[filmscanners] Re: Polaroid Dust and Scratch Removal

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

I'm not Constantine, and I'm sure he might have something to say, but I
addressed this issue once before when this query come up.

First I'll quote my old info, and then add some new stuff that I've
learned since.

====================

Diffused light is bounced around and comes from many directions and
angles, and as such, it enters the film base at different angles and is
defracted within the film base as well.  Even more so for the material
on the surfaces of the film.  The grain or color information, which is
what you do want to record, is kept in a very tight thin set of layers.
This is in part why thinner emulsions produce sharper enlargements.  It
is nearly two dimensional compared with the film base or dust and dirt.
Light therefore, regardless of the angle it enters those thin layers
scatters and bends considerably less.  In theory, this should cause a
few things to happen.  Surface dust and dirt should be diminished in
visibility on the projected image, as the light works itself around
those larger 3-d objects.  Also things within the film base, like
scratches, are diminished because again the light bounces around within
the film base mixing up and coming out of there all tumbled around.

The actual image layers, being quite thin, relative to the rest, do get
slight amounts of softening especially on edges of grain or dye clouds,
  but this can work for you in most cases, as it is a very slight type
of defocusing of the light.

If you have ever worked in a wet darkroom with a color head enlarger
which has a choice of condenser or diffusion light piping, or if you
have worked with a condenser enlarger in B&W which uses a cold cathode
lighting source rather than incandescent, you probably would have
noticed this.

==========================

OK, now to add to this.

There is no advantage, and in fact, there are several disadvantages to
capturing information beyond the frequency the CCD can capture without
creating sampling errors.  Using lenses or lighting which increase the
frequency beyond the Nyquist sampling limits just introduces errors
which cannot be filtered out after the fact without destruction of other
accurate data.  This becomes very obvious in 2400-2900 dpi scanners
which ends up blowing grain size up due to sampling errors.

I don't claim to fully understand the mathematical theory, but I do
understand that a well focused optical capture can provide data beyond
the ability for the digital sampling process to accurately represent
even at 4000 dpi.  The best way to fix this is to not capture or attempt
to digitize these excessive frequencies to begin with, rather than
trying to removing them after the fact.  The easiest way to do this is
by reducing the focus resolution to not allow discrete data of frequency
higher than the Nyquist sampling accuracy numbers can accurately capture.

Lastly, in looking at photomicrographic images of films which are
typically classified as very fine grain, I have noticed they tend to
have soft edged dye clouds.  This implies to me that we perceive lower
grain not only as a result of dye cloud size, but due to indistinct
borders, once again placing a diffused light source at an advantage,
since a very, very slight dye cloud edge softening would occur with
diffused light versus focused light that something like the Nikon (other
than the LS-8000) uses.

Art



Bob Frost wrote:

> Constantine,
>
> Don't the same considerations apply to the illumination of detail in the
> image? Doesn't the Polaroid cold cathode light provide a softer image of the
> details you want as well as of the details you don't want? If not, would you
> explain why not? If one type of illumination is harsh, and one is soft, I
> would have thought that applied to everything in the light path, not just
> dust. But I don't mind learning why this is not so.
>
> Bob Frost.
>
> ----- Original Message -----
> From: "Kapetanakis, Constantine" <KAPETAC@polaroid.com>
> To: <bob@frost.name>
> Sent: Wednesday, May 22, 2002 6:15 PM
> Subject: [filmscanners] RE: Polaroid Dust and Scratch Removal
>
>
> Absolutely right. The Nikon scanner is using a narrow and focused beam of
> light that will generate shadows even of the slightest film surface
> imperfections.
> The Polaroid scanner cold cathode light source, provides a softer more
> diffused light.
>
>
> -----Original Message-----
> From: Simon Lamb [mailto:simon@sclamb.com]
> Sent: Wednesday, May 22, 2002 1:04 PM
> To: KAPETAC@polaroid.com
> Subject: [filmscanners] Re: Polaroid Dust and Scratch Removal
>
>
> I think Art meant that the LED light source in the Nikon scanners is more
> susceptible to picking up the presence of dust than other scanners.
> Specifically, the cold cathode light source in the SS120 hardly picks up
> dust.
>
> Simon
>
> ----- Original Message -----
> From: "Maris V. Lidaka Sr." <mlidaka@ameritech.net>
> To: <simon@sclamb.com>
> Sent: Wednesday, May 22, 2002 4:16 PM
> Subject: [filmscanners] Re: Polaroid Dust and Scratch Removal
>
>
>
>>Art, you curmudgeon,
>>
>>I for one read and respect your posts, as they are intelligent if lengthy.
>>But I have to take issue with your statement that Nikon manifests problems
>>from dust and scratches - the dust and scratches are on the film, not in
>>
> the
>
>>Nikon scanner hardware, are they not?
>>
>>Maris
>>
>>----- Original Message -----
>>From: "Arthur Entlich" <artistic-1@shaw.ca>
>>To: <mlidaka@ameritech.net>
>>Sent: Wednesday, May 22, 2002 5:37 AM
>>Subject: [filmscanners] Re: Polaroid Dust and Scratch Removal
>>
>>
>>[snipped]
>>
>>This software is not a replacement for IR cleaning, but then again it
>>doesn't really need to be on a Polaroid scanner.  After all, Polaroid
>>scanner simply do not suffer from the problems that Nikon's manifest
>>from dust and scratches, so why should it be designed to fix Nikon's
>>problems?
>>
>>[remainder snipped]
>>
>


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