Almost everyone working in digital photography has a tale to tell about making the transition from the silver world. For our part, we had invested many years at Kodak designing and building new black-and-white materials—films, papers and chemicals—all firmly entrenched in the traditional world of silver imaging. But after retiring in the late 90s, our curiosity about the digital equivalent of the silver world grew. Eventually, we purchased some modestly priced gear and began what we thought was probably an absurdly trivial exercise: Pick a favorite black-and-white silver print, scan it and make an inkjet print that looks just like it. A piece of cake, no doubt. As an entry point to digital black-and-white, simply duping a silver print as inkjet seemed a can’t-miss first experiment.
Ha! Not even close! After scanning several prints, we found mismatches between the appearance of the originals and their associated images on the monitor that ranged from moderate to severe. Further, nothing that came out of the printer looked much like either the original silver image or what we’d seen on the monitor.
This prompted a lot of reading, web surfing and attending seminars by digital gurus. It sounded like our problems might be solved if we were willing to spend enough on top shelf equipment, screen-calibrating hardware, custom printer profiles, and other outside services. However, we weren’t out to see how much we could spend, and had no interest in black-box solutions. We wanted to understand what we were doing and why (or why not) it worked.
That understanding led to a pair of key transforms that were easy to design and implement. (This is explained in detail in Parts II and III of this series.) The first transform adjusts how the monitor displays what the scanner records. The second adjusts how the printer converts the monitor image into a print.
The first transform forces each density in an original print to display as a unique brightness value (between 0 and 255) on the monitor after it is scanned. For example, this transform ensures densities of 1.00 in an original always appear as a brightness value of 98 on the monitor. All density-brightness value pairings are selected so the tones on the screen look like—are visually interchangeable with—the same tones in the print. Our eyes then accept the appearance of a print and its screen image as visually equivalent.
The second transform is the reverse of the first. Continuing with the example, if you hit the Print button, all screenbrightness values of 98 are forced to print to density 1.00. The end result is twofold. We can always achieve a screen image that looks just like the original on the first scan. And, we can always produce an inkjet print that looks just like the screen image on the first attempt. If there was no intermediate manipulation of the screen image—just scan and print—densities in the original and its dupe will match exactly. Further, this second transform is easily customized to work with any combination of printer, ink and paper.
This ability to controllably scan and print from print originals (rather than negatives) has value in a number of circumstances (beyond that where we have a print but not a negative). Often, we simply wish to make duplicates of carefully prepared silver print originals without the fuss of going back to the conventional darkroom and dodging, burning, split-filtering, hot spotting, selectively bleaching, etc., each and every dupe.
Or, we may want to further improve a print original in Photoshop, especially when working with old prints in serious need of restoration. Here, there is value in exactly duping the original into Photoshop, then parking this dupe to the side of the screen as a reference while exploring digital improvements.
There is another benefit for those entering the digital black-and-white world. It took most of us a long time in a conventional darkroom to “calibrate” our eyes to silver negatives or their images projected on an easel in order to pre-visualize prints. If your starting point is the monitor’s display of a scanned negative, the mental calibration is different. But, if you begin with a print, you have preconceptions, or expectations, about how things “should” look on the screen and as the final output. Strictly in terms of the learning process, prints as originals make a far easier entry point to digital.
In this series of articles, we’ll describe how you can do everything we have done with your own equipment. You’ll learn to manage your scanner setup so what you see on the screen looks like the print you scanned. You’ll also learn to manage your printer setup so what you print looks like what you see on the screen, which, if you choose, will match exactly the print you scanned in the first place.
Some caveats: We’re focusing on successful scanning of print originals, not negatives (we’ll move on to black-and- white negative scanning, but that will be far easier with the tools for managing copy prints already in place). We are talking strictly black-and-white, not color. We are working with single black inks, not quad-inks or duotones. Our concern is only with managing tonal relationships, not sharpness nor any other aspect of image quality. We aren’t using very sophisticated (or pricey) equipment. Our scanners are Epson Perfection 1640SU Photo models. We have three printers—a pair of really old Epson Stylus Color 600 units, and a newer Epson Stylus Photo 820. Our procedures require the use of a densitometer, but we will explain how you can make your flatbed scanner function as an acceptable densitometer surrogate.
 |
 |
Some tips on monitor calibration
Before starting a printing session in the wet darkroom, you make sure you have a paper and chemicals that can yield prints with clean whites, strong blacks, and a smooth continuous range of grays. The equivalent in the digital lab is making sure your monitor exhibits all the same characteristics, i.e., it is properly calibrated. All calibration required in our simple black-and-white world was running the well-known Adobe Gamma routine, but with some important embellishments. We used Adobe Gamma to enter values, and to create and save the ICC profile for the monitor.
The entered values derived from a web resource at www.aimdtp.net/aim/photoshop/index.htm. From the index on the left of this site’s opening page, select “Monitor Calibration with Adobe Gamma.” Read this and all links it contains.
This web site explains how to set the monitor brightness (black point) relative to a portion of the screen that is not being scanned (it is “turned off” and as dark as it can get). This lets you select the darkest possible black point setting, confident you are not sacrificing the visibility of any near-black tones. A short downloadable program interrogates your system and returns your monitor manufacturer’s specifications for RGB phosphor x and y chromaticity values (often not included in the monitor’s technical literature), which you can enter in Adobe Gamma’s “custom” phosphor field. The site also provides larger and more complete dither patterns that you can place behind Adobe Gamma’s gamma slider screen for more precise, easy adjustment of gamma and neutral balance. Finally, it offers a more rigorous method for adjusting your monitor’s white point color temperature, provided you use a daylight balance.
Having done this, our monitors provide white whites, black blacks, near whites and near blacks that are not blocked, and grays devoid of any color cast. Most important, however, when our monitor receives properly scanned images, it is able to present them as very close approximations of the originals. We urge you to perform this type of calibration, then write down (and put in a safe place) all the entered values and the name you gave the resulting profile before proceeding with the techniques below and in the following articles.
 |
 |
The scanner as a densitometer
Figure 1 shows the Epson Twain window controlling our Epson 1640SU Photo Scanner. Selecting the indicated button opens the Image Controls panel (Figure 2). Note the eyedroppers on this screen—one each for highlights and shadows. These can be used to sample an image in the scanner’s image preview window. Herein lies the not-so-obvious densitometer function of the scanner. While you may not own the same scanner, the highlight and shadow eyedropper functions are common to many current scanners.
To use the eyedroppers, we made a step tablet on traditional silver paper— essentially, a “test strip” with increments fine enough that the strip exhibited a dozen or more tones between white and black. We used more than a dozen different papers to be sure the choice of paper (fiber or RC, silver or chromogenic, surface, tint, etc.) did not influence subsequent measurements. (It didn’t.) Each step on each paper was read with a densitometer. Then, each paper was individually preview-scanned, the Image Controls panel (Figure 2) opened, and “eyedropper values” measured on the same steps that density had been previously measured.
To make a measurement, select one of the eyedropper tools (click on it), position it in an area of interest in the preview scan, then click again. This will generate a number in the box next to the eyedropper in the Image Controls panel. The shadow dropper reports numbers from 0 to 60 (it can only “see” densities above 1.1), while the highlight dropper reports values from 61 to 490 (and works with all densities below about 1.1). Reading eyedropper values this way and plotting them against previously read densities furnishes the smooth relation shown (Figure 3). However, if the logarithms of the eyedropper values (termed “log ED”) rather than the raw numbers are plotted against the densities, the relation is almost linear, as shown in Figure 4. Specifically, log ED and density are connected by the equation:
log ED = 0.022D2 – .60D + 2.43
This equation probably will be different for your scanner; it is slightly different for our two 1640SU units. You can figure out the equation for your equipment if you borrow a densitometer and read your homemade strips for density and Log ED. Or, you can buy a calibrated step tablet, like Kodak’s Q-60 target, with densities already listed for each gray patch, scan it and derive your own equation. However, there is no need to know what density leads to a particular Log ED value. It’s sufficient to do a preview-scan of a print, measure an eyedropper value, look up the log of that number, and know you have a figure that can be used to represent a true density value. This technique works for everything we will describe in the following articles. Besides, it’s a lot cheaper than buying a new densitometer. It’s important to do this procedure only with the scanner’s eyedroppers, and not the tools in Photoshop or other post-scan software. The scanner’s eyedroppers allow you to make measurements on the raw image, exactly as the scanner receives it. By the time you get to the post-scan software, decisions on how to scan —Gamma, highlight and shadow setting, etc.—have been made that permanently alter the original image.
There are a couple of things to be aware of when using the scanner’s eyedropper sampling tools instead of a densitometer. Note in the equation above, the value of the multiplier is “-0.60”. The fact it is only 0.6 and not 1.00 means the eyedroppers are not as sensitive as a densitometer, but the difference is small. The fact it is negative indicates the density and Log ED scales move in opposite directions: as the sampled patches become darker, density values become numerically larger while Log ED values become numerically smaller. This doesn’t present any problems; it’s simply a quirk in how the scanner’s eyedropper scales were created.
As an aside, the Epson 1640SU Photo Scanner has a transparency unit that allows it to scan film as well as reflection material. The eyedroppers work the same way when previewing film scans. The equation relating log ED for film scans to film transmission density is slightly different, but the principle is the same. Log ED values from a film scan can be used for the same purposes as film density values. For example, you could build your own version of the Zone System based on Log ED values.
 |
A note of caution
The following articles will make frequent use of a densitometer or the scanner eyedropper tools to characterize prints. When making such measurements on an inkjet print, it is imperative to make them over a short period of time, preferably in one sitting. Inkjet prints can change over time, and if you measure part of a print one day and complete the measurements at a later date, there is a real possibility the print will have changed in the interim. We made an inkjet print with “non-archival” materials and left it on a shelf exposed to modest light and ambient atmospheric conditions for several weeks. Periodically, we measured the density of a specific portion of the print. The results (Figure 5) illustrate density changes (loss) with time can be rapid and substantial. You don’t want this kind of gremlin distorting your careful measurements.
Next steps
To scan a black-and-white print and achieve a visual match between the original and its image on the monitor requires two things.
• Establish the proper viewing conditions for both the print and the monitor.
•Force print densities(orLogEDvalues) to translate to Screen Brightness Values according to a transform unique to your equipment and working conditions.
Author’s note:
Part II of this series explains how to do this. The measuring and testing we encourage will take some time, but don’t lose sight of where you are going. In the end you will be able to scan any black-and-white print once to obtain a screen image in Photoshop that is tone-for-tone faithful to the print. In Part III, we’ll show you how you can print this black-and-white screen image with the printer, ink and inkjet paper of your choice, and get a tone-for-tone match to the screen image. The time invested in testing and calibrating will be repaid many times over in a quicker, less expensive, and far more predictable black-and-white digital workflow.
Dick Dickerson and Silvia Zawadzki are retired Kodak black-and-white product builders. Their PT articles include “The Genesis of Xtol” (Sept/Oct ’99), and “Testing Kodak’s (Remanufactured) Black-and-White Films” (March/April ’03). Through their company, Dick & Silvia’s B/W Toolbox, Inc., they offer consulting and teaching services in black-and-white photography.
