I recently tried a tweak to the documented HDR10 3D LUT workflow that has been discussed here. I did this on a LG C4 42" TV.
I did the full 22pt with Peak Brightness = Off. Then I profiled the TV with a WRGCYM patchlist that I generated (also pb off), and then uploaded a peak brightness value that was ~30-35 nits lower than I actually measured. This was done to push up the Peak Brightness Off range since there appears to be a tracking mismatch between Peak Brightness Off and Peak Brightness High. (you can have a perfect 22pt pre-cal with pb on, but once you turn pb off the tracking will be different, and vice versa.) The LUT was generated with 2.2 gamma. I used Map Space and Gamut Mapping disabled.
The typical workflow has you doing the full 22pt with Peak Brightness = High, profiling with pb off, then uploading a LUT with a gamma value lower than 2.2 (~2.1 - 2.17 can sometimes work.)
I think both methods can give similar results but I was just curious if anyone tried it my way, and if so, if you found those results better or worse than the typical workflow.
Additionally, I'd love if anyone else had any tweaks to the process that resulted in better validation for them. The HDR10 3D LUT calibrations on the LGs are tricky. My guess is there are other features in Colourspace worth trying that I haven't yet. | 