Digital BW, The Print

----- Original Message ----- 
From: "Alan Zinn" <AZinn@...>
To: <DigitalBlackandWhiteThePrint@yahoogroups.com>
Sent: Friday, February 20, 2004 2:20 AM
Subject: Re: [Digital BW] Scanning Negs


> I've made a spiral staircase of strips of slide glass 0.85 mm
> thick and on each step a brush stroke of grainy ink with some
> scratches through it. One side of the staircase starts at 0 mm
> and when turned over there's a 0.45 mm step first. So I can
test
> the sharpness per 0.40/0.45 mm step. Best focus is
approximately
> at 1.7 mm with my 3200. Others have found other distances from
> the glass bed.

> The scanner has three planes where it should be more or less in
> focus. At the glass bed for reflective scanning, at 1 mm for
the
> filmholder and at almost 3 mm for objects larger than the
> scannerbed (for reflective scanning again). Ideally it should
be
> in best focus on the film at 1 mm and the DOF of almost 3 mm
> should take care of the reflective scans that normally are not
> scanned higher than at 600 ppi. I guess Epson thinks that the
DOF
> will take care of the shifts in production tolerance as the
> actual focus can be higher or lower than the 1 mm above the
> glass. Finding the better focus isn't a dramatic quality
increase
> but it helps especially when the scan is improved in Phothoshop
> later on. Even more important is the planerity that I get by
wet
> mounting the film.
>
> See Norman Koren's pages for the focusing measurements.
>
> Ernst

>Ernst, Sean

I did a test with slide glass on the 3200 and it was sharpest
with
the emulsion right on the scanner glass. Raising it the thickness
of
the slide glass softened it. So much for DOF and neg holders!
Someone posted a test method of laying a ruler lengthwise on the
glass with one end raised a few mm .  The prime was the sharpest
point along the ruler. <

Alan, it is the same discussion I have with someone on the 3200
list. He has a 3200 that has its focus on approx. 0.5 mm above
the glassbed. That's very nice as he now can scan on the glassbed
itself. You seem to have an even more suitable focus for the
tasks you want to do. The distribution of focus per 2450/3200 is
however random and the DOF of about 3 mm is certainly there. Of
course half of your DOF will be underneath the glassbed if the
scanner is sharpest at the bed. In a sense it is a pity you
couldn't stick the ruler through the glassbed, there may be more
sharpness below the bed :-)

>SilverFast allows variable USM sharpening and no sharpening.  I
found
that with zero pre-sharpening I couldn't get as crisp an image
later
with PS USM as I could with auto sharpen.  One thing to bear in
mind -
 Epson 2400 & 3200 use a diffuse light source for film. This cuts
film grain and dust but effects sharpening. <

The 4870 adds ICE to that dust removing effect. Wet mounting if
properly done can reduce dirt as well. I sharpen in PS as I want
to know what is in the raw Vuescan scan. I will save that raw
scan at least till the first print is made.

>Another concern is that all scans are RGB before conversion to
256
grey. Does the pre sharpening effect each color differently?
Seems
like: (1) pre-sharpening would toss out a lot of data conversion
to
256 grey could use. (2) sharpening after converting would
be "cleaner" than doing three channels.   Looks like the trick
already suggested is to find a combo, pre AND post sharpening
technique, that suites the image.<

In the description of the Qimage sharpening filter (that's the
one you can use when a digital camera file is loaded, not the
sharpening that happens at printing stage) there's a way too
handle the R/G/B data separately as the frequencies of the hues
are different on the mosaic sensor. The influence of sharpening
is equalised. So B&W files made with digital cameras have a
problem there it seems. But an appropriate filter setting in
Qimage with the added custom B&W conversion would solve that.
Wonder how well this is done in "In Camera" conversions. The ARM
CPUs in cameras are capable enough but is the software that
advanced ?
For scans of B&W film the frequency of RGB sensors is the same. I
really don't know what is the best approach and whether it
matters. The question refined would be: does the sharpening of
the RGB file influence the grey levels and grey level detail
different to sharpening of the converted B&W image? There must be
someone who got the same doubts you have and tested it.

>For the experience I'd like to try wet scans.  What is the
chemical
substance in the fluid?  Would Pec-12 work?  Seems very time
consuming for occasional scans but for a large-volume scanner it
would save spotting. <

At first I thought that Kami SXL 2001 was just plain lighter
fluid or a slightly heavier distilate. It behaved very much alike
though the Kami evaporated slower. When I put the two together in
one bottle it didn't mix at all and Kami was at the bottom, so
they are different. I now use lighter fluid only as I didn't see
a difference in the scans itself.

!!!!!! It is "lighter" fluid, there are all kinds of dangers
included .......... carcinogenic, fire, can be swallowed, etc.
Hope that I will not be arrested when I go to NY in July. !!!!!!

>RE quality of scan - it seems to me that, assuming operator
skill is
good, the difference between a scan of a 6x6 or larger neg from a
E3200 and a N8000 would be tough to see in a  11 x 14  ink-jet
print.<

That's my opinion too. And with both wet mounted and the 3200
sharpened appropriately there's very little difference at even
larger sizes.

>I looked at the Epson 4870 at the store.  It has interesting
film
holders with milk-white plex above the film. The lamp travels the
length of the film like a Umax.  I have a like-new 3200 for sale
:-)<

Same thoughts here and a friend in Amsterdam has one already. ICE
has no influence on the scans he says. Think about the time you
gain on 4x5's.

The lamp travelling with the sensor most likely solves one of the
problems of the 2450/3200. The static lamp of the 3200 produces
more heat. To the extent that when I did two sample scans +/or
the "long exposure" of 4x5 with Vuescan I lost sharpness. With or
without wet mounting. The film expands and probably it even
shifts then with the wet mounting. By wetmounting underneath a
separate glass sheet that effect disappeared. The glass kept the
film cool enough. Given an even higher Dmax number quoted for the
4870 I guess they now could bring the effective light strength up
too. And possibly the evenness of the lighting. The C't test
showed that that is a problem with the 2450/3200 models. The
overall transparency is increased too. Two 4x5's in one scan.
Don't think I would use that ever but the centre of the unit will
be even better, so one wet mounted lengthwise in the middle will
be my first test.

Ernst