How to visually gauge absolute black in nature and tone map it to output.

[Tim Lookingbill]

I have to wonder why in the digital age, so many people have these issues when mankind has been creating art work for centuries before we even had man made lighting!

Because mankind by and large never had to edit or create their images on a transmissive display and certainly not with captures whose dynamic range contains tons of data we can now zoom in with a simple keystroke better than a loupe on a light table and massage all that detail out of Raw captures. Zoom out and look at the overall results and not note that we made the image either too dark or low contrast or high contrast or too sharp or saturated.

This has nothing to do with the discussion of editing images to compensate for various differences in display contrast and brightness for web viewing. This is about editing the image to a minimum brightness that can at least communicate what the scene is about in all normal light levels.

It doesn't change the image that much aesthetically by conducting a reality check like this. We're here for the print, RIGHT?, not what people see online on uncalibrated displays.

This is an old issue that goes back to my training preparing and designing images for mass production before digital imaging came along and made things so much easier with far more options. We should use those options and I'm just illustrating from my own experience how I have to be mindful when using these fantastic digital tools.

I don't know what 'normal room lighting' implies but if such environments have any kind of art work, and the people there want to see it represented as best as possible, they need adequate lighting.

The minimum light level I demonstrated in the image above. Anything darker is pointless. Those compensatory edits don't affect the overall image that much aesthetically when viewed in a controlled viewing booth situation.

[digitaldog]

Because mankind by and large never had to edit or create their images on a transmissive display and certainly not with captures whose dynamic range contains tons of data we can now zoom in with a simple keystroke better than a loupe on a light table and massage all that detail out of Raw captures.

Fair enough and I totally agree about the part using the emissive display, an intermediate part of the final process of which results in a print. But then how do you account for the print? People view the display, most likely incorrectly calibrated to match the print, ignore the mismatch believing the display is 'right' and don't look at the print? Doesn't make much sense. Do we have an issue with a disconnect between an emissive and reflective representation of an image, even with the best soft proofing the technology provides? Sure. But the proof is in the proof (or print). So what's the problem?

This has nothing to do with the discussion of editing images to compensate for various differences in display contrast and brightness for web viewing.

It does in the context that print viewing conditions and display viewing realities are all over the map and you can't hit a prefect middle ground.

[Tim Lookingbill]

Fair enough and I totally agree about the part using the emissive display, an intermediate part of the final process of which results in a print. But then how do you account for the print? People view the display, most likely incorrectly calibrated to match the print, ignore the mismatch believing the display is 'right' and don't look at the print? Doesn't make much sense. 

I'm not advocating adjusting the image or calibration for variances in web viewing. That's way too much work. Stick with a calibrated display and print matching viewing booth and forget what others see on uncalibrated displays. That's why this thread has nothing to do with the other.

Do we have an issue with a disconnect between an emissive and reflective representation of an image, even with the best soft proofing the technology provides? Sure. But the proof is in the proof (or print). So what's the problem? It does in the context that print viewing conditions and display viewing realities are all over the map and you can't hit a prefect middle ground.

The problem I've already illustrated in this thread and offered some type of solution which is to be mindful of what editing images on a transmissive display can do to the eyes and how it affects perception and brightness judgement. I've never had any issues or need for soft proofing because all my images, regardless if I made them too dark or too light or saturated, look exactly as they appear under my controlled print viewing environment.

I'm talking about initial editing of the overall image to a minimum brightness/contrast so the overall look of what's depicted in the image can be seen in minimum lighting environments. We're not talking about a major edit here.

For example prints of cityscapes at night aren't going to fair well in a dimly lit restaurant but then most wouldn't care if detail in the shadows is seen by patrons eating their steak anyway. Of course the reflections in foreground water from cityscape lights might benefit from lifting the shadows and or adding some clarity. That's the kind of mindful edits I'm referring.

[digitaldog]

IF the print, in whatever condition it's viewed is too dark, you only have three options I can think of:

1. Lighten the print.
2. Increase the illuminant.
3. A bit of both.

The display has nothing to do with the above.

[Isaac]

For example prints of cityscapes at night aren't going to fair well in a dimly lit restaurant but then most wouldn't care if detail in the shadows is seen by patrons eating their steak anyway.

That scenario seems unlikely to me, restaurants that show art work always seem to light the art work.

Or the artist takes on the responsibility -- "For places where track light wasn’t an option, I got some clip-on light fixtures…"

[digitaldog]

The display has nothing to do with the above.

And there were the days we had to shoot transparency film for reproduction on some kind of ink on paper that was carried all over the place in differing viewing conditions. Somehow, we managed to get that to work.

[PeterAit]

I don't need condescending responses that question my thinking. 

I was not being condescending, I was trying to be helpful. And if your thinking is beyond questioning, perhaps I can have your autograph? (As an FYI, that is not condescending, it is sarcastic).

[Schewe]

I'm talking about initial editing of the overall image to a minimum brightness/contrast so the overall look of what's depicted in the image can be seen in minimum lighting environments. We're not talking about a major edit here.

For example prints of cityscapes at night aren't going to fair well in a dimly lit restaurant but then most wouldn't care if detail in the shadows is seen by patrons eating their steak anyway. Of course the reflections in foreground water from cityscape lights might benefit from lifting the shadows and or adding some clarity. That's the kind of mindful edits I'm referring.

What you are talking about is a known problem (with work arounds to resolve them). The bottom line is that at dim light levels, your eyes don't perform as well as with brighter levels. You lose contrast and color saturation when viewing an image in dark illumination.

You can compensate by lightening and increasing contrast (in fact, Lightroom has such a function built in called Print Brightness–which Andrew hates, but was Thomas Knoll's solution to "why are my prints too dark") but doesn't address the loss of color saturation at dim levels...

All of this can be tested and adjusted based upon trial and error–if you want to go down that rabbit hole.

But the bottom line is, yes, if a print will be viewed under less than ideal light levels, you can make adjustments to mitigate how a print will be viewed under low lighting levels.

Which is actually a different thingie that what you first posted about..one of the things I see a lot of people struggle with is not crunching the blacks in images. While there is no such thing as a "perfect histogram". I see a lot of people shying away from clipping blacks (while also lightening shadows). A print pretty much always needs a "real black" otherwise the image tends to recede too much. Punching the blacks can actually help show shadows and midtones better...there's usually less issues dealing with highlights when viewing a print under dim light.

I actually wrote about this issue in The Digital Print. The best solution is, of course, to put more light on a print.

I recently went to NYC and went to the Whitney Museum to see a Steiglest show...the museum lighting is always very low level (to protect the prints) and of course, the prints look like shit. All you can do in that situation is to close your eyes and wait till your eyes adapt to darker conditions, then open and look at the dimly lit prints. It sucks...but the print curators, in an attempt to protect the prints from bright lights are very, very conservative. It's a trade off of viewing vs conservation...and I realize how important conservation is, so I've learned how to adapt my viewing to dim conditions. Sucks, but it is, what it is...

For my own work, I always display under bright lights :~)
(cause I'm not concerned about conservation of my prints)

[digitaldog]

I recently went to NYC and went to the Whitney Museum to see a Steiglest show...the museum lighting is always very low level (to protect the prints) and of course, the prints look like shit

The Andrew Smith gallery had a huge Ansel Adams show (http://www.andrewsmithgallery.com/arrington_collection.html). This gallery is massive, three stories, very, very old building with rooms with some limited window light to none. The provided lighting was just awful, many of the prints looked like shit.

Update: so does the web images 
http://www.andrewsmithgallery.com/exhibitions/anseladams/sierra_club/index.html

[Tim Lookingbill]

What you are talking about is a known problem (with work arounds to resolve them). The bottom line is that at dim light levels, your eyes don't perform as well as with brighter levels. You lose contrast and color saturation when viewing an image in dark illumination.

As long as what's depicted is recognizable will suffice was my point which only requires viewable areas of the image be brightened to a minimum enough to see. For example being able to recognize family members in DIY hobbyist individual and group portraitures comes to mind. These are exceptions of course compared to serious fine art photographers who can and are able to properly light their prints. However there are far more DIY hobbyists than fine art pro's who want their images viewable by those that might not have the proper lighting.

You can compensate by lightening and increasing contrast (in fact, Lightroom has such a function built in called Print Brightness–which Andrew hates, but was Thomas Knoll's solution to "why are my prints too dark") but doesn't address the loss of color saturation at dim levels...

Wasn't aware of that feature. Might have to check that out, but if it's anything as unreliable as Soft Proof's "Simulate Paper/Black Ink" setting I'll most likely side with Andrew. But thanks for mentioning that feature. I'm not really concerned with saturation levels. Just want to be able to see what the print is conveying. I thought my image samples of my framed prints viewed in the dimmest light made that clear. There's more to appreciate about an image than color. B&W street photographers can explain that point better than I.

But the bottom line is, yes, if a print will be viewed under less than ideal light levels, you can make adjustments to mitigate how a print will be viewed under low lighting levels.

I'm more for overriding the adaptive effect from editing on a brightly lit transmissive display to prevent me from making my renderings darker than they should be and so I'm establishing for my own purposes some minimum RGB number for shadows that I know will be viewable in reasonably  minimum lighting situations without destroying the aesthetics and dynamics of the image. Sort of like a zone system for minimum print viewing conditions. Just a reality check for certain captured/rendered scenes that I now know are going to be trouble viewed in dim light but also look good in a print matching booth.

...I see a lot of people struggle with not crunching the blacks in images. While there is no such thing as a "perfect histogram". I see a lot of people shying away from clipping blacks (while also lightening shadows). A print pretty much always needs a "real black" otherwise the image tends to recede too much. Punching the blacks can actually help show shadows and midtones better...there's usually less issues dealing with highlights when viewing a print under dim light.

I agree, I like max black, too, but it usually requires, for me at least, a custom shaped point curve to smoothly blend out of black light enough shadow detail to be seen while not making blacks look like posterized blobs in brighter viewing conditions. I can actually see shadow detail in prints pretty good in dim diffused light mainly from adaptation to those light levels but more so if the roll off of shadow detail out of max black isn't too mushy. That's the part I'm working on and trying to improve if I can just stop editing my images too dark. Also the diffused dim light provides its own smoothing effect as it is.

I recently went to NYC and went to the Whitney Museum to see a Steiglest show...the museum lighting is always very low level (to protect the prints)...

For my own work, I always display under bright lights :~)
(cause I'm not concerned about conservation of my prints)

Museums don't display inkjet copies of famous photographer's work? If so, that's a big waste of technology IMO.

Thanks for your feedback, Jeff. Also appreciate everyone's points and thoughts on this subject.

[Wayne Fox]

I'm more for overriding the adaptive effect from editing on a brightly lit transmissive display to prevent me from making my renderings darker than they should be and so I'm establishing for my own purposes some minimum RGB number for shadows that I know will be viewable in reasonably  minimum lighting situations without destroying the aesthetics and dynamics of the image. Sort of like a zone system for minimum print viewing conditions.

Sort of wondering if you are too worried about the numbers.  I don't think whether your blackest black is 5,5,5 or 2,2,2 or even 0,0,0 will make much difference in the overall tonality of your output. If your prints are too dark I don't think your blackest black has a lot to do with that.

It's been a long time since I worried about the "numbers".  It's something everyone use to worry about and work with and I remember a long time ago mapping my black and white points with a levels adjustment, but I haven't done that in years ... I think since ACR was introduced.   , With current post processing techniques as well as  inks/papers/profiles not sure if there is a real number to worry about.  Overall tonality/density yes, because that's the challenge of using a transmissive device to predict what a reflective print will look like, but I don't think that changes the blacks, that's more about overall density.

[Schewe]

Museums don't display inkjet copies of famous photographer's work? If so, that's a big waste of technology IMO.

Museums will display inkjet prints, if that's all they have. I've seen inkjet prints at all the major museums...but if they have vintage silver gelatin prints, (printed by the artist) that's what they prefer–even if a corrected inkjet would look far better. They still look down on inkjet prints when there's an alternative. And yes, I've seen a lot of really good images printed on silver gelatin that looked way too dark in the dim museum lighting...and these are the major museums around the world. Sad really, but it seems like curators and conservationists have dumbed down the display standards to as low as they can get away with. Yes, it might help preserve a print for some additional time. But it also makes looking at what should be really great prints less satisfying to the viewer :~(

But hey, if I had the money, I would buy the prints then blast them with light when I wanted to look at them (and keep it dark when I wasn't' looking :~)

[mouse]

I recently went to NYC and went to the Whitney Museum to see a Steiglest show...

Bit my tongue several times but I just couldn't resist.  Alfred Stieglitz is too much an icon to allow it to stand uncorrected.  Forgive me. 

[Schewe]

Thanks for the spelling correction...truth be told I get really confused between Stieglitz & Steichen–so, who was it that slept with Georgia O'Keeffe?

Yeah, ok, I had to look it up...it was Stieglitz who slept with Georgia. And I had to look up the friggin' show I saw, it was Steichen...

But, the prints were really dark in the museum lighting environment. (which was the point i was trying to make!)

:~)

Sorry, grey moment...(but I still can't keep the two of them straight).

[Eric Myrvaagnes]

(but I still can't keep the two of them straight).

It's easy: Stieglitz good; Steichen not so good (IMNSHO). 

[Tim Lookingbill]

Sort of wondering if you are too worried about the numbers.  I don't think whether your blackest black is 5,5,5 or 2,2,2 or even 0,0,0 will make much difference in the overall tonality of your output. If your prints are too dark I don't think your blackest black has a lot to do with that.

It's been a long time since I worried about the "numbers".  It's something everyone use to worry about and work with and I remember a long time ago mapping my black and white points with a levels adjustment, but I haven't done that in years ... I think since ACR was introduced.   , With current post processing techniques as well as  inks/papers/profiles not sure if there is a real number to worry about.  Overall tonality/density yes, because that's the challenge of using a transmissive device to predict what a reflective print will look like, but I don't think that changes the blacks, that's more about overall density.

Not worried at all, Wayne.

It's just one number I now go for in the shadow zone of images whose dynamics affects my perception into editing them too dark. The first sample I posted of the rocks/water I took note of the RGB numbers in ProPhotoRGB of the fine midtone gravel among the darker dead leaves, green algae and greenish flat stones whose darkest shadows were at around 3,5,20 black. The gravel which should've at least read 127RGB in the green channel (2.2 gamma encoding) was at 99. That scene was shot with the sun directly overhead which I'ld forgotten about and was later reminded after reworks by noticing the thin shadows along the big white rocks above the water. I had to question my perception of this image in deciding just how bright to make it while preserving the rich green/blue colors, dynamics and textures.

Surprisingly the brightened version still has a decent black point but it's not in the shadows of the debris in the water, it's in the small sliver of space in the cracks between the white rocks above water. I checked holding down "Option" and dragging the black slider. So now I figured the brighter the actual scene being shot the smaller the area absolute black will reside. But there's still some absolute black in any image, it's just that by the time I get the image on my computer to edit it I can no longer remember the character of light to determine how far to go in setting shadow densities especially if I ETTR or expose to preserve highlights.

It's a bright scene so why should it look darker to the point you can't make out detail in the print in dim light while other prints of scenes shot in dim overcast light look OK under the same dimly lit print viewing conditions such as the deer in my other posted sample. It's nice to be able to edit to taste on the display and not be concerned about numbers but at some point a pragmatic approach to overcome limits in human perception must come into play to prevent making dark prints.

BTW here's a quote from "Set Print Color Management" section LR4 User Guide page 176 on "Print Bright" settings Jeff mentioned before...

3.(Optional) To achieve colors in print that more closely resemble the bright and saturated look of onscreen colors in Lightroom, select Print Adjustment. Then, drag the Brightness and Contrast sliders.

Note: Dragging the Brightness and Contrast sliders produces tone curve adjustments. These adjustments do not preview onscreen. It may take some experimentation to determine what settings work best for your individual photos and your specific printer.

The part in bold makes me wonder if the author is addressing this perception issue I've been dealing with or the limits of screen to print matching.

[JRSmit]

IF the print, in whatever condition it's viewed is too dark, you only have three options I can think of:

1. Lighten the print.
2. Increase the illuminant.
3. A bit of both.

The display has nothing to do with the above.

Ditto. And take the Kruithof curve into account.

[Tim Lookingbill]

Ditto. And take the Kruithof curve into account.

Clearly you and Andrew aren't reading what I keep having to repeat about five times so far from having to read through unhelpful, over simplistic and irrelevant answers to my points which I think are valid.

I'll say it again for the obtuse crowd, I'm not talking about "pleasing" viewing conditions of prints.

[Frankomatic]

I've been struggling with the same issue as Tim.  I've read some articles, don't remember where, that allude to processing for placing a print in ambient living room lighting but most don't go into detail about that processing.  The best I could find in my search is to use a contrast mask.  I have the PhotoKit plugin, which can produce contrast masks, but I send files to a lab for printing and haven't yet ordered proofs.  Want to learn more before I do.

I also found this article titled Softening Contrast, Balancing Light, but haven't yet used the actions.
http://tonykuyper.wordpress.com/2010/04/25/softening-contrast-balancing-light/

I'm wondering what the more experienced members think of the above techniques.

Regards,
Frank

[digitaldog]

Clearly you and Andrew aren't reading what I keep having to repeat about five times so far from having to read through unhelpful, over simplistic and irrelevant answers to my points which I think are valid.

Cool, JR and I are dismissed from 'class' and now move on to better use of our time.