PDA

View Full Version : Best way to Manage ICC's For many paper stocks.



Troy76
07-24-2017, 11:22 AM
Hi There,

The company I work for provides a wide variety of digital stocks (30-40 plus a dozen or so compatible offset stocks) I've created calibration sets for the most commonly used papers (almost 30) But we all know: a profile's accuracy changes slightly with the machine's condition and paper lots, so over time the profiles need to be redone.

Not only do I think this may not be the best way to do things, I just don't have the time to be constantly rebuilding icc's on a regular basis to achieve the 'best' results all the time. So I'm looking for a better way.

I'm thinking of reducing my icc's to paper weight & coating and then build a device link for each matching stock. (e.g.. I'd use x1 icc for all my 80# plain covers and then have a device link for my Finch, Cougar & Soporset paper stocks) I'm thinking itinerating a device link every now and then will be faster then constantly re-profiling, but I'm not sure if this is how device link is intended to be used or if it will work this way at all?

Any suggestion on whether I'm barking up the right tree would be great? Or recommendations from what other people are doing when faced with this many stocks and a demand for consistant colour accuracy?

Thanks in advance

adam1991
07-24-2017, 05:23 PM
I honestly don't think device link profiles is what you're looking for. They have a purpose in life, but general shop work isn't it.

You're focusing too hard on the "iteration now and then is quick and easy" but you're ignoring the core functionality of the device link, which hard codes too many things together. I don't think it's the right tool for the job.

I think that you'd be well served by simple calibration sets unique to each paper. Make a single "plain cover" profile as you say, but create calibration sets for each stock, sets that use that plain cover profile.

It really depends on what "good" looks like to you.

Frankly, I dispute your core assumption that "profiles need to be redone". I understand paper lots changing, but a good machine will manage itself to present a consistent face to the outside world (it will manage itself well with regard to producing accurate AND consistent ink density on the sheet) and that takes away your concern about machine condition.

For complete accuracy, take the time to create custom profiles for each stock, and then use the Fiery calibration system to keep them in check.

Even if you did end up re-creating profiles, how often would you do that? Yearly? Twice a year? Yes, that's a concern--but with a modern autoscan spectro such as the KM FD-9, creating a very good profile can be as simple as using CPS's Express Profile feature. I call it "7 sheets of paper and a cup of coffee". 10-15 minutes and you're done. And once the profile patches are printed out, the press can go back to production while CPS does its thing.

Assuming worst case of 30 profiles twice a year, spread out during production using an autoscan spectro--that's quick and easy for what you get. But yes, it is a lot of work.

The real question is, do you need to do that for what your shop is expected to deliver? Or will simple calibration sets using one or two basic profiles do the job for you?

You can test all of this using Verifier.

oxident
07-25-2017, 07:02 AM
Very well explained. I totally agree!
In my experience, you could also skip the different calibration sets for different stock (of same coating) but yes, this depends on your needs. For me, that's sufficient.

adam1991
07-25-2017, 08:13 AM
yes, absolutely. It's a continuum of degrees of acceptability. Everyone's needs are different. All you can do is start at the bottom, create procedures and apply them consistently, and if they're not enough then go to the next level and keep doing that until you're happy.

Many people start with coated/uncoated custom profile/calibration sets, using good stocks, and never go further.

Troy76
07-25-2017, 08:30 AM
Thanks for the info,

I think you cleared up a few misconceptions I've had about profile management.

Currently I do have profiles for each stock, as I was having a problem with Cool & Warm white-points with various papers stocks causing unwanted colour shifts when printing unsaturated colours. I also notice density / colour changes with different weighted stocks using the same icc. This was my reasoning for profiling each stock. I was also told some shops rebuild their profiles with every lot# change or major servicing of the machine, this seemed overkill but I was hoping to find a happy middle.

I think your suggestion of Calibration sets for less profiles is a good way to go. I didn't even know that was an option or am I misunderstanding you. Are you suggesting create a profile on a stock (e.g.. Plain_Cougar Paper) and then just calibrate on any Cougar paper I load (text or cover) or is there and option to choose different calibration sets when I calibrate and icc for a stock. If the later, how do I do this or where can I find the info to do it?

As for my shop's expectations, it varies with every customer (and salesperson) sometimes they're easy going and sometimes highly demanding. As a result I aim to hit a bullseye on each job to not risk orders coming back to me. So my intention is a workflow that gives consistent and accurate colour but can be maintained easily without cutting into production time.

Thanks again for the reply

adam1991
07-25-2017, 02:42 PM
I also notice density / colour changes with different weighted stocks using the same icc.

Yep. Expected.


I think your suggestion of Calibration sets for less profiles is a good way to go. I didn't even know that was an option or am I misunderstanding you. Are you suggesting create a profile on a stock (e.g.. Plain_Cougar Paper) and then just calibrate on any Cougar paper I load (text or cover) or is there and option to choose different calibration sets when I calibrate and icc for a stock. If the later, how do I do this or where can I find the info to do it?

You could do your first idea, but that gets in the way of a media driven workflow where a job might call for two similar stocks that you really want differentiated.

Do you know how to create a new calibration? Do that--and at the very end of the process, Calibrator asks you to associate the calibration with an underlying output profile. You can create a custom one, or you can choose an existing profile that's on your system.

There. Now you have one profile that is underlying a wide variety of stocks, each with its own calibration. It's very similar to what you're doing today, but without having dozens of unique output profiles. Instead, you might have a small handful. Coated cover, coated text, uncoated cover, uncoated text. Make 4 custom profiles, then using Calibrator create individual stock calibrations that each calls one of those 4.


As for my shop's expectations, it varies with every customer (and salesperson) sometimes they're easy going and sometimes highly demanding. As a result I aim to hit a bullseye on each job to not risk orders coming back to me. So my intention is a workflow that gives consistent and accurate colour but can be maintained easily without cutting into production time.

Honestly, given that you will have to satisfy a demanding customer and that you want to hit the bullseye every time, you want unique custom output profiles per stock like you have today.

To get rid of your anxiety about managing them all, find out how your press works with regard to managing a consistent and accurate ink density across time. (It may not be able to do that. What press do you work with?) If your press can present accurate densities on paper and do it consistently, and can measure and adjust itself automatically on this basis, then you have no worries about machine aging--because the machine is maintaining itself in this regard, and it's presenting a consistent face to the outside world, to the Fiery that's running it. As long as the machine is consistent this way, your profiles won't age nearly like what you're thinking.

If your press struggles to provide consistent ink densities across time, you'll have to figure out what compromises you want to make with regard to hitting that bullseye.

And if you use a quality paper vendor/manufacturer, I think the paper lot thing is overblown. Not that it isn't an issue; sure, when you buy on price alone you get what you get. Me, I'm too poor to buy cheap. Buy quality paper from a quality vendor, and eliminate the worry. Now all you have to worry about is your press and its abilities.

Troy76
07-26-2017, 08:48 AM
You're just filled with good info... thanks.

I don't think I've created multiple calibrations for one profile before. I have noticed if I alter a profile in editor and re-save it, it creates a new icc & calibration linked to the original profile. Is this what you're talking about? or is there a better way of doing this?

I was also thinking there may be a way in Editor to compensate for any potential tonal differences among stocks with each new calibration, but all the editing options I see, seem to involve eyeballing (which I try to avoid as eyeballing usually works on a job per job basis) and there seems to be a lack of tutorials on that part of Profiler Suite. Is this worth looking in to?

As for my machine: it's a Ricoh Pro 7110x. Maintaining the density levels on it used to be really bad and I often called it a lemon :o but since I started following the maintenance schedule I came up with a year ago, it maintains really well now and I'm getting near press quality. This is probably why when I did test yesterday varifying an icc on different stocks (using same calibration set) I only averaged 1 off on the DeltaE, where as 1-1/2 years ago I averaged 7-8. But this may not hold true on all stocks which is why if there is a way to compensate for bigger shifts in Editor I'd like to know. But if the stocks only vary 1-4 points off, I think that is more than acceptable.

Thanks again for the time you've spent, it's been very VERY helpful.

adam1991
07-26-2017, 10:40 AM
You're just filled with good info... thanks.

I don't think I've created multiple calibrations for one profile before. I have noticed if I alter a profile in editor and re-save it, it creates a new icc & calibration linked to the original profile. Is this what you're talking about?

No.

Just start by creating a new calibration in Calibrator. At the very end of the process, it asks you to associate that calibration with a profile. Choose a profile. That's it. Now yoiu have a new "calibration set," which is nothing more than the calibration info and requirements bolted on top of a copy of the profile you chose.

So the calibration MUST be associated with an output profiles, but you get to choose just which output profile that is. It can be: (a) a new one you created at the end of the "create a new calibration" process; (b) an existing custom profile that you've already created; or (c) an existing factory profile.





I was also thinking there may be a way in Editor to compensate for any potential tonal differences among stocks with each new calibration, but all the editing options I see, seem to involve eyeballing (which I try to avoid as eyeballing usually works on a job per job basis) and there seems to be a lack of tutorials on that part of Profiler Suite. Is this worth looking in to?

At this point, you're going down a DEEP rabbit hole with very little return on investment--and you are WAY better off, from both a time AND accuracy standpoint, just making a custom profile for each stock type (brand, model, coating, weight) like you talked about at the beginning.

Question: is there anyone at your vendor who can discuss this with you?

Justin_dB
07-26-2017, 11:19 AM
At a minimum, every profile must be associated to 1 calibration.
You can use the same calibration for many profiles.
You can't (without creating duplicate profiles) create many calibrations for 1 profile.

This thread is interesting.

The strength of the Fiery is its flexibility - and I love in the world of color how there are many ways to achieve your goals - and just as many ways to go wrong.

Justin

adam1991
07-26-2017, 11:36 AM
Yes. Many, many paths to achieve a goal. That comes from a depth of engineering talent.

But that sometimes comes back to bite you. I always say the best thing about EFI is that they have twelve different engineering teams. And the worst thing about EFI is that they have twelve different engineering teams.

But I have to say, within the context of this thread, CPS and Fiery together make it super easy to just do the right thing--custom profile to the depth you think is important, and assign those calibration sets to whatever paper catalog entries you need.

But an autoscan spectro is the cherry on top of all that. I wouldn't approach CPS in anger without an autoscan spectro.

Troy76
07-26-2017, 12:36 PM
Thanks again,

I've only been using fiery on an advanced level for 2 years now. But it's definitely great how the software works in layers / various user levels and you can "go deeper down the rabbit hole" as you go. :) Which is a good thing because each shop has a different set of demands and workflows (if any) which I'm still trying to find the best solution for.

I do have a local "guru" I can speak with but he is often busy and where I've been jumping around with various ideas trying to figure out how I wanted to change my system, I figured the forum would give me the most view points and ideas.

My Current plan is to create profiles for each stock type and similar white point. This should consolidate me down to about a dozen profiles verse 3-4 dozen. While I'm creating and setting everything up I'll use Verifier to make sure the stocks work properly (or properly enough). I've also set-up a spreadsheet to monitor and track the stocks.

I've been thinking and I'm not sure why I need to create more than one calibration set for each profile?

When I did my test yesterday I did get a warning when I calibrated, but my density graph looked fine and Verifier gave it good match to the profiled stock. But if I can't reach a density when I calibrate on an alternative stock, I'm thinking I could just modify the PaperCatalog on my Ricoh to boost the Image Density of the problem colour and let fiery drop it down. (I'm probably going to need to bug my "Guru" for that one.)

Has anyone else tried this?

Thanks

malcolmc
07-27-2017, 10:11 AM
Some comments first:
You should not need to redo the ICC profiles that often unless the environment is varying significantly and the print engines are in non-consistent states. Regular calibration should be used in order to adjust for environmental factors.
Regards DeviceLinks: DL's are designed to take you from a source profile (either RGB or CMYK) directly to an output profile, or through intermediary profiles in a more advanced usage. Be careful implementing these because your source files may contain other color spaces such as Device-n, L*a*b* CIE, grayscale etc so DL's can be misleading as they won't handle these color spaces directly. The Fiery uses a mechanism whereby you use a source profile and output profile combination to trigger your device link. This is done so that you have individual profiles available in order to handle and manage color spaces that will not be handled by the DL. If your shop uses color spaces other than pure RGB and/or pure CMYK DL's may not be the best choice. Remember that Spot Colors are Device-n1 so an output profile will need to be used to properly apply the Spot-On feature of the Fiery.

And now for my advice:
Use the Fiery Paper Catalog to assign profiles that apply to the stock. You may not need one profile per stock (as you have indicated), so create a single calibration set/profile for a group of similar media that has similar white points (L*100, a*+/-1~3, b*+/-1~3), coating (uncoated, silk, gloss) and surface characteristics. Then create PaperCatalog entries for every stock using order numbers and descriptions to facilitate selection, and assign the output profile to the stock. When submitting a job use the paper Catalog pull down in Job Properties (or driver if BiDi is setup) to select the stock. Depending on the print engine you can assign paper catalog entries to each tray. If the print engine has a supported SLM (stock library manager) this function is done using the tools the engine provides.

DragonLeaves
08-21-2017, 01:25 PM
Variables affecting color are: toner age (you've been running 10K of black only for example), drums level, fuser age, temperature, humidity, machine dead cold or has been running, same machine or not?, calibration set used, paper surface texture, white point and gsm.

Always profile when the machine has been running a while, has fresh drums and good hardware status overwall, ensure humidity is at least 30% and temperature in 20-35 celsius max.

Variations over time will be seen based on your drums level mostly and humidity/temperature.

You shouldn't need a lot of profiles and calibration sets. With Xerox Versants, one 24lbs plain and one 80lbs gloss text is usually enough for each type. Maybe add a custom calibration set/profile for a cougar 350 but that's about it.

adam1991
08-21-2017, 04:08 PM
You shouldn't need a lot of profiles and calibration sets. With Xerox Versants, one 24lbs plain and one 80lbs gloss text is usually enough for each type. Maybe add a custom calibration set/profile for a cougar 350 but that's about it.

Disagree.

At least, as a blanket statement.

That may be fine for the work your shop puts out, but other shops need better. Fortunately, it's easy to provide better.

DragonLeaves
08-21-2017, 04:20 PM
To be fair,

There's a tiny fraction of shops that actually profile anything. Of those, again a fraction actually profile lots of stocks. I know this from years of experience dealing with shops filled with production grade colour devices. On Xerox, the fact that you can tweak the transfer on a per stock basis, and that you don't have inboard/outboard gradients, or lack of consistency between 1st sheet and nth sheet, etc. all these explain partly, partly, why I didn't see a lot of the shops I deal with profile lots of stocks.

That said, it makes those who profile that much easier to stand out !


Disagree.

At least, as a blanket statement.

That may be fine for the work your shop puts out, but other shops need better. Fortunately, it's easy to provide better.