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]

[filmscanners] Re: Fuji Superia in Vuescan

On Sat, 13 Apr 2002 17:58:01 +0200  Alex Zabrovsky (alexz@zoran.co.il)

> As I mentioned previously, I tried the Reala settings and regular
> White Balance unsuccessfully. Perhaps this is something to do with the
> issue
> of trial version against full one ?

No, I'm sure it is identical

> Or, the SS4000 support for Superia emulsion is eventually better then
> for IV
> ED in Vuescan for some reason ?

Unlikely. Possibly it is that the IV ED which doesn't cope as well, but I
wouldn't expect that to be a major factor.

The most likely explanation is that you are expecting a well-balanced and
colour-corrected scan to be produced by Vuescan, without further work. This
is a lot to ask of any colour negative + scanner, it just isn't the simple
translation of dye densities ->RGB values that tranny is because the neg is
far more mutable, it's the raw material for an image, and scanning or
printing is interpretive.

To pick just one parameter, density range. Colour neg can record a subject
brightness range well in excess of 10 stops, sometimes 12+ stops. Colour
neg is designed to massively compress those 10+ stops into a low-ODR dye
image as a halfway house to the even more compressed reflectance range of
paper prints.  With a half-decent scanner you can scan this so all of those
10 stops range are present within the scan, but on screen it will look
flat, dull and unsaturated (there will be colour casts as well, most
likely, too).

With tranny, the film itself can cope with usually rather less than 5 stops
range because it's designed to reproduce an ODR which the eye finds
acceptable in a projected image. Scan that and it will look acceptable,
because the brightness range of the tranny is already close to what looks
right on a monitor.

With colour neg you will need to adjust levels to get near to an
acceptable-looking image. If this is done automatically by software, the
software is making decisions about discarding shadow or highlight detail or

Vuescan's best use IME is to get it all so you can make those decisions
manually in PS. This it does very well with Superia 400, especially if you
scan to 16bit/ch. Colour correction is done only to the extent only that
further manual adjustment is easy because the mask and non-linearities of
the scanner/film combination have been adjusted out.

What it doesn't do especially well is produce a fine-looking, tonally
balanced, colour accurate scan from colour neg. which needs no further
work. If it did, I wouldn't use it, 'cos I want to do it to suit the image
and how I want it to look rather than have it fixed up according to some
hardwired algorithm.

If you like, email me one of your unhappy scans (please resample down to
7.5x5"@300ppi and Jpeg quality 6 to keep filesize ~.5-1Mb, 8bit/ch is fine
for this) and I'll have a poke around.


Tony Sleep
http://www.halftone.co.uk - Online portfolio & exhibit; + film scanner info
& comparisons
Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe 
or 'unsubscribe filmscanners_digest' (as appropriate) in the message title or 


Copyright © Lexa Software, 1996-2009.