I'd like to create a 3D LUT for an OLED panel (Flanders/LG WOLED EP/BP series) that can replicate color from an Apple Display P3 device (XDR display, Studio Display, iPhone/iPad). Many clients review work this way, and it would be a helpful option to see these colors on a mastering display. I've gone into an existing .bpd preset (sRGB) and modified the primaries to reflect the Display P3 primaries (based on https://en.wikipedia.org/wiki/DCI-P3), and kept gamma at 2.2. The latter as I understand it, should also match sRGB ("Pure Power"?), but it appears one can only specify 2.2.

The result (from a 17^3 raw profile on a 32BP95-E probe matched to CR300) is a LUT for which primaries more closely resemble AdobeRGB than they do P3/2.2: the reds are more pure red, and less saturated/skewing orange as they appear in Display P3. Have I gone about this correctly, and/or is there a better way to match? I could profile my MBP XDR screen, but I don't have offsets/EDR or a good spectro, so hoping for a better route.

Andrew Huebscher
Los Angeles, CA

From the website info on the 'Grading Calibration?' page.

"Apple have what they call displayP3 colour space, which incorrectly combines the P3 gamut with the encoding sRGB compound gamma - not even the correct display defined sRGB gamma, which is a power law 2.2 gamma."

The gamut is P3, but the EOTF appears to be using the incorrect Encoding sRGB values?

If your display has a different gamut to P3 D65 it is incorrect in some way.

Steve

Is this to say they have incorrectly stated how they arrive at that profile? Or is it that there is an inherent conflict in combining that gamut with EOTF?


Isn't my display's gamut overruled by CS profiling in raw and subsequent loading a LUT?

Apple P3 just uses an incorrect EOTF.
It's that simple.

So, Apple Display P3 is standard P3 Gamut, with what looks to be the sRGB Encoding EOTF.

Steve

To be a bit more specific:


The link to Wikipedia doesn't seem to work, so I have no idea what values you have used - But the correct values are within ColourSpace for P3 D65.
And I do not understand the comment "kept gamma at 2.2. The latter as I understand it, should also match sRGB ("Pure Power"?), but it appears one can only specify 2.2." What do you mean exactly?
But be aware of the above info on Apple using the wrong Encoding EOTF.


The native response from any wide gamut monitor will nearly always be a gamut that is not any standard colour space, and will be close to a 2.2 power law - all screens natively are a 2.2 power law EOTF - that is how they are manufactured.

Any further help?

Steve

Apologies - I copied the link wrong. This is the correct page: https://en.wikipedia.org/wiki/DCI-P3
I've verified the primaries shown there are a match to CS's P3 data.


What number would I need to specify for gamma in a CS preset that will mirror's Apple Display P3?


Is a "2.2 power law EOTF" the same as "2.2"?

As Apple use an incorrect compound Encoding EOTF, you would have to use the Parametric EOTF capability of ColourSpace to mimic that.

And yes, '2.2 power law EOTF' means '2.2'.
The alternative is the (incorrect) compound Encoding EOTF Apple appears to use.
(Or a PQ or HLG, or BT1886 with lifted black EOTF...)

Steve

Thank you, Steve. This clears the gamma question up.

I'm still unclear about why I'm seeing different reds between the units (ASD vs. LG BP95). My LG is out for service, so I'll have to wait to re-profile each to understand which is showing an incorrect red for P3 (my money is on the Apple).

Can you recommend a spectral file to use with an i1D3 when profiling the Apple Studio Display... GB-LED?

One monitor is therefore not calibrated correctly...

But, remember Apple has ColorSync always active, so messes with any images sent to connected displays.
That obviously includes TPGs.

And no, unfortunately I have no knowledge as to what Correlation File would be close to the Apple monitor.

Steve

Hi Steve,

I also would like to creat a Display P3 preset in ColourSpace that accurately targets Apple's Display P3 (which uses the sRGB compound transfer function) – but I noticed there isn't a native preset for this based on above posts. It seems that, despite Display P3's popularity, users are forced to manually "hack" it by tweaking parametric EOTF values that I haven't seen documented elsewhere.

Wouldn't it be easier if ColourSpace natively included an Apple Display P3 colour space preset or, at least, if the EOTF dropdown under "Target Gamut" were selectable? For example, once "DCI P3 D65" is chosen as the target gamut, an "sRGB" option would become available for the EOTF. Right now I can't seem to unlock that dropdown once a target gamut is selected (or maybe it's not possible at all).

After researching for a long time, I got some help from ChatGPT to figure out what values might be needed for the parametric adjustment. Here's what I ended up with – could you please confirm if the below write-up and values are correct, and if not, provide the correct values? I have also submitted a request for parametric examples, just in case this is already available. (Update: I received the parametric examples, none of them documents a srgb curve for Display P3).

Here's chatgpt's write-up and values:
To mimic Apple's Display‑P3—which uses this "compound" (i.e. piece‐wise) sRGB curve—you need your custom Display‑P3 colour space in ColourSpace to use a parametric EOTF that follows the same rule.

In practical terms, when you "add" a Parametric EOTF to your colour space preset, you can enter parameters that represent the inverse sRGB function (i.e. the display EOTF):
1. Breakpoint: 0.04045
– Below this, the function is linear.
2. Linear Segment:
– Slope = 1 ÷ 12.92 ≈ 0.0774
3. Non‑Linear Segment (for values > 0.04045):
– Use a power function with exponent 2.4
– Add an offset of 0.055 before scaling, so that:
output = ((value + 0.055) ÷ 1.055)^(2.4)

If your parametric editor uses a form like:
  y = a · (x + b)^g + f  (for x ≥ breakpoint)
then you can choose:
 · b = 0.055
 · g = 2.4
 · f = 0
 · a = 1 ÷ (1.055^2.4) ≈ 0.879

In summary, to have your custom Display‑P3 preset mimic the sRGB (compound) EOTF:
– Set the breakpoint at 0.04045
– Below that, use a linear slope of ~0.0774
– Above that, use the function:
  y = 0.879 · (x + 0.055)^2.4

Enter these values into ColourSpace's Parametric EOTF editor. This will cause your calibrated target to follow the same "incorrect" compound sRGB curve that Apple uses for Display‑P3.

Thanks for your help and any insights you can share, I really hope that the above suggested EOTF dropdown menu tweaks could be added so that the parametric EOTF "hack" isn't required in the future.

Thanks and looking forward to your reply.

As Apple P3 is a bastardised colour space, and not a standard, we do not include it.
Apple have also not actually defined exactly what the EOTF uses is.
If they truly use the Encoding sRGB EOTF, that is just wrong.
But as this has not been defined, we cannot state if you hack is accurate, or not.

Steve

Hi Steve,

Thanks for your response. I understand your concerns regarding Apple's Display P3 being a "bastardised" colour space and your reluctance to include something that isn't a formal standard. However, as you know, Apple's Display P3 is now used on millions of iOS devices and has been clearly defined by Apple. For example, Apple's own documentation states that Display P3 uses the sRGB transfer curve (approximately a 2.2 power law). You can see this on Apple's developer page here:
displayP3 – CGColorSpace

Additionally, the Wikipedia entry on Display P3 describes it as using the sRGB transfer curve and a D65 white point:
Display P3 – Wikipedia

Given that this profile is so widely deployed (starting with the iPhone 7) and that its EOTF is conclusively documented as using the sRGB curve, wouldn't it make sense for ColourSpace to include a native preset for Apple's Display P3? This would eliminate the need for users to manually "hack" the parametric EOTF values—which, as far as I can tell, should be:

 · Breakpoint: 0.04045
 · Linear slope: ≈ 0.0774 (i.e. 1 ÷ 12.92)
 · For x > 0.04045, y = 0.879 · (x + 0.055)^(2.4)

Including a preset or allowing the EOTF dropdown (so that when "DCI P3 D65" is selected the option for sRGB becomes available) would greatly help users target a colour space that is both widely used and well defined by Apple.

I'd appreciate it if you could reconsider this—instead of debating whether Apple's approach is "right" or "wrong," why not support the common colour space that millions of devices already use? As a ColourSpace paying customer, having a built‑in preset for Apple's Display P3 would simplify workflows, especially for clients who want their images to display correctly on Apple devices.

Thanks for your time, and I look forward to your thoughts.

Unfortunately, you have defined the issue yourself.

"Apple's own documentation states that Display P3 uses the sRGB transfer curve (approximately a 2.2 power law)."

That immediately proves the problem.
Is it sRGB's Encoding EOTF, or a power law 2.2 EOTF, or something else Apple have decided to use?
We have no idea.

But any user can define whatever target colour space they want, and save that as a User Preset - be it right or wrong.

Steve

Hi Steve, thanks for your insights again.

I realize now that including the word "approximately" in my last message was my own interpretation and wasn't warranted. Apple's documentation and Wikipedia entry clearly state that Display P3 uses the sRGB transfer curve, with no qualifiers suggesting it's only an approximation. By definition, Display P3 uses the exact sRGB EOTF.

Given this clarity, I'd like to circle back to my request for an Apple Display P3 preset. Since Display P3 is a well‑established standard used on millions of devices, it would be a valuable addition for Colourspace users like myself who need to target that colour space accurately.

Thanks for considering this.

Unfortunately, that still doesn't define the EOTF actually used...

Is it sRGB's Encoding EOTF, or a power law 2.2 EOTF, or something else Apple have decided to use?
We have no idea.

Steve