Mastering Digital Color CHECK OUT...
Digital printer manufacturers and imaging software companies make it look as easy as pie to get excellent prints from your digital images on your very first try. But their ads are misleading, as those of you who have bought a digital printer can attest. At best, it will take several prints to get it right. At worst, it can take several hours of experimentation, curve adjustments, fiddling with the monitor color balance, and offering sacrifices to the gods of digital before a really decent print comes out of the machine. What's going on? And why do colors on the monitor rarely match the colors on the print? There's really no simple answer, since the complexities of working with color in the digital realm are much greater than they are in the traditional darkroom. (It only seems easier because you can make fast adjustments and changes on your monitor using slider controls and dials.) Unless you know how digital color works, it will be difficult to set up your computer, scanner, digital camera, imaging programs, and printer software so that the image you see on your monitor matches the print you make. Here's why: As a rule, every input device "sees" color differently, and every output device has its own color "personality" (a term we used to apply to color slide film). Film scanners and most digital cameras use a light-sensitive CCD to capture the blue, green, and red components of a scene or scanned object (a scanner has to supply a light source that is either transmitted through a slide or reflected off a print into the sensor). The various levels of red, green, and blue are recorded as analog signals that have to be converted to digital data using an analog/digital (A/D) converter. This digital data is then analyzed, transformed, and usually compressed into a digital file that can be read by a computer. Monitor hardware then reconverts the digital data into analog signals that are used to drive its red, green, and blue guns in order to make a color image appear on the screen. Most monitors can display up to 256 shades of each R,G,B (Red, Green, Blue) color on their screens (although pure cyan and pure yellow can't be displayed accurately). A white (or neutral gray) color on a monitor is made up of equal parts of R,G, and B, with bright white consisting of the brightest intensities of each color (R=255, G=255, B=255). The monitor's darkest black would have the lowest illumination (R=0, G=0, B=0). By mixing the various components provided by the RGB data in every pixel, a color monitor can display up to 16.7 million colors. You need 8 bits of data to store 256 shades of gray, and there are three color channels, so 8 bits x 3 RGB channels = 24 bits or 16.7 million colors. Up until this point, the concepts are fairly easy to understand, but there are some other issues that make it difficult to maintain color accuracy in the digital darkroom. On the input side, the color sensitivity of CCDs vary, and there are even some visible colors that CCD sensors ignore. A/D converters in scanners and cameras, and even the image processing software in the computer, can translate color information differently. Monitors also differ in their ability to display colors correctly, with a wide range of display contrasts and brightness variations. And while most monitors can display 24-bit color, most decent scanners capture at least 36-bit color (which equals billions of color possibilities, many of which can't be displayed on a monitor).
These variations pale in comparison to those introduced in the printing process. Before printing, R,G,B data in a digital image file must be converted into the Cyan, Magenta, Yellow, and Black (CMYK) ink data used by all 4-color desktop printers and most commercial printing presses. CMYK dyes and inks absorb their complementary colors (cyan absorbs red, magenta absorbs green, and yellow absorbs blue), allowing others colors to be reflected back from the paper. Whereas maximum amounts of RGB colors produce a bright white on a monitor (an Additive Color System), the maximum amounts of CMY inks produce a muddy black on a print (a Subtractive Color System). Black (K) ink is then added to improve black density and reduce color ink usage in areas of overlap that produce gray. It wouldn't be a big problem converting RGB capture and display data to CMYK ink data if every printer used pure CMYK inks. But most ink and dyes (especially Cyan) contain color impurities, plus the whiteness and ink absorbency of the print media also affect the way colors appear. As a result, the same RGB image can wind up looking vastly different from one printer to another, or from one paper to another. Fortunately, we can map out the ways a digital device captures or reproduces color if we have the proper tools, such as a colorimeter or spectrophotometer. These can be used to determine the color gamut of a device (also referred to as a device's color space) and to generate International Color Consortium (ICC) profiles that can later be used in a color-managed workflow (more on this later). But these measurement device's are expensive and require training to operate, so most of us will have to rely on the scanner and printer profiles that are included with the device. Generally, higher-end scanners have a wider color gamut than low-end scanners, which means they can recognize more colors, especially in highlight and shadow areas. The same generally holds true for higher-priced digital cameras, 6-color inkjet printers, and very expensive monitors. In addition, an individual monitor's color gamut can vary based on room lighting, the age of the monitor, and whether it is warmed up.
Generally, higher-end scanners have a wider color gamut than low-end scanners, which means they can recognize more colors, especially in highlight and shadow areas. The same generally holds true for higher-priced digital cameras, 6-color inkjet printers, and very expensive monitors. In addition, an individual monitor's color gamut can vary based on room lighting, the age of the monitor, and whether it is warmed up. The color gamut of even the best monitor or printer is usually smaller than the color gamut of a decent scanner, so some colors will be lost or modified when going from a scan to a print. COLOR MANAGEMENT Ideally, profiles provided by the manufacturer of your scanner, monitor, or printer should be accurate enough to get you closer to a perfect print then you can get on your own without quite a bit of trial and error. But every device has its idiosyncrasies, so these profiles may vary in accuracy. And with digital cameras, profiling is especially difficult because of the automatic white balance built into the camera, which may not balance the scene lighting correctly. You're actually better off using a profile that is made under daylight (5500k) and then setting your camera manually to daylight balance. It's when you veer from the straight and narrow that color management stops working. Fine art printers, for example, will run into difficulty when using a fine art paper or ink set made by another manufacturer. That's because few "third party" papers and inks ship with profiles (Note: the paper profiles provided by printer manufacturers assume you are using their inks). If you want to use papers and inks that aren't made by the printer manufacturer, there's no easy way to implement color management. Your best bet is to experiment with some of the profiles provided by the manufacturer in hopes of finding one that delivers decent results on your paper/ink combination. For example, you could choose a manufacturer's profile for their watercolor paper when printing on someone else's watercolor paper, and then make color adjustments as needed. If you can't find a profile that works, check with the paper or ink manufacturer to see if they have any profiles for your combination. Then check with nearby color labs and service bureaus to see what they'll charge to generate a color profile of your specific combination (the cost will be worth it in the long run). For further information on color management, check out some of these web sites: http://www.agfa.com/, http://www.apple.com/, http://www.adobe.com/, and http://www.kodak.com/.
SHOULD YOU BECOME A DIGITAL PRINTING CONVERT? Several years later, after spending thousands of hours (and dollars) in various darkrooms, I advanced to the point where I could get an excellent color enlargement in two to three tries. And once in a blue moon, I achieved a perfect color balance on my very first print of a new photo. Those of you who have made your own color prints in a darkroom know how difficult it can be. Color printing is an art that takes years to master because it requires an understanding of how different colors interact with each other, and how different paper and processing combinations affect overall image quality. According to our most recent survey, most POP PHOTO readers are interested in making their own prints, but it's pretty likely that most of you don't have the time or darkroom space needed to develop the skills for making traditional color prints. So what can you do? GO DIGITAL! Today, the cost of purchasing a high-powered computer, imaging software, and a photographic-quality 8x10-inch printer (under $1,000 total) is equal to or less than the cost of setting up a film-based darkroom. And for only a few hundred dollars more, you can get a pigment-based inkjet printer that makes up to 13x44-inch prints that will probably outlast most silver-halide prints. There are now dozens of digital printers, and hundreds of combinations of papers and inks, that can be used to make beautiful prints. But the best thing about digital printing is that you can experiment on the look and impact of your image and manipulate reality without even making a trial print or exposure series. In the traditional darkroom, every major change requires a test print to help balance the system. A color-managed digital system is much more tolerant of change. | ||||||||||||