War Stories: Color House Tales

Midnight. Six people are huddled around a sink in the women’s restroom. Except for me, all are men. In this vast printing plant—ablaze with sulphur, neon and mercury lights—one pathetic 60-watt bulb is the only incandescent light we can find.

Is my Christmas catalog cover green in ordinary room light (as intended) or silver? My sales rep peers through the gloom at a just-printed sample in my hand. “I could convince myself that’s green,” he says.

Color-correct lights aren’t always the best for viewing color. They do ensure that everyone in the industry views proofs and printed samples under similar lighting conditions.

Unfortunately, the most important viewers (our customers) don’t have color correct lights at all. In fact, consumers often don’t even have incandescents. Fluorescents are appearing more and more in consumer homes, because they conserve electricity.

In this marketing context, it can be dangerous to view your color only under color-correct lighting, especially if you’re selling food, high-end apparel or other color-critical merchandise.

The point of color-correcting our catalogs is to offer images that arouse consumer expectations in ways that the actual product will satisfy (otherwise, returns will kill you).

Color-correct lights at color houses and printers generally are too bright and too white, compared to consumer lighting. That’s why many catalogers routinely take their final proofs and press samples outside for a last look before signing off, and check them under fluorescents and incandescents, too. (F.Y.I.: Women’s bathrooms still tend to have an incandescent bulb or two, so women checking their makeup won’t look like ghosts.)

Meanwhile, back at the plant, the expensive inkjet printer slowly printed out a test grid on very expensive inkjet paper using even more expensive inkjet cartridges. Nearby, our ridiculously expensive desktop color expert smiled as we compared the finished sheet with the color blocks on his test pattern. The match was perfect.

“Now for a real product,” I say. The designer, who had developed this desktop system to test whether we could move some expensive color correction in-house, begins printing out a picture of a green hunting coat. When the printout is done, I compare it with an actual product sample. “OK, is this even the same coat?” I ask.

Outguessing the System

Desktop color correction is an unfulfilled promise of the digital age. The current desktop state-of-the-art involves calibrating your desktop monitors, scanners and printers, so their combined output supposedly will match what you see press-side. But so far, these systems fall short of what a skilled color operator and high-end proofing system can accomplish at a color house.

A big part of the problem lies in the phrase, “skilled color operator.” Color monitors are RGB devices (they produce light in the same three primary colors that your eye detects and processes), and web presses are CMYK devices (the pages they produce reflect light through transparent inks, each of which blocks one and passes two of the RGB primaries; that’s why cyan and magenta are such ugly colors). The vast difference in how color is produced on a monitor and on a press creates equally vast differences between what you see on screen and what you see on paper, no matter how much calibration goes on.

Skilled color house operators learn to simply outguess the monitors. They look at the screen, and mentally translate what they see there into what their experience tells them will appear on the web press or high-end proofer. That kind of judgment takes time and practice to develop. Someone doing color correction in a color house naturally will become better at it than the part-time color corrector in your design shop.

So until the day arrives when the process is more highly automated, I think we’re still going to find ourselves paying significant dollars to color houses. Sad but true.

Going Digital

“OK, tell me which is which.” My color house rep hands me a dozen Kodak approvals. His color house recently had bought a photo studio that specialized in digital photography, and I’d resisted switching my high-end books to digital. So to prove to me how wrong I am, he prepares this test. Half the shots are digital, half are film. I have to tell the difference. I quickly make two piles. He checks them—and I’ve missed only one.

“OK, what technical glitch did you spot?” he asks.

“Nothing,” I reply. “I knew you’d have all the technical details covered, so I just asked myself, ‘Does this shot move me?’ If yes, I put it in the film pile. If no, I put it in the digital pile.”

Astronomy Magazine has taught me a lot about digital photography. Astronomers got into digital imaging long before catalogers, and they study it far more deeply. What interests me most about their struggles with digital imaging are the huge inaccuracies in the digital chips.

The word “digital” evokes an idea of mathematically precise ones and zeroes. But actually, a digital imaging chip isn’t very digital at all—it’s mostly an electronic sensor that uses physical effects to try to read light intensity. As a physical sensor, its accuracy is dramatically influenced by a variety of forces. One of these is the background current that it generates internally (if you cover the lens of a digital camera and take a picture, the resulting shot will be a vague haze).

Other influences include sensitivity fluctuations with temperature, and sensitivity variations with light intensity and color—in fact, all those nasty physical effects that we imagined we were leaving behind with film. The dirty little secret of digital photography is that digital imaging chips and film are much alike—both are mechanical contrivances that use brute physical effects to transform light intensity and color into something else.

The fact is, film manufacturers have 80 more years of experience in doing this with chemistry than digital chip manufacturers have doing it with transistors. This is why I generally find that a good shot on film can more easily move the soul than a good digital capture—canny film manufacturers have had longer to get the details right.

If your sales depend on moving the hearts and souls of your customer (rather than simply conveying visual information about your product), for anything from high-end jewelry to a luscious slice of country ham, I suggest doing your own digital versus film test before gambling on the current state of the art in digital photography. Such a test is easy to do, and if you find your digital shots aren’t moving you, they probably aren’t moving your customers either.

Outfox the procrastinators

It’s 1 a.m., and my color house rep and I are standing in the lobby, awaiting the door-to-door courier. As we wait, we check a pocket guide of airline schedules. “Seventeen more minutes and our package misses the last flight,” he says. I call the courier service, talk a moment, hang up.

“Doesn’t matter now,” I say, “the Postal Service has taken over the baggage compartment. No more courier packages can get on that last plane out.” We both instantly begin searching our airline guides for the first morning flight out.

When each catalog year ends, I make a list of all the vendors I’ve used and rank them by how much grief they’ve created for me. The No. 1 grief spot shifts year to year between data processing firms and color houses, but it’s usually a color house that has really worn me flat.

One of the most common excuses is that final files and proofs were delivered late. But I’ve found that this makes little difference. Even if I deliver client materials early, color houses still tend to deliver at the last possible moment. My solution combines fear and lying. I begin by warning my color house at the start of the season that if they don’t deliver on time, I’ll consider that very profoundly when deciding whether to use them the following year. Then, throughout the year I simply lie to them about delivery dates.

This requires skill. At first, when I gave false delivery dates (a few days before the actual drop-dead delivery date), the color house reps would routinely call the printer, discover the real date, and still deliver at the last minute. To circumvent this, I began working with my printers to make sure the color house couldn’t discover the real date. This was difficult—printers tend to have a lot of staff. But overall, with the enthusiastic cooperation of my color house rep (he didn’t like hanging around the lobby at 1 a.m. either), I’ve become skilled at keeping everyone at the color house in the dark about when the actual drop-dead date is for delivering film or files.

Is this silly? Probably, but it’s worth knowing if you value sleeping soundly at night, rather than standing around after midnight consulting your OAG (Official Airline Guide) to determine when the last flight out of town departs.

The CTP Shuffle

10 p.m., somewhere in the Midwest: the printer just handed me a laser proof of the CTP (computer-to-plate) files that my color house delivered. I check each page, then suddenly stop. “This product isn’t in the catalog,” I say. “It’s in the file you delivered,” the printer replies.

“How soon do we go to press?” I ask. “A few hours,” he replies. I dial my color house rep’s cell phone. He was sound asleep. “You have a problem,” I say.

When CTP printing first swept the printing industry, my fear was type reflows. I expected to arrive press-side and see my designs fractured, with no time to recover.

But that hasn’t happened to me. My biggest problem has been delivering wrong versions of files to the printer. As a cataloger, you realize that you have many versions of your images, and often many versions of your catalog, too.

But at the color house level, the versioning problem has become far worse. Your color house generates new versions of individual images each time you correct an image. The number of versions increases with each revision from your designer, each in-house type correction, and each output version for the ever-growing variety of proofing and CTP systems supported by most color houses.

In the old days of film, everything ended up on sheets of separated film from which a Matchprint or similar proof could be directly printed. You absolutely knew that what you saw on the proof was what you were delivering to the printer on the separated film.

Not so today. You still see a final proof, of course, which you deliver to your printer. But the problem is that one final proof has almost certainly not been produced directly from the final CTP files you’re shipping to your printer.

Your color house’s final proofing device isn’t the same machine as your printer’s CTP machine. To print that final color proof, your color house must process the files, which instantly opens the possibility of printing one version for your final proof, and delivering a different version to your printer. This happened to me more than a half-dozen times in my first year of CTP printing. I’ve solved the problem by requiring that some sort of proof be pulled directly from the CTP files that I’m delivering to the printer.

Generally, this proof won’t be high-end, because it’s pulled too late in the process to allow a high-resolution proof to be generated. Instead, you’ll probably see a low-res proof with inferior color, and your challenge will be to use that proof to make absolutely certain it matches your approved high-end proofs.

I’ve found that comparing file creation dates and times is pretty foolproof. If the rough proof pulled directly from the final CTP files contains exactly the same creation date and time as your high-end proofs, you’re almost certainly delivering the correct CTP files. Since I began using this system, I’ve occasionally caught a versioning error, but I’ve never delivered a wrong version of any page to a printer.

Susan McIntyre is president of McIntyre Direct, a catalog consulting company based in Portland, OR. She can be reached at (503) 735-9515.