Difficulty understanding verification results

[Carlitos]

I’m calibrating a 2017 iMac and an iMac from 2009 with a recently purchased Colormunki Display.  I regularly use FCPX, Davinci Resolve, Lightroom, Photoshop and After Effects to a lesser degree.  I have no experience calibrating a monitors.  To simplify things I tried to follow this article as closely as possible:

How to Calibrate iMac and iMac Pro Displays

So far the 2009 iMac is visibly much better but I really don’t understand the Measurement Report and  I’m getting the following less than ideal result.  All other results seem to fall within standards:

Measured vs. display profile whitepoint ΔE*00 of 2.03

I’ve uploaded a copy of the full report.

Some questions:

Does the article I referenced give good/accurate advice that will give me good results?

Is there better advice somewhere specific to my needs that is simple and easy to understand?

Why, after calibrating, am I still getting a less than ideal result?  What does it mean?  Is it worth addressing?

Thanks in advance.  I did search the forum but didn’t find exactly what I was looking for.

Attachments:

You must be logged in to view attached files.

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

[Vincent]

1- I’m not sure if that CCMX correction is what you want.
-Very old CCFL widegamut iMacs are like Dell U2711, so they need WGCCFL correction (CCSS) to i1d3 devices like yours
-Older LED sRGB iMacs need bundled WLED correction (CCSS) or an WLED-like CCSS correction from community. Since DIsplayCAL has now a built-in spectral distribution CCSS plot, use it (it should be WLED-like)
-Newer retina P3 iMacs need a P3 WLED PFS phosphor CCSS. There is one bundles in DisplayCAL for those Mac

2- “If you trust that CCMX, so you trust those measurements”, verification seems to be good: Daylight white (OK), neutral grey with low a*b* range (OK), display colorspace matches profile (OK)

3- White pint can be compared against to targets:
-profile whitepoint (white when it was created), 2.x dE means that it has drifted a little *if profle was made some time ago* …or that profile idealizations (single curve+matrix) needed to solve some macos bugs related to color management impose some restrictions on profile whitepoint (ask Florian).
-assumed white curve: it test if white looks like some “natural” warm or cold whites. Yours is OK.

[Carlitos]

Thank you Vincent.  Pretty much all of that is over my head but I dug around and I think I understood your suggestion.  Then do you suggest using the following Correction for the 2010 iMac (I had mistakingly posted that it was a 2009 iMac earlier but that doesn’t seem to matter in this regard) ?

Spectral: LCD White LED family (AC, LG, Samsung)

When you do a Verification does the measurement change according to which Correction you chose when you calibrated or is that information in the Basic Information simply as a note?  So, if the results of the Verification are good does that mean that they are good even if you chose the wrong settings (Correction and other settings) during calibration?

I’m attaching the new verification in case it’s helpful

Attachments:

You must be logged in to view attached files.

[Vincent]

IDNK what kind of panels Apple used in 2009. If they were LED and they have sRGB-like gamut… the one you choose will work.

 

When you do a Verification does the measurement change according to which Correction you chose when you calibrated or is that information in the Basic Information simply as a note?

Correction is used to correct measurements
-while you calibrate
-while you profile
-while you validate
etc

so they are not a “note”. Whatever you choose will have an impact on readings. The closer colorimeter is to an ideal CIE 1931 2º observer, the less important is which CCSS you choose.

So, if the results of the Verification are good does that mean that they are good even if you chose the wrong settings (Correction and other settings) during calibration?

By chance in your particular case, but yes.

Usually if you want to verify how a 3rd party program works you must use the same correction that program used, so you play with the same rules as that program (be it an accurate correction or not, then you can use the good one).

• This reply was modified 4 years, 11 months ago by Vincent.

[Carlitos]

Thank you very much Vincent.  I can’t say I understood everything but you’ve been very helpful.

[Author]

Posts