Using Device Link Profiles to improve color consistency

[Greg_Firestone]

Hi all,

I'm just wondering how many shops are using device link profiles or color servers to improve color consistency across various presses (excluding proofing systems). Or if you're not using them, do you not have the need. This stems from a conversation I had regarding printing to standards and how often shops can't match color easily between devices.

Thanks for your feedback.

Regards,
Greg

 

[TerryWyse]

Personal opinion of one who sets up device link profiles for my customers.....I believe that device link profiles can help you print more *accurately* to a print standard.....but better *consistency* is a myth. If you didn't have your process under control BEFORE you started using device link profiles, don't expect device links to magically improve your consistency....they won't. They could potentially MASK poor print consistency but they won't improve consistency in any real sense.

Regards,
Terry

 

[Lukas Engqvist]

Good point Terry. I find the biggest reason for using device or colour servers is too much ink. There in the Agfa Prepress this is referred to as re-purposing, and has many similarities with ink saving and technologies.

The tricky part is when there is a CMYK mix defined yet the customers order is ambiguous with regard to if this was a coated or uncoated recipe... Here a dialogue with the customer and returning to the source may be the only solution... to simply drop it on a colour server is convenient, but in my opinion a breach in professionalism. I think the client should have part in the decision to reprocess the images.
For clients that want to use out of the box technology, a service provider may need to implement device link to help their client interface to reality.

 

[David Milisock]

Consistency is about having all your color management process under control. A device link profile may be part of that but will have no positive affect on consistency if the other processes are out of control. Which unfortunately is the case in many situations.

 

[gordo]

I think this thread is an example of the all too frequent habit of misusing terminology, which in turn causes confusion in this industry.

Neither device link profiles nor color servers can improve color consistency across various presses since a device link profile is simply a conversion formula to transform an image from one specific profile to another. And a color server simply a computer server.

So, shouldn't the question be more like:

"I'm just wondering how many shops are reseparating incoming files and applying GCR techniques to improve printability (excluding proofing systems)."

Of course the statement "This stems from a conversation I had regarding printing to standards and how often shops can't match color easily between devices." without some clarification does not follow from the question.

So the poll questions might become:

Are you reseparating incoming files and applying GCR techniques?
Using DVL profiles with a Color Server?
Using DVL profiles integral in the workflow?
No, but have an interest in reseparating incoming files to improve printability
No, it wouldn't help in our shop

Or, maybe I'm misunderstanding the post.

best, gordo

 

[David Milisock]

I think this thread is an example of to all too frequent habit of misusing terminology, which in turn causes confusion in this industry.

Neither device link profiles nor color servers can improve color consistency across various presses since a device link profile is simply a conversion formula to transform an image from one specific profile to another. And a color server simply a computer server.

So, shouldn't the question be more like:

"I'm just wondering how many shops are reseparating incoming files and applying GCR techniques to improve printability (excluding proofing systems)."

Of course the statement "This stems from a conversation I had regarding printing to standards and how often shops can't match color easily between devices." without some clarification does not follow from the question.

So the poll questions might become:

Are you reseparating incoming files and applying GCR techniques?
Using DVL profiles with a Color Server?
Using DVL profiles integral in the workflow?
No, but have an interest in reseparating incoming files to improve printability
No, it wouldn't help in our shop

Or, maybe I'm misunderstanding the post.

best, gordo

Agreed! If you think this is bad try troubleshooting an entire postscript proecess or an applications complete color handling process from a forum.

The level of use of technology in graphics is embarrassing, ask gradutes or journeymen what the 3 core technologieds in which a graphic application has to be compliant and see if they even understand the question.

My computer division 30 days ago made a decision to no longer support printing companies, many are in such a mess technologically and have no money. Were still doing business and Cm consulting but we won't sell them computers.

 

[Greg_Firestone]

Thanks all for contributing to the thread.

@TerryWyse - I agree, "accuracy" could replace the word "consistency". The main idea is getting the same color if a job is printed on various presses. Obviously if the process control is out of whack, it won't help at all, but it was assumed process control was in order.

@Gordo - I wasn't implying GCR or printability with my question. I specifically wrote "can't match color easily between devices" but maybe it wasn't too clear. Think more about different color gamuts between presses.

Some examples:

Printing a job for a customer which requires two different jobs on two different press types (e.g. flexo & offset, or digital and offset) and achieving the same color.

Printing the same job on multiple presses (e.g. 3 different offset presses) in different plants and achieving the same color between presses

I'll change the question from " improve color consistency" to improve matching color between devices" I hope this helps clarify. If not, I'll try to provide more details.

- Greg

 

[Greg_Firestone]

Hmm. . . not sure if I can edit a poll.

 

[Lukas Engqvist]

@Greg
Consider this. I was working in a constalation of printers in Sweden. Over a decade ago, with the teaching of the time all printers went through a project that was to calibrate their presses. All printers worked out their maximum contrast and blackpoint and then made profiles for their presses. But we couldn't share jobs. We had adopted late comittment workflows, with images in RGB and that meant that photos, were reasonably consistent… but printed materials have som much more. No good having images look the same if logos and artwork were inconsistent, the short we couldn't share jobs.

All firms were independant and so no one was going to say what rip, CTP or press to buy, but we still needed the strength of the network. But every one "knew" that all presses printed differently (we'd seen the results when we posted our charts with test images on a wall), and so dreaming of sharing jobs met resistance. But ISO was being talked about, even if it was hard to understand and know how to interpret. In the end we were able to agree to try target the same TVI curve, even though some used film and some had possitve and others negative plates. We also adopted an ICC curve with a more aggressive GCR (limit around 300% ink rather than 350%). The results were that the jobs were less noticably inconsistent (grey component now more dependant on K rather than sensitive to CMY variation)... we had a clearer target and so were able to hit it more often... and we could share jobs between presses with much greater consistency. We could even do a prelimenary run in one site with a smaller press and then a major run in another facility (different RIP, different to-plate process and different press, even different vendor ink) with acceptable results.

The systems were all dependant on local quality control. We did find some errors where a proofer would run one RIP and then there would be a bug in the output systems RIP, or perhaps a bad configuration of one sites trapping engine…*but a colour server or device link wouldn't have solved that.

The key was using a standard target for TVI and that standard was to be measured on the printed sheet. Also I would not be fair if not mentioning that these were all sheet offset. For cross technology I do see that device link may have benefits, but one should be careful not to see the colour server as a bigger carpet to sweep inconsistencies under.

 

[Greg_Firestone]

Hi Lukas,

Thanks for the response. I appreciate the real-world example. When I had this conversation with my friend, I argued that if a printer was printing to a standard (e.g. ISO), they should be able to get reproducible color across different machines. But then he threw in the curve ball with different types of presses, ink sets, substrates, etc.

I believe most printers don't use device link profiles or color server software; that's why I created the poll. Your post is a great example illustrating how printers achieved similar color by improving their processes and not investing in additional technology.

Greg

 

[gordo]

@TerryWyse - I agree, "accuracy" could replace the word "consistency". The main idea is getting the same color if a job is printed on various presses.
@Gordo - I wasn't implying GCR or printability with my question. I specifically wrote "can't match color easily between devices" but maybe it wasn't too clear. Think more about different color gamuts between presses.

Some examples:

Printing a job for a customer which requires two different jobs on two different press types (e.g. flexo & offset, or digital and offset) and achieving the same color.

Printing the same job on multiple presses (e.g. 3 different offset presses) in different plants and achieving the same color between presses

That's much clearer now.

This question came up when FM screening started to become more popular in publications printing in the EU. Because FM screening has a larger gamut than AM/XM the FM presswork didn't match the ISO proof or the the previous AM/XM presswork. The solution would be to reseparate files in order to "dirty" the FM presswork. The majority of printers did not want to do this since they would lose the benefits of the larger gamut FM presswork.

I think that would still hold true. In order to align ("match" is too strong a word) color across printing methods you have to target the printing method with the smallest gamut - and that results in poor color on the printing process that has the larger gamut. I don't think that printers, and their customers would accept that compromise.

I'm not sure that you can get "the same color between presses" when printing the same job on multiple presses (e.g. 3 different offset presses) in the same shop, let alone at different plants. Even if they are running the same ink set. You can get close but saying "the same color" is a bit too absolute for me.

In the shop where I worked we could align color across our 9 presses (different vintages and number of units), since they all used the same ink hue sets, by simply using different plate curves to compensate for the different dot gains. No reseparation required.

Since you are a vendor, what information are you trying to understand from the poll? Because it might be better if you broke the poll into several polls. e.g. a poll to find out if printers are, or would, reseparate in order to align color across different print methods given the pros and cons. A different poll to find out if printers prefer a reseparation solution that is integral with their workflow vs one that requires a separate server, etc.

You can PM me if you want someone to review your poll before you post it.

best, gordo

 

[Greg_Firestone]

That's much clearer now.

This question came up when FM screening started to become more popular in publications printing in the EU. Because FM screening has a larger gamut than AM/XM the FM presswork didn't match the ISO proof or the the previous AM/XM presswork. The solution would be to reseparate files in order to "dirty" the FM presswork. The majority of printers did not want to do this since they would lose the benefits of the larger gamut FM presswork.

I think that would still hold true. In order to align ("match" is too strong a word) color across printing methods you have to target the printing method with the smallest gamut - and that results in poor color on the printing process that has the larger gamut. I don't think that printers, and their customers would accept that compromise.

I'm not sure that you can get "the same color between presses" when printing the same job on multiple presses (e.g. 3 different offset presses) in the same shop, let alone at different plants. Even if they are running the same ink set. You can get close but saying "the same color" is a bit too absolute for me.

In the shop where I worked we could align color across our 9 presses (different vintages and number of units), since they all used the same ink hue sets, by simply using different plate curves to compensate for the different dot gains. No reseparation required.

Since you are a vendor, what information are you trying to understand from the poll? Because it might be better if you broke the poll into several polls. e.g. a poll to find out if printers are, or would, reseparate in order to align color across different print methods given the pros and cons. A different poll to find out if printers prefer a reseparation solution that is integral with their workflow vs one that requires a separate server, etc.

best, gordo

Hi Gordo,

Interesting story about AM/FM in the EU. What happened to the FM printers that didn't want to re-separate? Did they lose the job or conform? I can see how the printers didn't want to lose out on the extra gamut, but if the customer approved an ISO proof, I would assume the printer's job is to match the proof. It'd be a bigger problem if they couldn't hit the gamut.

The poll is really just a curiosity question based on a conversation with a friend. I feel quite a bit is possible with process control and standards, he feels device link profiles or special color servers are needed for "similar" color on different devices. I know it's a bit broad because I'm not specifying similar output devices (e.g. multiple offset presses) versus different devices (offset & flexo) because I didn't want the poll to be too specific, though maybe it should have been. I do appreciate the responses that have come up and think they're useful for people to read, especially if they're dealing with similar issues at their shop.

Greg

 

[gordo]

Interesting story about AM/FM in the EU. What happened to the FM printers that didn't want to re-separate? Did they lose the job or conform? I can see how the printers didn't want to lose out on the extra gamut, but if the customer approved an ISO proof, I would assume the printer's job is to match the proof. It'd be a bigger problem if they couldn't hit the gamut.

In the EU, the larger circulation web printers largely abandoned FM because they couldn't always align their presswork color to the ISO proof while the smaller sheetfed publication printers stayed with FM because their customers appreciated the extra color.[/QUOTE]

The poll is really just a curiosity question based on a conversation with a friend. I feel quite a bit is possible with process control and standards, he feels device link profiles or special color servers are needed for "similar" color on different devices.

If the presses are printing to a specification and you have process control in the shop(s) and you control the separations coming in (or they are done to specification) then IMHO, you don't need to reseparate to achieve "similar" color on different devices of the same type (e.g. sheetfed offset). But if the separations vary quite a bit in how they were done then reseparating can normalize the different separations which, in turn, would make it easier to hit the presswork target for color. (e.g. separations done for sheetfed being printed on coldset web).
Aligning color across different printing methods (e.g. sheetfed and flexo) can be problematic because they have different industry specifications.
Device link profiles can be used within a workflow or on an external color server - there are pros and cons to each - but they both would use some kind of DVL.

Whichever way you work it you will need specifications (shop-based or industry) and process controls to be in place. DVLs aren't a substitute for specifications or process controls.

best, gordo

 

[Mike Strickler]

As Gordo says, there is no important question regarding DLs per se as they are required for any CMYK-CMYK conversion, typically to repurpose a file for a different print condition (e.g., HQ sheetfed to coldset web) or to impart higher GCR for a bit of ink savings and gray balance stability. Despite all the mysticism surrounding them, DLs are used for one big reason: They allow preservation of the black channel as well as primaries and/or certain 2-color overprints, etc, depending on the capability of the software. That's it.

Or not quite. PDF workflow tools can create DLs on the fly when converting files to a new destination profile. Even Acrobat Pro does this when the Convert to Color command has the pure black and CMYK primaries options checked. In these cases the user doesn't see the device link, but one is being used nonetheless. Here again, the DL isn't the point; the question is whether one is employing a GCR-type solution or not. Most of that issue will go away as nearly all CMYK will be created in higher-GCR color spaces to start with and the main function of DLs will be for repurposing files for other presses/substrates.

 

[J]

Can you use a device link profile to contaminate a color. E.g. a 50% magenta becomes a 45% magenta with 3% K and 2% C?

J

 

[TerryWyse]

Can you use a device link profile to contaminate a color. E.g. a 50% magenta becomes a 45% magenta with 3% K and 2% C?

If that's the result of the device link conversion (difference between the source/destination profiles you used when building the link) then, yes. With most device link profile creation software, you have the choice of keeping the primary and secondary colors pure or contaminated. "Contaminated" is usually the more accurate conversion, if there's a difference between, say, the hue of the magenta ink in the source profiles compared to the destination. If you allow contamination, you'd essentially be doing a standard ICC conversion. Most of the time with a device link, especially if it's for an offset press, the primary inks are kept pure and sometimes the secondary inks are kept pure as well. Preservation of purity (or not) can have unexpected consequences in areas of solid color such as color bars or special tint builds.

Terry

 

[Prepper]

Just a thought

Just a thought

@TerryWyse ....."but better *consistency* is a myth.....They could potentially MASK poor print consistency but they won't improve consistency in any real sense."

I understand what you mean but if they do MASK poor consistency then isn't that improving consistency?

I don't use DL's because have no need to, being in a totally in-house enviro where all our files are RGB and I get to convert them myself to CMYK. I would like to add that printing to a standard/spec, like G7, using curves to bring all devices to a common tonal curve, and using a high amount of GCR, has given us amazing results with color consistency.

I now see press checks that are close to proof way before the densities are leveled out. One color can be up 15-20, another down 10-15, and the color is still pretty close to proof, even at that. It's amazingly better than what we used to get using regular USWebCoatedSWOP, which has med GCR? built in I think, and we were using curves at that time to control dot gains to be within the SWOP specs and color varied way more at startup and during the runs.

Just our experience, but it has been an amazing difference. One of our designers was just commenting the other day at a press check that we don't have color problems like we used to and that we haven't had to adjust anything for color in a long time, other than minor tweaks at press.

 

[gordo]

[SNIP] I would like to add that printing to a standard/spec, like G7, [SNIP]

Just for the record "G7" is neither a standard nor a specification. It is a method.

best, gordo

 

[tmiller_iluvprinting]

Just for the record "G7" is neither a standard nor a specification. It is a method.

best, gordo

G7 is a specification, please see 'The G7 Specification 2008' final working draft.

Regards,
Todd

 

[rich apollo]

Couple of odd-duck thoughts to go with this thread.

1) I've recently been going over some measurement data from a press test that someone was generous enough to share. They ran AM and FM screening on the same sheet. Now, the ink and water requirements are vastly different for AM and FM screening, so with them being on the same sheet the operator can only optimize the conditions for one of the two. The surprise, though, was that while they hit the same color targets for the solids on each, the dot gain characteristics of FM screening caused plugging and hue shifting in saturated colors. The measurements from the AM screened target exhibited none of this - the measurements were beautiful. In essence, the FM screening seemed to actually limit the gamut because the saturated colors became uncontrollable. The data indicated that the operator was over-inking.

2) Device links might not be appropriate for items that extend across multiple processes. For example, the solids from a sheetfed press and a screen printer probably won't match. Preserving the purity of solids and overprints would actually create a mismatch across the processes. A color match would require the primaries and overprints to be contaminated.