Evaluating Color Calibration on Pro1000

[Doug Gray]

Canon's Pro1000, and others of that series, have calibration routines built into the printer. This should provide consistency as printers age without having to make new profiles. But does it also improve accuracy re the canned profiles Canon provides? This was a first look.

Testing using Canon's Plat. Pro.  indicates it does both. To test I calibrated using the Plat. Pro.

Selecting factory defaults is done by selecting "Disregard Value" in the driver's advanced settings. Selecting "Use Value" will use the results of the previous calibration with that paper type (Plat. Pro.)

First Test:
To look for profile improvements I made a test chart of 512, 8x8x8 grid spaced patches (rgb deltas of 36 with the high value set to 255) and added 343 centered patch grids such that each point was surrounded equidistant by the main grid.

The measured Lab values of the 8 surrounding points were averaged and the deltaEs compared to measured Lab value of the centered patches. The variation is a kind of measure of linearity in Lab space. It is also a rough indication of how many patches are needed to achieve good printing accuracy. Average dE of the calibrated patches was .99. Average dE of the factory default patches was 1.22. Prior experience with adjusting patch count suggests that the latter requires about 20 to 40% more patches to produce the same accuracy.

Second Test:
The second test measured 855 printed colors of the calibrated and factory default prints against the canned profile reported Lab values for the Plat. Pro. paper.

The factory default printed patches came in at 1.26 dE Ave. while the calibrated printed patches came in at 1.09 dE.

Summary:
I'm impressed how good the canned profile was for the Plat. Pro. paper. The canned profiles were made with a different spectro and there are often differences of 1 dE just from instrument variations. But here we have a printer that almost a year old as well.


My takeaway is that there is a benefit to calibrating the printer to the paper type being used. Especially if one uses the canned profiles. And, if one makes custom profiles it's best to calibrate the printer first, then make or have made, custom profiles. After that one can recalibrate the printer as it ages w/o as much need to re-profile.

The other thing Canon touts, is that calibration provides consistency between different printers. I can't test that as I have only one Pro1000 printer but it's a theoretically sound concept and consistent with these results.

[Stephen Ray]

My takeaway is that there is a benefit to calibrating the printer to the paper type being used. Especially if one uses the canned profiles. And, if one makes custom profiles it's best to calibrate the printer first, then make or have made, custom profiles. After that one can recalibrate the printer as it ages w/o as much need to re-profile.

This has always been the process in the professional world and with today's machines along with consistent materials (if lucky enough), profiles may be valid for a very, very long time.

Question Doug,

What exactly do you consider the printer to be calibrated to other than it falling into alignment to the canned profile? Neutral grays throughout with max color gamut but without excess inking?

[Doug Gray]

Question Doug,

What exactly do you consider the printer to be calibrated to other than it falling into alignment to the canned profile? Neutral grays throughout with max color gamut but without excess inking?

I'm not sure what the Pro1000 calibrates to nor how consistent the sensors are from device to device. My assumption is that the calibration process targets the same values as were used for making the canned profiles subject to device to device tolerance variations.

It has density sensors and, based on the calibration prints, it looks at the density of the CYM and additional colors R and B at different dot densities as well as the neutral Ks. The calibrated prints and factory settings both created the most dense blacks at L*=1.8 to 1.9 which is also what the canned profiles predict.

Interestingly, calibration made less difference on the neutrals than the colors which I supposed is to be expected with the large set of neutral inks.

I also ran tests of the dE deviation of neutrals for L*=2 to 94 in steps of 1 (printable range of paper) and compared them against the Canon canned (OEM) Plat Pro profile.

The custom profile made with the calibrated profile had an average dE of .97 and max dE (at L*=6) of 1.8. The custom profile made with the factory setting  had an average dE of 1.13 with a max dE (at L*=49) of 2.3.

So the calibration process improved the printer's conformance with neutrals as well.

[Stephen Ray]

I'm not sure what the Pro1000 calibrates to

Just a quick visual evaluation would be to print an image file such as the PDI target or another familiar reference to judge known neutrals and color checker swatches. *Some* mfrs seem to calibrate toward cool neutrals, generally. So, that bias influences one's b&w printing whether for favor or not. I'm interested to learn if you have an opinion of whether your particular machine is cool, neutral, or warm.

[Panagiotis]

My takeaway is that there is a benefit to calibrating the printer to the paper type being used. Especially if one uses the canned profiles. And, if one makes custom profiles it's best to calibrate the printer first, then make or have made, custom profiles. After that one can recalibrate the printer as it ages w/o as much need to re-profile.

Thanks for testing this. A problem is that we are not sure if the paper companies have calibrated their printers first before making their profiles.

[digitaldog]

This has always been the process in the professional world and with today's machines along with consistent materials (if lucky enough), profiles may be valid for a very, very long time.

Correct and nothing that new. When I built the ICC profiles for Epson (Exhibition Fiber), using 5000 color patches, we utilized a half dozen printers spread over the U.S. The following printers profiled are below. Multiple targets were then printed from each printer over the course of a few days. The average dE of all the data was well below a dE of 0.5 per unit.

[Stephen Ray]

A problem is that we are not sure if the paper companies have calibrated their printers first before making their profiles.

What's the method to examine the profile for the state of printer linearity while the profile was created?

[digitaldog]

Depending on the printers, there is absolutely nothing to calibrate.

[Stephen Ray]

Depending on the printers, there is absolutely nothing to calibrate.

Is Doug not calibrating his Canon Pro1000? He mentions before and after, sensors (built in?), etc.

[digitaldog]

I can't answer for Doug, I can again say, depending on the printer, there is absolutely no calibration required.

[Stephen Ray]

I can't answer for Doug, I can again say, depending on the printer, there is absolutely no calibration required.

So, if the particular printer has no user-access for calibration, the printer then relies upon factory settings for the life of the machine. However, if the factory neutral gray bias was noticeably *cool" as an example compared to a standard Kodak gray card or scale, one would need to rely upon either a canned profile or a custom profile in hopes the profiling results to a known neutral gray.

Is see at the Canon web site the imagePROGRAF PRO-1000 apparently does allow calibration.
"Performing Color Calibration Using the Operation Panel - imagePROGRAF PRO-1000
Article ID: ART163225   |   Date published: 10/22/2015   |   Date last updated: 04/11/2019"

[digitaldog]

So, if the particular printer has no user-access for calibration, the printer then relies upon factory settings for the life of the machine.

Yes. And again, at least with the Epson's I've profiled, that's how they roll.

[Doug Gray]

Yes. And again, at least with the Epson's I've profiled, that's how they roll.

That's certainly been the case with my Epson 9800. Very little change in the 10 years I owned it outside of a period where I had used off-brand inks which was a mistake corrected.

However, not having multiple 9800's I have no idea how consistent they were between each other. I'm not sure but Epson service may have a way to calibrate a printer after, for instance, a head change.

[Stephen Ray]

Good to learn from both of you your experiences with Epson machines.

[Doug Gray]

I compared the canned Canon plat pro profile's accuracy against colors printed with the factory defaults and after calibrating. Attached are the plots of dEs (1976 and 2000) for 855 evenly distributed color patches.

I also ran a test to see if custom profiles made after calibrating were more accurate than custom profiles made with the factory settings. Opposite to my previous expectation, there was no material difference and the averages for both color patches and near neutrals were within .01 dE. Gamut sizes were virtually identical well. My conclusion is that if one can easily make custom profiles there is no advantage to calibrating.

OTOH, if one buys custom profiling services it's best to calibrate first. Future calibration should compensate for printer aging w/o the need to get new profiles.

[henrikolsen]

I compared the canned Canon plat pro profile's accuracy against colors printed with the factory defaults and after calibrating. Attached are the plots of dEs (1976 and 2000) for 855 evenly distributed color patches.

I also ran a test to see if custom profiles made after calibrating were more accurate than custom profiles made with the factory settings. Opposite to my previous expectation, there was no material difference and the averages for both color patches and near neutrals were within .01 dE. Gamut sizes were virtually identical well. My conclusion is that if one can easily make custom profiles there is no advantage to calibrating.

OTOH, if one buys custom profiling services it's best to calibrate first. Future calibration should compensate for printer aging w/o the need to get new profiles.

Adding to the data. Similar results.

I've run similar curiosity-tests in late 2017 on the Pro-1000 and now some again. On smaller targets though.

Back then I compared Canon's OEM profile for Pro Luster with actual measurements, and confirmed that a Color Calibration brought the output closer to the expectations stated in their profile. So one can hope that at least Canon themselves assume the printer has been Color Calibrated when using their profiles, and one could hope other paper manufacturers do the same when issuing profiles, for best consistency.

Before (without) calibration I saw average CIEDE2000 errors of ~1.8 and peaks of ~4.7.

colverify -k -w -x target1-uden-cal.ti3 target1.ti2
Verify results:
  Total errors (CIEDE2000):     peak = 4.667316, avg = 1.825594
  Worst 10% errors (CIEDE2000): peak = 4.667316, avg = 3.015105
  Best  90% errors (CIEDE2000): peak = 2.493174, avg = 1.687279
  avg err X  0.011928, Y  0.012853, Z  0.013387
  avg err L* 1.430251, a* 1.170308, b* 2.192808

After calibration I saw average CIEDE2000 errors of ~0.8 and peaks of only ~2.2, and better results on "Worst 10%". Assuming profile is also smooth, this looks fine to me.

colverify -k -w -x target1-med-cal.ti3 target1.ti2
Verify results:
  Total errors (CIEDE2000):     peak = 2.177756, avg = 0.831824
  Worst 10% errors (CIEDE2000): peak = 2.177756, avg = 1.779103
  Best  90% errors (CIEDE2000): peak = 1.449286, avg = 0.721675
  avg err X  0.003210, Y  0.003624, Z  0.006529
  avg err L* 0.401963, a* 0.609139, b* 1.114283

That was back in December 2017.

As a sidenode, the OEM profile assumes high/highest printer driver quality setting. On standard it's a significantly different output.

Here around the start of 2020 / end of 2019 (around two years later) I ran a new Color Calibration on Canon's Pro Luster (which I use for generic Color Calibration - I haven't found use for custom Color Calibration). The change from before to after calibration was bigger than expected:

Average CIEDE2000 difference of ~1.3 and peaks of ~4.4.
  Total errors (CIEDE2000):     peak = 4.446026, avg = 1.299139
  Worst 10% errors (CIEDE2000): peak = 4.446026, avg = 3.168163
  Best  90% errors (CIEDE2000): peak = 2.559952, avg = 1.081811
  avg err X  0.007117, Y  0.006639, Z  0.009679
  avg err L* 0.632272, a* 1.046717, b* 1.642424

The total average is ok, but the peaks are a little bothersome to me, and the "worst 10%" are significantly off.

Even after re-calibration, I have a feeling my output is currently somewhat off compared to results from the earlier couple of years, as preliminary target tests after re-calibration doesn't seem to agree that well with old results (also after calibration) - which should be the whole point of calibration, to keep consistency. I'll dig in further. Perhaps something has unintentionally changed output slightly that even the calibration is missing, like inks (original), firmware and media types.