Script for Printer Calibration on MacOS (free to use and modify)

[knowing]

Hi everyone,

I recently got me a new and unused Colormunki Photo, as SpyderPrint was too much hazzle.

I found a farily simple script on github AutomatedArgyllPrinter, by Jintak Han. I used this as basis for creating a much more advanced version, see link below. My new script is for macOS and Linux, and aimed at Colormunki.

However, menu choice 2 to 5 can be used by other instruments supported by Argyll. Menu choice 1 in my script presents a set of pre-defined targets to generate, with Colormunki in mind, but choosing other than Colormunki as intrument allows to make target from custom arguments and a default setting. Many settings can be configured, and anyone is free to modify the script to their liking, but it would be cool if that work is shared on the forum, if anyone does.

Hope someone will test it and give feedback. Sorry, I do not have windows, so this is not for you that only have that.

The script works well for me now, but I am sure bugs exist (log file content is not ideal I noticed). There is a index.md file for installation help and user guidance, but a lot is self explanatory.

The project has now been published on github.com and can be downloaded from:
https://soul-traveller.github.io/Argyll_Printer_Profiler/

Cheers,
K

 

[The Hat]

Posting the same thing twice doesn’t help..

 

[knowing]

Posting the same thing twice doesn’t help..

Yes, sorry about that. Changed my mind where to post.

 

[knowing]

I have now updated the script to support Linux. That was doable, but I doubt I can support Windows without rewriting the whole script or porting to python, which I cannot do.

Updated the first post also.

Hope this allows more people to try it out. I have not teste on Linux, only verified quality, logic and syntax using AI.

https://soul-traveller.github.io/Argyll_Printer_Profiler/

Cheers,
K

 

[knowing]

I have now updated the script to support Linux. That was doable, but I doubt I can support Windows without rewriting the whole script or porting to python, which I cannot do.

Updated the first post also.

Hope this allows more people to try it out. I have not teste on Linux, only verified quality, logic and syntax using AI.

https://soul-traveller.github.io/Argyll_Printer_Profiler/

Cheers,
K

Hi,
I made another significant modification and release of the script, as the menu system was very rigid and with bugs. Now both Colormunki users and those with other instruments should be able to use it. I added support for defining some more arguments in the ini file, and there is now a choice under menu item 1 to chose custom target, thus parameters can also be specified freely (with help from a specified default value). I hope this is now working.

I also changed the targen and printarg values I had in the ini file (for colormunki choices), as I was experimenting with multi cube steps, and that seemed to make some bad profiles (I guess there is a reason few seems to use that option).

https://soul-traveller.github.io/Argyll_Printer_Profiler/

Cheers,
K

 

[knowing]

Hi,
I made another significant modification and release of the script, as the menu system was very rigid and with bugs. Now both Colormunki users and those with other instruments should be able to use it. I added support for defining some more arguments in the ini file, and there is now a choice under menu item 1 to chose custom target, thus parameters can also be specified freely (with help from a specified default value). I hope this is now working.

I also changed the targen and printarg values I had in the ini file (for colormunki choices), as I was experimenting with multi cube steps, and that seemed to make some bad profiles (I guess there is a reason few seems to use that option).

https://soul-traveller.github.io/Argyll_Printer_Profiler/

Cheers,
K

I forgot to mention,
a good feature is that the script allows configuring 6 different target setups, accessable by menu, for two instruments (Colormunki and one other, such as i1 Pro, i1Pro3+, DP20, SpectroScan etc.). The "other" setup will be available for all the instrument types other than Colormunki. In principle, if you have three different scanners/instruments that one uses frequently, the Configurable parameters (ini-file) could be configured so that

• One menu of 6 targets for colormunki
• One menu split into target choices for one or more instrument type (menu option text configurable).

Hope it is useful.

Cheers,
K

 

[knowing]

Hi All,

I made a significant update of the Argyll_Printer_Profiler/ script.

Version 1.2.0 has the following uptates:

1. Added many pre-made targets under folder Pre-made_Targets/ mainly for Colormunki. Many targets are from this forum, grouped according to patch width and ease of use.
2. I also made a Colormunki version of the TC9.18 target provided by @pharmacist, which I also attach here for availability, but is part of the release. I tested the A4 page random target version, which gave an excellent profile.
3. Added better existence checks for .ti1/.ti2/.ti3 and TIFFs,
4. Improve filename prompt/UI text and menus
5. Improved user-ability.
6. Fixed various bugs.
7. Added flag to turn on or off (default) auto opening of tif-target images in ColorSync Utility.
8. Update of user guide (index.md).

I expect that I will not work much more on this now, but please return with any issues found. Alternatively, anyone may improve the script too.

As it stands, It works quite nice for my own sake. Hope it helps others too.

Cheers,
K

 

[knowing]

Hi,

In my last release of the Argyll_Printer_Profiler/ script I added the blow reference table, as a guide to understand how accurate various printer types can be expected to be when a ICC profile is made. The information was given by ChatGPT, which it said was extracted from loads of product papers, tests and performed calibrations.

Any comments to this? Does this information seem reasonable to you, because you are really the people with the experience to say anything about it... My Laser printer would be classified within the "Professional laser printer", and my profiles seem to fit well within those numbers.

Thanks for your thoughts.

Cheers,
K

Here is the table I added in the script:

Below is an overview of expected accuracy of profiles.
If dE values are too large it is recommended to remeasure
patches or whole target.

───────────────────────────────
ΔE2000 Color Accuracy — Quick Reference (Profiled Printers)
───────────────────────────────

Printer type _______________________ Avg _____ 95% __ Max
───────────────────────────────
Professional photo inkjet___________0.5–1.2 _ 1.5–2.5 __ 3–5
Examples:
• Epson SureColor P700 / P900 / P9570
• Canon PRO-1000 / PRO-2100 / PRO-4100
• HP DesignJet Z9+ / Z6 PostScript

Prosumer / high-end inkjet _________ 0.8–1.8 _ 2.0–3.5 __ 4–7
Examples:
• Epson SureColor P600 / P800
• Canon PIXMA PRO-200 / PRO-300

Consumer home inkjet _____________ 1.5–3.0 _ 3.0–5.0 __ 6–10
Examples:
• Canon PIXMA TS / MG series
• Epson Expression / EcoTank series

Professional laser printer ___________ 1.5–2.5 _ 3.0–4.0 __ 5–7
Examples:
• Xerox Versant / PrimeLink
• Canon imagePRESS C series
• Ricoh Pro C series

Office / consumer laser _____________ 2.5–5.0 _ 4.0–7.0 __ 7–12+
Examples:
• HP Color LaserJet Pro
• Brother HL / MFC color series
• Canon i-SENSYS / MF color series
───────────────────────────────

Notes:
• Values assume proper ICC profiling and correct media settings
• Avg = overall accuracy, 95% = typical worst case, Max = outliers
• Lower ΔE = higher color accuracy
• ΔE < 1.0 is generally considered visually indistinguishable
• Source of these numbers: https://chatgpt.com

 

[knowing]

Yet a small update to improve the sanity check with Add ΔE analysis, version 1.2.1:

- parse ΔE values from the log, compute largest/smallest,
- calculate percentiles (99/98/95/90),
- calculate percentages of patches below ΔE thresholds (1.0/2.0/3.0).


The sanity check output for my old laser printer is then shown like this:

Delta E Range Analysis:
Largest ΔE: 5.644922
Smallest ΔE: 0.044647

Percentile Values:
99th percentile: 4.590486
98th percentile: 3.857946
95th percentile: 2.887379
90th percentile: 2.398526

Patch Count Analysis:
Percent of patches with ΔE<1.0: 48.9%
Percent of patches with ΔE<2.0: 82.7%
Percent of patches with ΔE<3.0: 95.5%

No of test patches = 1168
Profile check complete, errors(CIEDE2000): max. = 5.644922, avg. = 1.243752, RMS = 1.544925


Cheers,
K