software for building dot gain curves?

beanz

Member
i am planning on profiling a press, which needs me to read patches (95%,90%,85% etc.) in order to build a tvi curve.

I was wondering if there is a software and a test patch that would enable scanning of step charts for fast reading of dot gain or TVI (via my i1 pro2) values.

Thanks you for taking the time to answer my question.
 

gordo

Well-known member
i am planning on profiling a press, which needs me to read patches (95%,90%,85% etc.) in order to build a tvi curve.

I was wondering if there is a software and a test patch that would enable scanning of step charts for fast reading of dot gain or TVI (via my i1 pro2) values.

Thanks you for taking the time to answer my question.

Are you looking for a target to use on a dedicated calibration press run or on live production jobs?
Do you have target TVI curves that you want to bring your press TVIs to match?
Are you just wanting software to plot the TVI curves or to build a dot gain compensation curve?
If you use an iT8 target or an ICC profile you can use "ColorAnt" (free from ColorLogic Gmbh) to extract the TVI data and plot the curves. It won't be quite the same as plotting from densitometric data since the TVIs will be calculated from spectral values but it might do the job.
 

beanz

Member
1. this will be a dedicated calibration press run.

2. i will be targetting fogra39 tvi curves, thats why i wanted a quick solution to do the curves (via scanning using the i1 pro2, instead of using the single measurement of exact, since i will be doing 20 readings per sample.

3. i will use the information as input to my metadimension rip for it to generate the calibration curve to attain the fogra39 curves.

thanks for that tip sir Gordon, i will be looking into that software, i appreciate your very meaningful reply :)
 

Ben

Member
I would suggest color toolbox from heidelberg. You can easily measure, analyse en mean them, export and read them directly into meta color calbiration.
 

Vernon Roberts

Active member
I'm currently using Color Toolbox 2015 as well, and it is very powerful. It has the paper white calculator built in from IDE Alliance, so you can adapt a dataset to your paper white. You can iteratively correct proofer profiles. It calculates TVI curves based on an ICC profile, or a dataset. You can print nice color reports. It can perform an G7 calibration from P2P targets and exports the iteratively correcting TVI press curves to Calibration Manager. Create ICC profiles. You can create wet process standards from dry ones and they are shared on the network for press operators to use. You can collect data from press scans pretty much in real time to see trends. It's a beast. It's by far the best solution if you're already using the Prinect Workflow.

It's a big box of color tools. I'm always learning more things it can do.
 

beanz

Member
being a cheapskate myself, are there no open source/free trial versions of software that can do what i need (quickly output dot gain values of CMYK patches via scanning)
 

Vernon Roberts

Active member
Curves 3 has several buying into options, but is probably close to the same price as Color Toolbox. It has a demo mode in the trial, so you can learn the interface before you buy. I personally have no experience with the software, but I looked at it heavily before we added Color Toolbox.

I'm still trying to get my head around the virtual press run. I read about this first in Heidelberg documentation, and didn't really think I was understanding it correctly, but Curves 3 has a VPR module which is supposed to let you profile a press in one run.

VPR
Virtual Press Run
is an add-on module to Curve3. This module makes it possible to curve and profile a press with only one press run.

Without VPR, obtaining a press profile from a G7-calibrated press requires at least two press runs; one with null plate curves to calculate the G7 calibration curves, and a second to print the profiling target through the resulting plate curves to profile the press.
VPR typically eliminates the need for the second press run. The G7 curves calculated from the first run are applied mathematically to the profiling target measurements of the first run, producing measurements that appear as if they were produced on a second "virtual" run. The savings can be huge. The first press run can be used to print on a number of different paper types. If VPR eliminates the second runs for each paper, one press run might be all that's needed to G7-calibrate a group of papers.
It's interesting.
 

lazzz

Member
being a cheapskate myself, are there no open source/free trial versions of software that can do what i need (quickly output dot gain values of CMYK patches via scanning)
I've just started a new topic about such Freeware online software http://printplanet.com/forum/prepre...ctral-calculator-on-javascript-help-is-needed
Open the page http://rudtp.pp.ru/spectralcalc_en.php
Enter your measured data in the top text field. You can easily obtain measures with trial version of X-Rite's Measure Tool.
Then click "Color" button to check your solid densities. Correct them and make another press run if needed. You can do that using laptop right near press to save time and paper. But don't forget that ink fades after drying.
When you achieved correct solid ink densities from ISO 12647-2, you can click "TVI" button.
Then save TVI values by clicking the button "Save TVI for calculation of compensation". Open saved .txt file, copy the table with values for one particular ink to clipboard. For example cyan, like that:

100 0
98 2.46
95 4.25
90 6.53
85 7.70
80 9.35
75 10.94
70 11.70
60 14.35
55 12.16
50 12.61
40 11.43
30 8.85
25 7.88
20 7.12
15 5.52
10 3.03
7 2.05
5 1.51
3 1.32
2 0.74
0 0


Then open this page http://rudtp.pp.ru/dgcor/, insert the values into the left field. Then enter the target curve name in the small field below, called "ISO TVI at 50% ". For example if you wanna aim FOGRA 39 Cyan, you should enter "f39c".
If you need interpolation of the data you can enter desired "steps". In the example above we don't have 1% and 99% field (they weren't presented in the initial measured control strip). So we can enter such table:

100
99
98
95
90
85
80
75
70
60
55
50
40
30
25
20
15
10
7
5
3
2
0

Then click "Calculate" button (it's not translated yet, so click "Считать")


You will see the correction values in the right field.

You can plus them to linearisation correction values in your RIP.
Even first run provides good results. But for making your curves perfect you may need from 3 to 8 iterations.
Then you should check you dotgain values, measuring control strips on regular basis. Sometimes you will have to correct compensation curves.

I will copy this message to original post to have all the information in one place for those who need it.
 

KamilT

Active member
You can plus them to linearisation correction values in your RIP.
Even first run provides good results. But for making your curves perfect you may need from 3 to 8 iterations.
Then you should check you dotgain values, measuring control strips on regular basis. Sometimes you will have to correct compensation curves.
Hi Iazz,
interresting tools, but one note: if you will need more than 3 iterations, something is really going bad. In fact, iterations are provided by most applications, but in real, they are just some kind of placebo.

beanz: You can use ColorPort from X-Rite, you can easy compute dot% from densities (or yxz, if you prefer), as ColorPort report them all to txt file, which you can import to Excel/OpenOffice.

Regards,
Kamil
 

lazzz

Member
Actually, iterations are needed not cause of software, but cause of process fluctuations. If you have rock-solid stable process, i believe one or two iterations should be enough. In our case we need some averaging between test runs. And sometimes we have to make another iteration to compensate process changes.
 

gordo

Well-known member
Actually, iterations are needed not cause of software, but cause of process fluctuations. If you have rock-solid stable process, i believe one or two iterations should be enough. In our case we need some averaging between test runs. And sometimes we have to make another iteration to compensate process changes.

I totally agree with KamilT. I've built very many on site dot gain compensation curves and the process is very straightforward - do a press run to see the current dot gain response, build the compensation curve, do a confirmation press run. Job done - leave the building. No iterations needed.

If your process isn't stable, reliable, and consistent then you can't build curves - you'll just be chasing your tail. If the process isn't stable, reliable, and consistent then spend your time finding out where the issues are and make them right.
 

mihas

Member
you can import to Excel/OpenOffice.
Excel is a good tool! But calibration, based on dot gain or TVI, requires high-quality interpolation of the measured data. Calculate interpolation in Excel is difficult without special training. http://rudtp.pp.ru/dgcor/ calculator offers a choice of several interpolating algorithms and smoothing algorithm. Dot gain data can be obtained from the spectrum of different ways, one of the options - the free tool http://rudtp.pp.ru/spectralcalc_en.php
 

gordo

Well-known member
Excel is a good tool! But calibration, based on dot gain or TVI, requires high-quality interpolation of the measured data. Calculate interpolation in Excel is difficult without special training.
I'm not sure we're talking about the same thing because calculating dot gain compensation curves is very straightforward and simple.
It's explained here (skip the part about what is dot gain): http://the-print-guide.blogspot.ca/2010/02/principle-of-dot-gain-compensation.html
 

mihas

Member
Gordo, thank you for your blog, it really helps.
When calculating dot gain or TVI compensation curves i and some developers (MetaDimension, Apogee) guided Goldberg diagram (http://rudtp.pp.ru/dgcor/Goldberg_diagram_en.png) from book: Helmut Kipphan, Handbook of Print Media, p.480. Difference between print characteristic curve and target characteristic curve are not equal to the amendment dot on the plate. Primitive arithmetic here does not work, modification of dot is not equal to the difference TVI(reference)-TVI(real). To calculate the compensation of dot from TVI(reference)-TVI(real)-difference required interpolation: cube, Akima, Hermite, Lagrange or good some others. It is difficult in Excel.
 
Last edited:

lazzz

Member
Gordo, thank you for your opinion!
I do not mean that the way we work now is correct. Sure we have to invest in climate control and water preparation. That's what i'm currently working on. But now iterative calibration keeps us inside of ISO tolerances range. We don't need more than one dedicated test run (first one after plate linearization) - all farther iterations are close enough to the standard. So we just place step wedges in the center of regular layout and measure them periodically.
I've done another calibration cycle several days ago. When we measured MW3 after second run (first iteration), all the fields of report where "OK", average deltaE was somewhat about 1,6, maximum dE was near 4. But in highlights TVI curves still were a bit too high.
I don't know if it is ok for the first iteration or not.
If you want, i can upload measurements.
I'm just regular commercial printer and can be wrong.
I have not compared different software. Don't have much time for that. But i hope to try another method soon.
 

lazzz

Member
Could you please answer few questions, considering your article?
1. If we use just one curve for all colors we hardly can get into tolerances of ISO 2013 - they have one curve for black and cmy. But black ink has higher dotgain when required density (lightness) is achieved. But our current standard is mostly the same with iso12647-2:2004. There we have to aim curve tailored for film process, but CtP has different shape of curve. So highlights are the toughest part.
How can your method deal with such issues? Looks like we need too much fields to find right values without interpolation. Like 1-3% step.
2. Why did you say we don't need to apply calibration curves in 1-5 range? I have customer who is very sensitive for lightness of their whitish goods on pictures.
 

gordo

Well-known member
Gordo, thank you for your blog, it really helps.
When calculating dot gain or TVI compensation curves i and some developers (MetaDimension, Apogee) guided Goldberg diagram (http://rudtp.pp.ru/dgcor/Goldberg_diagram_en.png) from book: Helmut Kipphan, Handbook of Print Media, p.480. Difference between print characteristic curve and target characteristic curve are not equal to the amendment dot on the plate. Primitive arithmetic here does not work, modification of dot is not equal to the difference TVI(reference)-TVI(real). To calculate the compensation of dot from TVI(reference)-TVI(real)-difference required interpolation: cube, Akima, Hermite, Lagrange or good some others. It is difficult in Excel.

Thanks for the reference - I have that book. The description is incredibly (to me) unclear and includes film in the equation.
The method that I describe in my blog is the method used by Creo/Kodak and has been used in thousands of print sites. It works perfectly well if you want to build tone reproduction curves to change the tone values that you currently get to the tone values you want. It is also simple to understand.

First off it does not matter what the tone values on the plate is (Quadrant 2 & 3 ?). What is important is that the plate imaging in consistent (or film/plate imaging). That is because if the plate imaging is not consistent then you do not have a solid foundation for building curves.

Assuming you plate imaging is consistent (it does not have to be linearized).

Then you need a target curve - the dot gain curve that you want to get with your presswork (could be an ISO curve, previous presswork, etc.).

Then you need a current curve - the dot gain curve that you get from from your tone requests in the file.

With those two elements it is a straightforward remapping of tone requests in the RIP to get the desired tone response on the press.

Put another way...if I want my 50% tone request in the file to result in 68% on the press sheet but I only get 57% on the press sheet then I simply find what tone request in the file resulted in 68% on the press sheet. Then I simply remap the 50% tone request to the tone request that resulted in 68% on the press sheet.

That remapping of file requested tone values is done automatically in the RIP.

It's a simple, deterministic, process that works.
 

gordo

Well-known member
Could you please answer few questions, considering your article?
1. If we use just one curve for all colors we hardly can get into tolerances of ISO 2013 - they have one curve for black and cmy. But black ink has higher dotgain when required density (lightness) is achieved. But our current standard is mostly the same with iso12647-2:2004. There we have to aim curve tailored for film process, but CtP has different shape of curve. So highlights are the toughest part.
How can your method deal with such issues? Looks like we need too much fields to find right values without interpolation. Like 1-3% step.
2. Why did you say we don't need to apply calibration curves in 1-5 range? I have customer who is very sensitive for lightness of their whitish goods on pictures.
IMHO ISO 2013 and the issue of dot gain is a mess as far as dealing with dot gain.

1 - The data set for ISO 2013 was not made from plates that had been curved to equalize the CMY TVIs but instead adjustments were made to the 2006 characterization data sets in order to create the 2013 data sets. The TVI curves for CMY in ISO 12647-2 were equalized for mathematical nicety and, more importantly, propriety. The changes to GRACoL2006 were limited to slight tweaks in solid colorimetry and and equalization of TVI for CMY at 16% to agree with ISO 12647-2. This was done using mathematical tuning/smoothing procedures that left the G7 neutral scale intact. After CRPC6 was finalized, the TC130 WG3 committee decided to also specify TVI for K at 16% for in ISO 12647-2, despite the typical 3% difference between K and CMY TVI on most offset processes. The result of this is that FOGRA51, based on ISO 12647-2, paper type1, has all four TVI curves at 16%, thus giving a far-poorer-than-hoped alignment to GRACoL2013.

2 - Calibration curves aren't typically used in the 1-5% range because the dot gain of those tint percentages have little effect on the resulting tone value and you are also measuring tones that are within the error range of your instruments. But if you want to include those tones by all means do. In my experience you will see a greater shift in the "whiteness" of goods due to the impact of OBAs in the paper.
 

mihas

Member
That remapping of file requested tone values is done automatically in the RIP.
Yes, automatically. Only one iteration! Two iteration maximum in MetaDimension only and the second badly.
While the point administration allows a lot of iterations, mathematically perfect. The difference between the independent calculation of the size of the point, and automatically in the RIP is the number of iterations.
I translated now Compensation Calculator http://rudtp.pp.ru/dgcor/en.php but my English is bad. Colleagues help me edit!
It is registered algorithm of modern rip, Creo/Kodak including. You can calculate the point itself to here. But it is indisputable - you can calculate automatically in the RIP, but only one iteration! Modern RIPs also uses interpolation other than linear, and it is very difficult to replicate in Excel.
For example, offset, the first iteration. We want to do better halftone black. Second iteration will help. In TVI or dot gain can not count right second iteration, but surely in the amount of dot.
 

Automatically Autonomous Automation

Automatically Autonomous Automation
Although the autonomous car is not quite ready, a lights out print operation is something you can do right now if you have a comprehensive Print MIS (Management Information System). The advantages can put money on your bottom line. So what’s your next step? Link to Article

   
Top