Comparing two printers for delta E

[MachServTech]

If anyone is feeling generous and has the time I would like to know what the best way to take the CGATS data from (5) IT8 targets and average them together, then take the averaged IT8 CGATS and compare it to another set of averaged CGATS IT8 data to find the relevant differences in ∆E for C,M,Y,K

In other words:

5 duplicate IT8 targets printed on machine X.

5 duplicate IT8 targets printed on machine Y.

CGATS data form X is averaged = Xμ

CGATS data from Y is averaged = Yμ

data from Xμ is compared to Yμ to give the ∆E for C,M,Y,K (yielding the main differences in delta E between the color from machine x vs machine y)

Maybe I should use Profilemaker's Measure Tool to average and Colorthink to do the comparison, but I have not done this before and could use some advice.

Any suggestions?

Your help is greatly appreciated,

[MachServTech]

afternoon bump

[meddington]

Yep, MeasureTool will allow averaging and comparing. Colorthink doesn't average, but will compare datasets.

[Lukas Engqvist]

Why would you want that? What would you gain? A number?
I would say it would be more interesting to compare the two, primaries and relevant secondaries independantly. A "∆e" as a linear (actually not even sure if it has one dimension) number is very limited to describe the differences of two presses which must be a 3 dimensional difference.

Studying the two one could probably come up with that lowering some SID and/or modifying a curve a closer match could be made. Seeing that "this variation" in a primary affects the secondaries and tertiaries "thus", can have value in enhancing your understanding so that you can make the difficult compromises which by necessity are decisions that need to be made in attempt to home down on the differences. But a one time value?

There are several tools out there to compare two IT8 charts that will give you a visual on which patches are differing in a greyscale image where black will mean no ∆e and white will symbolise high ∆e. (ColorTune from Agfa is one of the tools I have used for this kind of comparison where you would use the one press as the reference, I am sure there are others) But I would still think the ∆e is not the only factor that is important, it is more important to qualify the ∆L, ∆a and ∆b for each colour.

[gordo]

And depending on how you calculate the DeltaE value you can get anything from an unacceptable to a perfect match. See point #2 under "Issues, concerns, and caveats when using CIE L*a*b* DeltaE tolerancing" here:

Quality In Print: Tolerancing color in presswork - CIE L*a*b* and DeltaE

best, gordon p

[Bloodsaler]

Some colors are sensitive to the eyes,so even the delta E value is small,but our eye can easily seperate those colors.
I ofter use measuretool to compare the two sheets of one press,but never compare the data between two press.
For the press condition limit,the delta E value didn`t give any accomendation on how to improve the gray balance between two presses.
I advise you use G7 method to compensate both machine,than you can have a color match~

[gordo]

I advise you use G7 method to compensate both machine,than you can have a color match~

G7 does not provide a color match. That is not what the method does.

best, gordon p

[Bloodsaler]

G7 does not provide a color match. That is not what the method does.

best, gordon p

So what G7 method does?Is controlling the gray balance?

[gordo]

So what G7 method does?Is controlling the gray balance?

Yes. Not for "controlling" gray balance but for making an output device gray balanced.

Quotes from the G7 literature:

"The G7 Specification is made up of formal colorimetric definitions for neutrality and tonality that together control the visual appearance of the grayscale component of an image. The GRACoL and SWOP specifications are both based on G7 gray balance and tonality."

"G7™ is a specification that defines the factors that can be used to provide consistent gray scale reproduction across multiple devices, processes and media"

best, gordon p

[MachServTech]

To be more precise I am comparing the differences in ink between two different ink manufacturers on an inkjet printer.