Curve adjustments

[Prepper]

We have always took the press time to print a set of targets, cut them up and measure, drop into Curve2 and then apply adjustments in the rip. The targets are printed by arranging P2P targets, 4 across a 25" sheet, flipping them top to bottom, and we have large take-off bars across the gripper and tail of 50% of each color CMYK to help even out and distribute the coverage.

I see there's software out there now that will let you take measurements from live jobs and average as many of those as you like, export curve adjustments from that data and import those directly into your plate rip.

We've tried it and it works.

My question is, do you think we get more accurate curves from targets arranged the same way each time with the same coverage on the sheet than averaging measurements from live jobs with varying degrees of coverage on them? Some may be heavy, some may be light, and what about where the patches you're measuring are in relationship to the coverage in line with them? Would the variation not be enough to worry about for adjusting curves?

This same software has pointed out the TVI variations that show up during our press runs, differences from light to heavy coverage, from morning to afternoon, from Thursday night to Monday morning, and started me asking this question, whether live jobs are a good average condition to update curves with or a standard targets sheet that's printing the same every time?

I believe the press is stable, but inconsistent thru a degree of variation due to physical characteristics of an offset press, but the coverage varies all the time also, from run to run, which has quite an effect on TVI from what I'm seeing here.

Thanks for any input

 

[chevalier]

There is no magic bullet to average everything else and get one set of curves to rule them all. Even if your press is completely laboratory grade vacuum sealed stable ink distribution across the press sheet is going to be the limiting factor. You can have many sets of curves and use the curve set that best reflects optimal production for that specific ink distribution for the press sheet. You'll have to do that manually and intuitively. It should absolutely be worth while if you have jobs that repeat over and over again with similar ink distribution.

Alternatively you can invest in software and hardware that can do it on a per job basis. There is at least one system out there designed for taking this on per job. Heidelberg Prinect can do this with a proper configuration, image control and proper marks on the press sheet. You plate a job get "up to color" scan the sheet(s) and then export the TVI data via the scanner to Prinect which then export curves and generate new plates that are matching to current press conditions and that specific press sheet ink distributions. It more or less can do this at the "press of a button" from the press console.

There is considerable overhead for all of this in software, hardware, training and management. You've also got multiple sets of plates per job, extended make-readies, etc. but if your run lengths are long enough I can see this system being worth the time, money and effort.

 

[Prepper]

and that the job we are printing is representative of the common printing we do.

How do you do that?

Ultimately, despite having the "more powerful software", we have returned to the old method. And it is far more stable we find.

By that do you mean you're not just chasing your tail then?

 

[Prepper]

BeauchampT:

Ruminated on this for a while and....surely you jest? Custom make curves for every run after its put on press and measured? Who could afford the time and effort to do that?
But alas, that is what our pressman desires here, for me to adjust curves run to run to fix the color for him, but at the same time he will be adjusting the color by eye to make it match the proof closer or look better to him. Anyone know an easy way to explain how and why that doesn't work, please help, I've tried for 10 years and this is still where we are.

 

[gordo]

But alas, that is what our pressman desires here, for me to adjust curves run to run to fix the color for him, but at the same time he will be adjusting the color by eye to make it match the proof closer or look better to him. Anyone know an easy way to explain how and why that doesn't work, please help, I've tried for 10 years and this is still where we are.

But it's been working for ten years! LOL

Seriously, a press is not a proofer. And their mechanical difference and function is critical to understand.

Your press operator should be running to shop established SIDs - and at that point the presswork should align to the proof, but will likely still require adjusting for optimum "match". And he will eyeball it to make the alignment closer.

That is standard procedure.

If there is any consistency in the way he has to run the press (e.g. the Magenta SIDs always have to be brought down) then there is an issue, or issues, (with many potential causes) that needs to be dealt with.

 

[chevalier]

Anyone know an easy way to explain how and why that doesn't work, please help, I've tried for 10 years and this is still where we are.

Because it costs too much! See my comment above about the Heidelberg Prinect methodology.

 

[Prepper]

Gordo, would like to clarify my earlier post a little, when I said "explain why that doesn't work" I was referring to changing curves from run to run, not the pressman adjusting, I realize he does have to adjust some from run to run somewhat.

I realize more than ever now that this is really just aiming at an average condition, when there really isn't any such thing and what is achievable is probably a lot broader target than I've been hoping for all along. That being said I do believe that color management and curves are narrowing down the range of what we can hit versus just saying, "here's a proof, run to some densities and adjust it to match best you can," and never aiming for something in the first place, like target densities based on Lab color, and curves to gray balance. I think that is getting him in the ballpark way earlier in the makeready and yes, each run may still require SID tweaks to dial it in closer to the proof if possible. However, at that point you are then adjusting by that specific pressman's eye to a proof, which can be held to a closer tolerance than a press sheet but is still a proof that's within a certain set of tolerances also.

My belief is that just running to the numbers, SIDs and gray balance, whether it's a perfect match to the proof or not, is more consistent and repeatable in the future than trying to reprint something 2 or 3 or 5 years from now by memory and how did I run it last time and is this proof even exactly the same as last time?

Take into consideration we are basically an in-plant printing our own work, so no customers demanding any certain thing, and we do a lot of reprinting book covers which have had the same design for the past 15 years now, which are a set and they go on our end-users shelves as such, and are displayed on shelves in our distribution libraries around the world, together, from different years and different printings, different inks and papers.

Thanks

 

[rich apollo]

A proof is a prediction of what will happen on press. If the press is performing in an inconsistent fashion, then you cannot predict what will happen.

Ask the pressman if he is offering to pay for the first set of plates, the second set, or both. Money wasted on double plating and unnecessary time is money that no one will see in his paycheck. This guy is trying to spend any and all future pay increases for everyone. That would upset me.

There will be variation in offset printing, but not so much that the proof cannot be matched acceptably. It seems like the system you describe is not performing as it should.

 

[Prepper]

A proof is a prediction of what will happen on press. If the press is performing in an inconsistent fashion, then you cannot predict what will happen.

Ask the pressman if he is offering to pay for the first set of plates, the second set, or both. Money wasted on double plating and unnecessary time is money that no one will see in his paycheck.

There will be variation in offset printing, but not so much that the proof cannot be matched acceptably.

Right, and "acceptably" here becomes the issue I think. Acceptable within a standard's tolerance or according to someone's eyes? Different approach trying to achieve the same result, closely simulate what the proof is predicting the press result will be. Our problem is, from my perspective anyway, is the pressman says, according to his 30 years experience that he can tell if he went to target densities and\or gray balance that it wouldn't match the proof anyway and so he adjusts it by eye first and never gives it a chance to work. That's been our battle, give me one good sheet, to the numbers, lets evaluate it over some time and see how well this process works. So far, we haven't been able to get that from him, he says he knows it won't most of the time, BUT at the same time wants our company to spend several thousand dollars on software, you know what I'm referring to Rich, so I can adjust the curves for him from run to run. We can't get him to understand that you really can't adjust that often or that if he isn't going to adjust to gray balance that the whole point of curves, made according to G7 with Curve software, to work at gray balance, is negated completely anyway.

Again, trying to achieve the same thing, but different approaches. It's like we speak different languages but we're both trying to get to the same point. I don't know how you do that with someone that's totally unwilling to budge.

 

[chevalier]

it sounds like rather than use common sense, your operator would prefer to use nonsense and try to push the blame around.

I totally agree.

 

[gordo]

Sorry, I can't let this go:

A proof is a prediction of what will happen on press.

With all due respect, and although this is commonly stated in the trade, the function of a proof is NOT to predict what will happen on press. The function of a proof is, color-wise, a target for the presswork to align to.

A proofer and a press are mechanically and functionally purposed too differently for a proof to ever be a predictor of what will happen on press. The only time a proof can be a predictor of what will happen on press is when the proofer is the same device that will print the actual final job.

 

[Prepper]

it sounds like rather than use common sense, your operator would prefer to use nonsense and try to push the blame around.

Not exactly, he's not blaming anyone, he's saying he's good, he's done it for 30 years, he can tell if it is or isn't going to match if he went to the target SIDs and gray balance, before it ever gets there, and takes a shortcut and adjusts it accordingly and that by doing that he's saving the company money. That he knows how the boss wants it to look and that's what he's doing. And it doesn't matter to him if we think he can't really tell whether it would be right or not, he knows it wouldn't be.

He just told us this morning, when we asked him to just give us one good sheet to the numbers from every run, then do whatever he wants from there, so we can evaluate the process, that he would do that and that he would write his predictions down of whether it would work or not beforehand and what he'd have to do to get it right, and that he would be right on all of them and then we could see that for ourselves.

WOW! That sounds even more ridiculous to me now, repeating it to other folks.

 

[chevalier]

Prepper,

This is the part while you laugh and cry simultaneously.

 

[gordo]

Not exactly, he's not blaming anyone, he's saying he's good, he's done it for 30 years, he can tell if it is or isn't going to match if he went to the target SIDs and gray balance, before it ever gets there, and takes a shortcut and adjusts it accordingly and that by doing that he's saving the company money. That he knows how the boss wants it to look and that's what he's doing.

I assume that you're not using ink key setting preset software. Could it be that he's talking about eyeballing the plate and doing the ink key settings according to what he sees on the plate (and possibly relating that to the proof)? What you may have is a simple failure to communicate.

Rather than asking him to give you "one good sheet to the numbers from every run" (which is a threatening request) take one example sheet (that the press guy has OK'd as the "match") of every job printed for a week. Then measure the SIDs (and possibly the over print traps) to see if there are any consistencies in what he is doing on press (don't worry about grey balance as a metric in the press sheets). Relate those sheets to the proofs to see if there is any relationship with those consistencies to the live image area of the proof.

What that does is position you as someone that is interested in learning from him rather than dictating or threatening. By putting him in the position of teaching and/or mentoring you through your questions you give him an opportunity to learn why you think the way you think about the process.

Or you could just fire him and try a different press operator.

 

[Deleted member 16349]

Sorry, I can't let this go:



With all due respect, and although this is commonly stated in the trade, the function of a proof is NOT to predict what will happen on press. The function of a proof is, color-wise, a target for the presswork to align to.

I hope the intension is for both the proof and the press to reproduce what the original colour target set by the customer.

 

[gordo]

I hope the intension is for both the proof and the press to reproduce what the original colour target set by the customer.

I guess that depends on what you mean by "the original colour target set by the customer".

 

[gordo]

Question, which one do you believe to be closest to the truth:

A) Digital CMYK and Pantone files are best represented by a digital proof.
B) Digital CMYK and Pantone files are best represented by an offset press sheet.

And do you think that since digital proofers are only imitating almost all Pantone colors, maybe one produces CMYK better and the other one produces Pantone better?

I don't quite understand what you mean.
By "Digital CMYK" do you mean screen tint percentages as, for example, specified in a page layout application? Do you mean a CMYK image placed, for example, in a page layout application?
By "Pantone files" do you mean a Pantone spot color as, for example, specified in a page layout application? Intended to print as a spot color?

When you say "And do you think that since digital proofers are only imitating almost all Pantone colors, maybe one produces CMYK better and the other one produces Pantone better?" From my experience RIP/inkjet printer combinations do vary in their ability to simulate CMYK and/or spot Pantone colors. Especially if the Pantone ink will be a spot color that will be halftone screened on press, and doubly especially if the spot Pantone is intended to overprint CMYK or another spot color on the final presswork. And that's not even considering the effects of another wrench in the monkey-works - the impact of the difference in the amount of optical brightening agents in the proofing paper and the press paper, nor the color effects of coatings and laminations on spot colors.

 

[Deleted member 16349]

I guess that depends on what you mean by "the original colour target set by the customer".

Maybe there is not a suitable answer to your question.

If a mechanical drawing is made for a part, that drawing gives all the information needed for someone on the other side of the planet, to know how the part should finally be. No need for interpretations of what a mm is. There normally is no need to describe how the part should be made unless there are some specific treatments. The actual manufacturing of the part is up to the people who manufacture it. As long as the part meets the specifications in the drawing, then it is good.

The drawing is the main document. If a manufacturer can not make the part to the drawing for some reason, he can say so right up front.

I don't see this kind of clear description of what is wanted in a print, in the printing industry. It can be done in that way but it has not evolved to be that way.

So when I say the "original colour target set by the customer", I am basically talking about the file of his image that he wants printed. I suspect that these files are not so clear in defining how the image should look.

If the files are separations, that makes no sense to me. Separations are a form of tooling in a manufacturing context. Customers virtually never specify tooling to a manufacturer. It is the responsibility of the manufacturer to determine the tooling that results in the part being made to spec.

A proof is like a prototype. A prototype for a part must match the drawing. It seems in the printing industry, the proof is not meant to match what the customer's file but is something slightly different. No wonder there is so much confusion.

My head hurts.

 

[gordo]

Maybe there is not a suitable answer to your question.

If a mechanical drawing is made for a part, that drawing gives all the information needed for someone on the other side of the planet, to know how the part should finally be. No need for interpretations of what a mm is. There normally is no need to describe how the part should be made unless there are some specific treatments. The actual manufacturing of the part is up to the people who manufacture it. As long as the part meets the specifications in the drawing, then it is good.

The drawing is the main document. If a manufacturer can not make the part to the drawing for some reason, he can say so right up front.

I don't see this kind of clear description of what is wanted in a print, in the printing industry. It can be done in that way but it has not evolved to be that way.

So when I say the "original colour target set by the customer", I am basically talking about the file of his image that he wants printed. I suspect that these files are not so clear in defining how the image should look.

If the files are separations, that makes no sense to me. Separations are a form of tooling in a manufacturing context. Customers virtually never specify tooling to a manufacturer. It is the responsibility of the manufacturer to determine the tooling that results in the part being made to spec.

A proof is like a prototype. A prototype for a part must match the drawing. It seems in the printing industry, the proof is not meant to match what the customer's file but is something slightly different. No wonder there is so much confusion.

My head hurts.

That makes perfect sense since, of course, the printing industry is a service industry - not a manufacturing industry.
(Thomson Reuters Business Classification: Industry 52203020 Commercial Printing Services, Industry group 522030 Commercial Services & Supplies, Business sector 5220 Industrial Services).
There is a small manufacturing component but print buyers aren't particularly interested in that part of the process - too dang confusing.

 

[Deleted member 16349]

That makes perfect sense since, of course, the printing industry is a service industry - not a manufacturing industry.
(Thomson Reuters Business Classification: Industry 52203020 Commercial Printing Services, Industry group 522030 Commercial Services & Supplies, Business sector 5220 Industrial Services).
There is a small manufacturing component but print buyers aren't particularly interested in that part of the process - too dang confusing.

Service industry. Oh. So it is more like ordering food in a fast food shop and not sure exactly what you are going to get and whether someone did something gross to your food order. OK, I got it now.