TVI curves from print sample

Dario

Well-known member
Hello there!!
A client gave us his file and his printed proof.
The design is built on 6 custom inks (or spot colours) - see image.
They are not the usual CMYK.
Luckily both print and file have the same color bar - 100-80-40-20%.
The goal is to print 6 plates to let our pressman do the job without too much pain.
Do you think I can get out some TVI curves from here?
Some hints for a fellow?
Schermata 2019-08-06 alle 14.00.36.jpg
 
How was the proof printed? Was it printed the same way that you’ll print the actual job? What print process are you using - offset, flexo, digital, etc?
 
We print offset, the proof should be digital printed on the same paper we are using and it simulates screens (maybe GMG).
 
Viele
We print offset, the proof should be digital printed on the same paper we are using and it simulates screens (maybe GMG).

If the proof is not made with special Proof-Paper it is worth nothing: Inks also running (?) different on different papers as offset on coated or uncoated!

(You can mesuare - just for curiosity - the screened proof from the customer and an own proof which must be simulate screens also of course and compare ..)

Next problem/question: are the patches in the proof(s) really screened? So: are there really „white“ dots between the colored dots in the patches 20, 40 and 80?

If not: AFAIK there is no exactly method to determine an increase from Lab-values. I remember a method via the xyz-values and the MurrayDavies-formula. With this method for example the profile inspecting Programm „colorthink/chromix“ determine generated TVI-curves for showing in a graph: And this shows exactly the limit of this method: cmy is different to each other in every inspected profile, but should be exactly the same!

I would ask the customer which increase he has chosen in photoshop as working space for spotcolors, dot gain 15% or 20%, that would be more help ;-)

I would not let the customer believe, that his claim is a professional. I understand his proof more as a wish than an expected aim. But you can offer and charge him for the work which is necessary to be done for more reliable prediction: A real offset print on stock paper for evaluating the TVI‘s for information and/or profiling a separation with his spotcolours.

Good to know Gordo on on your side in this topic ;-)

Ulrich
 
Last edited:
Since the colorbar is screened I thought to measure TVI values on it and to use these values to make tailored TVI curves - one for every ink.
So...
1. printing the same colorbar using my printing machines.
2. measuring the colorbar to make TVI curves.
3. applying new curves and print the job.
 
I would think getting the LAB values of the solids in the print as close as possible to the proof would be more important than the tvi. How many screens are in the art?
 
I do not know the design/layout.
If all the colors are used separately, you can of course try that. Please let us know after.

(However, I would have no confidence in the reproducibility of such a construct that may be based on six different dot gains for the same printing condition, although the similarity of four of these colors to the scale colors is so striking that they seemed are also intended to provide a "four-color" separation of a color image and such footage may have been converted to "CMYK", which in turn was assumed by identical dot gains via a “classic” profile like ISOcoatedv2 for cmy(k) ... But these are just guesses ...)

If I understand you correctly, you consider seriously to adapt the much more elaborate process (offset) with much more parameters are to controll to the result of a less elaborate process (digital proof).

That changes the direction: Normally the purpose of a proof is to simulate the more complex process (offset) ...

Completely outside you leave the density: Should all colors be printed with the same or those of CMYK correspondingly assigned and the light yellow with that of yellow and the brown with that of black or of cyan or magenta for reaching the optimum print contrast (I do not know the correct English term ...),?
Or is the best density that match the 100% patch visually nearest to the proof? (What could that mean to light and shadows?...)

For screened layouts is (maybe forgotten, but still important), in my humble opinion, to determine first the optimum print contrast. The ideal density for each color on a printed color series is to find regarding the result of screened patches (especially in lights and shadows) and printed with these densities leads you the respective dot gain for the single separation (for creating “normal” cmyk-profiles the average - c, m, y - dot gain is to determine, which is then the basis for the respective curve fittings in the RIP…).

I do not only know the design / layout, I can judge the claim of the customer only indirectly too (much want, less knowledge is a hint, that a might be high claim for e.g. a new brand on packaging is not developed by much experience...):

In short: I would communicate your necessary expenditure in any case to the customer and/or offer him a "low cost" alternative in treating the CMYK-like spot colors just like CMYK (unless it is about highly pigmented colors e.g. from Epple Anniva, then of course higher densities than usual / drive according to the manufacturer's specifications and previously convert to the profile from Epple/Anniva for that ...) and make a real print (offset on stock paper) with selected layouts from him and next step he has to correct his files, if necessary ...

But maybe your customer is happy with the result of your thought/stated method because he feels cared “enough”
 
Completely outside you leave the density: Should all colors be printed with the same or those of CMYK correspondingly assigned and the light yellow with that of yellow and the brown with that of black or of cyan or magenta for reaching the optimum print contrast (I do not know the correct English term ...),?
Or is the best density that match the 100% patch visually nearest to the proof? (What could that mean to light and shadows?...)

Talking about the 100% color patches we are supposed to print tailor-made inks so my thought was to reach the LAB values for those and then measuring TVI on screened patches.
I understand this could seem a weird procedure, but here we are - and here I am only with a file and a printed proof!
Sorry, I cannot post the design itself for company privacy-related rules, but you are right about your suppositions. Trust me: this is a really famous brand. I thought to be just me thinking bad things about this way of work! It seems this brand works this way worldwide and since it's a really big brand then this is the way things must stay.

Since I can't say the proof is right or wrong then my first step now is trying to get the CXF files about those special inks, so as to check the 100% patches correctness first. Then the rest.
 
Well... Do your homework generating sensible TVI curves according to the specs given (maybe add 5% to the color bars, too) and nail down your pressmen/-women and prepress to the agreements made and tolerances given.

For the proofing part, I'd strongly suggest having a very close look on GMG Opencolor (which really does some great "magic", just ask a good dealer or GMG themselves for a sample generated out of exactly this printed color bar after giving them just the strip from the press sheet; you'll be surprised!).
 
Here is what I've done to get what I wanted...

A first print-run using no TVI curves, measuring only 100% patches, using eXact eXrite "best match" feature to print as close as possible to the target, then I've measured screened patches, made custom TVI curves for every ink, made a second-run print and compared screened patches to be sure I was getting the right result.
 

PressWise

A 30-day Fix for Managed Chaos

As any print professional knows, printing can be managed chaos. Software that solves multiple problems and provides measurable and monetizable value has a direct impact on the bottom-line.

“We reduced order entry costs by about 40%.” Significant savings in a shop that turns about 500 jobs a month.


Learn how…….

   
Back
Top