Target to determine optimal patches required / linearity

[smilem]

Hello, I would not like to re-invent the wheel, so I would like to ask:

Is there a target / method so that the optimal patches required for certain printer could be determined?
Also this would be nice to determine printer stability how much dE one page differs from another (paper variablity etc.).

[MonsterBaby]

for what reason?

RGB profile? CMYK RIP printing? proofing.. or CMYK+RGB profile?..

[smilem]

for what reason?

To check whether the client printer is stable or not, to make sure the printer is stable from one sheet to the next.
I had cases where even EPSON 9800 and better printers were having problems with print stability from one sheet to the next.
Making profile for such setup means you make a profile that your client can't use, and without further investigation you can't understand what is the problem, client was printing with wrong settings, printer problem, etc. etc.

RGB profile? CMYK RIP printing? proofing.. or CMYK+RGB profile?..

Both, RGB and CMYK. I'm sure separate targets for each system must be created.

[MonsterBaby]

i dont really see what you are trying to accomplish...

if your clients dont have a proof setup.. then it wont be stable from printer to printer..

its a big thing for RIPs to let u calculate device link profiles so the prints will come out the same on different printers..

[smilem]

My client uses Kodak Prinergy with EPSON 9800. AFAIK this rip is simple pass trough.
They use device link profile calculated with separate software Alwan Linkprofiler.

The problem I'm having is that proof color does not match as expected (mesured with i1Pro), and that IT8.7-4 targets have large patch variability from sheet to sheet 2.4dE on some patches. So I say it's printer instability but need to confirm using tests of some sort.

[MonsterBaby]

ah oki. that makes more sense now .. ;-)

and you dont want to use an it8 target because its too large and a waste you think?

i am not familiar with prinergy. but i wouldnt use anything else but the program and printer he is having problems with.

cant you print some sort of teststrip with this rip?.. fogra wegde? idealiance??. and then you can use a veryfier to compare results printed within a week?

[MHMG]

The problem I'm having is that proof color does not match as expected (mesured with i1Pro), and that IT8.7-4 targets have large patch variability from sheet to sheet 2.4dE on some patches. So I say it's printer instability but need to confirm using tests of some sort.

max 2.4 dE on one patch after reading hundreds of patches with an i!Pro?  This sometimes happens just from instrument/measuring error. Subtle changes in contact with the surface during readout can introduce more variability than instrument specs suggest would happen. You might try reading the same target a few times and comparing your measurement repeatability. I'd wager some of the error is right there and not related to print-to-print errors.

good luck,
Mark
Marl

[smilem]

Here is the comparison,

We use i1isis not i1Pro (I'm modifying the iO table the Xrite design was a disaster dE 10x more than the ones I list here), same sheet variability (read 3 times) is very good:

With the 0.65x0.65 I see:
Maximum Total worst ~0.26 dE-2000
The Average total 0.04 0.08
      Best 90%   0.03 0.07
      Worst 10% 0.12 0.18

[digitaldog]

To check whether the client printer is stable or not, to make sure the printer is stable from one sheet to the next.

OK, to check process control and trending, not build a profile or anything like that?

First, I’d build a custom target where the defined colors will pretty much within the gamut of the printer (why measure colors that fall outside it’s gamut?). Next, the numbers are kind of up to you, that of course affects the average dE values. That said, I think you’ll probably see low averages, it is plotting the max values and where in color space that becomes useful. I use ColorThink Pro for this. You can sort visually the worst patches and see the Lab values (and where in the Lab values the delta’s are). I think 500 or so colors would work pretty well. You’ll want to sample through Lstar (paper white to pure black of course). You might want to repeat some patches on the target too (like whites, neutrals) so that you could in theory see how the dE’s across a page show. Not really a issue on an Epson or similar printer, a big issue on a press!

You can load RGB or CMYK values you create (again, thinking about having them within device gamut) in ColorPort from X-Rite and build a target.

Lastly, what you’ll want to do is make a reference of the profile as a color list in ColorThink. You can then compare that reference to the mesaured color you get though the profile. To get an idea of what I’m talking about, see: http://digitaldog.net/files/How_to_compare_deltaE_of_ColorLists.pdf

If you are really serious about trending for customers, let me suggest CHROMIX's Maxwell service. It will use the color lists described above or you can use their ColorShuttle on site to load this data into Maxwell.