Ben Goren wrote:
Hmmm...as I understand it, for any given illumination level and ISO, there is exactlyone corresponding EV (and vice-versa).
Right, but the information flow is from left to right :- you input the light level, combine with other assumptions (ISO, intended exposure == object of interests placement in resulting tonal curve + whether incident or reflected light is being measured) and compute an EV. The fact that by making certain assumptions you can backtrack and estimate the light levels from an EV, doesn't make them safe equivalents (ie. to put it scientifically, they have different units). So to add a correct EV output to spotread would entail adding options to input the other parameters. At the moment, I feel this goes a bit beyond it's intended use, given the command line tool constraints. It would make more sense in a GUI tool context, where the assumptions can be parameters. Graeme Gill.