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

[Tim Lookingbill]

This subject has had me puzzled for some time especially seeing web images of other photographer's work not having any blacks near 5 RGB. Some shadows in landscapes of shaded detail look murky/foggy with max black levels at around 20RGB, but the overall appearance doesn't seem to suffer with some having an added nuance or "soul" to scene ambiance.

I also realize this topic of discussion is going to be hard to answer due to its subjectivity since most shots edited in post don't allow the photographer to go back and make note of the max density of the darkest detail in the original scene outside of studio shoots. I used to take notes when I painted landscapes to block out blacks, mids & highlights. Thought digital cameras and proper exposure would not require this but I don't have paints to tell me RGB values in ProPhotoRGB shooting out in the field.

I just want to get some ideas on how others interpret max black and whether it's difficult and/or contributes to dark prints viewed in regular household light. I'm beginning to believe dark prints are a result from long edits on transmissive displays attempting to grab as much dynamics within the scene over allowing aesthetic tastes to render proper contrast brought about by NOT going for absolute black.

With all the edits under my belt shooting Raw I've always sought out the darkest part of the image and forced it to measure around 5RGB working in ProPhotoRGB output space either with the point curve, black slider and/or Parametric curve in ACR/LR and tone map up the tonal scale while bringing out all the modeling and texture of surfaces brightly lit and in shadow. To my disappointment what has resulted from this is somewhat over contrasty and dark images after a second look months to even years later. I actually started to suspect my display calibration was out of wack but that's not the case.

And another disappointment as a result from this obsession with mapping absolute black to 5,5,5RGB has been...you guessed it...DARK PRINTS! And here I thought I was immune to this. It seems it's not as simple as I thought.

I've found it's very difficult editing images on a transmissive display to override or even notice the affects of adaption spending too long pulling dynamic range detail out of both highlights and shadows in Raws that already start out flat, dark, colorless and murky. I do the before and after and I think I've achieved a bright enough rendering of the actual scene to be reasonably viewable in a print viewed in normal household lighting. They've done it with paintings in the past. I don't see why I and others are having an issue with inkjet prints.

I'm having to do reprints of images I worked on years ago due to the fact I tone mapped them in a way that brought out tons of detail but created darker prints than I realized. It has a lot to do with judging and gauging overall contrast that is tied to where absolute black falls within the entire tonal scale of the actual scene. I'm now actually having to force myself to edit shadow levels according to where the green channel reads 100 in ProPhotoRGB for detail I want viewable in normal household lighting.

The first image below demonstrates my attempt re-editing shadow detail gauging with the green channel to make it viewable in regular household lighting while allowing black to fall where it may so it blends into the viewable shadows. I really don't think this has anything to do with matching display luminance to print viewing booth brightness because I'm aware that you can't get exact matches for every lighting situations especially for those who don't display prints using proper gallery lighting. We should be able to get an overall impression of the scene photographed on a print viewed in normal household dim lighting.

The second image shows the re-edited brightened image with the weak blacks that still looks presentable under bright viewing booth conditions. You'll also note the actual scene was bright at the time of capture but because of the tone mapped first attempt at accentuating all the modeling and texture detail the judgement of overall brightness suffered.

[Alan Klein]

Tim:  Please summarize with a simple question. 

[ErikKaffehr]

Hi,

Just as a general note, the way I work with stuff in LR5 is the following:

- First I crop the image
- Second step is applying gradient to darken sky if needed
- Adjust exposure
- Reduce highlights for good tonal separation
- Adjust blacks so I just have some clipping
- Increase  shadows to get some decent to in the image
- Adjust vibrance and clarity

This tends to be an iterative process

For printing I use soft proof. I have my screen calibrated at 90 cd/m^2, soft proofing works mostly well for me.

Best regards
Erik

[Tim Lookingbill]

Erik, nice rundown on your LR workflow, but how do your prints look viewing them in normal household viewing light levels as most folks have done when we used hang paintings on our walls without spot halogens lighting them?

The main point I'm trying to make here is that the wider the dynamics of the image and the technology allows, the denser the blacks become, the darker the image is going to look to where we have to renegotiate what we want viewable in a print viewing in severely reduced dynamics settings with regard to light. Note the character of light (gradualness in diffusion) in my shot of the prints on my wall and what I had to to do to brighten the original source image. They don't match up. It's not the light level that's the problem. It's the character of the light we view prints with regard to contrast.

Mapping black to absolute densities and still retaining viewable detail in normal household viewing light can't work. In film prints and with paintings the densities of blacks and thus the scaling of shadow detail was much lighter and so we could get a general impression of the entire image viewing in dim light. With the expanded dynamics we're now capable of capturing and reproducing creates viewing problems under reduced dynamic environments. It's more complicated than saying my prints are too dark.

Alan, I don't know how to reduce what I'm getting at into a simple question. Sorry.

[Tim Lookingbill]

To illustrate more clearly what I'm getting at with regard to black densities causing dark prints from expanded DR below is a link to a photograph of a print of a painting my grandmother displayed in her guest room we stayed on my summer visits as a child. This room was lit by one single 60 watt bulb six feet away from this print. No matter how dark and dim the room got I could make out every element in the image and could tell what it depicted. I could see the bird, the folds of the girl's dress, the bench in the woods (huh?) and bark detail in the trees. It wasn't an eyepopping image to begin with as you can see from the link.

http://www.antiquemystique.com/images/7158_jpg.jpg

In fact this was the typical dynamic range of most of the pictures we had hanging on our walls 40 years ago. When I hang my inkjet prints with more contrast, black densities and detail they're much darker under the same light just 2 to 3 feet away unless I intentionally lighten them and reduce contrast as I've shown above.

If I was to edit that old print so the darkest part under the bench went to 5RGB black or even 20RGB (currently around 50RGB) the rest of the tonal scale would have to descend into darkness in order to blend into black such as the bird, the back hills, the girls hair and folds of the dress. If I bumped the lightest part, the skylight above the back hill to 250RGB the only thing that would be viewable would be the girl's face.

[Simon Garrett]

You've made a lot of statements about blacks, but I'm struggling a little to understand the precise issue you have. 

Is it, for example:

• The difficulty in getting a print that differentiates different near-black tones
• The difficulty in getting a print that works in different lighting levels
• Something that occurs now, but didn't in the past with pictures 40 years old
• Something intrinsic to digital

I'm not meaning to be critical, just to try to narrow down the problem you're outlining to something we can get our teeth into. 

[TonyW]

Not sure I fully understand as my experience of print viewing suggests a huge difference in how a print is perceived depending on the lighting level at the print surface.

Apologies in advance but just my thoughts and not trying to ‘teach granny to suck eggs’  

I would expect that as the lighting levels drop to less than ideal for the print we are likely to see a loss in shadow detail and less saturation in the colours.  If we then throw some bright display lighting at the print then it should lift tremendously in contrast deep blacks, plenty of shadow detail - assuming it is there in the first place  .  So in the case of a print that will be viewed under known (dimly lit ) conditions surely it would be of benefit to Increase the lightness of the low value tonalities and add some saturation to the colours, without being too concerned with taking too much notice of the RGB values other than being aware how our printers respond.

Is it possible that what you are observing could relate to the printing of images and the reliance on particular RGB values for the lowest tonalities when your printer may not actually be able to produce values below a certain level for the particular paper you are using?
 
An image probably gives a better idea of where I am coming from and the attachment shows a test image I occasionally use when trying to evaluate papers.  After printing and viewing in good lighting (daylight) I can discern tonality differences down to around the 12 point (=RGB values) and discern separation at 254 before paper white.  Moving the image into another area with only marginal (subjective) change of lighting I cannot see difference in tonality between max black and a value of 22 (of course no problem with the highlights).  So in this case a choice of either better print illumination or raising the black levels to compensate?

[Tim Lookingbill]

I can understand how my approach to describing this topic is getting difficult to grasp mainly because I'll admit it involves my lack of organization in lining it out, but in addition its a subject that straddles between the aesthetics behind the choices we photographers make in post reconstructing what we saw at the time of capture on a transmissive display and the technical hurdles from a perception standpoint of editing in the gargantuan dynamic space of Raw and all the detail we want (and maybe shouldn't) show that even our eyes didn't see in the original scene mainly in finding and establishing black densities.

I see from the responses where I'm making my mistakes in trying to get to the point, but Simon's bullet point of "something intrinsic with digital" gets pretty close in that editing digital images especially in Raw after capture now requires an additional conscious adjustment or compensatory preemptive decision to achieve an overall rendered look of an image so that it at least registers to the viewer "aesthetically" in a print viewed in all possible lighting situations.

This observation harkens back to my commercial art school training instructing on what paints and substrates to use so commercial press technicians could reproduce our artwork to emphasize communication over exact reproduction. For example 30 sheet billboard painters (before commercial presses were used) gave us future illustrators at this school a tip about using a lot of white paint to make the image lighter than it would appear on our smaller and up close illustration board renderings. This lightening for outdoor billboard display was to assure the elements could be seen and communicated on a 3 second drive by from onlookers on the interstate. It made the image less contrasty than lightened from what I could see at the time.

For me it's getting difficult to establish overall brightness level of an image that doesn't destroy the aesthetics of what I want to communicate. I liked the modeling and texture variations of the rocks/water/leaves image above which induced me to make the scene darker than it actually was at the time of capture. I couldn't figure out how to judge my decisions in my edits image to image each shot in various lighting situations and dynamic ranges.

I forced myself to shoot a rare foggy autumn scene outside my window (shown below) to see if there was an actual absolute black in the scene such as in the deep shadows of the green bush and whether I should force it or just let it hover around 20RGB and sure enough the darker I made black the less foggy the image looked. I don't shoot a lot of scenes like that most are either dimly lit or blasted with bright sunlight. The foggy look has to fit within the aesthetics of the actual scene. But still the finished image shown below is too dark overall for viewing in my living room light, but at least it's a lot easier to lighten without destroying the foggy ambiance, a look I fought tooth and nail to maintain for hours attempting to expand the dynamics.

[Alan Klein]

Wouldn't you want to adjust your print to look best in the light you will be viewing it?  Of course there's a tradeoff against detail.  Beyond that, you really have no control if it's shown in brighter or darker light circumstance somewhere else.

[Tim Lookingbill]

Wouldn't you want to adjust your print to look best in the light you will be viewing it?  Of course there's a tradeoff against detail.  Beyond that, you really have no control if it's shown in brighter or darker light circumstance somewhere else.

The only control is rendering the overall tonality at the beginning in post in order for the aesthetic intent of the photographer to register what is being communicated according to elements that can be made reasonably recognizable in the dimmest lighting situation as in a bar or at home.

I'm assuming you understand I'm not talking about screen to print matches under controlled lighting. I'm talking more about adjusting contrast/brightness in post to a level that still maintains reasonable aesthetics so the image can be seen in dim light. And I'm also not talking about viewing in dark rooms. We all know what a nice comfortably but dimly lit living room looks like. We don't live in a gallery or museum. Of course those who can afford a track lighting setup don't have to be concerned, but that's the exception.

A visual and psychological experiment came to mind in that I bet if someone taped a $100 bill onto an 11x14in. framed inkjet print hung in a dimly lit local bar depicting a beautiful landscape of mountains spot lit by an intense pinkish orange sunset with a naked lady lounging in the overcast darkly shaded foreground barely viewable because the photographer didn't lift the shadows enough for viewing in the dimmest light, the viewer would immediately zero in and recognize Ben Franklin's portrait which is at a significantly smaller and harder to see size.

IOW this is a different side of the coin concerning "My Prints Are Too Dark" that hasn't really been discussed if it hasn't already been lost in past discussions concerned with the technicalities of insuring screen to print matches.

[Bart_van_der_Wolf]

I see from the responses where I'm making my mistakes in trying to get to the point, but Simon's bullet point of "something intrinsic with digital" gets pretty close in that editing digital images especially in Raw after capture now requires an additional conscious adjustment or compensatory preemptive decision to achieve an overall rendered look of an image so that it at least registers to the viewer "aesthetically" in a print viewed in all possible lighting situations.

Hi Tim,

If I'm interpreting your question as you may have intended it to, I'm pretty sure you would want to take a look at Topaz Labs Clarity.

Also your example of the foggy atmospheric background with less foggy foreground can be tweaked quite well with Clarity, one can increase of decrease the fog. Don't think it does what the LR or ACR versions of Clarity do, it's much more powerful, including edge aware masking for local control.

Cheers,
Bart

[ErikKaffehr]

Hi,

I got my workflow funktional by going down to 90 cd/m^2. Ideally the white on your screen should match the weight in the viewing environment to get a good soft proof, AFAIK.

What I normally do is to make a small test print and check out in light similar to viewing condition. I am fortunate enough not to have my images to hang in really dark rooms.

What I do for dark viewing is to raise exposure little add some clarity and perhaps a little more of s-curve.

But, yes, making prints to hang in bad light is difficult.

I never tried it but there is a print adjustment in the print dialog in Lightroom. Perhaps it could help? Just adjust image optimally and find a print adjustment that works for weak illumination? Just an idea…

Best regards
Erik

Erik, nice rundown on your LR workflow, but how do your prints look viewing them in normal household viewing light levels as most folks have done when we used hang paintings on our walls without spot halogens lighting them?

The main point I'm trying to make here is that the wider the dynamics of the image and the technology allows, the denser the blacks become, the darker the image is going to look to where we have to renegotiate what we want viewable in a print viewing in severely reduced dynamics settings with regard to light. Note the character of light (gradualness in diffusion) in my shot of the prints on my wall and what I had to to do to brighten the original source image. They don't match up. It's not the light level that's the problem. It's the character of the light we view prints with regard to contrast.

Mapping black to absolute densities and still retaining viewable detail in normal household viewing light can't work. In film prints and with paintings the densities of blacks and thus the scaling of shadow detail was much lighter and so we could get a general impression of the entire image viewing in dim light. With the expanded dynamics we're now capable of capturing and reproducing creates viewing problems under reduced dynamic environments. It's more complicated than saying my prints are too dark.

Alan, I don't know how to reduce what I'm getting at into a simple question. Sorry.

[Tim Lookingbill]

Quote from Erik...

I am fortunate enough not to have my images to hang in really dark rooms.

But, yes, making prints to hang in bad light is difficult.

I would think this would be of more concern for those who want to sell their prints (it can happen) to customers or friends/families who will be viewing in regular living room light which isn't dark at all. No one lives in the dark and if the lights are off, most likely they're not looking at prints on their wall, maybe watching TV or other activities one does in the dark if you get my drift.

I just want to render an image so those that live in normally lit living room environments such as myself don't have guests look across the room and squint or walk by the print and have to say..."What's that a picture of?...can't make it out...ya' got a flashlight?"

After investigating this issue and adjusting the tone in certain images to establish the lowest possible shadow levels in RGB numbers that will be viewable under the dimmest lights, I don't have this problem NOW. The certain images in question I edited to look too dark are usually of scenes either having extremely wide dynamics like sunsets, brightly lit rocks next to dark water and foggy scenes (where black point is in question), all having default ACR/LR previews that start out way too dark due to exposing for highlights.

Hi Bart,

The Topaz Labs Clarity appears to be a very powerful and interesting piece of software. I'll have to give it a look see. Going by the Before/After samples in that linked site which are too subtle to make a difference to dimly lit prints, I can already make those kind of tonal adjustments with my current ACR/LR setup though it took me quite a while to figure out especially in how and where the sliders affect the tonal scale of the entire image. But thanks for giving me a heads up option to consider.

[Tim Lookingbill]

Below is my present display of 8x10 prints lit under normal to very dim light that I can tolerate in my apartment where I can still see what the image is about. The top is lit by one 850 lumen LED bulb behind my flatscreen tv about 2-3 feet beneath the row of framed prints.

The bottom is another example of a re-edit of a dark print in order to suit a much dimmer lighting setup with the same bulb only six feet away and closer to the wall. For some reason I can see the image more clearly than expected with less light than I'ld expect from a simple edit, but I think the spread of the diffused light from the other bulb on the adjacent wall may be helping out. Or maybe my eyes are adjusting better to less glare from staring directly at the harsher light from the other wall.

[PeterAit]

I honestly think that you are way over-thinking this. The viewer's reaction to a print is subjective, and when I am processing an image I do it subjectively. This is exactly as it should be! Thus, black in an image is "right" because it looks right to you and not because it is mapped to 5,5,5 or 8,8,8 or whatever. Yes, there's a lot of technical stuff in today's photography, and it is easy to get bogged down in that while losing sight of the real goal.

[digitaldog]

I honestly think that you are way over-thinking this. The viewer's reaction to a print is subjective, and when I am processing an image I do it subjectively. This is exactly as it should be! Thus, black in an image is "right" because it looks right to you and not because it is mapped to 5,5,5 or 8,8,8 or whatever. Yes, there's a lot of technical stuff in today's photography, and it is easy to get bogged down in that while losing sight of the real goal.

+1 on that.

[Tim Lookingbill]

I honestly think that you are way over-thinking this. The viewer's reaction to a print is subjective, and when I am processing an image I do it subjectively. This is exactly as it should be! Thus, black in an image is "right" because it looks right to you and not because it is mapped to 5,5,5 or 8,8,8 or whatever. Yes, there's a lot of technical stuff in today's photography, and it is easy to get bogged down in that while losing sight of the real goal.

I'm not talking about subjectivity of a print.

I'm talking about rendering scenes of various dynamic range and contrast characteristics in post long after the memory of the original lighting characteristics has faded where fighting the adaptive effects on the eyes in attempts to reconstruct the character of light causes misjudgements in brightness that wind up making the print too dark for the dimmest lighting situation.

Here's another one I reworked today where I thought the first attempt done a month ago was bright enough because I wanted to bring out all the detail in the texture of the petals only to go back to discover it was WAY too dark. It took me quite a while to arrive at the brighter version on the right in order to maintain the modeling in the texture of the petals. I REALLY thought the one on the left one was bright enough. I'm glad I didn't waste a print. I don't have this problem on most my images, but I've found this is a real problem and thought others experienced the same and found other solutions to overcome it.

Now you're not going to notice much difference between those two edits because you are viewing them on a transmissive display but when they are printed the differences are amplified viewed under dim lighting. That's my point.

I don't need condescending responses that question my thinking. That helps no one.

[digitaldog]

Now you're not going to notice much difference between those two edits because you are viewing them on a transmissive display but when they are printed the differences are amplified viewed under dim lighting. That's my point.

Avoid dim lighting. Not good for your eyes either 

[Tim Lookingbill]

Avoid dim lighting. Not good for your eyes either 

Normal living room lighting that encompass most of the homes of regular folk (I've visited) who don't have track lighting and big bay windows is dim similar to my living room. Same for restaurants I've visited with even dimmer light but I can still see what's depicted in the paintings and prints hanging on the walls.

Thanks for the well thought out responses BTW.

[digitaldog]

Normal living room lighting that encompass most of the homes of regular folk (I've visited) who don't have track lighting and big bay windows is dim similar to my living room. Same for restaurants I've visited with even dimmer light but I can still see what's depicted in the paintings and prints hanging on the walls.

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. Simple as that. I don't know why a digital print hanging there would be any different from an analog print, painting, or similar in that respect. If someone is more concerned with dim lighting for mood than ideal lighting for the artwork (the restaurant scenario), so be it. Got no problem with that.

This isn't much different from the recent discussion about showing images on the web and adjusting them because some people have displays that are too bright. The viewing conditions are all over the planet and you simply can't produce anything that is ideal for all conditions so adjust as best as possible for something in the middle (not a 6 watt nigh light blub, not a 1000 watt lamp 5 feet from the image).

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!