Puzzled with Color Match Issue

[dizzydev]

Hello all - first post on the forum, but after lurking for quite a while I figured I'd finally say hi ... and ask the experts and collective expertise in the forum to help me with a puzzler.

I recently picked up an Xrite i1Pro 2, and to test it out I've been re-calibrating printers and paper.  Generally, this has gone quite well, and across a reasonably wide range of media on three printers (iPF8100 and 8300, and my old battleaxe Canon Pro9500), I'm getting solid color fidelity with very low fuss.

As I have been working through this, I had a couple of papers where I only had boxes of sheets handy, and started to work these through each of the three printers as well.  This is where the puzzler happened.  I calibrated several roll papers between the 8100 and 8300, and they matched fine.  However, when I've calibrated sheet paper, I can get the 8300 and 9500 to match spot on - but the 8100 is consistently coming out visibly mis-calibrated.  I've re-profiled and tried a couple different papers, and I'm consistently getting the same mis-calibrated result.

I've tried to identify printer, printer driver, i1 Profiler, or other settings that would cause this, but as far as I can tell the profiling workflow is identical, and the printer/driver settings are identical between roll and sheet.

Any thoughts on where to look to start diagnosing this?  Appreciate any wisdom anyone might have.

Thanks,

Mike

[digitaldog]

Could it be the paper itself (poor QC)?

[dizzydev]

Could it be the paper itself (poor QC)?

I thought this at first myself.  When I recalibrated the first paper after the initial off-color result and got the same result again, I suspected this might be the case - so I pulled out another box from a different manufacturer and tried it.  It came out the same.  These are reasonably fresh boxes of Museo Silver Rag and Canson Platine, so I'd be mildly surprised if both failed QC.

[digitaldog]

I’m not familiar with this calibration process you speak of for the Canon.

What you can do is print targets from each and compare the deltaE’s and see WHERE the differences lie if you try using the Data Analysis part of i1Profiler (or better, ColorThink Pro). Knowing how large the dE is and as importantly where in color space might lead you to the cause.

[dizzydev]

What you can do is print targets from each and compare the deltaE’s and see WHERE the differences lie if you try using the Data Analysis part of i1Profiler (or better, ColorThink Pro). Knowing how large the dE is and as importantly where in color space might lead you to the cause.

Great idea.  I think my thought process had been focusing on the printer driver/printer setup (i.e., an overall configuration problem) rather than inside the profile, which might point my elsewhere.  I should have time over the weekend to stare at this, will let you know what I come up with.  Appreciate it.

[digitaldog]

This test doesn’t necessary have to do anything with the profile. You’d just use the target on both papers and that target has color patches and reference values. You could compare each to the profile but at this point, the idea is simply to compare the two papers to each other. It doesn’t tell us which one is ‘right’ but tells us how each is different. So say you compare the two and plot the dE values by worst to best. In this sort, you see the biggest dE is white. Bingo, it is the paper! Or you see it is dark blues or blacks etc. Could be the paper or maybe some media setting affecting how the ink is being used. You can compare those values to the profile itself (at least in ColorThink). You find paper A correlates better to the profile than paper B. As to WHY, well that isn’t clear just yet. Running the same test on another batch of the questionable paper would be the next test. If they have tiny dE differences, that points to good QC. If the values are off, it points to a problem with the paper QC being inconsistent. Or maybe it is a dry down issue. You could measure the target as you usually do, then a few days later. The values should be tiny (average less than dE of 1). Maybe you find the dE is high. That points to a stability issue or a drying issue. At this point it could be just about anything. But the idea is to use the same target, print differing tests and compare two sets of measurements. This tells you how different they are, where in color space and that may help pinpoint the error.