Re: [APML]: SprintScan 35

[Jim Fakatselis <jimfak@netusa.net>]

Have been "lurking" in the background following the thread on scanners. I have
evaluated flatbed scanners for use as fingerprint scan input devices for the FBI.

We commonly used grayscale (or Kodak color) targets and sine pattern resolution
targets to achieve resolution and MTF of the scannner. These are available to
everyone, so negatives do not have to mailed around. Image processing programs
allow you to look at slices of each image so you could get an idea of the scanner
response vs line frequency of the targets.

Jim Fakatselis


A common process to evaluate scanners

Jerry Lodriguss wrote:

>
> >I found that the ability to control the gamma curve shape
> >is VERY useful.  And the really useful thing to adjust the
> >curve is that you can run the mouse over any part of the
> >preview image and a little dot moves up and down the gamma
> >curve to show you the intensity of the part of the image
> >you are pointing at.   This helps me a lot to decide to
> >boost or lower contrast on that part of the curve.
>
> Oh, yes!  It is a terriffic tool.
>
> >Regarding these comparisons, I guess if you are willing to
> >do these tests, I'd vote for the test film to go to you first
> >since you have a number of scanners available, plus you
> >seem have mastered the art of scanning, judging by the
> >pictures on your web site.
>
> It's funny, but I don't feel like I have enough experience with the Nikon
> and Polaroid to do the tests.
>
> I have already tried to do tests comparing the Kodak RSF 2035 to the
> Polaroid Sprintscan to the Nikon LS-1000.  Each is different in the prescan
> software.  The trouble is, you have to use them a LOT to master the
> subtleties.
>
> I have spent the last five years using the Kodak every day, and it was only
> about a year ago that I learned from Chuck that it was better to adjust the
> prescan to optimize the data. Before that, I labored under the impression
> that I was preserving the most data for archiving purposes by doing a
> straight unmanipulated scan.  So, what the hell do I know!
>
> >By the way, the scanner choice may be obvious to you, but
> >I'm sure there are many people wondering if the quality of
> >scan really is (linearly) proportional to cost.  I'm
> >certainly wondering if my $1600 Nikon is really 3 times
> >better than a $500 HP.  Some of the images on the web pages
> >around look pretty good to me.  The problem is that we
> >can't compare apples to apples unless we have the same input.
>
> I agree, I would really like to see someone (CHUCK!!!!???) try out that HP...
>
> Jerry