I'm using scanned film negatives to make a film emulation LUT. I'm trying to figure out if I should be using the base curve / filmic RGB modules if I want to just capture the captured densities and characteristic curve of the film negative. The seemingly-obvious answer would be no, but I wonder if the fact that I scanned it using a DSLR means that I somehow do need a tonemapping module to kind of revert whatever linearization the digital sensor is doing. Can anybody disambiguate?
Bonus second question: I'm particularly interested in emulating negatives for motion picture (and of course the proceeding print positive) and would be very keen to hear people's opinions on how best to use the negadoctor math to this end. Like, recreating a flat 10-bit DPX scan that an ArriScan would provide, but using Darktable.