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: Tips needed on difficult scan

> Hi.
> I posted a query in regard of LS-40 (IV ED) performance for the scanner
> users, however nobody answered yet. (I even though no LS-40 users on the
> List yet)
> Would appreciate if you will give your opinion about this scanner.
> I mainly care about his true dynamic range for the ability to
> pull out dark
> details (I shoot a lot of night scenic) against his noise level

Well, you caught me, I own an LS40.

I get the impression that noise in this scanner is *very low*, i.e. not
relevant.  Multi-scanning is possible (only with Vuescan) and in my opinion
it gives *no improvement* in image quality.  Others disagree.  There is a
difference - I perceive this difference as no improvement, that's all.

Do you shoot negs for your night scenes?  You will not have any problems
with noise.  BUT, if the film is relatively grainy, the LS40's 2900dpi
resolution *will* tend to exaggerate the grain in the film.  It's called
"grain aliasing" and can make the grain in the scan look significantly more
obvious than it looks in a print.  The LS40's scanner software, Nikon Scan
comes with an option called GEM (some acronym to do with grain, I forget the
precise name) which is tweakable.  I find that a setting of 2 reduces the
graininess very nicely while having no detrimental effect on detail in the
scan.  Some of my films have aged badly (I suspect mostly due to my lack of
care before processing) and show high levels of grain, requiring me to use
GEM at the max setting.  (Maybe I've misunderstood the kind of damage one
can do to a film by storing it badly.)

As far as I can tell, scanners with a resolution in the range 2400-3000dpi
are the scanners that will tend to exaggerate grain the most.

Additionally, the type of light in the LS40 (three LEDs that turn on and off
rapidly so that only one colour, red, green or blue, is shining at any time)
apparently tends to exaggerate grain (and defects such as scratches and
dust).  There is an argument for saying that if you run software that
corrects for grain (GEM) or corrects for defects (ICE) then the scanner
should resolve these features clearly - i.e. it makes the software's job
easier, since the problems are "clearer".  Dunno if this argument is true.
Dunno how much of a difference in image quality one actually sees when
comparing scanners of the same resolution with the same software but with
different lights.

> and the
> Nikon's known AF (insufficient DOF) problem from which 4000ED (and
> apparently LS-40) are reported to suffer from.
> Did you experienced DOF problem with either mounted/unmounted slides and
> negative strips ?

No.  I just can't be bothered looking and testing.  Life is too short.  Note
I used a "crappy" 28-200mm Vivitar lens from the mid 80s and a Canon P&S
zoom camera, so my lenses are hardly the most demanding.  If you were the
average member of this list you probably would notice that the LS40 has a
small DOF, though.  I freely admit to being fairly un-picky  - which is why
I didn't want to respond to your original email, since I think my views are
very much biased towards the lenient end of the scale.

> Does the LS-40 allows multi-pass or multi-sampling ?

Only with Vuescan.  You can get Vuescan to perform upto 16 passes.  You can
get Vuescan to performa a long-exposure pass (in an effort to see into the
dark parts of the film).  As I said before, I firmly believe these two
options do not provide an increase in image quality, under any circumstances
(slide or neg, good or bad exposures).

As for dynamic range, I personally believe that Astia is just within the
capabilities of the LS40, Provia F 100 is marginal and Velvia is just
outside.  Since I don't normally shoot slide film, my opinion isn't worth a
huge amount - but I have tested this using a number of slides for each film

A while back I posted the following:

http://www.cupidity.force9.co.uk/Scanners/LS40/tests.htm (yeah I know, life
wasn't too short, I got geeky)

which you might find interesting.  I have decided, on reflection, that some
of my comments on these pages relating to colour are unfair.  Not all of
them, mind.  My advice, therefore, is to ignore my comments on colour.

One other note: I used Auto Exposure off, when scanning with Nikon Scan
(hmm, I think I did...).  This has a slightly beneficial impact on colour
(in my opinion - though others on this list disagree).

It's also worth noting that Vuescan has been updated significantly since
these tests.  Vuescan will produce different looking results.  But I suggest
it will not affect the overall conclusion: multi-scanning and long exposure
are a waste of time with the LS40.

Have to say, looking back at these images, the Nikon Scan images strike me
as massively better than the Vuescan images in colour and contrast.

Have you searched the archive?  Can I be bothered to root out the web
addresses for the archive - no, not now...


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.