Ivan Tsyba wrote: > So, first of all, target curve is offsetted to match display and target > curve black points. Then, if -g option selected, dispcal try to fit actual > display curve to the target curve by fitting actual output at 50% input (by > the changing it curvature without touching black and white points). When > it is done, actual gamma of display will be close, but not the same as > target gamma value. > Right? Yes. The -g option attempts to give the same overall visual response as the chosen gamma would on an ideal zero black system. Graeme Gill.