.

George

That certainly is a mystery, George. Could it be that Capture "knows" that a JPEG file has already had exposure-compensation applied in-camera, so it doesn't allow it to be done again?

I can't think of another benefit of the UNI-WB then to explore the behaviour of the sensor. I don't see its relevance for an image.

You must not have read or understood this link, which has been posted here several times already:

http://www.guillermoluijk.com/tutori...b/index_en.htm

There, the relevance of UniWB to "an image" is fully explained. Since it is a Google translation from Spanish, I will be happy to explain any part that you don't understand.

[george013]

That certainly is a mystery, George. Could it be that Capture "knows" that a JPEG file has already had exposure-compensation applied in-camera, so it doesn't allow it to be done again?

I think it's a part of the conversion as Urban said. I still don't know where to place it. But also with whitebalance algorithmes.

George

[xpatUSA]

. . . I thought have seen that RT is using percentages in stead of values in their histogram. I must say I don't know how to recall that histogram anymore. . . .

George

It's not in the histogram, it's just below the Navigator and RGB is in % as you say. The "accuracy", as you put it, is 0.1% - one in a thousand - ten times 'better' than 0-255, eh?

[ajohnw]

There is no essentiel difference between 8,10,12,14,16 bits. Only a difference in accuracy. Any value within the range you can replace with a percentage. I thought have seen that RT is using percentages in stead of values in their histogram. I must say I don't know how to recall that histogram anymore. Using floating point numbers is the same. Only a difference in accuracy and programming technique. The number 254 in 8 bits is 0.996....%, in 12 bits it is 4080. Both are representing exactly the same color and luminance.

When I'm talking of a rasterfile I mean that file that's the most important in digital photography but just doesn't has an own name. It's the file created out of the RAW-info and written in memory as a RGB-rasterfile. It's the file that's written to disk or to printer. A JPG-file or TIFF-file is a DISK-file.

I can't think of another benefit of the UNI-WB then to explore the behaviour of the sensor. I don't see its relevance for an image.

George

Different sensors and cameras have different dynamic ranges George. Over and out.

John
-

[Inkanyezi]

This subject has led to misunderstandings many times before, and I think it is essential to try understanding the implications. There are several ways to cope with saturation and clipping. There is also a particular order to do things when we process. Some operations should be done in RAW conversion to the "raster image", while other operations can be done afterwards. Tweaking the colour gamut to fit into a more narrow space can be done in Photoshop or any other image processing program that does 16 bit colour and the Prophoto colour space. But the essential white balancing or correction for exposure should be done in RAW.

When using the in-camera processing to jpeg, disregarding RAW, possibilities are more limited, and essentially, the only way to hold back exposure is to expose less. RAW will permit filling all three channels with useful data, and thus you can often expose much more and still get good results. That is the virtue of UNI-WB and ETTR. If you can use two stops more exposure, you will have lots of more data in the darker parts of the image.

Whether that means anything to you, or helps your creativity, is another matter.

[george013]

Different sensors and cameras have different dynamic ranges George. Over and out.

John
-

Nobody mentioned dynamic range. It's getting time you think about it, what is dynamic range.

Ted,
For if you didn't understand, showing a percentage doesn't say anything of the tonal depth. From my former post. Forget the percentage sign, it's already in the number, max.1.

The number 254 in 8 bits is 0.996....%, in 12 bits it is 4080. Both are representing exactly the same color and luminance.

George

[george013]

That is the virtue of UNI-WB and ETTR. If you can use two stops more exposure, you will have lots of more data in the darker parts of the image.

That's exactly what ETTR means. Overexposing if possible and correcting in PP. But that is just what you denied in a former thread.
George

[Inkanyezi]

That's exactly what ETTR means. Overexposing if possible and correcting in PP. But that is just what you denied in a former thread.
George

Hmmm...

I think I still deny it. In fact, the criterion I have for "over-exposure" is to clip a colour channel in RAW. We obviously must have different definitions of "over-exposure".

[ajohnw]

Nobody mentioned dynamic range. It's getting time you think about it, what is dynamic range.

Ted,
For if you didn't understand, showing a percentage doesn't say anything of the tonal depth. From my former post. Forget the percentage sign, it's already in the number, max.1.


George

You clearly don't George otherwise much of this wouldn't be causing you any confusions which it pretty obviously is.

John
-

[xpatUSA]

Ted,
For if you didn't understand, showing a percentage doesn't say anything of the tonal depth.
George

I understood 100.0% - accurate to one decimal place in fact

[george013]

Hmmm...

I think I still deny it. In fact, the criterion I have for "over-exposure" is to clip a colour channel in RAW. We obviously must have different definitions of "over-exposure".

These are your words, "how far to the right are you" post 13. Bold I added.

The idea of ETTR and "other concepts like that" is NOT to over-expose. The idea is to expose correctly for the workflow that is chosen. ETTR in no way promotes over-exposure. That is a total misconception of the idea. ETTR means that you place the white point when you expose the image.

George

[Inkanyezi]

Instead of talking in riddles George, why do you think that I suggest over-exposure, and what do you mean by over-exposure. If you explain, it will give us a more solid ground for discussion. I have already told you, that my definition of over-exposure is to blow out any colour channel. Yours evidently is different. Please explain what you mean by over-exposure. The quoted passages do not suggest over-exposure as I define it.

[george013]

Instead of talking in riddles George, why do you think that I suggest over-exposure, and what do you mean by over-exposure. If you explain, it will give us a more solid ground for discussion. I have already told you, that my definition of over-exposure is to blow out any colour channel. Yours evidently is different. Please explain what you mean by over-exposure. The quoted passages do not suggest over-exposure as I define it.

I don't understand you. You want me te explain your words? I can't.

That is the virtue of UNI-WB and ETTR. If you can use two stops more exposure, you will have lots of more data in the darker parts of the image.

The idea of ETTR and "other concepts like that" is NOT to over-expose. The idea is to expose correctly for the workflow that is chosen. ETTR in no way promotes over-exposure. That is a total misconception of the idea. ETTR means that you place the white point when you expose the image.

And now coming to the level of what I or you mean with overexposure? Has been discussed so many times and by so many different people. But I'll try it once more. Overexposure is more exposure as the lightmeter would suggest. Please, don't start a discussion on how to use the lightmeter. And in ETTR you use the eventual left space on the highlightside of the histogram, without clipping. And correcting the overexposure in PP.

George

[ajohnw]

That's exactly what ETTR means. Overexposing if possible and correcting in PP. But that is just what you denied in a former thread.
George

What you should say George is that ETTR MAY involve selecting a BRIGHTER exposure. May because some shots might do that without any exposure correction and brighter because it might be done via either speed or aperture settings.

In short exposure is set by viewing a histogram of the shot and always ensuring there is little or no space left at the right end no matter how dark the shot appears according to the histogram.

So to use it many people have to take a shot and then use the preview's histogram, learn what it's appearance means in terms of stops, reset the exposure accordingly and take the shot again this time exposing to the right. Some have a histogram in the viewfinder so can actually set the exposure directly but these are so small and indistinct that it's easy to miss low highlight area counts. One of my cameras changes the colour when clipping occurs which helps a lot with that aspect. Generally the resolution of these histograms is rather coarse. It is also possible to take a shot that shows zero tone at some level - flat line in the histogram - followed by a blip of highlights. A metered exposure is likely to miss those so they wont appear in the histogram - fix - preview again checking for blinkies or assume that all will be ok in the raw file.

ETTR is often associated with fiddling with a cameras jpg setting. The blinkies and histogram are based round a jpg's dynamic range. Raw files hopefully have a larger one. Compacts - often very little extra, 14 bit D/A cameras significantly more. One troublesome aspect of this is that increasing ISO also at some point decreases dynamic range by an amount set by the ISO level setting. Despite this people fiddle with jpg settings to try and make the blinkies and or histogram show raw file dynamic range. A common comment from people who find out about this area at some point is why don't the manufacturers show raw file dynamic range exposure information. Some of the possible reasons have already been mentioned. ISO changes are another complication. Jpg clipping levels also vary from on manufacturer to another as does the tone curve used to produce them. This even happens model to model from the same manufacturer. It's an area that is often fiddled with to try and produce excellent jpg's all of the time what ever is in front of the camera.

John
-

[Inkanyezi]

I don't understand you. You want me te explain your words? I can't.

In no way did I ask you to explain my words. I asked you what you mean by "over-exposure".

And now coming to the level of what I or you mean with overexposure? Has been discussed so many times and by so many different people. But I'll try it once more. Overexposure is more exposure as the lightmeter would suggest. Please, don't start a discussion on how to use the lightmeter. And in ETTR you use the eventual left space on the highlightside of the histogram, without clipping. And correcting the overexposure in PP.

George

So, you have a flatworldly attitude toward the light meter, and you do not understand what the light meter does, and then you say that I propose over-exposure? When you don't even know how the light meter works and refuse to accept my usage of the term over-exposure, even though I explained it in a way that should be easy to understand.

And then you have the nerve to ask that I should not explain how to measure light.

Well, we can keep that explanation outside the thread if you wish. You simply don't want to learn, and I respect your choice. But your post clarifies without any doubt, that you use the term "over-exposure" different, that we have different criteria for when a digital file will show signs of over-exposure. As we quite clearly do not speak the same language, there is no ground for discussion.

[george013]

What you should say George is that ETTR MAY involve selecting a BRIGHTER exposure. May because some shots might do that without any exposure correction and brighter because it might be done via either speed or aperture settings.
-

Some basics.
With exposure you control the amount of light that hits the sensor/film. It's done by a combination of shutterspeed and aperture diameter. Very simple and basic.
The lightmeter inicates what settings should be used for the existing light and sensibility of the sensor/film.
Overexposure is using a higher exposure as the lightmeter suggest.

George

[pnodrog]

It is a mater of semantics or definition of over exposure.

Urban (and I) define over exposure as an exposure where the light falling on the sensor exceeds the capcity of pixels on the sensor so information is lost - the bucket is overflowing. The highlight detail can no longer be recovered as we have no way of determining how much light was "spilled over".

You (and many others) seem to refer to over exposure as increasing the exposure beyond that indicated by the metering system. Many photographers use the the term a stop over exposed when refering to one stop of positive exposure compensation. It is a common phrase but if the highlights remain unclipped (your 254 or less for 8 bits) it has not been truly over exposed - the exposure has just been increased but the resultant data can still be validly adjusted mathmatically to produce PP manipulations that retains a controlled relationship between the lightest tone and the darker tones.

ETTR when used correctly should never cause true over exposure and the resultant irretrievable loss of highlight detail.

[george013]

It is a mater of semantics or definition of over exposure.

Urban (and I) define over exposure as an exposure where the light falling on the sensor exceeds the capcity of pixels on the sensor so information is lost - the bucket is overflowing. The highlight detail can no longer be recovered as we have no way of determining how much light was "spilled over".

You (and many others) seem to refer to over exposure as increasing the exposure beyond that indicated by the metering system. Many photographers use the the term a stop over exposed when refering to one stop of positive exposure compensation. It is a common phrase but if the highlights remain unclipped (your 254 or less for 8 bits) it has not been truly over exposed - the exposure has just been increased but the resultant data can still be validly adjusted mathmatically to produce PP manipulations that retains a controlled relationship between the lightest tone and the darker tones.

ETTR when used correctly should never cause true over exposure and the resultant irretrievable loss of highlight detail.

In that case you might have problems with underexposure. To me it's a reference to the "right"exposure, that's what gives a good exposed image. And if you're going to far at the high site, you get clipping, will it be in the JPG or the RAW.

George

[pnodrog]

No I do not have problems with under exposing. If the scene exceeds the dynamic range of my camera I make an informed choice as to whether to sacrifice detail in the highlights (allow some clipping) and retain good shadow detail or adjust the exposure to ensure no clipping in the highlights but risk losing some shadow detail with a possible increase in noise. I am skilful and experienced enough to set the exposure correctly for which ever choice I make.

[george013]

No I do not have problems with under exposing. If the scene exceeds the dynamic range of my camera I make an informed choice as to whether to sacrifice detail in the highlights (allow some clipping) and retain good shadow detail or adjust the exposure to ensure no clipping in the highlights but risk losing some shadow detail with a possible increase in noise. I am skilful and experienced enough to set the exposure correctly for which ever choice I make.

I mean a definition of under exposure, not under exposure as a choice.

The right exposure is the exposure you get from the lightmeter. And relative from that there is an under or over exposure. And in the extreme clipping can occure.

George