What's new

Apple XDR Display Calibration & Calibrite Display Plus HL

Cody Ranaldo

Cody Ranaldo

I haven't seen this specifically mentioned here, but it was posted on another forum-- since mac os Tahoe 26.3.1 you can now use the Calibrite Display Plus HL to do a full recalibration in the display calibrator utility.

I got one for myself and confirmed this (also confirmed that it would not recognize any of my older i1d3 variants). Takes ~2 hours and recalibrates all the factory modes. I personally used it on my M1 14" macbook pro display. I'm assuming that there's some kind of internal spectral correction file being applied seamlessly. I haven't done any confirmation measurements of the actual profiling, but Czornyj on luminious landscape forum compared it to his Jeti Specbos 1211 and found the calibrations to be well aligned.

source: https://forum.luminous-landscape.com/index.php?topic=145578.0
 
hmm, indeed he does, yet in the photo he's clearly using the "Plus". I personally got the Plus version, but I'd be willing to wager that it would work with both the Pro and Plus HL devices since both are supposed to be able to support the maximum brightness of any current apple pro display

Pro HL = 3000nits
Plus HL = 10,000nits

Studio Display XDR = 2000nits HDR
Pro Display XDR = 1600nits HDR
 
Thanks. I have the Pro HL and a Macbook Pro XDR, once I update to Tahoe I can check, It can still take a while though.

But great to see that it's now possible with a cheap colorimeter, I didn't know.
 
Last edited:
Cody Ranaldo said:
I haven't seen this specifically mentioned here, but it was posted on another forum-- since mac os Tahoe 26.3.1 you can now use the Calibrite Display Plus HL to do a full recalibration in the display calibrator utility.

I got one for myself and confirmed this (also confirmed that it would not recognize any of my older i1d3 variants). Takes ~2 hours and recalibrates all the factory modes. I personally used it on my M1 14" macbook pro display. I'm assuming that there's some kind of internal spectral correction file being applied seamlessly. I haven't done any confirmation measurements of the actual profiling, but Czornyj on luminious landscape forum compared it to his Jeti Specbos 1211 and found the calibrations to be well aligned.

source: https://forum.luminous-landscape.com/index.php?topic=145578.0
Click to expand...
Cody,

Does the calibration software you used produce a detailed report listing Delta E 2000 and Delta ITP metrics? If so can you share the Delta metrics?

What test pattern generator did you use?

Did you calibrate both SDR and HDR?
 
It's the built in calibration routine that you can get to in the menus for changing reference modes. It's all totally automatic, does both SDR and HDR, uses apple's built in TPG, and gives you no such options for reportage, in typical apple fashion. It recalibrates all of the reference modes as well as the standard default modes. It would need to be verified with 3rd party software, I know there have been a few threads on the accuracy of the laptop screens OOTB here on the forums-- Not much else to say about it, but it would be interesting to dig in and try to find some logs... but then again this is not a newly available calibration routine, they just finally extended it to be usable with a decently priced meter. I think previously the cheapest supported meter might have been the Klein...
 
Cody Ranaldo said:
since mac os Tahoe 26.3.1 you can now use the Calibrite Display Plus HL to do a full recalibration in the display calibrator utility.
Click to expand...
It supposedly works on all of the Apple "Reference Preset" displays with Tahoe 26.3.1, but the new Studio Display XDR requires Tahoe 26.4 and display firmware version 26.4. See below for a list of those monitors.

I tested it using an M1 Ultra Mac Studio running Tahoe 26.4, and it seemed to work, but I haven't yet validated it using DisplayCAL or ColourSpace.

I got a "calibration failed" error a couple of times, but I had SIP (System Integrity Protection) disabled. I re-enabled that and changed the monitor refresh rate from Adaptive (47-60 Hertz) to 60 Hertz, then it worked.

As you said, the ability to perform hardware calibration on those Apple monitors has previously existed, but it required one of these expensive probes: Photo Research PR-740, PR-745, PR-788, or PR-1050, or Colorimetry Research CR-300, CR-250, or CR-100. It can now be done with the Calibrite Display Plus HL.

MacBook Pros with reference mode:

- MacBook Pro 14-inch (2021, M1 Pro/Max)
- MacBook Pro 16-inch (2021, M1 Pro/Max)
- MacBook Pro 14-inch (2023, M2 Pro/Max)
- MacBook Pro 16-inch (2023, M2 Pro/Max)
- MacBook Pro 14-inch (late-2023, M3 / M3 Pro / M3 Max)
- MacBook Pro 16-inch (late-2023 M3 Pro / M3 Max)
- MacBook Pro 14-inch (2024, M4 / M4 Pro / M4 Max)
- MacBook Pro 16-inch (2024 M4 Pro / M4 Max)
- MacBook Pro 14-inch & 16-inch (2026, M5 / M5 Pro / M5 Max)

Other Apple monitors with reference mode:

- 31-inch Pro Display XDR
- 27-inch Apple Studio Display (2022)
- 27-inch Apple Studio Display (2026)
- 27-inch Studio Display XDR (2026)
 
Cody Ranaldo said:
I haven't seen this specifically mentioned here, but it was posted on another forum-- since mac os Tahoe 26.3.1 you can now use the Calibrite Display Plus HL to do a full recalibration in the display calibrator utility.

I got one for myself and confirmed this (also confirmed that it would not recognize any of my older i1d3 variants). Takes ~2 hours and recalibrates all the factory modes. I personally used it on my M1 14" macbook pro display. I'm assuming that there's some kind of internal spectral correction file being applied seamlessly. I haven't done any confirmation measurements of the actual profiling, but Czornyj on luminious landscape forum compared it to his Jeti Specbos 1211 and found the calibrations to be well aligned.

source: https://forum.luminous-landscape.com/index.php?topic=145578.0
Click to expand...
Any luck finding the EDR, CCSS or CSV file Apple is using?

IDNK if Apple has some system utilities like Windows where you can choose a running executable and list all file descriptors that executable has opened.

There seems to be a few sqlite files with spectral data for a "HID" meter (which is what i1d3 colorimeter is, like a mouse or keryboard regarding communication):
/System/Library/HIDPlugins/ColourSensorFilterPlugin.plugin/Contents/Resources
but info table says 2017 and current QLED from Apple seems to be younger.
 
Last edited:
Nevermind, its True Tone controller, not i1d3.

Obvious (facepalm), it was here: "/System/Library/CoreServices/Pro Display Calibrator.app". There seems to be no EDR or user updatable files. Executable mentions matrix colorimeter correction, but not sure if this about higher end colorimeters like kleins.... but it may be tacking about i1d3 too (like default behavior for ColorNavigator) and that wikll be disapointing since it's not "portable" between colorimeters
 
Last edited:
I looked at this and can give a few details, but I mainly found ideas for a later deeper investigation. It probably only writes the calibration data at the end of the 2-hr run, and today I didn't have time to wait that long. So I just ran it for a brief period to capture what I could using XCode Instruments and other tools. Observations and findings:

/System/Library/PrivateFrameworks/SpectroKit.framework/Versions/A/SpectroKit: drives the colorimeter probe (Calibrite Display Plus HL identifies as i1Display 3 family; the framework’s resource folder contains I1D3Mapping.txt.

Analysis of fs_usage file access data when running Pro Display Calibrator:

/System/Library/PrivateFrameworks/SpectroKit.framework: typeA.edr is the panel-family spectral correction selected for Apple displays.

~/Library/Containers/com.apple.displaycalibrator/Data/.../Resources/CIEXYZ1931_2deg.json: the 2° standard observer for converting spectral measurements to XYZ.

Code analysis indicates there is a PAYLOAD_CALIBRATION_CCM.bin and CSV measurement dumps; I can't yet tell where. These are likely relegated to diagnostic or debugging purposes only. In normal use, they are not present.

It appears the calibrator produces and ships to the display a bundle of info, particular to each reference preset. This includes: CCM, a 3×3 float matrix (36 bytes). However, these bundles seem to be shipped directly to the monitor. There is no persistent Mac-side database or plist.

PRC / PR-CLUT is a primary-correction lookup table. It may be written directly to the display's flash memory.

Gamma LUT: a per-channel transfer-function correction LUT.

BLU LUT: backlight uniformity/luminance correction

CalibrationInfo envelope: Includes name, description, primariesData, whitepointData, headroom, luminance, temperature, calibrationType, timestamp, minorVersion.

It appears that all of these are pushed across the cable to the Apple monitor, and under normal operation, nothing is written to the disk.

~/Library/Containers/com.apple.displaycalibrator/Data/Library/Preferences/com.apple.displaycalibrator.plist holds only behavior toggles, not the calibration. The calibration itself is on the display.

There is apparently some debug method whereby some calibration files are written to /Downloads/calibratorFiles/<MMMdd.HHmmss>/, but I don't know what invokes that.

However, it is possible to obtain other debug and status information using the terminal's log show command. E.g:

sudo log stream --predicate 'process == "Pro Display Calibrator"'

But doing this incurs several problems, so I recommend this not be done. Issues:

- The resultant info is only useful if you have extensive internal understanding of the macOS frameworks.
- Once begun, the Pro Display Calibrator "Full" mode cannot be aborted from the keyboard, and it runs for nearly 2 hours. It is possible to abort if you connect via SSH from another machine, and assuming you have all the syntax and commands at hand.
- When I tried it, there was a storm of extraneous messages from IOKit, so the s/n ratio was about 97% noise.
- It appears there is a sidecar sidecar XPC service named com.apple.xpc.displaycalibratorXPCService. Pro Display Calibrator communicates with this separate process, and that is where all the action happens. Any target for logging or debugging must focus on that process. I may do that next time, provided I have the time for such explorations.
 
type B,C,D,G,H = WLED PFS ("old" P3 displays)
type E,F = QLED (newer XDR, wider gamut but not the same AdobeRGB coverage as current WLED PFS, Apple has prioritized full P3 coverage)
type I = WLED PFS 9x% P3, like typical "cheap" multimedia monitors nowadays (equivalent to Xrite's Panasonic VVX17P051J00.edr from years ago)

Howto: copy EDR to some folder, ArgyllCMS "oeminst -v -c *.edr", then "specplot typeB.ccss" and such.

Mistery solved. Apple calibration app has all ingredients to be able to properly measure its displays. Also, theoretically, i1d3 could be able to use Apple CMF as std observer instead of CIE 1931 2 deg.
 
Last edited:
Wish they would open this to the M4 iPad Pro as it has an Ultra Retina XDR display. In reference mode it matches a calibrated display pretty well even w/o proper calibration.
 
Clark Bierbaum said:
Wish they would open this to the M4 iPad Pro as it has an Ultra Retina XDR display. In reference mode it matches a calibrated display pretty well even w/o proper calibration.
Click to expand...
With the M4 iPad Pro, you can do a "fine tune" calibration with a probe, then enter the measured and white point x/y values. That adjusts for luminance and white point drift. The results are written to the iPad hardware. It's not a full calibration, but it's better than nothing. The procedure is shown here:
 
Remember that Xrite/Calibrite software do not support any of the backlights of modern Apple displays with i1DisplayPro of colorimeters, excluding i1Profiler 3.7.1 (& older) with the "type I" display (new Neo macbooks?), so despite their claims including "ambassador" videos... it's fake support.
When you set "PFS phosphor" or "Mini LED" in Calibrite Profiler it is using RG_phosphor EDR (GB-LED backlight) under the hood.

To take an accurate reading of white you'll need other tool, like DisplayCAL or if you are not afraif of command-line, get a big 255 white patch on screen, then run ArgyllCMS command spotread with -X param and the path to the CCSS (equivalent to EDR in message #11), place the colorimeter in the white patch & measure.
Calman & LightIllusion have support for PFS & QLED backlight, but LightIllusion suport is more flexible using CSV equivalent to ArgyllCMS. I do not remember how to take remote measurements with these apps (specially if you have only windows versions and you ned to take a "remote" measurement), but certainly you can do it with them too.
 
Thank you Andrei for chiming in and clearing it up, very helpful.

Will there be future support for Pro HL too and would that be Apple's or Calibrite's decision? Thanks.
 
Calibrite Display Plus HL Is now $80 off at Amazon and B and H, $259.00 in the US. I ordered from B and H, will be interested in the results.
 
You must log in or register to reply here.
Back
Top