Art wrote:
> But the SS4000+ has a very clean blue channel. The Minolta Dual II has
> a harsh blue channel. It posterizes the middle tones and shadows with
> yellow. Basically, it can't handle the full dynamic range the other two
> channels carry. Looking at just the blue channel in greyscale, the
> higher middle tones and highlights seem relatively smooth, but the lower
> middle tones and shadows are a mess.
For anyone that's interested, I've posted a comparison of the shadow details
from the Elite and
the Elite II at:
(Crop of full frame for reference and full resolution extract from each scanner
- 470KB)
(Separate channels of full resolution extract from each scanner - 307KB)
Raw scans were made from each scanner using the Minolta software with
autoexeposure off and
processed using exactly the same settings in Vuescan. (From memory, a VS
brightness setting
of 1.5 or 1.6 but no blackpoint adjustment). The only other difference was the
amount of
multiscanning needed to bring noise down to minimal levels - x8 on the Elite
but only x4 on the
Elite II.
I made these scans to show Minolta the issues with the Elite II but (if you
ignore these) it also
gives a reasonable idea of its potential shadow performance. As Art commented
about the Dual
II, the blue channel is also the weakest on the Elite and Elite II as well.
On paper at least, the Elite II should have a much greater dynamic range than
the Elite (16bit
output as opposed to 12 bit output). The blue channel on the Elite II shows
slightly less grain but,
other than that, I can't see much evidence of the increased dynamic range in
this example.
Maybe it's just a bad example!
The Elite II does exhibit less noise than the Elite but some multisampling is
still required to retain
maximum shadow detail. If anyone is interested, I can always post the same
example with no
multiscanning to show the noise levels.
> there appears to be some non-optical
> sharpening that goes on, perhaps in firmware, or in the electronics.
> This causes the noise levels to be more obvious, and makes unsharp
> masking more difficult to apply without artifacts
I don't know whether the Elite or Elite II does this but even modest sharpening
tends to pull out
noise and grain. The Elite II software now has an unsharp mask option but I
tend to use John
Brownlow's smart sharpen action instead in Photoshop as the only practical way
of not making
noise/grain too obvious.
> using dICE on
> the Minolta Scan (via Elite II version) might eliminate this problem.
Not particularly. GEM, however, does smooth out grain and noise nicely but can
introduce some
unnatural tonal effects. Like grain reduction in the Vuescan, the downside is
that GEM adds a
HUGE processing overhead (I only have a humble PII 350). I suspect that any
additional cloning
etc needed on the Polaroid to get rid of any dust spots would take less time
than waiting for GEM
to finish on the Elite II. Of course, this isn't helped by the fact that I'm
using the USB
connection.....
ICE is excellent at recovering damaged images that might otherwise be
unscannable but
thankfully I don't have too many of these. I'm finding more and more that the
elimination of
grain/noise is a bigger issue in the way of getting final images I like.
I'll have to wait and see what Minolta suggest on the faulty Elite II before I
decide whether to stick
with Minolta or not.
Al Bond
----------------------------------------------------------------------------------------
Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe
filmscanners'
or 'unsubscribe filmscanners_digest' (as appropriate) in the message title or
body
