On Thu, 28 Jun 2001 12:37:33 +0200 Tomasz Zakrzewski
(tomzakrz@ka.onet.pl) wrote:
> But I've just read a review od the 4000ED in German magazine "digit"
> ftp://ftp.lasersoft.com.pl/SFPrasa2001/Digit_3-2001.pdf which says,
> that the
> true Dmax of this scanner is 2,3! It was even worse than with Coolscan
> LS-2000 which had 2,6. What't this???
> It means no details in shadows. The reviewers say that this low Dmax is
> the
> consequence of the increase in resolution at the cost of the
> light-sensitive
> area of the sensor (whatever it means) which causes too low
> light-sensitivity.
I guess they must mean that the smaller pixel area restricts the
light-gathering ability of each pixel, hence sensitivity. FWIW the LS2000
wasn't too clever in the shadows, which tended to be submerged in noise,
though multiscanning addressed that very successfully. I think there's a
fair amount of circumstantial evidence that Nikon's LED lightsource is not
all that bright, which will mean the CCD is forced to operate nearer the
noise floor - and integrating exposure over time to compensate will also
accumulate noise.
> Test scans at www.imaging-resource.com also show that only after some
> tweaking in the sanning program scans with good tonal separation in
> shadows
> can be obtained.
Not unusual - CCD's generally compress the shadows a little. Up to a point
you can regain separation by editing the curve.
> I'm puzzled. Can you comment on this Dmax matter?
Not without getting my hands on one. But the usual caveats about mfrs OD
range specifications apply - there's no standard way of measuring them,
and the marketing dept is likely to be in control.
> I thought that the
> 4000ED
> might be such a machine, but now...
> Apart from this Dmax thing, this scanner's real resolution is 3240 ppi
> effective.
> The 4000ED doesn't differentiates reds good enough, too.
Yet a lot of people seem to be happy with it.
> Why is it so that I can't buy a flatbed scanner in $ 1000 price range
> that
> would show no noise in shadows of scans of reflective media (prints) and
> it's impossible to find a well designed film scanner in the $ 2000 price
> range?
Scanning reflective artwork is a doddle compared to film, which exhibits
much more severe OD range, and also your demands for colour accuracy and
optical excellence are likely to be higher.
> Right now my conclusion is that the home filmscannig technology is still
> immature.
Up to a point, although hardware is now vastly better than it was 4-5
years ago. Software, useability and colour management are now to the fore.
There's still a steep learning curve. But it's perfectly possible to get
very good results indeed, with occasional limitations and problems, most
of which can be worked around.
Regards
Tony Sleep
- Online portfolio & exhibit; + film scanner
info & comparisons
