Brian, these are very informative videos. Thank you for your work! However, if you have the time and you're keen to find out yourself, could you possibly redo the first test of D-Cinelike, None and Art, using -2/0/0?
That way we would see which of these modes renders the most dynamic range. The clipping in your previous videos was due to Contrast being -2.
You could of course put Saturation to whatever you would like, in fact it would be very interesting to see the difference in color spectrum from -2 to 0, but Contrast should stay at 0, if I have understood the matter correctly.
Adjusting contrast will, similiarly to using d-log, put the recorded footage into a narrower dynamic range, with the same gamma curve, resulting in less possible gradiation and DR in the spectrum available to us. I think the actual info degrades with about 8% per stop from 0 -> -3.