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: Vuescan: "device RGB"

On Tue, 27 Mar 2001 19:15:35 -0800  shAf (michael@shaffer.net) wrote:

>  but while we all recognize with kudos the
> advantages of VS, we need to also recognize its weakness and lack of
> scanner characterization.

Vuescan uses a hard-coded tristimulus transform derived from empirical testing 
of each scanner supported, though this is presumably not the case for scanners 
which happen to be supported just 'cos they understand SCSI commands for 
another model. He hasn't hived the matrix off into a profile as he doesn't want 
competitors nicking it, at least that is what he has said. 

I don't find any problem with this in practice - what emerges from Vuescan is 
tagged as being in a selected colour space, and it is, as the image data has 
been run through the transform and then into the tagged output space. It is 
just that how it got to be there cannot be reverse-engineered without delving 
into his code. 

The VS workflow goes
- raw scan (scan+n.tif, if you opt to write it to disk)
- apply hardcoded scanner transform
- apply selected output profile
= output file Crop_n.tif, tagged with output profile

AFAICS you only *need* the scanner profile if you want to work with the first, 
raw scan. And you'd only want to do that if you had some means of 
characterising your own, personal scanner and making your own superior profile 
for it.


Tony Sleep
http://www.halftone.co.uk - Online portfolio & exhibit; + film scanner info & 


Copyright © Lexa Software, 1996-2009.