Re: Scanner Training Advice (was: UCR vs GCR...or Neither - RAW CMYK?)


samarsh@...
 

I originally wrote:

I was wondering if you and perhaps the list have any
thoughts on
how I could maximise the training. I know many on this list
attend
specific training or seminars - are there any tips from all
those
people who have 'been there - done that'.
Martin replies:

I have no experience of CreoScitex training (apart from
LeafCapture on the
Volare camera). However, I have been to a number of scanner
demos and would
offer the following advice:
Thanks for the reply Martin. So, did the LEAF training simply take
you through the printed exercise book, using the half dozen or so
examples or whatever?

I have a gut feeling that my training will not be too extreme.


Take your own originals rather than allowing the
demonstrators to use their
own standard demonstration images. This will enable you to
see how the
demonstrator copes with an image from scratch rather than
have them rattle
through a set routine that they have no doubt done hundreds of
times before.
Very good point.

The scanner comes with some exercise originals...

Trannys, negs, colour reflective, BW ref, Halftone print.

I presume that these will be used, in conjunction with the
exercises which I have used in the manual anyway...but I will see
on the day I guess.

Make sure that your originals are tough scanning propositions,
for example:

1. An *arty* image that has no natural looking highlight,
shadow or gray
midtones. One that has been shot using coloured gels with
vibrant RGB
colours. This should demonstrate how well the demonstrator
can cope with
numbers only -- or whether the scanning software has profiles
for standard
film stocks.
Yes this is a good test, and all the supplied scanner exercise
images do have endpoints...


2. A tranny that has had a hard life. See how the demonstrators
cope with
scratched and pitted originals. If the demo is on drum
scanners then oil
mounting will fill in most of the damage. If it's on a flatbed is it
still
possible to oil mount?
The EverSmart Supreme can oil mount - but this is optional.

CreoScitex also offer an optional Photoshop plug to help
automate dust/scratch removal.

But this would be very good to see, since this scanner seems
very unforgiving of originals quality (also the six surfaces in
scanning trans can be a real dust spotting problem).


3. Under- and over-exposed transparencies. Take one of each
and see how the
demonstrator and scanner hardware cope with pulling out
shadow and highlight
detail.
The range of this scanner is very good, not drum but very very
good.


4. Black and white prints. Take one of these along as a test of
the
scanner's light source.
Are you referring to native scanner RGB grey balance?

I did a test using a Kodak Greyscale - and each patch was within
one or two RGB points of each other, all over the scale. This
scanner, with nothing more than a canned profile - is near grey
linear balanced. It is so close to R=G=B - but you expect that for
the insane price tag!


5. A negative. Don't know what CreoScitex's position on
negative is? Most
conventional repro houses refuse to go anywhere near them. If
the software
is able to produce positives from neg, watch the adjustments
that the
demonstrator carries out.
It's a bit like getting a poor RGB, with no profile. Most repro
places would not like to go near these too!

The scanner has various neg tables for filmtype makes/models -
and this may help up to a point, but is not a holy grail (too many
variables for canned profiles).

It is all memory colours.

This is a strength of CMYK by the numbers editing.

I will attempt to push for this, as I have not got around to playing
with the supplied exercise neg yet. I have inverted/removed the
orange mask before using Photoshop - but not in the scanner,
on raw data before the image is finally acquired.


While your originals are being scanned, take a detailed look at
the USM
functions of the scanner software. Scanner USM is usually far
superior to
Photoshop. However, this usually leads to a fairly complicated
interface
that you tend to forget after the demo and switch to "auto" when
you get
back to the shop :-)
The controls for USM are very good, and you can even use filters
in the scan.

The big drawback is that the scanner does not do this on a live
interactive way.

You change the setting, then perform a new hi res prescan of the
selected area.

Each time you check a new area of the image, its a new 'max
detail' scan. This can take about 30 seconds just to evaluate the
USM.

Dan wrote to me that Photoshop is better. Now I understand
what he meant.

No matter how many options, or how good the tool - if it is hard
or impractical to use, the results may be poorer than
Photoshops 'poor' USM.

Likewise, with the separation settings, the scanner
manufacturers usually
tell you that profile x is best for this and profile y for that. Get as
much
information on the profiles as you can. Find out how easy it is
to edit the
CMYK profiles and whether the same profiles can be used in
Photoshop.
It is my guess that the basic CMYK table will be used.

I will PUSH for an explanation of their crazy OFF/UCR/GCR
function.

There is a weird table called 'LUT.id' which I cant find any
mention of - but I presume it is something to do with calibration.

All other files are true ICC profiles for input/output.

There are also 'device link profiles' which are a proprietary
combination of tables/profiles/scanner settings which are
presaved into a 'profile'.

This does not seem to be a true ICC profile - but a proprietary
one.

It is my guess that ICC scanning, RGB output and the tabled
DT/SOOM will be ignored...but again I will shortly see this for
myself.

That should be plenty for one day!
Martin, you have helped me more than you can know - and with
luck others as well. Cheers!

Stephen Marsh.

Join colortheory@groups.io to automatically receive all group messages.