RE: filmscanners: Sprintscan 120 and new negative proile scheme

["Austin Franklin" <darkroom@ix.netcom.com>]

> With one film term for transparencies and color management, individual
> film characteristics is exactly what you do get.  *Effective* film
> terms for color negative films will get closer to a specific films'
> characteristics, not further away, and the problem to solve is
> ineffective film terms.

What do you believe "film terms" are?  There are two issues here (well three
actually).  One is the film it self, two is the image on the film, and three
is the scanner.  Of course, I want to color correct for the image on the
film, due to lighting or whatever...and I want to color correct for the
scanner.  Setpoints and tonal curves are not film dependant, they are image
dependant, and one setpoint/tonal curve for one image may not be the correct
setpoint/tonal curve for another...even on the same strip of film.

> The Leaf was designed before practical color management.  Scans from a
> correctly calibrated and color managed scanner will look very much
> like the original when you first bring it into PS unless you've worked
> on it in the scan software.  Who wouldn't want that?

I get that with the Leaf now, with no scanner color management.  I am the
scanner color management!  Scanner color management is somewhat dubious,
IMO.  Monitor, I agree with, printer, paper, ink, yes, those are all
somewhat consistent...more so than film!

I do not believe you can characterize a film such that you are color
managing it in the same way you are with the monitor/printer etc.  Those are
all deterministic.  Film is image dependant, and is far from deterministic.
Too many variables, lighting, exposure, development etc.

Unless you truly profile/characterize a film/system (which I do BTW) for a
consistent set of conditions (or include a color chart on every frame), I
believe it just can't "work".  There is far more to it than providing one
film profile for everyone to use!