LensWork Technology Blog

Shorpy Tones 2

First, some objective data. Here are four representative images and their corresponding histograms from Photoshop.

Very interesting. Note that there are true, dead blacks in every image, just not vary many. Same with whites. Medians at 130, 142, 143, 111.

Compared to an image from one of my recent blog posts.

Median = 90. Is this just the way I am processing them, or something in the digital capture? In particular, note that the histogram in Shorpy #4 and my image are similarly shaped, through the lowest 1/3 of the tonal scale. In the remaining 2/3, however, look at all those light tones in the Shorpy image! Wow.

More research and some experimenting needed . . .

Epson Signature Worthy Sample Pack

I just received the Epson sample pack and ran a few test prints yesterday. I'm glad I did. With only one exception (the Premium Luster is a dog, IMHO) I found all of these to be outstanding papers — at least as far as the overall look, density, paper color, texture.

   

This sample pack contains 2 Sheets of the following:

• Hot Press Bright
• Hot Press Natural
• Exhibition Fiber
• Cold Press Bright
• Cold Press Natural
• Velvet Fine Art
• Premium Luster

Why is this important? It wasn't that long ago when we were all looking for one good paper, especially a glossy alternative that didn't bronze or feel like crappy RC paper from our darkroom days. When I last tested papers extensively, I found Harman FB AL Glossy the only paper I was comfortable printing on for that look, and Hahnemühle Photo Rag the only paper I was comfortable using for a matte look. Two papers, that was it. That is scary. Sure, we all have a "favorite," but when your favorite is selected because it is the one and only paper that meets your standards, it's a position of some tenuous vulnerability should the manufacturer stop producing it — or even if they simply cannot meet demand and it's out of stock.

If either of my first choice papers were to drop off the planet, I would not hesitate to use one of these Epson papers. In fact, I might just start using them for specific projects — particularly the Velvet Fine Art or the Cold Press versions. Lovely surface textures. Beautiful colors. It's also seductive that they are just a bit less expensive than HPR.

In particular, the Exhibition Fiber was a very pleasant surprise. I still prefer the slightly smoother surface texture of the Harman FB AL. This latter, BTW, is now being marketed by arrangement with Hahnemühle under a new name, Harman by Hahnemühle Gloss Baryta, but as far as I can tell, it's exactly the same paper we've used for a couple of years now. The Epson Exhibition Fiber is the only paper I've seen that measures a black density that rivals the Harman. It does not bronze with my 4880 inks, and the paper has a nice, heavyweight feel. I like it a lot.

So, if you are like me and don't mind a little testing, this Epson Signature Worthy Sample Pack is a great way to see if any of these papers meet your expectations.

Shorpy Tones

I love the daily entertainment of the Shorpy website. Do you?

One thing I've noticed that I am fascinated with are the relative tonal quality of those old images.

Notice the overall lightness of tone. Sure, there are good, deep blacks, but something in those old films or in his reprocessing of the old images gives them a lightness of being that is very different than what we see in today's aesthetic, or perhaps the way our equipment responds to light.

Indoors or outdoors, I see the same thing. Look at the shadows of the following . . .

Is it the very slow film? Is it the extremely gentle shoulder that keeps the highlights from blasting out? What is it about these images that seem so airy and illuminated? And — I guess more to the point — can we achieve this look with today's materials? I love this look, so I'm off to do some experiments.

Textured paper and inkjet printing

I've had a number of people ask me about my use of textured backgrounds in my inkjet prints. The natural question is am I printing on textured paper? As much as that would be a seductive idea, I can't imagine that an inkjet printer could succeed very well if the paper was extremely textured, primarily because of the simple physical problems of both coaching the paper and spray of those tiny inkjet droplet.

   

   

Instead, I simply create an image that is the textured background. In my recent work, the Living Landscape exhibition for China, I used some textures that I made some 10 years ago. I started by scanning Japanese paper cut designs as high-resolution black-and-white line drawings. In Photoshop, I then converted them to grayscale images and used a level control adjustment to turn them into medium-gray-and-white images. Once that was complete, I was ready to create the texture — the part of the process is where things get really interesting.

   

   

In the spirit of true confession, this is the one and only practical use I've ever found for the software I originally purchased as Fractal Painter. (I think it's now owned by Corel and is simply known as Painter.) It's a fascinating and seductive piece of software. I just never found a use for it in my photography other than its ability to create these wonderful textures. When I purchased it years ago, it included dozens of paper textures that can be applied to any image. Most importantly, the paper texture can be controlled as to size, depth of texture, strength of lighting, direction of lighting, and a half-dozen other characteristics that allow you to create, essentially, any texturized paper/image you can imagine.

After considerable experimenting, I found the right combination of paper texture and depth that I thought would make wonderful backdrops for photographs. At the time, I never did find a photographic project that I could use this technique with, so they've languished on my hard drive now for some 10 years. From time to time I resurrect them to see if they'll work with some project I'm currently working on. This new exhibition that combines my fan shaped images with classical Chinese poetry was the perfect project to employ these old paper texture images. I was able to resurrect half a dozen of these and finally find a use for them that fit the project perfectly.

   

   

Each piece in the new exhibition is constructed of half a dozen components. First, the paper texture image creates the background. On top of that, I placed a simple box with a fill color. By changing the fill color and its opacity, I was able to create just the right density and hue for the backdrop that complemented the image without overwhelming it.

Next I imported the photographic image and modified its container shape so that the photograph becomes this fan-shaped presentation. (This is the same idea I used in the Uchiwa-e series, although I modified the shape for this project.) To this fan-shaped InDesign object I apply a subtle drop shadow.

Next I bring in the Chinese poem into its own text box and apply a paragraph style to determine its typeface characteristics. This text box also has an object style that employees rounded corners and a white fill that is throttled back to about 40% opacity. This gives the text that lighter backdrop that helps its readability and makes it a distinct item on the page. By using an opacity setting, the paper texture is still visible behind the text box and makes it almost appear that the text is on a tissue sheet. Finally, the text and graphic on the bottom of the page round out the design. These latter elements are actually included on an InDesign master page so their positioning is uniform throughout the project.

This kind of project is a lot more intense in terms of its design elements than the standard photographic project of my youth. We used to simply slap the image in a beveled window-cut, white mat board. The kind of design work I've included in this project would have been unheard of back them, but I find it a much more interesting creative pursuit to add these design elements where they are appropriate for the project. It expands my role beyond "photographer" and deep into the realm of designer. For me, this is exciting, particularly because I know that I have a safety net — I can always employ a professional graphic designer in any project in which I feel I'm in over my head.

Measuring Density, Part 2

Measuring typically is some sort of comparison to a known standard: here is a standard ruler, there is a widget — let's compare them so we can find out how long the widget is. Measuring density is not quite like this and perhaps that's why it's a little less straightforward about how to do it.

It's still a comparison, but in this case a comparison between how much light falls on an object compared to how much is reflected back from it. We first measure the amount of light that falls on the subject (not an easy thing to do) and then compare that to how much we see reflected back (not an easy thing to do either). In essence, we can think of it as a percentage — although that's not how it's actually stated by a typical densitometer.

Rather than percentages, density is expressed in logarithmic terms. If 100 units of light fall onto an object and it reflects 1% of it back, we would say that it's density is 2. In practice, the scale of reflectance density runs from 0.1 to about 3 or less. The blackest black I've ever measured is 2.45 which is roughly 0.5% reflectance.

But, here is where we find our first complication. The stated "optical reflectance density" of an object will vary depending on what wavelength of light you measure. Some densitometers measure only a vary narrow range of color; others measure the full spectrum. So, you and I might measure the same patch of black and report different readings for its density if we are using different machines. This is why density measurements reported on the web, for instance, can be so misleading. Unless you know how the measurement was made with which kind of densitometer, you really don't know anything when someone reports a density measurement.

The second complication is one of understanding, particularly for us photographers. If I tell you a certain gray tone has a reflectance density of 0.8 on the printed page, what shade of gray is that in your Photoshop file? In Photoshop, we measure the "density" with the eye dropper tool which reports to us either an RGB value or (if we set the tool to report K values) a percentage number. Comparing these to the numbers in a densitometer reading is like trying to measure a musical tone with a tape measure.

Before we resolve this, we need to know precisely what Photoshop is telling us. Actually, the K number is nothing more than the RGB number translated to a percent of black. (To change the way Photoshop is reporting the eye dropper tool, use the tiny triangle fly-out menu just below the eye dropper in the Info panel. Hold down on the fly-out and you'll see lots of options.) In the Photoshop world (8-bit, anyway) the scale of black to white runs from 0 to 255 (which is 28 = 256 shades of tone). When Photoshop reports a shade of gray is 50%, it is reporting that the luminance value of this tone is half-way between 0 and 255, a luminance of 127. You can find any luminance value in terms of the 256-step scale by simply multiplying the converse percentage by 256. So, a 14% patch (one that is 14% of the way to full black) is a luminance value of 220. (1.00-.14*256 = 220.16). In the screen captures above, note that none of the three color channels is actually 220; it is the luminance that is 220, a sort of average mix of the three colors. Close enough. But neither RGB 220, nor 14% are numbers that we use in the densitometer.

We need a translator.

Enter a couple of pairs of scientists. First, a guy named Murray and a guy named Davies got together and developed the Murray-Davies curve — a translation for converting density measurements into percentage of reflectance. This mathematical formula was later modified by a couple of other guys named Yule and Nielsen into the "Yule-Nielsen optical dot gain curve." (You can't make this stuff up. Just in case you want to torture yourself with math, here is more.) There are densitometers that have the Yule-Nielsen curve built into them so you can actually measure density and the machine reports back a Photoshop-type percentage. Now we're talkin'.

So, we have two ways of expressing density — the density number itself, and the Yule-Nielsen percentage. We use density to state how black is black — useful for comparing, say, the maximum black of a gelatin silver print to the maximum black of the ink in a book. We use the Yule-Nielsen percentage to compare, say, the tonality of a middle gray tone on the printed paper to the same area of tone in the Photoshop file. This way we can know if our printing device is accurately reproducing what's in the Photoshop file or whether it is warping the tones in the printing process.

There is lots more theory to go into, but not really necessary in the world of the pragmatic photographer — unless you want to. So, class dismissed for now.

In the next post on measuring density, I'll share some practical experiences I've had with measuring density and some thoughts on the tools to do so.

Measuring Density, Part 1

It's been a flurry around LensWork-land the last few days — deadlining LensWork #88 last Friday, launching our new website day before yesterday, and yesterday the press check up at Hemlock. And in the midst of all this we are starting the final video tests for the new Visual Workshops. I feel a bit guilty about not adding a new post since the 17th, but whew! I'm looking forward to taking this weekend off and starting the printing of a new photography project of my personal work.

Speaking of the press check, I was razzed yesterday at Hemlock for being the one and only customer who ever brings his own densitometer. This may be true, but I must admit, I'm still surprised how any serious photographer can neglect this most important tool.

As photographers, we live and breathe tonalities. We may strive for an emotional and fuzzy concept called "life" in our photographs, but we do so with technological means that are numeric and precise. We measure light with our meters, exposures with our histograms, Photoshop eyedropper readings in our digital files, Zone calculations for our analog negatives and prints — you name it — we are measuring, number-crunching, precision-driven artists. Why is it then that we don't regularly measure how black is black?

I suspect because the numeric measurement of black (or any other shade of gray) is conceptually more complex than, say, reading a ruler. Also, it requires another, sometimes expensive tool. Besides, I can hear the counter refrain as I type this, "Black is black and nothing more need be known."

Hogwash.

True, one can argue that ultimately the maximum density of a print will not be the ruling criteria of whether or not it is a good print. Some use this kind of argument to sidestep the issue of measuring tonalities. But, the same could be said of f/stops or shutter speeds; in and of themselves, they don't determine a good or bad photograph, but few photographers would therefore decide to ignore them.

Measuring density is one of the foundations of understanding the thousand and one steps we take to arrive at a finished photographic image — whether that is a traditional wet-darkroom print, a digital print, or a commercially printed book. Along with all the other aspects of technology we try to master in our pursuit of the fine photographer, I would propose that all of us photographers need pay a lot more attention to the numbers involved in measuring tones — density in particular.

In my next post, I'll cover a couple preliminaries about measuring density. Finally, in a third installment, I'll share my methods and tools.

Preserving a Body of Work

Robert Goulet, 1965, queue the music — "To dream the impossible dream . . ."

Some things are simply not possible, or at least so close to impossible as to be indistinguishable from impossible. Preserving an entire body of work would be one of these.

I'm back from Houston FotoFest where lots of interesting discussions were had. I attended a lecture by Anne Tucker (Museum of Fine Arts, Houston) and Ed Robinson (Los Angeles County Museum of Art), two of the most important curators in the country. One of the things that caught my attention during their talk was that major museums are not likely to want more than one or two pieces of work from a project. They simply don't have the room, resources, or funds to collect whole bodies of work. That got me to thinking: then who does?

I find there are few photographers these days not working with projects. Virtually every review I've done in the last dozen years has been a project. The days of the single "greatest hits" approach to photography is simply not being done by anyone that I've seen — not in ages. So, if everyone is working in projects of 10-200 images, how are these being preserved as a whole for future generations? How will we be able to see the project in its entirety?

This is not easy. Consider the alternatives …

Books are the kneejerk reaction, but books are usually produced in relatively small numbers; often considered "reproductions" of the artwork, not the artwork itself; very expensive to produce; problematic to market; often much smaller in scale than the original prints; require layout decisions that are often foreign to the original artwork (e.g., facing pages, gutter jumps); ink methods of reproduction often cannot match the full gamut of the original artwork. Books excel at preserving large numbers of images and their sequencing, but that is their primary virtue for preserving whole bodies of work.

PDFs and all the other digital media come to mind, but there is the looming question of long-term access both to the PDF file format and to the digital medium it is stored on. PDFs are inexpensive, preserve sequencing, can include other media components like audio and video, and are easy to reproduce. Great. They are also not the original medium, rely on an ever changing technology to be seen, present innumerable problems about reproducing color accurately, and require compatible hardware to use them. Short term viability is assured, but beyond a dozen years or so gets very dicey.

Collectors (those beneficent, well-heeled mystery class who are curiously difficult to locate in the flesh) might be the answer. But how many of them are there, how do you convince them that your project is worth preserving in its entirety? How will they preserve work for a larger audience? Do they collect whole projects, or only the investment-worthy 5-star images from a project? Hmmm …

There is your closet and the stack of Light Impressions boxes you keep there. Your project can live with complete integrity there, in the dark, all alone, with all your other projects as its only company. [Soft violins in the background.] Sort of like a benevolent jail. And then your heirs can figure out what to do with all that work after you take up residence in that great darkroom in the sky. [Sniffles and hankies. For the poor suffering artwork, that is.]

I can't help but think that the only answer is to not play the game. The entire concept of preservation is based on the assumption that someone in the future will care about our work. Perhaps. More likely not. Instead, why not put all our energy into finding an audience for our work today? If we can't interest folks in our work today, what makes us think folks a hundred years from now will care about it? From whence do we imagine all this fascination for our work will bubble up if we can even approach a gentle simmer of interest in the here and now?

Seems to me the best strategy for preserving our work is carpe diem. If we are lucky and deserving, perhaps our work will live in the imaginations of those we touch right now. As to future, in the memorable words of that noble philosopher Alfred E. Newman, What? Me Worry?

Image Sharpening, Served ala carte

For my repast this morning, I shall breakfast on an appropriate viand. One large plate of crow, please.

In podcast #276 A Book I Won't Be Purchasing, I poked fun at the book Image Sharpening by Bruce Fraser and Jeff Schewe. With disdain, I proposed that it just couldn't be that complex of a topic to require and entire book to explain. I mean, really, 345 pages on focusing?

Let me here and now formally apologize to the late Bruce Fraser and to Jeff Schewe. I was wrong. With a little prodding from Mike Johnston, I purchased the book. Curiosity, if nothing else.

I've just spent the weekend testing their theories of multiple pass sharpening — a workflow that involves what they refer to as capture sharpening, creative sharpening, and output sharpening. By golly, it works. Really well. It is not a replacement for a solid tripod, a fast shutter speed, and critical lens focusing, but their ideas do make the most of whatever you've captured via your digital camera exposure or a scan of your negatives. I now have about 200 sheets of Hahnemüle HPR that prove it.

Using the workflow in the book, I was able to improve the visual sharpness of my latest work from Japan, the Uchiwa-e prints project. Dramatically? No. But, but enough to convince myself that it is a worthwhile approach. Making art is, at least for me, the pursuit of excellence. Because this workflow is better than the single pass sharpening I've been using, it opens the door to a new level of excellence I am more than willing to walk through — even if that means leaving my pride and ego at the threshold. New tricks and old dogs do, occasionally, go together.

Thanks, Bruce and Jeff. Sorry for doubting you.

Monitor versus Print, Additional thoughts

It occurs to me that using the "two-state" strategy I discussed in the previous post there are some fascinating implications.

• If I make visual judgments that determine the master file based on my monitor view, wouldn't that version of my artwork be considered "the original"?
• If this is true, what happens to my "original" if I change monitors, say from CRT to LCD, etc.?
• The print is always different — and rarely better. There is always a greater dynamic range in the monitor view, often more shadow detail, and definitely richer colors, always. So, isn't the print a compromise? I'm very uncomfortable with this because I was trained that the print is highest quality presentation of my photographic artwork — higher than, say, the reproductions in a poster or a book. But, if the print lacks the full color and dynamic range of the monitor view, it is unquestionably a compromise. This jars me, and for some time I've been struggling with the implications.
• Yes, but, the print might have more detail than the on-screen image, I say to myself. But in a PDF, the encoding of the image can allow for high resolution images that reveal more and more detail as we zoom in to look more closely — something that we can only do with severe limitations when viewing a print. So does a print really have more detail, or is it an illusion based on the assumption of only a single viewing distance/magnification?
• I console myself that most people who see my prints won't see the (better) monitor view that I do when I see my images on-screen. What they don't know they are missing won't be missed, right? But what about web images, PDFs, etc.? In fact, they may indeed see screen versions of my images if they download my PDFs or shop my online store. What are the implications of trying to shop for and purchase photography online?
• Technology is constantly changing. Will there be even better monitors in the future that will substantially change the way our images appear? Are we now producing digital artwork for a medium we cannot now see or predict? Think posters of Monet paintings or websites (and T-shirts) with images of the Mona Lisa and you'll understand my point.
• "Monitors" are not a single, unified thing. CRT, LCD, iPhone screens, e-Readers, netbooks, HDTVs, yowza. And on the horizon are new view-screen technologies that will undoubtedly supplant these old familiar versions. As an image-producer, for which technology should I tweak my image tonalities?
• As I produce various on-screen image presentation, e.g., PDFs, how do I deal with the fact that some of the viewers of my work will be doing so on uncalibrated monitors? I suppose this is no different than people viewing my prints under various lighting scenarios, but for some reason it's more bothersome to me. I can't control it. And I want to. Letting go of this need to control the viewing situation is mandatory in the digital world, but that doesn't mean I have to like it.

  In the end, I get frustrated when all these thoughts run around in my cranial squirrel cage, so I just give up, do the best I can, return to producing artwork — and having fun. What else can we do?

     

     

Monitor versus Print

I tend to prefer darker imagery with color richness and density, or dark B&W. When I process the pictures for use online they look great, but when I print those same pictures via my Epson printer they look too dark and to an extent the richness is lost. I attribute this to a difference in backlit on screen viewing versus reflective print viewing, among other things. In your workflow, do you process two versions of some pictures, one for online and one for print? I'm beginning to work through how to get in print what I see on screen, and I'll need to keep track of each pair. 

— Andrew

First, are you calibrating your monitor with hardware calibration (e.g., Colorvision Spyder3) or just by guess and by golly? Without a hardware calibrated monitor, you will be fighting a no-win battle. Hardware calibration is the first step and simply cannot be avoided for critical photographic work. Period.

One of the holy grails of digital photography is the so-called "soft proof." We would ideally like to make aesthetic decisions visually based on what we see in the monitor and then have that exactly translated to the print. Boy, would we!

Can't happen. The two media (monitor and print) are simply too different. Electronic versus analog, backlit versus reflectance, contrast ranges, definition of "white" — it goes on and on. You could spend $10,000 on a Barco reference monitor like the most critical prepress and film production houses do, but even then there are so many variables it simply isn't going to match.

Instead, most photographers employ a simpler strategy: Get the monitor and print to come as close to each other as possible, but ultimately judge each medium on its own merits. So, yes, when it is critical to do so, I create a version of an image for on-screen presentations and a different one for print. I always start with a master that is the monitor version.

Specifically, my strategy is to begin by making the best looking, visually judged image on-screen. I consider that my "master." I know my printed photograph will not look like this, but it should come fairly close as a starting point. I then tweak subsequent prints by adding adjustment layers in Photoshop (or additional work in Lightroom) that are intended to compensate for the print's differences and maximize the printed tonalities within that medium.

I work from the monitor master as the starting point because I know the necessary print adjustments will vary depending on the paper choices and printer I use. The adjustment layers, therefore, are specific for a given printing routine. If I change the routine, I can simply turn off the layers for a previous printing routine and apply new layers for the current printing routine. The master image stays the same — sort of a control state, if you will.

In truth, however, I rarely do all this in practice. Most images don't require such fussing. Those with critical deep shadow detail are the most likely to need this level of tweaking, so I can see where your images might need this approach more than mine do.