Apache-Talk @lexa.ru 

Inet-Admins @info.east.ru 

Filmscanners @halftone.co.uk 

Security-alerts @yandex-team.ru 

nginx-ru @sysoev.ru 




      :: Filmscanners
Filmscanners mailing list archive (filmscanners@halftone.co.uk)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: filmscanners: Sprintscan 120 and new negative proile scheme

> Austin,
> All scanning software characterises film in some way as an attempt to get
> you near where you want to be. You can still use your individual artistic
> talents to effect the final product.
> In no scanner software of which I am aware will give you by
> default the raw
> data from the ccd.

David, raw data has nothing to do with "film profiling".  Setpoints have
nothing to do with film profiling.

> The raw data fom the scanner is processed through a
> matrix filter or profile.

What is a matrix filter?  The raw data from the scanner is thresholded with
the setpoints, then run through a LUT to correct for the non-linearity of
the CCD, then LUT'd again for the tonal curve adjustments you make.  You can
do the non-linearity correction before or after the setpoints are applied,
it doesn't matter.  This is all done on high bit data.  If you are getting 8
bit data, then the data is decimated from the full span of the data between
the setpoints, down to 8 bit data.

> The goal of these profiles and matrix filters is to recover correctly as
> much information from the film as possible, removes the base, do general
> corrections based on what it knows about the ccd/scanner system and film.

Er, right.  But you don't have to profile the film to do that.  The CCD is
already "profiled" in the firmware of the scanner.

I still disagree with film profiling.  How come the Leafscan has given
perfect scans for the past 10+ years with no film profiles?


Copyright © Lexa Software, 1996-2009.