- This topic has 83 replies, 5 voices, and was last updated 2 years, 3 months ago by
.
-
Posts
-
Your ICC profile created by displaycal is present in the presets in the red frame ?(photo attached)
Your mac like many other good monitor had neutral grey and more or less gamma on track (2.2)… so you just need to fix white from that cool white* to CIE 1931 2 degree D65.
Since you can edit those presets, copy one like in #32315
Then choose as white point custom xy and with trial & error figure which setting gives you D65. IMHO the fastest way is to open Pages or GIMP with a white big patch, then open console and run manually Argyllcms “spotread” with “-X macbook.CCSS” parameter (spectral correction, put the path where you have that CCSS stored). Measures are made by pressing any key on console window. try, move, try move… Start raising a little X coord and raising just a tiny bit Y coord.Once white is fixed, run measurement report (with WLED PFS CCSS) against default profile. If all is OK (excluding measured vs display profile due to macos bug about WTPT x CHAD) save that preset as the final everyday use D65.
It would be like “HW cal” although I’m 95% sure that such correction is running on M1 GPU somehow, like windows night mode.
*maybe D65 CIE 2012 2 degree, can be verified on command line with spotread changing observer, but you may ignore this.
- Offline
Your ICC profile created by displaycal is present in the presets in the red frame ?(photo attached)
Yes, but it’s for the external display it do seem like the internal display can’t get profiles I’m guessing that you should use the corrections there according to Apple you can correct it in that window.
My guess is that this will get fixed in next release / fix from apple.Hello everyone,
I’m following the discussions here and on Macrumors. I don’t have a new MacBook Pro here to test, but have been trying to gather as much information as possible and corresponding with other users in order to find if a traditional color managed workflow is still possible on those displays.
One of my contacts tried to measure the display using the latest ccProfiler and both an i1Display Plus and an i1Pro 3 Plus spectrophotometer and both profiles show a much smaller gamut than DCI P3. I’m puzzled by that. I was expecting the at least the spectro would work fine with those new MiniLED displays.
So my question for you is: have you been able to achieve the full display gamut on your resulting profiles? And which measurement device and correction have you used? If you could share the actual profile that would be great too.
Hello everyone,
I’m following the discussions here and on Macrumors. I don’t have a new MacBook Pro here to test, but have been trying to gather as much information as possible and corresponding with other users in order to find if a traditional color managed workflow is still possible on those displays.
Default ICC profile (which laptop display seems to be tracking in a relative white point white), fix white on preferences as instructed in #32350
One of my contacts tried to measure the display using the latest ccProfiler and both an i1Display Plus and an i1Pro 3 Plus spectrophotometer and both profiles show a much smaller gamut than DCI P3. I’m puzzled by that. I was expecting the at least the spectro would work fine with those new MiniLED displays.
i1Pro3 (or 2 or 1) using ccProfiler is taking 10nm measurements, it is not measuring actual red channel. It is pointless to use an i1ProX spectro on an WLED PFS display with Xrite drivers. Argyll at 3nm can… more or less, but those Xrite’s spectros are not really suited for narrow spectral porwer distributions (WLED PFS, laser projectors and such)… they are more oriented for printing than aimed to measure newer displays.
Using ccProfiler and i1Displaypro colorimeter you can choose WLED PFS correction (95% P3 gamming displays). Green is slightly different but it’s the closer one to actual backlight between your limited xrite options.
Using same colorimeter in DIsplayCAL you can use actual backlight CCSS, same as any other P3 mac. (green channel modification from Xrite WLED PFS)Also there is the possibility of user error or misconfiguration. Use ArgyllCMS instead of ccProfiler. If something seems off try to use another computer + ArgyllCMS commandline “spotread” to measure full R , G & B saturation patches full screen in that macbook in a DisplayP3 image (so even on a color managed workflow like PS should show close to native gamut colors)
So my question for you is: have you been able to achieve the full display gamut on your resulting profiles? And which measurement device and correction have you used? If you could share the actual profile that would be great too.
Take a look on 1st page report with WLED PFS, if you do not care about a little cool whitepoint (CIE 1931 2degree) it is more or less on spot.
Calibrite Display Pro HL on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.Excellent, Vincent. Thanks for sharing.
It seems clear to me that Apple is pushing users towards a non color managed workflow that relies on a reference display, not dissimilar to what most colorists use.
What I’m trying to find out is a) how to correctly profile those new displays and b) what happens in the background when the user selects a display preset. How is the OS informing color managed apps about those gamut and white point changes if the ICC profile assigned to the display is a generic one? Does it change when the presets are altered?
So, for example, the generic display profile describes a DCI P3 gamut and D65 white point, while the preset is clamping it down in hardware to sRGB. Did Apple break the color management chain here, or am I missing something?
I’d love to have the hardware to test it here, but it’ll take quite a few months to arrive in my country and the price will be exorbitant.
Thanks again,
Excellent, Vincent. Thanks for sharing.
It seems clear to me that Apple is pushing users towards a non color managed workflow that relies on a reference display, not dissimilar to what most colorists use.
It’s a color managed workflow… that asumes that display is “exactly DisplayP3” on mac Displays so they can oversimplify even more than what they are doing now (with all those issues with detailed 3TRC or XYZLUT profiles). It is more or less what they have been doing for several years.
If you can tune xy coordinates of WP on display preset like someone show a few post earlier, all other calibration parameters seem to match default Display ICC.I mean: you can open an sRGB image or an PhotoRGB image and within rendering intent limitations it will show in gamut colors properly. It’s a color managed workflow.
Colorist expect that display (or display+ a software or HW LUT3D) matches content, for every content type (mostly Rec709 so it does not matter). That is a totally different game.What I’m trying to find out is a) how to correctly profile those new displays and b) what happens in the background when the user selects a display preset. How is the OS informing color managed apps about those gamut and white point changes if the ICC profile assigned to the display is a generic one? Does it change when the presets are altered?
All those preset are DisplayP3 (so default ICC profile applies) with sRGB/HDTV Rec709 exception => use preset run mearutemet report (which relies on default ICC display profile for that screen on OS)
I assume that those sRGB presets are for full screen non color managed apps (no colro managed video players or games for new M1 platform)So, for example, the generic display profile describes a DCI P3 gamut and D65 white point, while the preset is clamping it down in hardware to sRGB. Did Apple break the color management chain here, or am I missing something?
If it works like I assume, yes it’ll be broken like in displays with HW cal when you switch to a srGB/Rec709 calibration to play games, but you’ll have to run a measurement report (you or any other who reads this).
I’d love to have the hardware to test it here, but it’ll take quite a few months to arrive in my country and the price will be exorbitant.
Thanks again,
- Offline
there you go
It says nothing, it’s a piontless review… actually they say that its pattern generator app is color managed…and at the same time apply a preset that is supposed to limit colorspace (akin to displaycal simulate colorspace without applying simulated profile to display => it does not matter which preset is set, it will work in the same way). Same mistakes that are made in rtings reviews and useless Calman generated ICC profiles.
Matck06’s review is more informative (although setting a higer brightness will be more helpful)
https://hub-assets.displaycal.net/wp-content/uploads/users/matck06/2021/10/28/Measurement-Report-3.8.9.3-%E2%80%94-Built-in-Retina-Display-@-0-0-1728×1117-%E2%80%94-2021-10-27-22-56.htmlI mean, that review avoids answering the questions asked by Fabio Pili in message #32367, they are ussing a clone of Displaycal measurement report + simulate colorspace (CHECKED) + set simulatred colorspace as display profile (UNCHECKED).
What Fabio & others want to know is how fullscreen and not fullscreen apps will behave on sRGB-like presets, if there is true colospace limitation or it does nothing (relying on native gamut + display p3 icc profile + “tagged” content + Apple color management.. which makes those sRGB presets useless => they add no new functionality, you had that in previous macbooks)
If you do not have a measurement device you can try “seeing” it, use commandline “dispwin” using non color managed output (parameterless) on an sRGB-like preset. Easy to spot by eye if it is P3 or sRGB.
Exactly! I’ve written about color management and often receive questions asking for setup recommendations. I love fiddling with my own displays, measuring, finding the best hardware settings, etc, so I feel at home here. But few people want to invest time to understand color management and monitor calibration.
From an end user perspective, most people would love a simple preset that can get the monitor close to their intended use and then forget about any other intricacies. Apple likes to hide complexity and they know it’s an important selling point, as much as advanced users that fall outside the normal curve, like us, would complain.
What concerns me is if this would break color management at the OS level by not syncing the presets to the generic ICC profile that describes the monitor. As Vincent mentioned, if the monitor can closely match a known standard and we have a canned profile to describe that standard, we’re good for most use cases.
So if we choose sRGB or a D50 preset and the ICC profile doesn’t change, color managed apps won’t be able to properly understand and map gamut since the canned profile is still set at DCI P3 and D65.
I’m corresponding with an user and he tried my browser color management test on Safari. According to his results, when selecting the sRGB profile the ProPhotoRGB vs. sRGB bars on the first set are clamped to sRGB and appear mostly seamless. Any possible error aside, this indicates that the display gamut is actually being reduced in hardware.
What I’m failing to grasp is what happens in the background if the generic display profile doesn’t change? Where in the color management pipeline this information is being passed to the apps, Safari in this case, if not by using an ICC profile? Or maybe the profile changes, but the filename is still the same and my contact didn’t notice the changes.
In other words: if the user selects a sRGB preset, does Photoshop know that the monitor is now sRGB and can map gamut properly, or it would still think of it as a P3 gamut display and do an incorrect gamut mapping?
If someone in this thread has access to hardware, I think the best way would be to follow Vincent’s suggestion and generate a DisplayCAL verification report for the various presets. And maybe also check with the ICC Profile Info utility if the generic profile information changes in any way to match the presets.
So if we choose sRGB or a D50 preset and the ICC profile doesn’t change, color managed apps won’t be able to properly understand and map gamut since the canned profile is still set at DCI P3 and D65.
Not exactly, only for sRGB and HDTV Rec709 (sRGB-like).
All the other non HDR preset will work with the same Display P3 profile without issue because rendering from app to display always is relative to whitepoint, no matter if it is D50 or D65. Thats why you open a prophoto rgb (D50 WTPT) image in Photoshop with white and it looks with the same color as your display white (D65 for example), no mater if you are on win or macOS.
I’m corresponding with an user and he tried my browser color management test on Safari. According to his results, when selecting the sRGB profile the ProPhotoRGB vs. sRGB bars on the first set are clamped to sRGB and appear mostly seamless. Any possible error aside, this indicates that the display gamut is actually being reduced in hardware.
It’s a valid test but try to run a patch non color managed like “dispwin” from command line while in sRGB prest, no need to attach a measurent device, it will show a series of patterns, non color managed. If green looks sRGB green while dispwin is skipping al color managenent… it is actually emulating sRGB (on GPU I suppose like AMD’s GPU can do).
What I’m failing to grasp is what happens in the background if the generic display profile doesn’t change? Where in the color management pipeline this information is being passed to the apps, Safari in this case, if not by using an ICC profile? Or maybe the profile changes, but the filename is still the same and my contact didn’t notice the changes.
In other words: if the user selects a sRGB preset, does Photoshop know that the monitor is now sRGB and can map gamut properly, or it would still think of it as a P3 gamut display and do an incorrect gamut mapping?
Those are different tests, Safari is relying on Apple CM (which could run all these color transformation in GPU without app to notice it only relying on native gamut ICC and preset informartion, like AMD’s GPU sRGB emulation in MS Windows, returning to app the clipped RGB values although overall display stays P3 and profile remains P3)… while Photoshop it’s using its own CM based on display ICC.
Testing a 255 red & 255 green on a DisplayP3 tagged image, and also a 255 red and 255 green in an sRGB tagged image, in Photoshop while being on sRGB preset will be a good test of this because unless PS is updated for these “presets”, older Photohop versions using Adobe CM engine will rely on default display ICC alone => if it works as inteded (sRGB 255 patches show as inteded without further desaturation, P3 2555 patches are clipped to sRGB) those presets are actually changing RGB primaries in ICC profile .
An additional test would be changing display preset AFTER Photosop has started, AFAIK PS reads ICC profile on start (at least on WIndows), so you need to restart PS if you change defaut ICC.
- Offline
So did you figure something out how to add your custom ICC?
External displays are not a issue it’s the internal one that is. And it’s off also so this is something that I want to do.
