Calibrating the Kodak XP4 Half-tone Proofer…

by admin on March 30, 2008

The difference between a halftone proofer and the type of proofers you see at most shops (HP or Epson continous tone inkjet) is that with a halftone proofer you can check the dots. With some type of jobs, this is important if you use AM screening and need to check for moire.

Also, some high-end customers insist on halftone proofs and are willing to pay for them. And nobody in this field of business says no to high-end clients willing to pay preminium rates for a halftone proof.

Chris Riley of Dixon Direct Corporation and a regular long-time reader of this blog (yes, I do have some regular readers) contacted me to ask some advice about calibrating a Kodak XP4 halftone proofer to SWOP 5. Enormously flattered, I proceeded to give him some bad advice. Oh well, can't be perfect all the time.

Here is the advice I gave him.

1. Linearize the XP4 (that is to say, ensure a 50% dot specified is a 50% dot on output)
2. Apply the calibration curves specified to hit SWOP 5.
3. Ignore profiling, except to ensure that the proper target is set in the Prinergy refiner process plan.

Well the correct thing to do to hit SWOP 5 according to Kodak.

1. Ensure the densities are correct (yeah that makes sense).
2. Download the standard curves (Ditto, but they don't say to linearize).
3. Apply a DVL (device link profile) to the output process plan (At which point I said, are you kidding).

Basically, to hit SWOP 5 or Gracol, you need a DVL as the gamut of the ink paper & combo of the XP 4 is not big enough without it, or so I understand. Mind you, applying a DVL means that sometimes extra colors are added so you lose your dot-for-dot veracity to the plate, in my opinion at least. I mean, it defeats the purpose of using a halftone proofer if you employ DVLs.

Anyways, the last time I emailed Chris he was making some sort of progress matching to SWOP 5 but he was going to try to match his XP4 to some XP4s from TWO other sister plants. And considering he was working with curves & DVL I thought I would give him some time before checking up to see how he was doing. I mean, I don't think he was going to get everything working perfectly in a matter of days.


Dave Kauffman March 31, 2008 at 10:18 pm

Well, I’m not exactly an expert on this color stuff, but just ignorant enough to dare post on this thread!

I think it’s fair to say that some of us think of proofing as a range between two extremes. On one side is the dot-for-dot proof, where you can compare the proof and press-sheet under a loupe and see the same dot structure. If your donors have the same hue as your inks, this should be a good contract proof. That’s how Chromalins, and Matchprints, and Approvals have been used for the most part.

On the other end of the range is color managed proofiing, as DJ alludes, say using DVLs. On this end of the range, you are giving up some dot structure, and probably allowing contaminants in your pure tones, but you do this to get a better overall color match.

Between the two, curves allow you to make this balance between dot integrity and color match.

So the way I see it, which you choose depends on your customer’s expectations. If they are the type to whip out the loupe and match dots, or do they put it into the match booth and stand back to compare color match. As usual, and with luck, adjusting the technology to match customer expectations is usually the profitable path…

Chris Riley April 3, 2008 at 8:26 am


thank you for your help and insight on this matter. your advice was very helpful, although, as you said, it wasnt exactly mirroring the advice i was getting from kodak. even that was helpful, though, in that it got me thinking about our entire approach on this project.

since last you heard from me, we have made some good progress, and have since created 3 different proofing media configurations (SWOP 3, SWOP 5, and GRACoL). all 3 of the specifications were built according to the protocol described by Kodak, with the DVL’s downloaded from their website. Here are the steps i took:

• output a proof (the Kodak Approval XP4 Dot Gain Calibrator pdf that was provided to us by Kodak) with no calibration curves, DVL’s, or profiles of any sort, to insure that the proofers was in fact proofing linear. while it was not perfect (and everyone knows we should strive for perfection), it was within the realm of tolerance for us at that point (+/- 2%).

• created a media config using the Kodak Media Config Manager, and adjusted the clicks to the specified density for the standard (as provided on the ADS downloaded for each specification from the website).

• output a proof to density specifications, and read the swatches proof into the Dot Gain Manager using the calibrated Spectrophotometer.

• used the numbers generated from Dot Gain Manager to create “Current” curve in Harmony. Created a Target curve in Harmony using the target numbers provided on the ADS for each specification. Used the Current and Target curves to create a Calibration Curve.

• in Prinergy, i used the downloaded DVL for each specification, and applied it in the ColorConvert tab of the Process Template. i also applied the Calibration Curve in the “Calibration & Screening” tab.

• i output another proof, using these settings, and had a proof that meets both the dot gain AND the density settings as reflected in the ADS for each spec.

now i KNOW what some of you are saying, and i am sure some of it has to do with fingerprinting the presses (we have 8 web presses here), and i AGREE with you. that is the goal, and it will be achieved eventually. for now, however, we have a proof that we have had pretty good success matching. we have only begun this project, and have lots of work to do.

i still have questions about what exactly a DVL profile does for me in this process. I use it because that is what Kodak recommends, and i am beginning to see why i would need it when fingerprinting presses and other proofing devices, but what EXACTLY do they bring to the table?

and one last thought. we still have the challenge of matching 3 Kodak XP4’s ahead of us. i feel pretty good about the proofs we are outputting, but getting everyone together on the same page from each plant is going to be a challenge. we are making progress, but each plant seems to have ideas about how this is to be approached, due to press conditions and so forth. using the method i mentioned above seems to be a pretty no-nonsense way of getting them close, but perhaps this is where DVLs come into play?

thanks to dj for helping a stranger out. this is a great site, and i continue to get great info from this blog. and thanks in advance to anyone else who has further insight on this, id love to hear it.

Comments on this entry are closed.