Too many data points?

tmiller_iluvprinting

Well-known member
After reading some of the latest color management posts, I am wondering if Rampage is using too many control points to create plate curves. I am using Curve 2 software, set to Rampage to create our desired plate curves. I am wondering what everyone else is doing? My thought is maybe my results would be improved if I were to plot 4 control points with Curve 2 and then create the desired curves in Excel, or Photoshop to smooth the results? Any and all arguments are welcome!
Regards,
Todd
 
It would be nice if Curve would allow you to use one set of fulcrum points to manipulate the curve and another set for exporting -- to enter into Rampage to fulfill the 25 entry points requirement.

I have made G7 curves with Rampage approximately 20 times and have good results. I print the KCMYRGB & 3-c ramps and if they are smooth, the NPDC is in check and gray balance is good relative to the gray decisions I've made, then I am happy. Send me a private message if you would like to see the data.

We have an Apogee system in a sister plant and just to keep things simple, I use the Rampage fulcrum points. It works fine there too. Apogee actually has some smoothing functions built into the RIP which cannot be turned off so far as I can tell. If the smoothing has helped or hurt, the outcome has been subtle beyond detection.

I suspect we would get good results entering 100 fulcrum points into Apogee. I understand the fewer the better argument and it is convincing, but I have yet to witness too many points causing a problem.

As with anything else, the answer depends on the context.

Matt Louis
 
Matt-Thank you for the response. I too have good results for every ink and paper combo.....except when profiling the press for Segundo screening. When I do the G7 calibration run I am hitting the Lab targets, but the resulting plate curves are pretty bumpy. I am wondering if maybe I should not try to achieve gray balance in the 100% CMY patch and sacrifice the Lab targets for Segundo during the cal run? I have also considered a higher tac ink set for Segundo to improve performance. I haven't done any iterative press runs, but maybe I need to? We would like to use Segundo on more live work, but to be truthful, the proof to press match leaves a little to be desired.The press prints full in the quartertones compared to the proof, and I haven't figured out why as of yet?
Regards,
Todd
 
G7 assumes ab*0,0 at the CMY endpoint so the further form this the worse when forcing graybalance all the way through P2P, row 5.

I would not allow overprint hue angles to veer from proofing reference or you will be in worse trouble than you think you are now. Gray errors can be largely hidden with aggressive GCR. I have had dE 10+ on the 300CMY but still made press sheets to match Gracol under these conditions. I mean, there's always THAT one job that will cause grief, but day in a life the matches are good.

As for am screening chroma hump in the 1/4s, I think Gordo has an illustration on this forum illustrating the phenomenon in ColorThinkPro. With FM compared to AM, a greater percentage of reflected light is filtered through the inks, making the light more colorful in lighter tones than with FM screening. If by full you mean too dark, then you have an NPDC problem --undoubtedly the press is printing different after the initial calibration run.

There also FM G7 threads on this forum (I think). So I hear, G7 gray balance is more difficult to achieve. To print FM and have the best possible match to proof, you may need to consider making a private characterization data set that is "based on" Gracol but with FM idiosyncrasies. On one hand I am jealous you get to print FM but I do not envy your proofing problem.
 
[snip]

As for am screening chroma hump in the 1/4s, I think Gordo has an illustration on this forum illustrating the phenomenon in ColorThinkPro. With FM compared to AM, a greater percentage of reflected light is filtered through the inks, making the light more colorful in lighter tones than with FM screening. If by full you mean too dark, then you have an NPDC problem --undoubtedly the press is printing different after the initial calibration run.

There also FM G7 threads on this forum (I think). So I hear, G7 gray balance is more difficult to achieve. To print FM and have the best possible match to proof, you may need to consider making a private characterization data set that is "based on" Gracol but with FM idiosyncrasies. On one hand I am jealous you get to print FM but I do not envy your proofing problem.

A bit of clarification...20 micron FM screening has a larger gamut than AM/XM screening up to about 400lpi.
The GRACoL data set is based on 175 lpi AM/XM screening which has a smaller gamut. Therefore the color delivered by FM/Segundo screening will not match the GRACoL color. The difference will be seen in one and two color screen tint builds. Vissually the difference is most obvious in screen values from about 60%-10%.

Here's a link to gamut projections and a movie of the two gamuts compared:
Quality In Print: AM and FM gamuts compared - part 1 of 2

An explanation as to why this happens is here:
Quality In Print: AM and FM gamuts compared - part 2 of 2

So, presswork using FM screening that tonally matches the GRACoL data set and that is grey balanced and is using the proper base process colors will not match a GRACoL 7 proof. At least not in one and two color screen tint build colors.

One option is to not worry about it (what many printers do). The other is to re-separate incoming files to "dirty" them up to eliminate the extra gamut (don't know anybody that does that). Another is to fine tume your proof based on the FM print characteristic rather than the generic GRACoL 7 (what some printers do).

Grey balance should not be any harder with FM screening - it's just that the screen percentages that deliver it may not be the same as they are for AM/XM.

To Todd:

Yes, too many data points can create bumpy curves. But, it is a good idea to first check the uncalibrated response curve of Segundo. This is to help ensure that there are no inherent ink transfer issues with your ink/screening combination. If the inherent response curve is "bumpy" - then it is dangerous to try and build plate curves. Better to work with your ink vendor on fixing any issues first. You probably should NOT go to a higher ink tac as this can lead to ink transfer and other problems.

I don't know what you mean by quarter tones appearing "fuller" than the proof. Usually the extra gamut with FM makes the quarter tones appear lighter (because of the extra chroma) even if there is a tonal match.

You might also check the OBA content of your proofing paper vs your press paper. Typically press paper has a high OBA content while proofing paper does not. Since ink acts to block UV, the difference in ink distribution between AM and FM screens might exaggerate a visual difference between AM/XM and FM in the quarter tones.
Go here: Quality In Print: Optical Brightners
for a 5 part explanation on the issue of OBAs (skip the first posting)

best, gordon p
 
Last edited:

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