is this right?

[xpatUSA]

Sorry. Digitized analogue sensel values, in Dave's example 12 or 14 bits.

George

OK.

[george013]

Just reading the RawDigger man, all those digitized analogue values are called pixels.

Just parsing through the manual I can't find any reference on how may stops one can correct. Beside personal interpretation of the histogram.
It seems I had it installed in this pc before. Can't install the trial again.

George

[xpatUSA]

Just reading the RawDigger man, all those digitized analogue values are called pixels.

So I see: "Overexposure and Underexposure Statistics will show the percentage of overexposed and underexposed pixels."

Oops. George, I am so accustomed to Foveon stuff; I forgot that each color channel from a CFA sensor is indeed called a "pixel" ... sorry!

Just parsing through the manual I can't find any reference on how may stops one can correct.

Sorry, I don't understand that at all.

[dje]

Here we are George, some higher exposures, similar story as before. Not quite as linear by the looks of things but still approx double DN's for 1 stop exposure increase. (It could also be inaccuracy in shutter speeds). First two are 14 bit, second two 12 bit.

[george013]

Here we are George, some higher exposures, similar story as before. Not quite as linear by the looks of things but still approx double DN's for 1 stop exposure increase. (It could also be inaccuracy in shutter speeds). First two are 14 bit, second two 12 bit.

If the histogram on your camera is peaking somewhere in the middle than I guess you've 2 or 3 stops left before clipping. I tried it on my camera. In RawDigger only one. I wonder what happens when you take a shot with max exposure, no clipping. Or just a little bit. Sorry, I can't do it.
Still not sure what that EV division means.
Another thing I'm curious of is how the histogram in the editor looks.

George

[dje]

If the histogram on your camera is peaking somewhere in the middle than I guess you've 2 or 3 stops left before clipping. I tried it on my camera. In RawDigger only one. I wonder what happens when you take a shot with max exposure, no clipping. Or just a little bit. Sorry, I can't do it.
Still not sure what that EV division means.
Another thing I'm curious of is how the histogram in the editor looks.

George

The EV scale is just a log (base 2) conversion of the DN numbers. Think of it as a relative scale though as you can adjust the EV0 point to whatever you like.

When I look at the histogram of the first shot of this latest set in ACR, after WB fine tuning I get just one peak at about 180. This is using a camera profile with a linear tone curve. However gamma 2.2 is still applied to the histogram.

[xpatUSA]

If the histogram on your camera is peaking somewhere in the middle than I guess you've 2 or 3 stops left before clipping.

Good guess!

In theory, for a peak of 127/255:

(127/255)^2.2 = 0.216 (21.576% exposure).

to clip at 1 (100%) requires 1/0.216 = 4.635 times the exposure = 2.212 stops.

[george013]

Good guess!

In theory, for a peak of 127/255:

(127/255)^2.2 = 0.216 (21.576% exposure).

to clip at 1 (100%) requires 1/0.216 = 4.635 times the exposure = 2.212 stops.

I didn't know you could be friendly too!

Serious. It does explain a lot, both the background, formula and practical experience.
But when RawDigger is showing a peaking in the middle it means I've one stop left before clipping. And that one stop uncorrected exposure will be equal with 2.212 stop "corrected exposure". Exposure is done with shutter speed an f-number. With 1 stop I consume all the free capacity in that raw histogram.

George

[xpatUSA]

I didn't know you could be friendly too!

Serious. It does explain a lot, both the background, formula and practical experience.
But when RawDigger is showing a peaking in the middle [of the horizontal linear scale relative to the saturation level], it means I've one stop left before clipping. <> Exposure is done with shutter speed and f-number. With 1 stop, I consume all the free capacity in that raw histogram.

All correct.

[royent]

Isn't it marvelous? I've been taking adequate photographs for nearly 65 years without knowing any of this
Roy

[ccphoto]

Isn't it marvelous? I've been taking adequate photographs for nearly 65 years without knowing any of this
Roy

It's not that I didn't at least know the basics of all these formulations, it's more like Roy is saying, SAY WHAT! Without a good eye for composition, a firm knowledge of shutter/aperture/ISO and EV connections with an added great sense of tonal evaluation both when shooting and processing, all these number don't mean doodley squat. I understand that what I do is heavily saturated in the above formulations, but I tend to rely on the old brain in making on the spot calculations based on experience and the basics. I am happy they make you guys happy but there isn't a thing you've written that will change a thing about the way I calculate my exposures when I shoot. Like Roy, I do pretty well with just understanding the simple stuff.

Happy Trails

[DanK]

Personally, although I like to understand the technical underpinnings of what I am doing with the camera, this thread went beyond what I found useful. One reason is the nature of the questions.

Regardless, I found the last two postings annoying. If you don't care to understand the technical underpinnings of digital photography, that's fine, but it is no reason to disparage people who do want to understand or threads in which they discuss them.

And while this thread may have gone beyond what is useful, a lot of the technical information provided by people who post on this forum is practically useful. I have certainly profited from it. In addition, I've given people instruction in night photography, macro photography, and focus stacking--classes in the last instance--and in all of these cases, understanding some technical issues allows people to take better images. They could learn rules by rote instead, of course, but then they wouldn't have the flexibility to adapt.

A weak analogy is jazz. When most people think about jazz, they think of improvisation. What they may not think of is the rich knowledge of music theory that underlies being able to improvise flexibly. I once heard a recording in which Herbie Hancock, the jazz pianist, accompanied two young Irish vocalists. I found it beautiful and haunting, but not knowing much music theory, I had no idea how they created it. Then I heard a recording of Hancock praising the two others. He explained how unusual the female vocalist's improvisations were--describing them precisely in terms of specific chord sequences and arpeggios. And the successful ones have learned these "changes" well enough that they can use them with some automaticity, just as experienced photographers can make some adjustments on the fly. Wynton Marsalis has sometimes discussed this. At one lecture/performance he gave here a year or two ago, he and members of his quintet discussed this. They explained that they needed all those changes to be so familiar that they could draw on them without stopping to think. Then one of them said that of course, they will sometimes make mistakes. Another laughed and said, 'you [the audience] won't hear them as mistakes, but we will.'

[george013]

My English is far from good. So I'm not always able to feel sarcasm. Maybe I should be glad for that.
My first question was can I calculate with the pixel values? So far there're 2 appearances of those values: the Raw channel values and the pixel values of the raster image. The pixel values of the raster image are the result of the demosaicing process and the gamma correction. One can't calculate with those.
But can it be done on the Raw values? I still have my doubts. It would mean I can create an ETTR image on the backhand. It would also mean I can change the ISO on the backhand.
There must be more.

George

[ccphoto]

"Regardless, I found the last two postings annoying. If you don't care to understand the technical underpinnings of digital photography, that's fine, but it is no reason to disparage people who do want to understand or threads in which they discuss them."

I did not feel I disparaged anyone's point of view and if I did, I certainly apologize as that was not my intent. I only voiced an opinion that knowing all these formulations don't help anyone if they don't first have a firm foundation in the basics of aperture, shutter, ISO and EV and in knowing these foundations, wondered how much this depth of understanding would help the average pretty good photographer.

[DanK]

"Regardless, I found the last two postings annoying. If you don't care to understand the technical underpinnings of digital photography, that's fine, but it is no reason to disparage people who do want to understand or threads in which they discuss them."

I did not feel I disparaged anyone's point of view and if I did, I certainly apologize as that was not my intent. I only voiced an opinion that knowing all these formulations don't help anyone if they don't first have a firm foundation in the basics of aperture, shutter, ISO and EV and in knowing these foundations, wondered how much this depth of understanding would help the average pretty good photographer.

Sorry that I misunderstood you. I certainly agree that this stuff is no substitute for knowing the basics.

[royent]

Dan and George. I too feel the same as Chris, in that I had no intention of disparaging anybody. My comment was not meant to be sarcastic either. Personally I find that I'm better knowing how to do something rather than why I'm doing it. I suppose that I'm practical rather than theoretical.
Roy

[DanK]

Roy, sorry again that I misinterpreted. That's a risk of electronic communication.

[xpatUSA]

My English is far from good. So I'm not always able to feel sarcasm. Maybe I should be glad for that.
My first question was can I calculate with the pixel values? So far there're 2 appearances of those values: the Raw channel values and the pixel values of the raster image. The pixel values of the raster image are the result of the demosaicing process and the gamma correction. One can't calculate with those.
But can it be done on the Raw values? I still have my doubts. It would mean I can create an ETTR image on the backhand. It would also mean I can change the ISO on the backhand.
There must be more.

George

What is meant by "on the backhand"? I don't understand ...

[george013]

What is meant by "on the backhand"? I don't understand ...

Afterwards. In the ETTR question I would also multiply the noise. So that wouldn't be helpful. But what about changing the ISO afterwards? Just trying to find an answer on my question if one can calculate with pixel values.

George

[pnodrog]

Afterwards. In the ETTR question I would also multiply the noise. So that wouldn't be helpful. But what about changing the ISO afterwards? Just trying to find an answer on my question if one can calculate with pixel values.

George

The whole basis of ETTR is as you increase the amount of light (exposure) the noise ratio decreases. Adjusting ISO after exposure amplifies the noise as well as the effective brightness level of the captured charge. It will never match the lower noise achieved by accumulating a bigger charge by correct or prudent (ETTR) over exposure.

I trust someone will state this more eloquently.