This is something that's been troubling me. I can convert a legal Rec709 (64-960) through various steps into linear, that's fine. But what i'm struggling with a bit is in that conversion to linear how is dynamic range handled?
So from a cam with 6 stops of range the REC709 curve will be applied, but the same curve would be applied to a camera with 10 stops of range. Then within linear how do i peg the range to actual linear values so when i adjust exposure it reacts as it would in real life? So i have a shot which has around 10 stops of real range, it's shot through a picture profile, so the first thing is to profile that camera so i can find out what the actual range is and the curve that it is adding to the linear real world source to record it. Then i can produce a reverse curve which would be like a modified Rec709 because the picture profile has adjusted knee and shoulder. I can draw a LUT in nuke, but then how do i specify that the brightest point is 10 stops over the 0 point? I might be missing something really simple here. If i just take the footage, whack a REC709 on it then reduce exposure the scene does not react the way i would expect if the scene was converted into linear properly. What i see as being white on the footage should still be white if i take it down a few stops but it becomes grey. This to me seems that the linear conversion applied to it says that dynamic value of white is not the same as it was in the real world. I hope this makes sense? many thanks Paul
_______________________________________________ Nuke-users mailing list Nuke-users@support.thefoundry.co.uk, http://forums.thefoundry.co.uk/ http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
