I could really use some help deciding on the preferred way of working
with this scanner.

I do most of my adjustments in Nikonscan on PC, with CMS on and output
to "Nikon Adobe 1998". I use no custom profiles and am not worried
about color management, generally... scans look good and noone has
complained.

After checking around, I see most people turn CMS off. Why?- is it a
quality issue? If there are advantages to doing this, I would need to
select "Scanner RGB" to allow me to use the Nikonscan tools. Who uses
this space, and are there any issues with doing so?

When the TIFFs come into PS7, I do a few more adjustments and work in
the Adobe 1998 RGB space. If these files are destined for agencies or
clients, is there any problem leaving the profile as "Nikon Adobe
1998"? If I decide to use "Scanner RGB" instead, then what do you
recommend for the Assign- and Convert to- profile options?

Thanks.

EdV

On 13 Dec 2003 15:08:00 -0800, verkaik@sympatico.ca (Ed Verkaik)
wrote:


Of all the critical steps in an ICC-color-managed
workflow, having a scanner profile is by most
accounts the least critical and also the most
problematic.

I mean -- if you trust your monitor's profile and
your printer's profile, what's the harm in tweaking
the scan a bit to get it looking the way you want it?



rafe b.
http://www.terrapinphoto.com

I can't believe that you are getting reasonable colors with NikonScan!

I just got the LS 5000 ED (successor to the 4000) and am having massive
color problems. I attribute them to the scanning software, or should I
blame the hardware?

I have read that IT8 color profiles can enhance color reproduction very
much. But you need to use other scanning software such as Vuescan or
Silverfast since NikonScan does not support IT8 profiles.

Olaf

Ed Verkaik wrote:

On Sun, 14 Dec 2003 01:59:45 +0100, Olaf Meyer
<flatpicker_no_spam_@gmx.de> wrote:


There's no such thing as an "IT8 Profile."

The modern standard for color management is more
or less governed by the ICC, so there are indeed such
a thing as ICC profiles.

The Kodak IT-8 is a standard test target often used
for color calibration -- and indeed for the creation of
ICC profiles of scanners and printers. As a chrome
(that can be scanned by a film scanner) an IT-8
slide can also be used to create ICC profiles for
film scanners -- including the Nikon -- given the
appropriate software. You can do this with or
without NikonScan. You *cannot* profile the
scanner for negatives (print film.)

NikonScan is a very cabable scanner driver
program. I use it almost daily and have scanned
many hundreds of slides and negatives with it,
producing many saleable prints from the scans.

What sorts of problems are you having with your
LS-5000? Have you used other film scanners
before, or is this your first?

Do you know what a histogram is, and how to
"read" it and how to fix it if it's not right?

Are you scanning slides or negatives?

Have you read the manual?


rafe b.
http://www.terrapinphoto.com

Rafe,

as you can see I'm quite new to this topic and I'm trying to get my
share of knowledge by reading newsgroups and looking for any other
information on the web ;-)

The Coolscan 5000 is my first scanner! And I'm sure I don't really know
how to operate it properly. But so far I have not found NikonScan of any
great help. I've read through the Nikon manual, but it is very terse and
does not go into any details, but details are what I need at this stage.
They do not mention how you need to tweak the histogram in order to
change the color reproduction ...

I ordered Wayne Fulton's scan tips book from http://www.scantips.com .
Hopefully it will give me a head start ;-)

As you recognized I've mixed up color profiles and calibration. I was
referring to IT8 color calibration. NikonScan does not support this, or
does it? You mention that I can create ICC profiles for my scanner with
appropriate software. I have heard from other people that this is one of
the real benefits from Vuescan and Silverfast scanning software. Do you
have anything in mind besides Vuescan and Silverfast?

I am scanning color slides mainly Fuji Sensia, Velvia and Kodak E100VS.
I plan to scan several thousand slides and cannot afford to correct the
color on each slide. My idea is to calibrate the scanner for the
different films I'm using via IT8 calibration in order to get reasonable
colors out of the box. I mean if this is going to be OK for 90% of the
slides this already is a real advantage.

Here is a scan of one of my Velvia slides with NikonScan4. I've just
enabled ICE (normal) but turned off all other optimizations. I've used
16 bit depth for colors on a single pass.

http://www.olafmeyer.de/Anchohuma.jpg

Here's a touched up version (that is much closer to the actual slide):

http://www.olafmeyer.de/Ancohuma-corrected.jpg

Olaf

Rafe B. wrote:

In article <brhjje$hd9$1@ulysses.news.tiscali.de>, Olaf Meyer
<flatpicker_no_spam_@gmx.de> writes
before you start any scanning at all, it is important that you
at least adjust your monitor to the correct brightness, contrast and
colour balance. If you don't do this then all of your work adjusting
your scans will actually be changing them to some false basis, resulting
in further problems downstream when you print them or send them to other
users. There are a variety of software packages for setting up your
monitor, some free. After you have done this then you can adjust the
colour to get the results on the monitor close to the original slide.

Obviously I cannot judge which of the images you reference above is
closest to the original, but the post adjustment looks very pink to me,
whilst the pre adjustment looks slightly green. Both images look like
the wrong gamma has been used though and are particularly dark. A
histogram of both images shows that the lightest part of these images is
less than 50% of full scale (126 max in the corrected scan and 100 max
in the raw scan).

What application are you using for your image manipulation and what
settings do you have set for gamma and/or Nikon Colour Management in the
Preferences setting? Do you have autoexposure checked for the image
type you are using here (single/batch and positive/negative film)?

You can view the histogram of the raw data if you pull up the Tools
Palette and open the Curves option in NikonScan. This defaults to show
the weighted RGB average, but can also show individual colour channels.
The vertical axis is the output range of the processing, the horizontal
axis is the raw scanner data. If necessary, brightness, contrast and
gamma adjustments can be made here prior to scanning. The option also
provides eye droppers to set the white and black points in the image.
Similarly, colour balance adjustments can be made at this stage as well.
After you have a preferred setting for a particular type of film then
that can be saved from the main NS window under the Settings|Save
Settings option, for subsequent use. If you are transferring the entire
image to your application in 16-bit format then these adjustments may be
as well performed in the application, however using preset film profiles
like this will get you pretty good results without further optimisation.
--
Kennedy
Yes, Socrates himself is particularly missed;
A lovely little thinker, but a bugger when he's pissed.
Python Philosophers (replace 'nospam' with 'kennedym' when replying)

Kennedy McEwen wrote:

Kennedy,

I've calibrated my Monitor using "Adobe Gamma", but I'm not sure it I
did this 100% correct. Among other things I've chosen 6500K for the
white balance (which seems a bit dull compared to my previous 9300K).
What can I do with the ICC profie that I created with "Adobe Gamma"? Do
I need to load the profile to NikonScan or any other image processing
application (e.g. Photoshop)?

Here are some of the settings that I'm using within NikonScan: The Gamma
is set to 2.2. I actually cannot adjust it, it is greyed out (disabled).
Do you have any idea why it may be disabled? I have autoexposure enabled
for the (positive) images I'm scanning (as well for single as for batch
scan). I've noticed that the preview usually looks considerably diffrent
from the actuall scan. The preview is much brighter. Is this a NikonScan
feature?

I've played arround with the histogram a bit, but don't really know what
makes the most sense. There are some many different ways of modifying
the curves. Is it true that an IT8 target to calibrate the scanner to a
specific film can provide some assitance? I just have the impression
that I as complete "digital imaging" amateur don't really have a chance
of finding out a reasonable histogramm by trial and error. What may be
good for one slide may be bad for another and using IT

I've used Photoshop elements for manipulating the scanned images. First
of all I'm confused that Photoshop only handles 24bit TIFF images and
cannot read the 48 bit images that my scanner delivers. Why should I
scan with 16 bit, if postprocessing is limited to 8bit (per channel)?

I really do appreciate you taking your time to give me some advice.

Thanks a lot,
Olaf

Rafe B. <rafe.bustin@verizon.net> wrote in message news:<n6hntv07dt77roh8n0b5367633so6bs15k@4ax.com>. ..
Rafe,
Maybe you could tell us whether you have CMS on/off, what profile you
scan into, and what you do with that profile in PS?
Thanks.
EdV

On 14 Dec 2003 09:19:00 -0800, verkaik@sympatico.ca (Ed Verkaik)
wrote:


I have Nikon CMS turned off.

I've been scanning film for the last five years,
and for most of that time selling prints made
from these scans. Been using a Nikon film
scanner for 2.5 of those 5 years or so.

So far, I've tried but have had no luck using
ICC-based color management. I use an LS-8000
with NikonScan, at 24 bits per pixel and do 100%
of my editing in Photoshop 4, which is not even
ICC-aware.

"Color management" is what ICC profiles do,
but it's not absolutely necessary, and several
pros that I have talked to don't bother with it
at all -- they are rather dismissive of it, in fact.

"Color correction" is necessary. For that, read
almost anything by Dan Margulis or Michael
Kieran, or any good Photoshop book that
discusses the Curves tool (eg. Photoshop
Artistry by Barry Haynes and Wendy Crumpler.)

Oh yeah, and listen to Kennedy, because he
knows his sh*t.

I realize that my avoidance of ICC based CM
puts me in a minority position. This is not from
lack of trying, and in fact I'm gearing up for
another try. IMO, the cheap profiling gear can
be far more trouble than it's worth.


rafe b.
http://www.terrapinphoto.com

On Sun, 14 Dec 2003 17:34:34 +0100, Olaf Meyer
<flatpicker_no_spam_@gmx.de> wrote:



I would predict that the histograms for these two, in
Nikonscan, show huge peaks near the left hand
edge of the scale, and a flat line at zero near the
right. Photoshop shows no pixels at all above
100 or so (out of 255.)

In a word: seriously underexposed in the scanner
driver. You are using less than 1/2 of the tonal range
available, on an image that seriously needs all of it.

Start by cranking up the exposure, say 0.5 to 1 unit.

Or just let NikonScan auto-expose the slide, it certainly
won't be any worse than what you've posted.

Eventually you will see the problems with auto-
exposure; for example on this image, it will take
some care to make out the details in the snowy
part of that snow-capped peak.

I do expect this to be a diifcult slide to scan; it
seems to have a good deal of contrast. Mabye
not the best place to start.

May I suggest a slide that's a bit flatter to begin
with, maybe something more pastel or shot on a
cloudy day.


rafe b.
http://www.terrapinphoto.com

Rafe B. wrote:
Well you are right, thats pretty much what the histograms look like.

I've actually used NikonScans auto-exposure. I cannot find any dialog in
NikonScan4 where I can adjust the exposure manually. Is there such a dialog?

That's what I'll do. I've acually also just ordered a book by Michael
Kieran (one of the authors you suggested for Photoshop essentials).

Am I right that I could actually apply all the corrections that I do in
NikonScan also in Photoshop after I have scanned the images?

Olaf

I agree with almost everything you said. Apart from Adobe Gamma, I use
no special CM tools or files and get good results. But you still
havebn't mentioned whether you use Nikonscan's CMS or not. Do you
keep the profiler assigned within the software or assign a new one in
PS? Or, do you leave files untagged (which is a bad idea if others
will be seeing them)? That was the root question of my original post.

I wish Nikon hadn't been so egocentric by including "Nikon" in their
"Adobe RGB 1998" tag! They appear to be identical but some may think
the file is in a nonstandard colorspace as a result.

EdV

On Sun, 14 Dec 2003 23:53:54 +0100, Olaf Meyer
<flatpicker_no_spam_@gmx.de> wrote:

I'm using NikonScan 3.12. There's a pane in the
Tools Palette called Analog Gain. That's it.


That's an old question. I suggest trying to get
it mostly right (if not perfect) in NikonScan first.

If you scan in 16 bits and import the image into
Photoshop with 16 bit per channel color depth,
then the answer to your question is mostly yes --
you can do the corrections in Photoshop.

OTOH, if you take proper care in NikonScan,
you don't need 16 bits per color.

I think Wayne Fulton's website (scantips.com)
discusses the Histogram and Levels tool in
some detail, and NikonScan's tool is fairly
standard - no surprises.


rafe b.
http://www.terrapinphoto.com

On 14 Dec 2003 15:22:12 -0800, verkaik@sympatico.ca (Ed Verkaik)
wrote:


I use Photoshop 4 for 100% of my image editing,
and PS4 is not ICC-aware. So any profile tags
inserted by NikonScan would be ignored anyway.

As far as I know, I have NikonScan's CMS turned off.


rafe b.
http://www.terrapinphoto.com

In article <bri3ei$m7b$1@ulysses.news.tiscali.de>, Olaf Meyer
<flatpicker_no_spam_@gmx.de> writes
Sorry about not getting back to you earlier, but I was a bit tied up
with other things yesterday.

This sounds like you have a mismatch between the application colour
space and the scanner, but starting at the beginning of your comments
and working through:

Don't worry too much about the colour temperature that you have selected
in Adobe Gamma - that is determined by the environment you use to view
your monitor in, and it just changes the colours produced on the monitor
to approximately the same colour temperature as the environment, so that
you see the image it presents fairly naturally. 6500K is roughly the
same temperature as normal daylight - it is the surface temperature of
the sun's photosphere. If you are viewing the monitor in an environment
lit by certain fluorescent lights then 9300K would be appear better
matched to the surrounding environment, whilst if you use a normal
tungsten light in a room without natural daylight then 5500K would be
more appropriate.

The most important step on using Adobe Gamma is setting up the monitor
brightness and contrast in the first place. Any errors, particularly
brightness, can result in a seriously wrong gamma being selected. It is
worth going round this several times to make sure that you have
consistent results - and checking the resultant black and white points
on an independent grey scale, such as that of Nokia Monitor Test, or
here: http://www.uni-mannheim.de/fakul/psy...mma/index.html
Particularly check the low and high intensity levels on the sub pages.

Anyway, once Adobe Gamma has been used to get your profile then that
should be picked up by the operating system automatically (I think - its
a long time ago since I did this!). You should be able to check the
correct profile is used under Display Properties (in the Control Panel),
select the Settings Tab, then the Advanced button and the appropriate
monitor profile should be listed and selected under Color Management.
(How you get to this point may vary by OS issue - I am using W98SE).

Once the monitor is calibrated, the profile is handled by Windows colour
management in the background and should also be loaded into Nikonscan as
the monitor profile under Preferences|Color Management|Use Custom
Monitor Profile. However, before messing around with the Adobe Gamma, I
would use the standard default setting for your monitor in both Windows
and NikonScan - at least with that you know that you haven't completely
misprofiled the monitor and only need to adjust the brightness and
contrast correctly. It is possible to get into a real mess with the
colour management and have gamma applied twice, causing very poor
results, so keeping it simple until you get the main problems ironed out
is worthwhile. Note that you need to restart NikonScan for any changes
to the colour management to take effect.

The Gamma setting in NikonScan is greyed out because you have selected
to use Nikon Color Management - or at least, I hope that is why it is
greyed out - if you haven't then, "Houston we have a problem". :-)

The reason that your preview window looks different from the final scan
is probably because you are using a different colour space for the
scanner/application and, possibly, the operating system. Windows, by
default, uses the sRGB workspace and, if your application is not Colour
Management Enabled (CME), you should select this under the RGB tab of
the NikonScan Preferences. If your application is CME then choose the
same colour space that your application is using. I don't know if PS
Elements is CME, and others with direct experience of it may wish to
comment. I use PS7.01, which is CME, and use the Adobe RGB (1998)
workspace in both. Due to similarities with Adobe products, you might
be able to find the PS Elements workspace by pressing Ctrl-Shift-K, if
not, assume that it is sRGB and select the RGB workspace on NikonScan
accordingly.

When you have this correct, the preview and the final scan should look
IDENTICAL (ie. not even the slightest difference in colour, contrast or
brightness) - if they are not then there is a major colour management
problem in your system. If this is the case then I would go back as far
as you can and switch all colour management off - including Adode Gamma,
use the default profile for the monitor and work from there. Once you
have established the problem, then you should be able to go back through
and turn these parts on piece by piece - observing the improvement at
each stage. If any individual step makes it worse then you have
identified the problem. ;-)

The histogram adjustments are really just a matter of taste - I would
advise the minimum adjustments here. However I note in your response to
Rafe that the histogram indicates a very low density level. I assume
that you mean the histogram viewed in the Curves window of NikonScan's
Tool Palette, as opposed to PS Elements (where the image may already be
corrupted due to the CM issues). If this is the case then make sure
that your white point target is set correctly to around 255 in the
Preferences|Advanced Color setting in NikonScan. Check the black target
is around 0 at the same time. Also, make sure that you have a
reasonably low percentage exclusion set for the auto contrast
calculations. This should default to 0.5% of the histogram, and
determines how much of the histogram tail at either end will be ignored
to set the black and white points. A smaller number means that extremes
of the histogram will be included, which could be spurious elements in
the image such as dust or just a glitch of some cause. This will result
in the histogram only being partially filled (as in your case). A
larger number means that more of the extremes of the density
distribution is ignored and permitted to saturate - resulting in a
histogram with large values at either end of the distribution, and
saturated images. The sample point just defines the area of the preview
which is used by the histogram eyedroppers for setting the black and
white points manually in the Curves and other windows.

With these Advanced Color settings in NikonScan, the histogram in the
Curves window should occupy all of the available range before you make
any modifications to it - click on the three diagonal lines icon at the
left to switch any adjustments you might have made off. In addition,
nowhere near the highest and lowest point of the histogram should show a
dramatic increases in population from the rest of the histogram, if it
does then it indicates that the CCD has saturated and the raw data in
the scanner is clipping.

If the histogram is not complete or significant clipping is evident,
then the exposure may need adjusting - although this is very unusual if
Autoexposure is selected for the Preview Setting. The exposure is
controlled by the Analog Gain setting on the Tool Palette and can be
increased or decreased from the default by up to 2 stops in either
direction. Note that if clipping is occurring then changing the
exposure will not change the range of the histogram at all, just stretch
part of it out within that range - the black and white point targets
will override the raw data of the ADC.

Unless you can get reasonably good results with the standard profiles
then you would be wasting your time going for an IT-8 based profiling
system. These are ideal when you require absolute accuracy, but are
really unnecessary for most users, and certainly someone starting out
with their first scanner. They should really only be providing the
final minor tweaks to profile mismatches throughout the system, and if
used when you have major profile mismatches (as you seem to) they will
are more likely to make your problems worse rather than better. At
best, they will cover up an underlying problem in your system - so get
the basics sorted out before you even think of going down that route.
Only consider it once you are getting consistent scan results that have
only very minor differences from the original but which, when viewed
without reference to the original look perfectly acceptable.

The issue of 16bpc versus 8bpc is something which continuously raises
its head here and is unlikely to ever be resolved to everyone's
acceptance. Basically, 16bpc retains all of the scanner produced
information (far more than you can see in the image) so that the user
can select the data range they need at a later stage, in the post scan
processing and make extreme adjustments if necessary. Suffice to say
that if you have a reasonably acceptable preview of the image (and that
matches the final scan) then you should only ever need to make minor
adjustments in the application, in terms of image brightness, contrast,
gamma and colour balance. In this case you will find that 8bpc is
perfectly adequate for all of your processing purposes. If you scan
without satisfactory previews or scanner settings then the chances are
that you will need to implement major level adjustments on the image in
the application and will consequently need an application (such as the
full version of PS) that can handle 16bpc processing. However, getting
a decent scan in the first place is what the scanner driver is all about
- and NikonScan is a very good driver that delivers consistently
acceptable results when set up correctly. Using NikonScan you should
get close enough to your final desired image on every occasion, making
16bpc only really necessary for handling batch scans of very difficult
and variable subject matter.

Hope that helps.
--
Kennedy
Yes, Socrates himself is particularly missed;
A lovely little thinker, but a bugger when he's pissed.
Python Philosophers (replace 'nospam' with 'kennedym' when replying)

On Mon, 15 Dec 2003 06:43:39 +0000, Kennedy McEwen
<rkm@nospam.demon.co.uk> wrote:



One thing to be aware of with NikonScan's histogram tool,
or Photoshop's for that matter.

If there is no selection rectangle within the preview window,
the histogram includes everthing contained within the
preview window. For most formats and especially 35 mm
that will include more than the film image itself.

So for a reasonable and accurate histogram, you do need
to create a cropping rectangle, and set it carefully so that it
is comfortably within the actual image area.

If you fail to do this properly, your histogram will be skewed
towards the black end -- ie., it will be measuring and including
all those black or near-black pixels outside the actual image.


rafe b.
http://www.terrapinphoto.com

Just curious, there seems to be lots of talk of NikonScan
here. I just got a Nikon 8000 (haven't even unpacked the box)
and I was kinda planning to use VueScan, since I'm familiar
with and like it. Is there some reason I should consider
using NikonScan instead ?

Thanks, John

In article <brl1mv$7i8$1@capefear.cs.unc.edu>, John Eyles
<jge@cs.unc.edu> writes

A lot of people like Vuescan, and a common user interface across all of
your scanners is another factor for many users too. However, I have
tried the demo version on many occasions throughout its evolution and
simply don't like the user interface. However I recognise that the
opposite is true for many other people, possibly you too. Then again, I
have used Nikon scanners for years - before Vuescan existed, so I have
"grown up" with NS.

When Vuescan first appeared, it was quite difficult to achieve
consistent balanced colour from any film scanning package, including NS
- you really needed to understand what was going on in the scanner
hardware, and a working knowledge of colour darkroom procedures probably
helped as well. Vuescan's big feature was the ease of getting near
perfect colour first time from a range of film types just by selecting
the appropriate film profile from a menu - no mean feet across a range
of scanner types! Whilst it worked admirably, the user interface was
really cludgy and, in my opinion, some of that still remains to this day
- but it has improved a lot. I find the NS interface much more
logical, easier to understand and operate and, meanwhile, the ease of
getting a good scan first time with NS has also improved as that product
has evolved.

Also, being the native package for the Nikon scanner range, NS
implements many hardware functions, such as ICE3, as designed, whilst
Vuescan implements "similar" algorithms. Similarly, Vuescan provides
some features that are not available in some scanners, such as
multiscanning - using a multipass process for scanners which were not
really designed to offer that facility at all. For a fully featured
scanner like the LS-8000 this is not such an issue.

Being comfortable with the software is probably quite a big step to
getting successful results from it. If you are already using Vuescan,
there is no need to change, but you should try the NS software too -
they are not mutually exclusive, and you might get on better with NS
once you set have it up.

I cannot say one package performs better than the other because, to be
frank, I have not made enough scans with Vuescan to give a fair
comparison however I have made several thousand perfectly acceptable
scans on a range of films (and Nikon scanners) with Nikonscan over the
years and know it yields as much as the hardware can deliver.
--
Kennedy
Yes, Socrates himself is particularly missed;
A lovely little thinker, but a bugger when he's pissed.
Python Philosophers (replace 'nospam' with 'kennedym' when replying)

Kennedy,

thanks for your really detailed reply. I've read it very carefully and
followed some of the steps you've described. Find more anwsers below ...

Yes that's the reason ;-)

As you suggested I've turned CM off for now and I am using the standard
settings instead. The preview scan and the actual scan still look
considerably different although both are displayed in NikonScan. And as
I'm writing this I just found the reason! It's the way NS displays high
resolution TIFF images. the 4000 dpi scan is displayed considerably
darker, it actually looks strange (at 16% of it actual size). There are
black vertical bars in the picture. Just out of curiosity I've also done
a 1000 dpi scan (which is displayed at 50% of its original size) and
this one actually looks very similar to the prescan!

As Rafe suggested I've used a another slide (E100VS) that does not have
as high contrast but the colors that I get are still pretty bad. My
Advanced color settings are exacly as you describe them: My white point
target is 255, black point 0 and the exclusion percentige is 0.5% (for
white and black). Again I've used auto exposure and have not modified
the analog gain settings (all values here are at zero). As far as I have
understood so far I can only correct the colors by editing the
histogram. Please correct me if I am wrong.

You can actually view the scanned image and a combined RGB as well as a
RED histogram from NS here:

http://www.olafmeyer.de/sorata-ns.jpg
http://www.olafmeyer.de/histogram-rgb.jpg
http://www.olafmeyer.de/histogram-red.jpg

Out of frustration I've tried a demo version of vuescan. The colors I
get here are so much better. I'm still not going to give up on mastering
NS, since I want to take advantage of the ICE features ... But see the
difference your self. Here's the result with viewscan:

http://www.olafmeyer.de/sorata-vs.jpg

It seems that I really need to do some major adjustments in NS. The
results so far are really poor. I'm a little clueless where to continue
from here with NS in order to get consistent scan results.

The starteling point for me here is, that vuescan delivers reasonable
results while NS doesn't with the same general system setup.

--
Olaf

(remove '_no_spam_' from email when replying)

On 15 Dec 2003 14:23:11 -0500, jge@cs.unc.edu (John Eyles) wrote:


Aside from the fact that it comes with the scanner
and is written by the same outfit that designed
the scanner?

OTOH, if you already own and like VueScan,
you should be in fat city, as they say.


rafe b.
http://www.terrapinphoto.com