[filmscanners] Re: Black and white scans on LS4000 EDandotherissues

["Arthur Entlich" <artistic-1@shaw.ca>]

Actually, the issue isn't as clear cut as opinion or choice.

Diffused light does act more on diminishing the three dimensional stuff,
like dust, dirt and scratches moreso than grain.  This is because
diffused light sources bounce the photons around, and so they enter the
film at many angles, which literally allows them to get "around" small
objects like dust and dirt and even film base scratches.  The narrower
the layer of 'information" the less effect the diffused light will have.
  Therefore, the grain or dye cloud layers, which compromise a small
percentage of the full film thickness, especially when considering the
dust and dirt laying on top or beneath, will have much less softening
occurring.  So diffused light acts much less on grain than it does on
dust dirt or scratches.  SO, that's benefit one.

Secondly, collimated light amplifies aliasing.  That's not real grain we
are speaking of, but the grittiness created by aliasing due to the extra
  contrast and lack of anti-aliasing which diffused light can causes
around each grain.


What many report as "grain" is actually aliasing which, if anything, is
more of an error or defect, than the minor softening diffused light creates.

The small softening as a result of diffused lighting can be regained
with USM.  However, aliasing is not really removable once it becomes
part of the data.

The collimated light source the Nikon uses is not only creating problems
with defects showing up more (and in black and white films, since dICE
is not usable, this is a real chore to clean up) but it also (in the
case of the Nikon) also seems to be a factor in why their scanners have
such poor depth of field, not allowing for the full frame to be in focus
from center to edge.

I do not believe that the way collimated lighting is used in the Nikon
improves the final scan result in any manner.  I believe it adds to
grain aliasing (in both color and black and white, but moreso with the
later) adds to problem of limited DOF, forces one to use dICE (when one
can) (and possibly GEM, depending upon the degree of aliasing that
occurs) with many scans, and makes true black and white film nearly
impossible to scan without major cleanup work.  It also ends up with a
considerably more expensive scanner.

This is not just a trade off as you state with enlargers.  There are
many side-effects of collimated lighting, as described above.  My
enlarger came with double condenser and two different forms of diffused
lighting chambers, and I used them all, so I am not unfamiliar with the
choices and results.

If you were to state "some people prefer grain aliasing and doing a lot
of clean up work on black and white", then I'd say, yes, its a matter of
opinion, style and choice, but if you are pegging diffused lighting
(soft?) against Collimated (sharp?) images, I disagree, because I do not
believe the Nikon scans are actually sharper (once aliasing, which can
give a false perception of "sharper", has been removed).

I have provided several people with a non-corrected and non-spotted
black and white scan from the Polaroid 4000, to people who own or have
used Nikons with collimated light source, and the comments typically are:

It's sharp throughout the frame, unlike my Nikon

The tonality is very smooth

The grain isn't exaggerated like with my scanner

And that's often before any USM has been applied, which is needed with a
diffused and anti-aliased or defocused image.

PS: this is not an offer to send this sample file out, my IP provider
limits my upload speed and attachment size, and it was a real pain (it's
a big file)...

Art


Bob Frost wrote:

> Art,
>
> I'm still struggling to understand your repeated dislike of collimated light
> sources, and particularly of Nikon scanners.
>
> The only 'advantage' I can see in using diffuse light is that all sorts of
> unwanted detail, such as grot on the film surface and grain in the film will
> be less clearly resolved. However, EVERYTHING will be less clearly
> resolved - I simply don't believe that diffuse light can selectively reduce
> some detail such as grot on the film, and not all the other detail in the
> film.
>
> Isn't it the same as photographing an object in bright sunshine (parallel
> light) or bright cloud (diffuse light)? One gives sharp shadows, the other
> soft shadows.
>
> Enlargers can have interchangeable diffuse light sources and parallel light
> sources. The former give soft images with less contrast, while the latter
> give sharper images with higher contrast.
>
> I came across another review the other day that said the same thing:
>
> " The Nikon CoolScans use an LED light source that produces a very highly
> "collimated" beam. Other scanners use more diffuse fluorescent light
> sources. We've found that the collimated light source used in the CoolScans
> reveals more fine detail in scanned images than do the diffuse lighting of
> other units. Note though, that you may or may not like this: Part of the
> detail that the collimated LED light will reveal is the film grain itself,
> which some photographers would rather suppress. We personally tend to prefer
> the results obtained with the collimated lighting, but recognize that others
> may not. Compare the sample images from the various scanners on this site,
> to see which suits you best."
>
> So why can't you accept that some like it sharp, and some like it soft? Some
> like grain, and some don't. We all have our likes and dislikes, so let's all
> agree that you like diffuse light scanners, but that others may not!
> Collimated light scanners are not per se badly designed; they are just
> differently designed - to suit a different idea of how to achieve a good
> scan.
>
> Bob Frost.
>
> ----- Original Message -----
> From: "Arthur Entlich" <artistic-1@shaw.ca>
>
> I guess Nikon has never heard of the potential benefits of diffused
> light sources.
>


----------------------------------------------------------------------------------------
Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe 
filmscanners'
or 'unsubscribe filmscanners_digest' (as appropriate) in the message title or 
body