On 4/21/11 at 6:38 PM, (Francois Lord) wrote: > You're not supposed to get significant data under 95. If > you do, I think your scans were badly done.
hmm, i'm not sure.. actually i seem to remember reading somewhere (was it steve wright's book?), ie that that 95 was chosen to allow for grain chatter below that. i'm looking at scans here from an arriscan that have grain chatter as low as 61. if i set the blackpoint in the loglin to 95, the scans look visually accurate with the black level around the place i'm expecting them to be, but some of the grain has negative values.. if i set it to 61, i dont have any negative values, but the shadows now look washed out and i get significant noise in the blacks. so as hugo said, we need to preserve those values, and setting the black point to 60 makes it hard to match it with other elements. i understand that the math gets wonky with some of the operations if we allow negative values, so i'm not sure how to best deal with that. btw, just a thought, i could imagine that on the original specs this was not a problem because a lot of operations happened in native log space on the cineon machines ? ++ chris _______________________________________________ Nuke-users mailing list [email protected] http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
