I'm currently shooting in RAW format (Nikon D70). I do know that the preview that shows up on the LCD screen is already processed with curves applied even when you're shooting raw. I was wondering if the resulting histogram displayed on the cameras screen is of the processed image or if it's of the actual raw data. My suspicion is it's that of the processed image because the on camera preview looks well exposed, but when I open the raw file in Photoshop the image looks underexposed (when the exposure setting is at zero in the raw conversion box). If I assumed right is there a way of getting an in-camera histogram of the actual raw image rather than of the preview image with curves applied?

The histogram is of the process jpg preview embedded within the raw file. Not the raw file itself and there is no way to get a histo of the raw data on the camera.

The fact that the image looks well exposed is a result of not inaccurate histograms but the fact that camera makers design the LCD to display the image brighter than it is; about 1 stop or so brighter. Never trust the LCD preview image for color or exposure.

Thanks for the tip, I must say using the histogram has made a world of difference.

It would indeed be great to be able to get actual channel clipping information directly from the RAW file. This is a completely valid request, and a capability that many users would like to have... let's hope that the brands are listening.

The readability and usefulness of a full RAW linear histogram is probably less obvious.

Regards,
Bernard

I use the jpg settings to effect the histogram and preview depending on the kind of shooting I'm doing. That way I get an idea from the histogram of the clipping that will be caused by the type of post processing I usually do to that kind of photo. For example for wedding work I discovered that I PP with more contrast than I had set in camera and was running into problems with the contrast of the red channel (faces). Now it is adjusted higher in camera and I am seeing the results on screen which give me a better indication of where I stand vis a vis exposure.

It would be even better histogram wise if I would take a custom WB in camera so that the histogram would be accurate within the WB that the photo will eventually be set at, a blue WB will show underexposure compared to a yellow/warm one, all at the same exposure as can easily be seen when adjusting in the RAW converter. Trouble is that the majority of the time I don't have the time to shoot both a custom white balance, and a Whibal in the field, especially when shooting events.

It doesn't really matter if the histogram on the back is from the raw or jpg version of the image...the camera doesn't alter the levels in processing.

I shoot with a D70s in raw+jpg and the histogram is all I rely on because it's the only thing close to accurate. Realize it's an average of all three channels, but granted your white balance is fairly accurate it's close enough. If you want a good idea of how the image looks, drop the LCD brightness down two notches and then invision what it would look like if you could drop it down one more

I also use an original D70 and a D100 and they all act generally the same in this regard.

That's not quite accurate. The jpg has had contrast settings applied and has been converted from the raw data into a smaller color gamut, smaller tonal range and a different gamma than what is available from a raw file. The histogram will show clipping on the jpg before clipping would normally occur on a raw file.

To get a histogram that most accurately represents the raw file you should reduce saturation and contrast settings to 0 and set the color space to AdobeRGB. Even this is more constrained than raw data but it gets you as close to it as possible.

I was wrong, but for pratical purposes there isnt enough of a difference to really matter. Granted your exposing properly on a jpg, your raw's will come out fine. That's my experience anyways.

Sure, your raws will be fine, but if you want to maximize the potential of a raw file then having a histo as accurate as possible is essintial. The difference in histograms can vary from camera to camera and how noticeable it is will vary depending on one's workflow and desires. For me, the difference is very noticeable.

Exactly!

Regards,
Bernard

I don't agree. Yes the information may be there, but there is very little way that you are going to have all that information in the final processed screen/version print. Unless you are a landscape photographer using multiple 'versions' of the same file, set your jpg to reflect the contrast/gamut restraints that you will be outputting the file to so that your histogram will give you a more accurate picture of how much play you have within your constraints, especially as far as DR is concerned. Technically there may be ways of converting the file at a later stage (years hence with newer software) that will allow you to maximise all the information the RAW only hitogram would show you but for the present it would make more sense to shoot with the constraints of your output medium in mind.

For example, you may set the contrast and histogram to '0' in camera, and the histogram may show you that you have protected your highlights sufficiently. Problem is that you are not going to process with zero contrast and saturation, you don't want a totally flat image. Although it will be possible to increase both without blowing the channels, it will need careful work and won't be ideal by any means. A histogram which shows you values closer to how you will post process will have shown you that you are underexposing and when you back off half a stop or whatever, you will get a perfectly exposed photograph, even after applying the appropriate contrast curve to give an image with sufficient 'pop'.

Don't set your histogram to show you a false sense of complacency in your exposures, be realistic unless you enjoy having to extensively work each image to maximise all the information the histogram told you that you had.

You have a valid point pom, but for my tastes setting the camera the way I previously described gives me the best idea of what I have to work with in post production. Since all of my work is done in 16-bit/channel in ProPhoto RGB, there is considerable amount that I can do to an image with tone and color that no jpg preview could ever remotely come close to representing (especially with programs like Lightroom...).

For example, you may set the contrast and histogram to '0' in camera, and the histogram may show you that you have protected your highlights sufficiently. Problem is that you are not going to process with zero contrast and saturation, you don't want a totally flat image. Although it will be possible to increase both without blowing the channels, it will need careful work and won't be ideal by any means. A histogram which shows you values closer to how you will post process will have shown you that you are underexposing and when you back off half a stop or whatever, you will get a perfectly exposed photograph, even after applying the appropriate contrast curve to give an image with sufficient 'pop'.



What I'm getting is that setting in-camera contrast, saturation and sharpening to full will yield more room to play with these variables in post proccessing?

Nopt, that would be just as bad, you would end up with an underexposed image in all likelyhood. Match to how you usually post process, shoot a RAW file, then shoot a bunch of jpgs at different settings. Process the RAW file to the level of 'pop' that you usually print at and find the jpg that is closest. I do set sharpening to '5' myself as I often have to check sharpness on screen but it's up to you.

What 61Dynamic is saying makes sense. If you process with the largest gamut and in 16 bit, shooting landscapes, etc, then you would want to know how much information you have though I still wouldn't suggest shooting everything with the settings at zero, if you expose to the right then you will find yourself in trouble.

As for 61Dynamic's premise that he has the room based on the gamut/bit space, keep in mind that your screen can only show sRGB and 8 bit, and unless you are printing in that huge prophoto space and in 16 bit, your information is likely to be compressed back into the printable spaces. I print my wedding work on a frontier whose colour space is close to sRGB and 8bit. Therefore it makes sense for me to keep that in mind from the moment I shoot, my histogram should reflect that so I don't get misled. My Landscape work is printed in Adobe '98 16 bit so I can let my histogram show that. If I relied on the histogram of '98 and 16 bit in the raw converter when I was printing in sRBG then I would be editing based on a false histogram (remember we can't see more than sRBG). I prefer to know what's happening right from the beginning rather than finding out I'm in trouble when going to print.

What 61Dynamic is saying makes sense. If you process with the largest gamut and in 16 bit, shooting landscapes, etc, then you would want to know how much information you have though I still wouldn't suggest shooting everything with the settings at zero, if you expose to the right then you will find yourself in trouble.

As for 61Dynamic's premise that he has the room based on the gamut/bit space, keep in mind that your screen can only show sRGB and 8 bit, and unless you are printing in that huge prophoto space and in 16 bit, your information is likely to be compressed back into the printable spaces.


Seeing as the histogram shown on camera is of the 8 bit jpeg preview. How do I best decipher it as a representation of my raw exposure? I'd rather have an optimally exposed 16bit RAW file that I can downsample as needed rather than the other way around.

Check out MRs tutorial http://luminous-landscape.com/tutorials/expose-right.shtml which also contain a reference to another good link "Understanding Histograms".

While i think it would be nice to have a histogram from the RAW data, I would rather see multi-channel histograms available universally. This should be standard on any camera capable of displaying a histogram.


--
Daniel Bell

I won't say no to that, I can't understand how I managed without the RGB histo that I now use on the 5D, when shooting people, flesh tones are almost all red channel, it gives a far more accurate picture of where my exposure is.

What is Lightroom? Are there any processing programs you would suggest other than PS CS2 (as in they are better)? I don't care if they only do one action or a few, and I'm working on a Mac.

This is LightRoom. Michael Has a couple articles on it.

MR's essay is what introduced many of us to the advantages of exposing to the right (ETTR) and subsequent comments Thomas Knoll and Bruce Lindbloom were very helpful.

From my reading of MR's essay, he is relying on the histogram and flashing highlights to determine when highlight clipping occurs. For most work this is probably sufficient, but with most cameras clipping in the RAW channels does not occur until exposure is 0.5 to 1EV over that indicated by the above methods.

If you want to carry out ETTR fully, it is necessary to take a spot meter reading of the highlight and apply an exposure compensation that has been determined by previous testing to place the highlight just short of clipping in the RAW channels. This is often about +2.3EV over the metered reading. This is discussed at length in a thread on the Adobe ACR forum:

http://www.adobeforums.com/cgi-bin/webx?13...5c.3bbb579c/141

Of course, if you miscalculate, you risk blowing the highlights and losing data. IMHO, if the shadow portion of the histogram is not clipped, the camera is capturing the full dynamic range of the scene. ETTR will give less shadow noise and more levels in the shadows, but this may not be worth possible loss of highlight detail. Moreover, most raw converters roll of the shadows anyway, Nikon Capture more than Adobe Camera Raw.

Bill Janes

Since the contrast merely applies an S curve to the data, the quarter tones are most affected, but the actual highlights are little affected and I don't think that the histogram would change that much.

Here is a graphical output of a Stouffer step wedge (0.1 density steps or 0.3EV) of Nikon Capture renderings at normal, low and high contrast. If all the grid lines are not visible, view at a larger size.

Bill Janes

Just by a light Sekonic meter and forget about the histogram, or bracket everything. Blah blah.