Re: 10 bit logarithmic format vs. 16 bit linear format

not strictly true mathemetically so not sure about log and linear then being similar quality. suppose what I quoting here relates to film rather than video a lot more as darks in a cinema are much more noticeable

I am no means an expert on this, just thought the explanation was simplifying it a bit


the relative brightness change between linear as it works across dark ranges is anything but even and can move into more perceptible banding. basically as you move towards blacks it wont record as much information unless outputting to a articially large range (using look up tables for gamma correction on monitor) for later outputting in log.

the relative brightness change threshold needs to remain below 1.00 for there to not be any imperceptible banding, this is not the case with 8bit linear formats and with 16 bit formats it is "just moving the problem around - it wouldn't change the fundamental nature of it all" p270 digital compositing for film and video

logarithmic colour spaces have equal relative brightness changes and have none of these problems in the darks with banding after colour corrections

my advice is read that book, has great explanation of it.