Correct way to perform calibrations? (Xerox V180 + EFI Fiery + CWS)

[donster]

Hi, we recently got our V180 and since I did calibrations for the papers that we use, it seems that the printer has smaller dynamic range (whites and blacks get clipped very soon). Could you please take a look at my procedure and confirm if I'm doing it right? And if so, what is the cause of worse results with own calibrations?

Let's say I want to create a profile for 300gsm uncoated paper (Xerox Colotech+ Uncoated)

1. In CWS I go to Device Center -> Tools -> Calibrate and highlight the calibration setting, that is closest to the paper I want to use (in this case "Uncoated 220gsm") and click on "Create New..." (see Image 1). Here are my first doubts: the default calibration set that I am starting from (Uncoated 220gsm) has no calibration performed (Default measurements). Is this right? Or should I first make a calibration for it? If so, on what kind of paper?

2. I will call my calibration set for example "UNCOATED 300 POKUS", add a note about what kind of paper it is and go into properties. It allows me to change everything, which is kind of strange, but I will change only paper size to SRA3, 257-300gsm uncoated, 3 copies and change Printer Screen Mode to "200 dot" (because that is the settings we use on all our jobs).

3. On the next screen I will confirm that I want to print calibration strips. I will take the 3rd copy, put a blank paper below it and perform the calibration using ES-1000. When all this is done, I get a window asking me to associate this calibration set with an output profile. Well I'm not really sure about this step, but I select "Fiery Xerox VP180 Colotech Uncoated 220gsm v1F"...and it takes that output profile, combines it with my calibration and creates a new output profile called "UNCOATED 300 POKUS".

So now, if I take the paper (Xerox Colotech 300gsm Uncoated) and print something using output profile UNCOATED 300 POKUS, I should get the best possible result, right? Well as I mentioned in the beginning - the result is not that good. Brightness is OK, color balance as well, but the dynamic range is poor. As if the profile couldn't print anything between ~0-10% of gray and between ~90-100%. There are no details in highlights nor shadows.

Am I doing something wrong?

 

[DYP]

You need to make a new output profile for your paper on top on the calibration.

 

[donster]

That's what I did in step 3...

 

[DYP]

That's what I did in step 3...

All you said was you associated it with already existing output profile. I don't see anything about printing a profile target and creating a new profile.

Also why are you using an old ES-1000? Are you sure the old ES-1000 is any good? I would have though with a new V180 you would have received a brand new ES-2000.

Did the V180 come with an In Line Spectro?

 

[donster]

Oh so you meant I should create my own output profile by measuring those hundreds/thosuands of small patches? Well, I can't do that, because I have no software that would allow me to create my own output profiles. My boss didn't want to spend more money on the ES-2000 + Color Profiler Suite, as he thinks the ES-1000 is good enough. Not sure what In Line Spectro is? But my guess is that we don't have it, since I haven't heard about it.

But from what we've been told by number of Xerox people, output profiles are not that extra important and more or less they've recommended us to stick to the default profiles and just keep doing the calibrations (linearizations). Especially since the default profiles are made for the same paper that we use (Colotech coated/uncoated).

But I have just found this PDF: FS200 How to: Calibrate printer with EFI ES-2000 spectrophotomer and after reading it - I think I can say that my procedure is correct.

So now I don't really know where could be the problem with the small dynamic range?

 

[pippip]

I'm following this thread closely to, never feel i have the correct grasp on all this.

The InLine Spectrometer is where the machine prints about 22pages of patches and reads/calibrates itself. When you do the calibration the last dropdown menu asks what measurement method to use, see if ILS is there?

 

[donster]

Nah, only ES-1000 or ColorCal to choose from.

I just did the calibrations again and am attaching the measurements images - if that is any help for someone? I don't really know how to read into them, but I guess that if the thin line (measured) is BELOW the target line - that is actually a problem, isn't it?

 

[Correct Color]

When all this is done, I get a window asking me to associate this calibration set with an output profile. Well I'm not really sure about this step, but I select "Fiery Xerox VP180 Colotech Uncoated 220gsm v1F"...and it takes that output profile, combines it with my calibration and creates a new output profile called "UNCOATED 300 POKUS".

Well, what it sounds like is happening is that you're doing a calibration perhaps in a wrong section or with a technique for making an entirely new media profile.

If all you're doing is a "recalibration" then you should be working within a particular media profile and you should not be able to change any settings or to assign an ICC profile at the end. All ICC profiles were built with a machine reproducing color in some state, and if you alter the machine state, they're no longer valid. It sounds to me like you're assigning an ICC that has some different characteristics than what you built into your "new profile" and that's creating your issue.

But from what we've been told by number of Xerox people, output profiles are not that extra important and more or less they've recommended us to stick to the default profiles and just keep doing the calibrations (linearizations).

There is lots of really, really, really bad color management information available out there from all sorts of supposedly well-informed people.

Oh so you meant I should create my own output profile by measuring those hundreds/thosuands of small patches? Well, I can't do that, because I have no software that would allow me to create my own output profiles.

Yes, that's what's needed if you're serious about color. Of course, buying tools doesn't buy the knowledge to use them. And both are required in equal measure.



Mike Adams
Correct Color

 

[donster]

Well, what it sounds like is happening is that you're doing a calibration perhaps in a wrong section or with a technique for making an entirely new media profile.

Well I am doing what we've been taught to do. And also the same process is described in the "How to" PDF which I mentioned earlier, here is the link:
https://www.google.com/url?sa=t&rct...00_en_us.pdf&usg=AOvVaw3ZotrFr97zug00nzdWPeWK

I don't think I am creating any new output profiles itself...The problem with doing calibrations for the default output profiles is that these calibrations have predefined Printer Screen Mode as "Enhanced text" and we've been told to use 200 Cluster dot for our jobs - meaning we also have to calibrate with these settings. Meaning we have to create our own calibration sets. And when you do that, the last step is to associate it with output profile (as described in 3rd paragraph in my first post).

 

[Correct Color]

Note that the term "profile" in this industry has somewhat of dual meanings. It can either mean a media profile, or it can mean an ICC profile.

A media profile defines a device printing in a certain state. That state will typically include media white point, single channel chroma values and ink limits, linearization (calibration) and multi-channel ink limits. An ICC profile is then a characterization of the printer printing in that state.

If you create a new machine state -- which it appears you are doing -- you cannot characterize it with an existing ICC profile -- which it also appears you are trying to do. You must create a new ICC profile to characterize the machine state you have just made.

Attaching some already existing ICC profile made for some other machine state will usually cause just the problems you describe.

I don't think I am creating any new output profiles itself...The problem with doing calibrations for the default output profiles is that these calibrations have predefined Printer Screen Mode as "Enhanced text" and we've been told to use 200 Cluster dot for our jobs - meaning we also have to calibrate with these settings. Meaning we have to create our own calibration sets. And when you do that, the last step is to associate it with output profile (as described in 3rd paragraph in my first post).

It's quite possible that whoever told you what to do might be in error. It's also possible that I'm missing something here.

But from what I read of the little manual you provided, if you're doing the longer procedure outlined on the first four pages, you need to make an ICC profile to finish the process. Note the last instruction:

17. To learn more about creating custom output profiles, review the How-To Guide: Achieve Accurate and Consistent Color with Printer Profiles.

If all you want to do is a recalibration of an existing media, you should follow the "Perform job-based calibration" procedure.


Mike

 

[donster]

Thanks, I really appreciate the time you put into me.

The process I described in my first post, is of course when doing only the first calibration - or to be more exact - creating a new calibration settings. Which I need to do (if I'm not mistaken) since I want to use 200 cluster dot on all of my jobs - but the default calibration sets are made for Enhanced text. Further calibrations are made by either "Perform job based calibration" method or via Device Center -> Tools -> Calibrate (and just choose for which media).

And the problem is, you cannot create your own calibration settings without associating it with output profile. What I'm basically doing is taking one of the default output profiles that is most similar to my media, creating a copy of it and associating calibration settings to it. When you think about it - it is no different than if you took default output profile and did a calibration for it (altough, as described in previous paragarph - it would be calibrating for Enhanced text, even though the jobs would later use 200 cluster dot - which is wrong).

So I believe, that if I took the default output profile and did a calibration for it - it would give me pretty much the same result as described in first post - smaller dynamic range. I don't know, could there be a problem in the spectrophotometer maybe?

I am attaching 2 test prints - one is OKAYish and the other is really bad (see the artefacts on the nose and face of the black haired lady for example). Or the color targets at bottom right - of course this is hard to judge from a scan, but i think the gradients should be really smoother....

 

[DYP]

Without the tools necessary to create a new calibration and a new output profile your going to have to live with the results that you have. Your boss is going to have to suffer the consequences of that poor decision.

 

[donster]

Well that leaves us with only solution (if we rule out getting ES-2000 + Fiery Color Profiler Suite): do not touch the default profiles and do not perform ANY calibrations (since calibration only makes the results worse)...

 

[pippip]

When it was installed did you receive any training? Did they indicate the ES-1000 would be suitable?

 

[DYP]

Well that leaves us with only solution (if we rule out getting ES-2000 + Fiery Color Profiler Suite): do not touch the default profiles and do not perform ANY calibrations (since calibration only makes the results worse)...

How did you acquire the ES-1000? Maybe an old version of FCP that is is licensed for. May be able to create a new calibration with ColorWise and then use the PDF method to create a new output profile on top of that calibration.

 

[donster]

Yeah, we had a training and also was on a "workshop"...which is what leads me to believe that I am doing it correctly! They said ES-1000 would be OK for calibrations (meaning linearizations) but obviously not for output profile creations (which didn't seem like a big problem - as the default profiles do more or less match our papers).

I don't really know how did we aquire the ES-1000, it's been here longer than me (more than 10 years). I don't think FCP existed back then

EDIT: maybe the ES-1000 is simply "wrong"? And should be repaired/recalibrated? Any way to verify if it is performing well?

 

[Correct Color]

Dot pattern is definitely a profileable (ICC profileable) aspect of machine state. If you want to use a dot pattern for which you have no existing complete media profiles -- including ICC -- then you have to complete the process.

And the problem is, you cannot create your own calibration settings without associating it with output profile. What I'm basically doing is taking one of the default output profiles that is most similar to my media, creating a copy of it and associating calibration settings to it. When you think about it - it is no different than if you took default output profile and did a calibration for it...

Don't think; it can only hurt the ballclub."

--Crash Davis

You can think of it that way, and it might even appear to make sense, but if it worked, you wouldn't be having this issue.



Mike

 

[donster]

OK, sounds reasonable. So you think that in my situation, it would be better to stick to the default output profiles, default calibration sets (made for Enhanced text) and print all the jobs with Enhanced text instead? Plus do regular calibrations for the default calibration sets (so with Enhanced text selected)?

 

[wonderings]

OK, sounds reasonable. So you think that in my situation, it would be better to stick to the default output profiles, default calibration sets (made for Enhanced text) and print all the jobs with Enhanced text instead? Plus do regular calibrations for the default calibration sets (so with Enhanced text selected)?

We use the profile that was set for us with our V2100. I just do a calibration with that and go about my business. I cannot imagine having to calibrate for each stock I use. Our 2100 colour has been good and consistent. The only colour profiles I have are adjustments I make for pantones or spot colours.

I do calibrations every day when the first colour job goes through. I will sometimes do a new calibration if it is a brand new customer file and I have to make adjustments to pantone or spot colours to match something they have.

 

[donster]

So you are using only ONE output profile and only do calibrations for that?