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: Re: Nikonscan v VueScan



Yes I've tried curves of all kinds.  I just wish I hadn't pursued the "flat
is best" school, because it was a very serious waste of time (well, OK, I
learnt to reject it).

Your point about Nikon and Canon performing "image processing" that matches
what D&P stores achieve is partially true, in my view.  The stores clip very
heavily (if need be) and show a depressing lack of intelligence.  Prints,
fundamentally, show much less dynamic range than a good monitor.  Nikon Scan
scans, even when they are clipped, are not suffering to the degree one
normally gets from the store.  Nikon and others are, I would guess,
targetting their output at what a monitor can do (i.e. around 6-7 stops as
far as I can tell).

I seems ironic to me that negative films, when scanned, come out of
something like Nikon Scan showing the same range of scene f-stops as the
lower contrast slide films.  And, viewed on a monitor, an image whose source
scene has been captured at around 7 stops has a kind of vitality that a
flat, S-shape-curved image that is holding more dynamic range from the scene
(say 9 or more) just doesn't have.  It's the forensic-versus-pleasing thing.

I'd just like to encourage some peeps not to think forensically about
dynamic range when they're scanning.  Once you de-prioritise the flat-scan,
there is a chance you'll like the "default" images that some scanner
software is capable of providing a lot more than half an hour's work on a
single image with a flat scan from Vuescan can render.

Sure negatives can capture more dynamic range than you can see on your
monitor, but in squeezing that range onto your monitor all the punch, life,
vitality seems to vanish.  I'm sure it's bad enough that when printing one
has to squeeze the dynamic range even more...

This whole thing is similar to comparing, say, Provia 100F projected versus
a print of the same scene...

My favourite image viewing method is slides on the end of a loupe against a
light table.  But that's completely impractical...  And it misses out on the
"scale" thing that I like...

Jawed

> -----Original Message-----
> From: owner-filmscanners@halftone.co.uk
> [mailto:owner-filmscanners@halftone.co.uk]On Behalf Of Colin Maddock
> Sent: 22 November 2001 07:12
> To: filmscanners@halftone.co.uk
> Subject: filmscanners: Re: Nikonscan v VueScan
>
>
> >Jawed wrote:.........By vitality I don't merely mean
> contrast/black-point/white-point.  I also
> >mean the nature of the tonality of the image.  Something related to the
> >question of "gamma" and also the inherent S-shaped response that
> all films
> >have (so far as I know).  So, all the effort I put into
> obtaining the full
> >tonal range in a negative (in the form of a flat scan) is wasted
> because I
> >get distinctly more pleasing images from Nikon Scan.
>
> Have you tried an S correction, stretching the mid-tones of a
> VueScan scan? This definitely will put back the missing
> "vitality", in my experience.
>
> I too find that Vuescan gives a flatter looking image than my
> scanner's own software, Canoscan, but by experimenting with the
> S-correction in Curves, and perhaps increasing the black and
> white clipping, a Vuescan scan can be made to look identical to a
> Canoscan scan, if that is what one wants.  It seems that writers
> of filmscanner software (Nikon and Canon at least) have decided
> that the "default" scan needs some stretching in the mids, and a
> degree of black and white clipping. That is what we get from most
> commercial D&P stores, going by the prints, so they have gone
> some way towards matching that look.
>
> At least withVuescan, both these factors are under our control.
>
> Colin Maddock
>
>
>
>




 




Copyright © Lexa Software, 1996-2009.