User login

Taking Control with Color Management

(May 2006) posted on Tue May 23, 2006

This article examines the benefits of color management and uses an actual job to demonstrate how color can be controlled.


By Rick Auterson

click an image below to view slideshow

Most imagesetters, direct-to-screen systems, and digital presses these days are sophisticated enough to apply input and output profiles. A color-managed workflow is possible without a workflow program like iQueue as long as a RIP that applies the profiles resides somewhere in the process. We find it easier to keep all of our output devices linear and apply the profiles in iQueue. Even our VUTEk 2600 and Inca Eagle inkjet printers are linearized.

How do you measure success?

We used to measure color bars on proofs and prints to evaluate our quality. We even went through a stage where we tweaked our profiles to produce the proper values on the color bars. But the process didn't seem quite right—it was too image-dependent.

Tweaking profiles can turn a 50% yellow into red and leave the 49% yellow a perfectly good yellow. That's a bad tweak, of course, but it shows what can happen. We experimented with a variety of methods for making profiles better, but we were missing the point. A profile is just a big list of builds and L*a*b* values. It's not as complicated as it might seem. If builds cease to produce the colors you expect, you need to quickly act to find out what changed. The builds produced those colors when the profile was made, and that's the best information you've got. Don't tweak your profiles; tighten your process.

Color bars contain extreme colors at the outermost edge of the gamut: pure cyans, magentas, and yellows. Our experience has taught us that tweaking a profile to get better color bars will cause other, less obvious, builds to be off-color.

So now that I've trashed tweaking and color bars, you may be wondering how Pratt measures success. We use the same color target that created the profiles. If we put a color target through our workflow, it should have the same colors when it prints as the one we made from a MatchPrint. A color target contains nearly a thousand builds—the perfect thing for someone who needs to know about color. Put a color target through the workflow—your digital proofer, for example—and read it on a spectrophotometer. The measurement data are written into a text file.


Terms:

Did you enjoy this article? Click here to subscribe to the magazine.