Help! Still Making Color Keys

Coming from a job with a fully automated Kodak workflow in the offset world to a flexo shop using application trapping, outputting film with a Harlequin rip, and making color keys seems like a stupid move, but the challenge of rebuilding a workflow was the attractor.
I'm trying to find a proofing system to replace our color keys. I understand there are printers and rips out there that show a dot, but are they accurate representations of the film?
We are looking at a EFI Colorproof XF bundle with dot creator, onebit tiff, file export and an Epson 4800 17". I think we can get great color matches with this, but we have lots of issues with hard lines and dot gain in hi-lite areas on press and I am not sure that this printer will accurately reproduce these issues.
The biggest hurdle of mine is that digital plates are out of the question. We need to match a proof with our analog plates. Anyone have any ideas? Color key is becoming hard to find, we need to find a solution.

ZH
 
Re: Help! Still Making Color Keys

We use Black Magic from Serendipity (Aus) running from an HqRip and using Real Dot approach and drive Epson and HP plotters. In general, we are very happy and even make color mockups for sign-off.

When I used to do flexo, I would make a litho proof first and then pass huge flexo curves (that singly, matched the presses) over the job for the actual 7 thou matte film used for Cyrel plates. As I recall, 5% became 8-11, 8 became 23-25, 50 became 80% etc.
We often had to run mono or duo color for highlight detail such as beer foam (yellow and black); the 4col foam was far too heavy. Using a rudimentary workflow as you are is actually a bonus because with a separated workflow, you have half a chance to get something good and you are in total control! But you also need operator skill for sure!

I suspect that the inkjet proofs as composites might be a problem and you might have to make progressive inkjet proofs too to give you the same result as your color keys.

If you have lots of money, you may want to look at Latran or Approval or Final proof but with all of these you still require skill to adjust for your crazy (but useful) print
process

John W
 
Re: Help! Still Making Color Keys

Brad,

We went from computer-to-film-to-plate to CTP, and when we did, we set up the CTP so that the resulting print showed the same color as the old workflow. So instead of going computer-to-film-to-plate where the linear film would get about 4 dot gain when the plate was made (50 in file = 54 on plate), we set the CTP with a bump curve so that 50 in file = 54 on plate.

Now I've tested and worked with GRACoL2006_Coated1v2 (I haven't used G7 method, but it is recommended), and it matches very well overall with what we've been printing for years. So now, I can choose to print to a U.S. specification of the ISO 12647 standard (which also matches closely with ECI's ISOcoatedv2) and my printing will match closely with what is printed other places as well.

As far as proofs, since printing flexo, will need to set up press using G7 method, and will need to make your own ICC profiles for each paper type, and then set those as source in the proofing rip, and set the custom proofer profile (inkjet) as destination, and use Absolute Colorimetric Intent.

Don
 
Re: Help! Still Making Color Keys

Thanks!
I've looked into Black Magic and on the surface, it seems right up my alley. I think the "rip once, output many" idea is more what I'm looking for. I spend a lot of time in Photoshop manually adjusting images, especially highlight areas because a 2% CMYK is more like a warm gray. Our 1% dots gain from 5-7, and different rips regardless of the profiles may output those areas differently, and that is what I'm trying to avoid.
Flexo is making a big surge into the world of high quality process work due to the advancements in anilox rollers, and I'm just trying to keep up.

Brad
 
Re: Help! Still Making Color Keys

Brad
If in any of your proofing operations, you are using ICC profiles, then forget about realistic rendering of the highlight areas. Look-up tables in ICC profiles will render on a linear manner from 10 to 0. I have added the notion of “first dot” in every output strategy and for proofing it consists of adding a bump-up curve related with such “first printed dot” value. May I suggest to use 2 Cyan,
1 yellow and 1 magenta as a white point. This will give the highlights a “clean” look. I proof my riped files on an Epson 9800 using Flexproof proof Downloader, it is kind of tricky to color manage but we see the actual plate dots. Doing so on clear substrate imitates the good old color key some plate mounters still cant live without.
 
Re: Help! Still Making Color Keys

> If in any of your proofing operations, you are using ICC profiles, then forget about realistic rendering of the highlight areas. Look-up tables in ICC profiles will render on a linear manner from 10 to 0.

Luc,

I don't follow you here. ICC profiles don't "render on a linear manner", but are reflective of the processes involved in generating the characterization data from which the profiles are created.

rich
 
Re: Help! Still Making Color Keys

Danny,

You've got to be kidding. The below info is for anyone that doesn't understand ICC profiles.

ICC profile is the description file, describing the output device, that is embedded in an image or document.

When one converts from RGB to CMYK (or builds in CMYK), the CMYK ICC profile used in the document give meaning to the numbers. That's why images look a certain way in your program. It's the ICC profile that gives meaning/context to these CMYK numbers. Without an ICC profile embedded, we wouldn't know how the image is intended to look. The appearance changes depending on output device.

The cool thing about ICC profiles is that if converted/build to the output ICC profile, the print will look as the designer intended (assuming the monitor is also profiled with a solution such as X-Rite's Eye-One Display2 or PANTONE's Huey - although I haven't tried Huey - to make sure what the designer is seeing is correct).

Every device in the workflow (scanner, camera, monitor, proofer, press) can be calibrated and profiled to make sure the device is displaying accurate colors (within limitations of the device).

When a designer makes sure everything is in the correct output profile when building (which most use the default SWOP profile now, which shares the same NPDCs with GRACoL2006_Coated1v2 and prints looking natural on GRACoL2006_Coated1v2, but I'm talking about using the actual GRACoL2006_Coated1v2 profile if designing for printing on #1 coated paper or #2 matte paper, and GRACoL2006_Coated1v2 is within tolerances of the international standard), and a printer proofs and prints to that same profile (using ICC profiles in proofing, and setting press up using G7 method or ISO 12647-2 documentation), then colors come out as expected. Simple as that. It's called everyone getting on the same page, and this industry has a hard time accomplishing this for some reason (communication breakdown, of which I have in my company I work for, and I too am part of the problem for not being able to simplify this stuff well enough for my boss to not have "deer-in-the-headlights" look on his face when it's brought up). That's why we proof customer's SWOP CMYK numbers and if they have a problem with the proof, we talk about things then, but usually there are no problems, and U have found that we've actually been printing close to the international standard for years. But having a profile available for us U.S. people now to use puts a stake in the ground and says that yes, the U.S. now has officially adopted the international standard and everyone can know what the aim is. Now we all can hit a defined target.

If different profiles are used thoughout the workflow, then matching becomes harder. But for designer's sakes, can't we all conform to the international standard that has been in effect for more than 10 years, that we now have an ICC profile defining that appearance, and now that we know that the U.S. and Europe's printing (and China's too, since it adopted the G7 method of IDEAlliance/GRACoL) is so close now that we can actually all use standard profiles and get accurate color no matter where the job is printing (Note: Europe's TVI/Dot Gain is about 2-3% less than GRACoL2006_Coated1v2 using G7 method, but they're so close in appearance most people wouldn't be able to tell - only the most color critical. But if printing in Europe, one could always use the profiles downloadble from www.eci.org)? Either way, I see that the appearance is only going to get better defined (mapping of out-of-gamut colors into the printing condition is where more work needs done), but we can start using now. It's that well defined what the appearance should be.

Don
 

PressWise

A 30-day Fix for Managed Chaos

As any print professional knows, printing can be managed chaos. Software that solves multiple problems and provides measurable and monetizable value has a direct impact on the bottom-line.

“We reduced order entry costs by about 40%.” Significant savings in a shop that turns about 500 jobs a month.


Learn how…….

   
Back
Top