Hi Titis,
I took a look at your two scans (I'm biting my tongue not to make a
comment about the rhyme between "two scans" and your last name.) ;-)
Anyway, from what I can see, these two samples show two different problems.
You'll note the Scanwit lines are somewhat random in position and
density, while the ones from the Nikon LS8000 are regular and of similar
density.
From the samples, I cannot tell if the images are horizontal or
verticals. Assuming the samples are on the web page in their proper
positioning relative to the full image, I'm guessing that both scans are
from horizontal images, but that the Nikon was scanned with the 6cm as
the width and the 4.5cm as the length.
The Acer (Benq) Scanwit scan, shows a few interesting things. After
brightening up the image a bit to see the streaking better, I note that
the thin line about 3/8" from the bottom shows up on all three channels.
I suspect this might be a thin long scratch in the film base.
The blue channel is showing noise and streaking, but the jpeg artifacts
make it hard to determine what is going on exactly. The Green channel
shows banding and streaking, and the red channel a bit of both as well.
But these bands and streaks take place in different locations along
the CCD lines, making a somewhat colored streaking. This can be a
mixture of events. Some of it might be CCD noise, although that usually
is more random speckling, some are just calibration errors in each CCD
line of elements. Some might even be due to random dust on the
individual CCD lines causing incomplete or inaccurate calibration of
individual sensor elements. I doubt if there is much to be done about it.
The Nikon is more interesting. The banding is more regular. It seems
to only be in the Red and green channels, and it appears to be the same
locations in both. Yet, the blue looks fairly clean. It may be that the
jpeg compression has obscured the banding in the blue channel. Also, the
image is mainly deep blue in that area, which may mean the blue channel
is at a higher light intensity than the red or green channels. It
seems, in general, miscalibration in scanners is more obvious in the
darker areas.
If this banding is indeed the notorious Nikon LS-8000 banding at play,
the way they fixed it was to rescan your image using the scanner in
superfine mode which uses only one of the three CCD lines, making for a
more consistent calibration. It takes about 3 times as long to scan
that way.
Art
Titus Tucan wrote:
> Hi everybody,
>
> I've scanned a 35mm slide, a night shot, on my Scanwit. In the lower part of
> the slide where it was supposed to be black or almost black I got a pattern
> of horizontal lines as seen here on my web site
>
> I have tried the Scanwit software MiraPhoto and Vuescan and got the same
> results. Looking at the slide trough a projector, there is no trace of
> theses line.
>
> Recenty, I went to my photo shop to scan another Velvia night shot MF
> (6x4.5). They used a Nikon 8000 and ICE.
> When I went to pick up my CD and checked the file, the whole slide had a
> similar horizontal line pattern from scanning:
>
>
> I complained and they scanned it again for me. The second time it was
> perfect. !
>
> Now here comes the question?
> What is happening with these patterns and how did they get rid of it on the
> Nikon 8000.
>
> thanks a lot,
>
> Titus
>
>
>
----------------------------------------------------------------------------------------
Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe
filmscanners'
or 'unsubscribe filmscanners_digest' (as appropriate) in the message title or
body
