From reports I have received, the problem with the banding is a design
flaw, and one in firmware.
Most CCD based scanners use a tri-line CCD sensor which has each line
represent one of the standard color separations. In other words, one is
filtered red, one green and one blue. Since each captures only that one
separation, it is very important that each line be calibrated before the
scan is made so that there is no general inconsistency between the
lines, and little inconsistency between individual sensor elements
within the line.
The Nikon film scanners don't work this way. Rather than a tri-line
sensor each line capturing one separation and using a white light
source, the Nikons use an LED light source which switches between red,
green and blue (and Infra red for the dICE) As a result, Nikon still
uses the tri-line sensor, but each can act independently capturing all
three separations, as the LED light changes. In theory, this should
speed up the process quite a bit.
In actuality, it causes banding. Why? Because Nikon decided that they
only need to calibrate one of the sensor lines, leaving two
uncalibrated. As most people know, CCD sensors response is not that
accurately maintained. The sensors elements don't come out of the
factory perfectly calibrated, and they don't stay calibrated over time
either. Why Nikon would decide to make this design in this manner is
the subject of much discussion.
Can you get rid of the banding? Yes, but it requires you to slow the
scanning down by using just one line, the calibrated one. They call
this super fine scanning mode or something, and it obviously slows the
process considerable since you are taking 1/3rd of the "width" of
information with each "pulse".
The soft edge problem is something that became more obvious with the
higher resolution scanners, although it has existed for a number of
earlier Nikon models as well. Conjecture is that the problem is the LED
lighting source not being bright enough to allow for a well stopped down
lens to work within the scanner. This causes them to use a very wide
apertured lens and ultimately very shallow depth of field meaning bowed
film, such as often results in paper mounted 35mm slides, and some negs
and unmounted works which were overheated during the drying process tend
to have soft edges. I have been told that it is more obvious on 35mm
than other formats, but it may be a lesser problem overall with the
medium format scanner.
Pretty much all CCD scanners clip highlights and have some noise in
shadow regions. The answer is typically to consider a drum scanner, at
considerable higher costs and maintenance. Most ain't pretty, but they
do a nice job. They are more complex pieces of equipment.
If a drum scanner isn't in your future, consider the Polaroid 120 for
medium format films. Reports I have received indicate it neither
suffers from banding or soft edges. Some people here could tell you
more about their experiences with it. However, it does not have dICE or
any clone of it, so if you treat your film poorly, you might find it more
labor intensive without the use of an IR channel and IR defect removal.
Art
doogle@doogle.com wrote:
>>On 2/21/02 12:23 AM, "doogle@doogle.com" <doogle@doogle.com> wrote:
>>
>>
>>>How 'bout quick list of the probs?
>>>TIA,
>>>Mac
>>>
>>These are the issues I found.
>>
>>
>>Banding:
>>See
>>
>
> Wow. Very noticeable. This was consistent, eh?
>
>
>>Soft focus at edges: don¹t have an example posted but might get around to
>>it.
>>
>
> 35mm, 120? Both?
>
>
>>Nikon scan clips highlights, compresses tonal ranges.
>>See
>>
>
>>Lawrence W. Smith Photography
>>
>
> Ever try VueScan with it? That shoulda given you better tonal range
to start
> with.
>
> The banding looks like fatal flaw if that was a consistent prob,
> though...wonder if this is one of main reasons it's "not available"
for the most
> part right now?
>
> Thanks for your input...
>
> Mac
>
>
>
> Mac McDougald -- DOOGLE DIGITAL
> 500 Prestwick Ridge Way # 39 - Knoxville, TN 37919
> doogle@doogle.com 865-540-1308
>
>
----------------------------------------------------------------------------------------
> Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe
filmscanners'
> or 'unsubscribe filmscanners_digest' (as appropriate) in the message
title or body
>
>
----------------------------------------------------------------------------------------
Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe
filmscanners'
or 'unsubscribe filmscanners_digest' (as appropriate) in the message title or
body
