Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Thanks for taking the time to publish this.
BTW, the Delta-E on my 16" MBP came in at 0.8 with very low standard deviation.
Also I have some update to the guide so I'll probably do another video update soon. Would you mind sharing with me the icc profile, I would like to check it. Thank you!
 
Also I have some update to the guide so I'll probably do another video update soon. Would you mind sharing with me the icc profile, I would like to check it. Thank you!
I used your video but with displaycal with the correction matrix "pfs wled 99% p3 macbook to generate the icc profile from the" D65 photograph p3 "and also to measure the white point, on the other hand on displaycal it is absolutely necessary to leave as "as measured" in the tab "etalonage" (attached photo of my settings)

for measuring white point on displaycal, using the pre-calibration step that happens in the beginning of the calibration to measure white point and customize the x and y coordinates. Always use the PFS Phosphor correction for all measurements. This is optional.

I compared the results compared to my macbook air m1 and Gygabyte aero 15 oled I get exactly the same colors.

on the other hand hats off to Apple because the d65 p3 photography profile is already damn good.
 

Attachments

  • Capture.PNG
    Capture.PNG
    15.6 KB · Views: 230
  • Screenshot_20211109-143959_Samsung Internet.jpg
    Screenshot_20211109-143959_Samsung Internet.jpg
    332.9 KB · Views: 228
  • Capture d’écran 2021-11-06 à 22.09.17.png
    Capture d’écran 2021-11-06 à 22.09.17.png
    957.1 KB · Views: 207
  • Capture d’écran 2021-11-06 à 22.08.21.png
    Capture d’écran 2021-11-06 à 22.08.21.png
    1,021.8 KB · Views: 262
That is about the standard deviation. This is why we do the fine-tune.
We’ll see that was the problem, if the deviation of the cdm was too far, the fine tune would give me an error. So I wanted 120cdm, but actual was 110cdm, fine tune wouldn’t allow for correction. So I ended up setting SDR to 135cdm, to get roughly 124cdm and then fine tune allowed for correction.

I’m not being critical of your findings, just trying to get an understanding of the limitations of fine tune.

Also, I received an email from Calibrite, they’re claiming to calibrate from the default Photography setting. Not sure I agree with that either based on your final remarks in your video.
 
We’ll see that was the problem, if the deviation of the cdm was too far, the fine tune would give me an error. So I wanted 120cdm, but actual was 110cdm, fine tune wouldn’t allow for correction. So I ended up setting SDR to 135cdm, to get roughly 124cdm and then fine tune allowed for correction.

I’m not being critical of your findings, just trying to get an understanding of the limitations of fine tune.

Also, I received an email from Calibrite, they’re claiming to calibrate from the default Photography setting. Not sure I agree with that either based on your final remarks in your video.
I am working with calibrite & x-rite engineers, at the moment. Their knowledge based is limited for these machines. Oh and you are correct the fine tune would only allow you do to it with in a range. So did you create a custom preset already with 120 nits? I have tested this on numerous machines already and the values are usually really close.
 
So if I have say a rec2020 HDR file in ProRes or MKV and I want to play it back on the new MBP M1 Pro and get reasonably accurate colour. What software do I use to play the file with correct HDR colour and what do I set the Mac colour options to? At the moment any Rec2020 files I play look very yellow tinted.
 
Hello @David Abrams!
The Patterns Test Generator app looks great, I wish it was more affordable to test it with my own display calibration software. Does it have an open API for other programs to make it display the color desired?


For everyone else, I just released a tool to tune your MacBook XDR display white point without losing peak 1600 nits HDR luminance or adjustable automatic brightness.
I created it to fix the bad white point on my unit and color-match it with my other calibrated displays.

It's called XDR tuner, free and open source.

Discuss on this forum:

Get it on github:
 
  • Like
Reactions: sydlow
Here's a link to the video on how to set up a custom reference mode, fine tune calibration and then do a full calibration along with reasons why you might not want to do that yet
it is longer so use the timestamp to skep to calibration section
I don’t get it.
First you fine tune it for .3127 / .329 that I get.
But then when your using ccpeofiker your doing the profile with the white point for sRGB (.313 / 0.329) and not DCI-P3 D65 shouldn’t that be wrong?

Edit;
Ccprofiler got a new version but no release notes so don’t know what’s new
 
Can anyone comment as to why Rec2020 HDR files playback on the new MBPros with a really strong yellow tint?
 
It's not from a camera, it's rec2020 HDR graded files from Resolve. iPhone HDR plays back fine, rec2020 grades look very yellow when played back from ProRes or MKV, but they look fine on the Dolby monitor in the grading suite.
 
I am working with calibrite & x-rite engineers, at the moment.
Have you got any update about your work with the engineers?
There's a new version of calibrite v1.1, just wondering if it include any improvements related to this issue.

Very little info in the Read Me. Just says "Adds support for macOS 12 (Monterey)"
but this in the Known Issues:
"Table-based display profiles (including ICCMax Compatible profiles) do not work correctly on M1 Macs and have been disabled. This feature will be restored when table-based display profiles are supported on M1 Macs."

And, the app has almost tripled in size from 320MB to 860MB.
That's a pretty huge increase for apparently no new functionality.
 

Attachments

  • Screen Shot 2022-03-08 at 11.21.40 am.jpg
    Screen Shot 2022-03-08 at 11.21.40 am.jpg
    142 KB · Views: 109
Last edited:
I don’t get it.
First you fine tune it for .3127 / .329 that I get.
But then when your using ccpeofiker your doing the profile with the white point for sRGB (.313 / 0.329) and not DCI-P3 D65 shouldn’t that be wrong?

Edit;
Ccprofiler got a new version but no release notes so don’t know what’s new
This is the updated video with out further white point adjustment from ccProfiler / i1Profiler
 
This is the updated video with out further white point adjustment from ccProfiler / i1Profiler
I’ll watch it later, so my toughest was right that it was wrong to run 1profiler with sRGB etc white point as you have fine tune with DCI-P3 D65 before
 
I’ll watch it later, so my toughest was right that it was wrong to run 1profiler with sRGB etc white point as you have fine tune with DCI-P3 D65 before
Not quite. The white point that gave in the video was for Display P3, sRGB and Adobe RGB D65. In the guide I didn't fine tuned it to DCI-P3 D65. However, in the revised guide for i1profiler, my recommendation was to set the white point to native rather than defining it because if you defined it, the program will adjust it, even if the original was right on. Either way, matrix data is not out yet and calibration is still not recommended.
 
Not quite. The white point that gave in the video was for Display P3, sRGB and Adobe RGB D65. In the guide I didn't fine tuned it to DCI-P3 D65. However, in the revised guide for i1profiler, my recommendation was to set the white point to native rather than defining it because if you defined it, the program will adjust it, even if the original was right on. Either way, matrix data is not out yet and calibration is still not recommended.
I understand, but just to clarify the whitepoint for P3 D65 should be .3127 x .329 and sRGB should be .313 x .329 right?
 
I understand, but just to clarify the whitepoint for P3 D65 should be .3127 x .329 and sRGB should be .313 x .329 right?
x=0.3127, y=0.329 is for sRGB, Adobe RGB, and Display P3. @D65
x=0.3140, y=0.3510 for DCI-P3 (theater)
"sRGB should be .313 x .329 right?" That is a round up value. And it is important to distinguished P3 color gamut to either Display P3 or DCI-P3, they have different white point.
 
  • Like
Reactions: paalb
I wonder why there is no option to calibrate the display using an iPhone, like what tvOS can do?
 
I wonder why there is no option to calibrate the display using an iPhone, like what tvOS can do?
The premise of these display with reference mode, implies that these panel are pre calibrated from the factory already and that the user does not have the need to calibrate it further. Just choose the reference mode that one like to use and it is done. This obviously skips over the fine tune portion of things but that is not the goal here. Having an iPhone calibration like tvOS can be done but it would not make much sense, the calibration on tvOS lack many of the nuances of a good calibration, it is good for TV - content consumption but not for these laptop displays.
 
I wonder why there is no option to calibrate the display using an iPhone, like what tvOS can do?
Because the iphone isn't a colorimeter - it's adequate for TV viewing, but not for true color managed workflow.

The OOTB calibration/profile on the new laptops is really good for general viewing, but is definitely off from a true calibration.
 
The premise of these display with reference mode, implies that these panel are pre calibrated from the factory already and that the user does not have the need to calibrate it further. Just choose the reference mode that one like to use and it is done. This obviously skips over the fine tune portion of things but that is not the goal here. Having an iPhone calibration like tvOS can be done but it would not make much sense, the calibration on tvOS lack many of the nuances of a good calibration, it is good for TV - content consumption but not for these laptop displays.
In any case, it'll do wonders for external displays and display-less macs. It's darn way easier than the manual calibration that's currently available.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.