Any interest in a "film negative" feature in RT ?

I much prefer scanners which don’t conduct the preliminary and secondary expansion you’ve described. The output image from the scanner should be densitometric in my opinion. Because it means on set exposure can correlate to a predictive output code value, and it’s simply less destructive. I’m particularly referring to normalising min and max density to min and max code values on a per shot basis, it’s a destructive process because the exposure may not contain d-min and d-max, therefore the relationship between scene exposure and code values would be all over the place. This process would only make sense if calibration is made on a per roll basis, using the exposed leader as D-max and unexposed perfs for D-min.

If the goal is to encode the color negative to the perceptual characterisation of a reflection print, then it may not be the case that D-min should sit at a code value of 0. For example, in the case of motion picture projection print, it makes more sense to sit D-min just above pure black, as when viewing D-max projected on a screen in a theatre, the surrounding light around the projection will always be relatively darker than the screen itself. So the experience of pure black in a theatre would likely translate to a 10-bit code value of around 32 or a bit lower.