Color Management

[miconut]

New to Digital, background in offset and flexo. I have ESKO Color Pilot and a few HP indigo's we're working with. I was assuming that color would be managed similar to what the normal process is. I have a printer profile, I have a target profile, I build curves to match my printer to the target. Now that I'm digging through things It looks like color can be managed at both the ESKO side and the HP side resulting in a few different options. What are people actually doing?

Options I can see -
1. I calibrate the press, use this as my printer profile. Load this into color pilot and set my curve to match GRACoL 2006, output as normal
2. Output from ESKO unmanaged, have HP calibrated and match my curve to GRACoL.

I'm sure I'm missing something but I'm only two days into looking at this.

 

[Mike Strickler]

Because you mention Color Pilot I assume that your presses are the roll-fed models. Setting aside the matter of spot colors, which are converted in a different path from RGB, gray, or CMYK objects, my observation is that most users employ a workflow similar to yours. The Indigo's primaries and standard EXP05 curves yield an "offset-like" result without color-managing further. You can fine-tune gray balance using curves alone with the usual G7 tools, though most shops don't, because they lack the time, perceived need, or the knowledge to do it. However, even if curve corrections are diligently applied, overprints tend to look different from those printed by conventional presses owing to their different trapping behavior and perhaps ink opacity as well. A more comprehensive approach is to actively color-manage the workflow by profiling the press and employing the profile as the output space in Color Pilot, or alternatively in a color server such as ColorLogic ZePrA with CM disabled in Color Pilot, as it is by default. CP has certain limitations--lack of perceptual conversions, use of non-ICC CMYK+N profiles, lack of good device-link options or convenient profile updates with small charts, etc. A strong argument can be made that conventional press jobs can benefit from effective color management, particularly when printed on a variety of substrates. The argument is stronger for digital print, if only because files are normally prepared by default for conventional web or sheetfed offset and without converting the color it won't match the former very well (depending on your standards, of course). Bottom line: No, you're not doing as much as you could, but most of your peers aren't either.

 

[miconut]

Bottom line: No, you're not doing as much as you could, but most of your peers aren't either.

Thanks for the reply! I've been at five different plants, and nobody has had an active color management process before I started so I'm excited to tackle this in a new print medium.

The biggest issues I currently see are predictability and consistency.

Predictability I've narrowed down to no tie between what are are outputting and what we are actually trying to target. There isn't anything in place right now for profiles or conversions so every time something prints it is matched visually by hand.

consistency I've narrowed down to a few things. If prepress isn't available the press operator will adjust at press on the fly, causing issues the next time something comes in for a reprint. Huge opportunity there. It also looks like the three presses all calibrate individually to some target - unknown what exactly but not to each other but all strategies out of color pilot are set exactly the same.

So my goals currently are:
1. Get each press to print and calibrate to a consistent baseline across the board.

2. Use that baseline as press profile, and use color pilot to set my target output intent to GRACoL. I don't imagine the entire gamut will be printable, but it will likely result in a closer acceptable output than we are currently at.

Is there anything that sounds incorrect or not possible with what I have outlined above?

 

[Mike Strickler]

Predictability and consistency are two sides of the same coin--the somewhat less stable output of this type of press. There is both short-cycle and longer-cycle color drift. Other than machine maintenance there is little you can do about the former; consider it the noise floor of the system. Longer-cycle drift can be addressed, of course, but some methods are better than others. Curve adjustments at the press computer have the disadvantage of messy housekeeping, which you've alluded to. The correction is for dot gain or loss, and by extension gray balance and contrast. Anything more, for example, a change in overprints as a consequence of a substrate change, would not be addressed effectively. For that a "3D" reprofiling would be better, and much is gained by keeping operators out of the curve tools. EXP05 is a good preset--you can leave it there and profile the press.

GRACoL CRPC6 is actually mostly in-gamut for an Indigo with CMYK inks printing on coated paper, so it is in fact a reasonable target for normalization of files destined for printing on similar substrates. CRPC3 is a good general-purpose reference for uncoated stocks. The output profile, assuming you're color-managing, is whatever the press profile is made on the selected substrate for the job at hand. So, this means fixed curves, a profiled press with profile set as output in the DFE our color server, and some means of updating the press profile as color drifts. If the procedure takes longer tan 5 or 6 minutes you can be pretty sure the operators will avoid it, as they have work to get out.

Once this reasoning is accepted you need to consider the quality of the color conversions. A number of third-party tools are available, though they differ greatly in capabilities in handling complex PDFs, and this includes not just image quality but also spot color conversions, transparencies, and many other tricky matters.

We haven't discussed spot color conversions, per se, but matching spot colors actually absorbs the greatest amount of operator time in a label shop, as variances in these colors are immediately noticed. Accurate press profiles are essential, but so is a rational method for characterizing these colors, and for most operators this means, wrongly, specific device values (e.g., CMYK) rather than colorimetric ones. Another handicap is the failure to exploit the extra Indichrome OGV primaries when needed to make the match. The usual reason given is the impact on press speeds and click charges, but just as important is the failure to understand management of multicolor spaces. This is not part of standard operator training.

I've tried to keep my comments general because these are the prerequisite ideas for understanding the problem. If you need more specific guidance you can find me offline.

Best of luck,

Mike Strickler
MSP Graphic Services