How to handle the G7 compensation curve with scheduling of production

[David Milisock]

On the other hand...I've not been to a shop yet that's using Postscript color management to drive their platesetter. Frankly, I thought PS color management died a long time ago and was effectively replaced by tagged PDF workflows. Different strokes I guess.

"...through a series of calibrations and internal software prosesses [sic]..." ?
Sounds good....but doesn't give me much to go on.

And, yes, I have an ICC mindset....it's my job.

I guess at this point we're just talking past each other David......but I would like to hear the answer to a question that's been asked of you a couple of times without an answer that I'm aware of: when you say "linearize a press", do you mean an original 50% value in the file prints and measures as a 50% dot on paper? I would really like to get a direct answer to that question....a 50% dot printing as 50% on paper is an interesting concept that I would like to hear more about.

Terry

It's more complicated than that. Yes you create a test plate and yes you print the plate and adjust the software for what the 50% on the plate reads on the paper. However through this process the software also has had entered into it ink limts for each color as well as the curve for each test swatch. One product required that a test be done util the final destination on paper did equal not just 50% bu matvh as many of the swatchedas posible. This was uses as below.

This creates the color space of the press. The RIP/setter uses a theoritical CMYK 100 per channel LAB equal as the soruce to convert to the created CMYK for the presses LAB equal as destination. PS CM easy and simple.

Now if your process does notinterject itseld in the RIP/Setterprocess then it's just PS CM midified ith an ICC profile.

 

[gordo]

.but I would like to hear the answer to a question that's been asked of you a couple of times without an answer that I'm aware of: when you say "linearize a press", do you mean an original 50% value in the file prints and measures as a 50% dot on paper? I would really like to get a direct answer to that question....a 50% dot printing as 50% on paper is an interesting concept that I would like to hear more about.

Terry

David answered that in another thread:

gordo: So you start the process with a linear plate (50% in the file is 50% on the plate). And your target is to end up with 50% in the file resulting in 50% on the press sheet?

David Milisock: Yes as close as possible across the entire curve from 1% to 100%, [snip] I've never seen it work any other way with any plate setter.

best, gordo

 

[meddington]

If Fraser, Bunting and Murphy didn't get it, than I have no chance. I just have to say that what I've seen described so far seems unecessarily complex. Printing to a standard is not that hard.

 

[gordo]

If Fraser, Bunting and Murphy didn't get it, than I have no chance. I just have to say that what I've seen described so far seems unecessarily complex. Printing to a standard is not that hard.

And as you well know, printing to a standard has other benefits to both print buyers and their suppliers. That being said, CtP and digital proofing do allow the print shop to offer more than one print characteristic - if that provides value to the project at hand. Unfortunately, linear presswork, at least the samples that I've seen, doesn't provide any value.

best, gordo

 

[Bloodsaler]

Sorry I`m confused by all of the replies,some gentleman may deviate what I want to ask~
I mean how to balance the compensation curve and the production schedule,if the presses not in a same condition.
I know,make the presses lineared will help.,but it need time and money.So if I truely have four completely different presses at present,how should I compensate them and make it work well,not conflict with production schedule~
Now the way I choose,I compensate them neither,but I notice the data of them everyday,but this is not a good way I know~

 

[meddington]

Hi Blood,

I mean how to balance the compensation curve and the production schedule,if the presses not in a same condition.

I think this was addressed in the first few replies to this thread.

I know,make the presses lineared will help.

Actually, it won't help at all. Inherent differences in your presses mean you will have to, one way or another, customize data sent to that press. Once plated, you're committed to a particular press. Forcing the press into any state, linear or otherwise is customization. Using linear, or uncompensated plates will allow plates to be interchanged between presses, but doesn't address the presses inherent differences. Only other option is to minimize the differences between the presses by homogenizing consumables, and process control. With different makes of presses, this could be challenging.

 

[meddington]

And as you well know, printing to a standard has other benefits to both print buyers and their suppliers. That being said, CtP and digital proofing do allow the print shop to offer more than one print characteristic - if that provides value to the project at hand. Unfortunately, linear presswork, at least the samples that I've seen, doesn't provide any value.

best, gordo

Sorry, poor choice of wording. Press calibration, while challenging, should not be this difficult to comprehend.

 

[arkay_desai]

hi guys, it can be make possible by preparing the plate "PLATES ON DEMAND"? Means lining up the jobs. preparing the next job's plate after finalizing the machine. We have the same process since pre-press and presses are running round the clock. We are preparing the plates for next 1 or 2 jobs for particular machine. So that we can maintain the consistency and save the plate cost and time

 

[David Milisock]

hi guys, it can be make possible by preparing the plate "PLATES ON DEMAND"? Means lining up the jobs. preparing the next job's plate after finalizing the machine. We have the same process since pre-press and presses are running round the clock. We are preparing the plates for next 1 or 2 jobs for particular machine. So that we can maintain the consistency and save the plate cost and time

Sounds great. Do your pressmen actually call up the plates to be made or is there still a dedicated plate setter operator?

 

[David Milisock]

If Fraser, Bunting and Murphy didn't get it, than I have no chance. I just have to say that what I've seen described so far seems unecessarily complex. Printing to a standard is not that hard.

Guys getting out your RIP and plate setter manual and linearizing and (for a lack of a better term) footprinting a press is PRINT SHOP 101.

This has been done since the beginning of the technology. It creates a PS color managed process that produces the maximum gamut for the press/paper combination. For those who like the ICC profile process, THE LINEARIZATION STILL HAS TO BE DONE. There is no getting around this process. If the ICC process is not interjected into the linearization software hardware process then all you're doing with an ICC process is modifying this process with another one. Why in the world one would want to limit the gamut of a press with a standard that reduces the gamut is beyond me.

It may if hardware/software allows it, be suplanted by an ICC controlled process but the linearization still must take place.

This is print shop 101.

 

[buckeye]

Oh boy.....here we go again.

 

[meddington]

Guys getting out your RIP and plate setter manual and linearizing and (for a lack of a better term) footprinting a press is PRINT SHOP 101.

This has been done since the beginning of the technology. It creates a PS color managed process that produces the maximum gamut for the press/paper combination. For those who like the ICC profile process, THE LINEARIZATION STILL HAS TO BE DONE. There is no getting around this process. If the ICC process is not interjected into the linearization software hardware process then all you're doing with an ICC process is modifying this process with another one. Why in the world one would want to limit the gamut of a press with a standard that reduces the gamut is beyond me.

It may if hardware/software allows it, be suplanted by an ICC controlled process but the linearization still must take place.

This is print shop 101.

Ok, I've got my manual (Harlequin Workflow):

"Linearization is a special case of calibration, commonly used when producing output on an imagesetter.
Linearization is the process of adjusting values on output so that the result is proportional to the values
requested, in some suitable measuring system. "

would be used for film or plate setter...

"It is an established industry practice to accept non-linearity in the transfer from film to printing press.
Frequently, whoever performs the scanning or layout design uses their judgement to introduce a compensatory
adjustment in the scanner settings or choice of tints, basing the compensation on the kind of
printing press to be used.
Calibration is still a valuable process in this scheme but the purpose is to produce a known and repeatable
non-linear transfer."

in other words...why bother to linearize the press.

Maybe the confusion is that your using the term "linearization" and we're interpreting that to mean a literal linearization of the press's tone response...forcing printed output dot percent to equal file input dot percent (which you've apparently confirmed that this is your intent, but I'm still not sure). For the record, this does NOT have to be done, and I can't imagine why you would want to. Measuring the tone response of the press, i.e.fingerprinting, calibration, etc, is necessary, yes. Maybe its semantics we're arguing about, but I can't follow your train of thought.

 

[David Milisock]

Ok, I've got my manual (Harlequin Workflow):

"Linearization is a special case of calibration, commonly used when producing output on an imagesetter.
Linearization is the process of adjusting values on output so that the result is proportional to the values
requested, in some suitable measuring system. "

would be used for film or plate setter...

"It is an established industry practice to accept non-linearity in the transfer from film to printing press.
Frequently, whoever performs the scanning or layout design uses their judgement to introduce a compensatory
adjustment in the scanner settings or choice of tints, basing the compensation on the kind of
printing press to be used.
Calibration is still a valuable process in this scheme but the purpose is to produce a known and repeatable
non-linear transfer."

in other words...why bother to linearize the press.

Maybe the confusion is that your using the term "linearization" and we're interpreting that to mean a literal linearization of the press's tone response...forcing printed output dot percent to equal file input dot percent (which you've apparently confirmed that this is your intent, but I'm still not sure). For the record, this does NOT have to be done, and I can't imagine why you would want to. Measuring the tone response of the press, i.e.fingerprinting, calibration, etc, is necessary, yes. Maybe its semantics we're arguing about, but I can't follow your train of thought.

Who uses film? I haven't seen any film in over 5 years, the concept of linarization for film has little bearing on CTP due to the inherent issues with plating.

Follow this, a linearized process, A in = A out. Color Standard in = Color Standard Out

Applications (Adobe What ever) uses a gracol proifle and creates a PDF or EPS that is gracold CMYK, it passes through a linear process to be gracol on paper.

I don't use gracol but all I do is color manage in the application, build the color for the destination and pass it through a linear process with exceptional repeatability. Press wear and pressman error is the only real issue.

Presses are easy, ink jets or UV flat beds with 50 different media are tough.

 

[David Milisock]

Oh boy.....here we go again.

Hey I like these guys they're sharp. Of course they haven't explained how we managed to do it right before ICC profiles came along but IMO they're A Ok. Some of us were laying ink down 20 years before the ICC was a gleam in someones eye.

With that said there's a great need for building media profiles and for using good application based color management practices. I'm just not sold on the need for an ICC profile for a press, certainly the color I see out there in print has not gotten any better since the ICC implementation.

 

[gordo]

Hey I like these guys they're sharp. Of course they haven't explained how we managed to do it right before ICC profiles came along but IMO they're A Ok. Some of us were laying ink down 20 years before the ICC was a gleam in someones eye.

With that said there's a great need for building media profiles and for using good application based color management practices. I'm just not sold on the need for an ICC profile for a press, certainly the color I see out there in print has not gotten any better since the ICC implementation.

Since I'm probably older than you and I've been in the prepress/print business for over 40 years, I can tell you how it was done.

I'll just go back to the late 80s and 90s - rather than the 70s and earlier.

Basically....

Linear film was output from a film image setter. Linear film was the standard file interchange method back then.
The same linear film was used for both plating and proofing.
The plates exposed from film were no longer linear - but in those days the plates weren't directly measured.
The film was exposed to proofing media which was designed to simulate the tone/dot gain response of a press - typically heatset publication web - SWOP. The color of the proof was decided by the proofing media manufacturer. There was Matchprint color and Fuji color and Chromalin color. The color the proofs represented was similar, but not the same. It was, according to SWOP, the printer's responsibility - by whatever means required - to "match" the supplied proof that had been made according to the proofing manufacturer's specification.

Color was determined by the proofing media.

With desktop publishing one could create color documents, even though one was working with a black and white or greyscale computer display by specifying CMYK screen tint percentages.

Along comes CtP.

That broke the film link between plate and proof. They are now independent.

The plate is no longer required to be linearized since no one will see the plate. What was/is required is that the plate be consistently imaged, and that the tone values on the plate will generate the required tone response on press at the appropriate solid ink densities.

At first, the files were exposed on the proofing media using the same device as the device used to image plates (e.g. Creo Spectrum proofer). Again the color was determined by the proofing media vendor. The proofs were imaged using different curves from those used to image the plate in order to deliver a proof that the press would align to.

As inkjet proofing became more viable it, for the most part, replaced halftone proofing.
In order to have the inkjet proofer represent color correctly, ICC profiles, must be used to color manage the out put of the ink jet. The profiles can be shop/press specific, or industry specified.

ICC profiles are also used to permit the computer display to represent the anticipated press or proof color which allows for earlier and less expensive color proofing.

End of history.

Linearizing plates in a CtP workflow is redundant and in the vast majority of cases offers no benefit. Linearizing the press (50% in file = 50% on the press sheet), means that the printer is no longer printing to an industry standard or specification. Which is fine if that print characteristic is appropriate for the project or print buyer. But it is certainly unusual. Printing linear is not the norm.

best, gordo

 

[meddington]

Who uses film? *I haven't seen any film in over 5 years, the concept of linarization for film has little bearing on CTP due to the inherent issues with plating.

I agree...linearization isn't necessary for ctp...or for press calibration.

Follow this, a linearized process, A in = A out. * Color Standard in = Color Standard Out

What you're suggesting it color standard in = linearized out.

Applications (Adobe What ever) uses a gracol proifle and creates a PDF or EPS that is gracold CMYK, it passes through a linear process to be gracol on paper

A gracol ICC profile has TVI of around 17% built in...you realize that right? A 50% dot is intended to gain. You'd want to align your press, via compensation curves, with similar gains, NOT linear output.

All this done without any ICC profile for the destination press, just one dimensional tone curves.

Can I ask what TVI you are measuring on the press sheet after all this?*

 

[tmiller_iluvprinting]

I tire of this

I tire of this

Why in the world one would want to limit the gamut of a press with a standard that reduces the gamut is beyond me.

Dave,
Linearizing press output will clip the press gamut substantially more than proper press calibration and printing to a standard. I am sure people thought Einstein was an idiot at first too. You may be a genius, maybe it's not what you are saying, but how you are saying it. You haven't sold me, and I kind of feel sorry for those that buy what you're selling.
Regards,
Todd

 

[TerryWyse]

Linearizing plates in a CtP workflow is redundant and in the vast majority of cases offers no benefit. Linearizing the press (50% in file = 50% on the press sheet), means that the printer is no longer printing to an industry standard or specification. Which is fine if that print characteristic is appropriate for the project or print buyer. But it is certainly unusual. Printing linear is not the norm.

You're being too kind gordo .....linearizing at the *press sheet* is just plain wrong-headed thinking if what is meant by linearization is that a 50% value in the file prints as 50% on the press sheet as measured with a densitometer set for Murray-Davies dot area (I'm feeling like I can't assume ANYTHING in this discussion so I'll state what should be obvious or understood). It's wrong-headed thinking because, like you say, it leaves the press in a state that is UNLIKE any known or "standard" printing condition....but it's also wrong because it puts the press in a very NON-linear state with respect to colorimetry. I dare say that if you were to plot this sort of "linearity" on 2D or 3D Lab or Lch gamut plot, you'd see extremely non-linear tonal distribution of the respective ink channels....not good!

I guess we also have to be careful about the term "linearization" itself. Unless the metric used for linearization is clearly stated (we assume 50%=50% is "linear"), the term itself is ambiguous at best. In my work with inkjet and press calibration, I use different linearization methods.....in the case of a press, I generally use a method called "G7"....for an inkjet device, I've cooked up my own "linearization" method that is all about achieving 'linearity" of the primary/secondary ink channels in a 3D Lab space. So there's dot area linearity, density linearity, L* linearity, "delta e" linearity....you get the idea.

Back to the original point......linearizing a press sheet is simply a Bad Idea on so many fronts that I can't fathom why one would go that route....UNLESS the end game was to first establish linearity and then build back in a controlled amount of dot gain for a specific target printing condition....but as far as I know, that's not what's being proposed here...it certainly sounds to me like the press is calibrated to be linear at the press sheet and then LEFT in that state.

Terry

 

[TerryWyse]

With that said there's a great need for building media profiles and for using good application based color management practices. I'm just not sold on the need for an ICC profile for a press, certainly the color I see out there in print has not gotten any better since the ICC implementation.

I think you're putting words in people's mouth David.....I haven't heard anybody in this discussion and others explicitly advocating using ICC profiles for *pressroom* color management (although it DOES have it's proponents and benefits). I think what most of us are trying to understand is what your "vision" of color management, calibration, press control, et al, is all about. So far, for me at least, it seems to be cloaked behind obfuscation and, dare I say, poor grammar to the point that it's getting in the way of your message.

You seem to be STRONGLY advocating Postscript color management. I would dare say this concept has been mostly rejected everywhere except outside of your consulting bubble. If it works for you, so be it. But I think ICC-based color management has proven to be at least as robust and much more flexible than PS-based color management. ICC CM can be applied in any number of workflow scenarios be it PDF, raster-based, etc.....IOW, it can be implemented both inside and outside of Postrscript environments...I would dare say that pure Postscript workflows are slowly on the way out and are being replaced by native PDF and proprietary raster workflows.

I'd love to sit down and debate point-by-point many of your other statements in this and other posts but frankly I've already burned too much bandwidth and need to move on to paying work.

Good Luck, Good Night and may God Bless...

tw

 

[Rick@IPCO]

The answer to the actual original question is easy... for each job, just make a set of plates for every press. Extremely expensive and wasteful, but easy. Maybe you'd waste enough plates and money that they'd eventually realize that it's not a plating issue, it's a scheduling issue.