[filmscanners] Re: LS-2000 and Kodachrome

["Arthur Entlich" <artistic-1@shaw.ca>]

A quick comment.

Kodachrome films are unique n their processing, in that they are almost
a dye transfer process.  They start out as black and white films with a
number of extra filtering layers to make the color separations.  The
actual color is added during the processing, with dyes.

What that means is that the colors can be changed as "simply" as using a
different dye solution on one of more of the dye batches.  Although I
don't know how many times such changes occurred, but Kodak changed the
colors in Kodachrome man times in its lifespan.  Profiling will
certainly help accuracy, but be prepared for variations based upon age,
fade, and more importantly dye lot and Kodak's specs at the time.

Art

Paul D. DeRocco wrote:

> I came up against the dreaded blue cast when attempting to scan some
> Kodachrome in my LS-2000. On Usenet, people have been fighting with this,
> trying to find some easy tweak to the controls the controls to fix this. I
> decided to buy a Kodachrome IT8 target and just profile the damn thing. I
> found the slide at B&H, and the data file on the Kodak FTP site, and built a
> profile.
>
> When I assigned the new profile to the scan of the IT8 target, it looked
> very nice, but I was surprised to see that there was still a blue cast in
> the light grays. After puzzling over this for a while, I decided to actually
> look at the slide, and saw to my surprise that it actually had a visible
> blue cast in that area. So the profile works.
>
> I also tried boosting the master analog gain on a dark slide, and comparing
> the result to boosting the levels using PS Curves. I was relieved to find
> that, while the former produced a cleaner result, the colors were identical.
> In other words, the scanner is linear enough that the you don't need
> different profiles for different master analog gain settings.
>
> If you want to correct for the blue cast without profiling, you can't do it
> with the analog gain controls, because the problem is that the scanner is
> misreading the density of the yellow dye, and there is no yellow dye left in
> white areas. So if you cut back on the blue, whites will turn yellow. There
> is no alternative but to apply some sort of curve. But of course, that's
> what a profile does.
>
> By the way, I was doing all my tests with NikonScan. I'm now trying to
> figure out how to build a profile that works with VueScan. The built-in
> profiling in VueScan is very primitive, producing a profile that only has
> some primary colors and simple curves--it's not a complete 3D profile. So
> I'm trying to figure out how to make a real profile that has the right
> response to work with VueScan. If anyone has done this, I'd appreciate any
> suggestions.
>
> --
>
> Ciao,               Paul D. DeRocco
> Paul                mailto:pderocco@ix.netcom.com
>

----------------------------------------------------------------------------------------
Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe 
filmscanners'
or 'unsubscribe filmscanners_digest' (as appropriate) in the message title or 
body