Verification report: all color distance values = 0

[Andrea Maino]

Hello to everyone!
I have been using DisplayCal for many years to calibrate my monitors, and I consider it a great tool for those who use graphic programs.  Every two months, I run a calibration and check the icc profile with the verification tool. Some days ago I redid my routine, icc was ok, but the verification report was…TOO OK (ΔL*00, ΔC’00,  ΔH’00, ΔE00 = 0), and I think it’s not possible. I checked two different monitors (EIZO CG277 and an old CG241), but the result was the same. I use Linux Mint 22.3 Zena 64-bit. I attached the reports I obtained.

Thank you very much for your help and for the great work of this community!

Attachments:

You must be logged in to view attached files.

[Ben]

My only guess is the python version.   The latest version of python is not supported.   Keep up with the github.   There has been lots of fixs.  https://github.com/eoyilmaz/displaycal-py3   .   Bug reports are in the github.

[Ben]

I just reread guides to calibration.  Now my 20 points work better.   Everything was to low not being exactly on center off screen.    https://www.avsforum.com/threads/the-certainly-not-complete-user-guide-to-get-to-know-and-calibrate-your-tv.1636306/

[Ben]

Oops.  I forgot to say I put a windows pattern up at the center of the screen. He said use a grid analysis pattern.

[Vincent]

In addition to Ben’s 1st post I see no reason to use 1D GPU calibration on those Coloredge CG. Use an old windows laptop and make use of hardware calibration with ColorNavigator (although ypu may need to go back to ColorNavigator6 for the 24″).
Also even for some reason you want todo the 1D LUT calibration in GPU:
-do not use XYZLUT profiles, just single curve matrix
-do not use ccmx correction unless you made it by yourself. CG277 is a GB-LED, U2413 EDR/CCSS correction will do the job.

[Andrea Maino]

Thank you for your reply.

For the CG277, I can use the internal calibration (without any external OS/software). For the CG241, however, it is different: it does not have an internal colorimeter, and ColorNavigator 6 does not support my X-Rite i1Display Studio.

I used to have an old DTP94B (which was probably rebranded by Eizo for the CG241), and it was supported by ColorNavigator 5, but I can no longer find that version.

I also have an Eye-One Display 2. ColorNavigator 6 recognizes it and it could work; however, the organic filters in this type of colorimeter degrade over time, so the ICC profiles I create show a strong color cast.

I would like to point out that the profiles for both monitors appear to be good. The only issue was with the verification report, which was probably caused by the Python version. I tend to install every update proposed by my Linux system without checking for possible incompatibilities—I will try downgrading, if my system allows it.

Thank you for your help. If I resolve the problem, I will post the solution.

Calibrite Display SL on Amazon  
Disclosure: As an Amazon Associate I earn from qualifying purchases.

[Andrea Maino]

First of all, happy easter! Then, about my previous thread, SOLVED!

I used the version from the Mint repositories (3.9.11). I then compiled the latest one (3.9.17), and verification is now OK.

Now I only have a small problem when creating a colorimetric correction:

‘utf-8’ codec can’t decode byte 0xaf in position 0: invalid start byte

Traceback (most recent call last):
File “/home/xxxxx/venv-displaycal/lib/python3.12/site-packages/DisplayCAL/display_cal.py”, line 13020, in create_colorimeter_correction_handler
check_last_ccxx_ti3(dlg.measurement_mode_reference)
File “/home/xxxxx/venv-displaycal/lib/python3.12/site-packages/DisplayCAL/display_cal.py”, line 12645, in check_last_ccxx_ti3
cgats = CGATS(ti3)
^^^^^^^^^^
File “/home/xxxxx/venv-displaycal/lib/python3.12/site-packages/DisplayCAL/cgats.py”, line 411, in __init__
key, value = values[0].decode(), values[1]
^^^^^^^^^^^^^^^^^^
UnicodeDecodeError: ‘utf-8’ codec can’t decode byte 0xaf in position 0: invalid start byte

This seems to be a Python problem, but now I’ll try to uninstall the old version and recompile the latest one.

[Andrea Maino]

After deleting the previous  ~/.local/share/DisplayCAL/storage (which contained a corrupted .ti3 file), the “Create colorimeter correction” function works correctly.

[Ben]

Good detective work.   Thanks for posting.   Reminds me of when I tryed to debug php website.   Starting clean and fresh is the easy way.    Backs ups are great.  Checking files formatting.   Checking the code.   Lots of work.

[Author]

Posts