Hi Steve,

I'm able to verify a LUT generated with a custom PQ colour space with modified nits fine without issue.
I select it as the Target Gamut and use Active LUT. It validates fine.

However if I try to do the same using a colourspace that has a multiplier (projection HDR), the target values shown in the EOTF graph when verifying against it are all unexpectedly low.

Is this a known limitation when using the multiplier function?
Is there a way validation can be approached that gives valid results?

Thanks for taking a look.

Take care,
-Jonathon

Validation example.

Attached file: Custom colourspace

Attached file: Validation

If your target luma is set at 250 (the one you have declared in your ST2084), the validation comes correct:


Anyhow it looks like a bug .... as if I re-select your user colourspace, CS set it at 130, that looks wrong.

PS: your mulplixer value is slightly form the one CS calculates automatically selecting the profile: 1822 vs. 1820

Thanks ebr9999!

Setting the target to 250 makes sense as that's the target of the custom colour space.

What is obvious in the morning is concealed at night.

: )

Thanks again!

The 130 it shows makes sense as that is the actual measured peak white nits of the profile.

I'll flag that as a bug.
As far as I am aware, the Multiplier should be taken into account.

Steve

I seem to remember that when you chose a PQ colour space as the target then this target peak luma was then statically set for the profiling. That always made chasing down a perfect verification a bit tricky as the peak was locked into the space. In a way, having it unlocked is useful, though having it default to the colour space target as a starting point makes sense as well.