After years with Viewscan, I decided to try and run my Nikon Coolscan 5000
on Nikon's own Nikonscan software. So far so good, except for one problem: I
can't figure out how to scan several frames with identical settings --
essential for stitching a panorama. I quickly discovered that using color
management made the software auto-everything, resulting with each frame
being radically different. So I turned CM off. But even though I am using
exactly the same settings for each frame and turned off the automatic
histogram equalization, there are differences. Seems that there is still
some post-processing going on over which I have no control. This is negative
film, so perhaps it has something to do with film base color compensation,
or perhaps with something else that's not on any menu. Is there a way to
coerce Nikoscan to produce a set of scans with identical exposure and
processing? Can anyone help?

Happy Traveler wrote:
settings? If you shoot panarama and the camera is set to auto exposure,
the shots will be exposed for correct exposure for each frame, and if
you're shooting a very wide angle setting (and what else would we be
shooting with panorama?) then the exposures will probably vary between
shots. Been there, done that mistake.

I have not tried autostitch, maybe that's the ticket if you do a lot of
pano stuff.

http://www.cs.ubc.ca/~mbrown/autostitch/autostitch.html

The autopano pro talks about doing exposure and color compensation

JD

Thanks for your kind advice. I am using Autopano Pro (believe that it's a
commercial derivative of Autostitch). Indeed, it has some features for
mitigating exposure and color variations; even claims to correct vignetting.
Did not do a perfect job, perhaps because I am still learning how to use it
effectively, but certainly helped.
As for the source negatives, they were all taken on manual exposure. The
advice I am seeking is about using Nikonscan effectively. I recently
switched from Vuescan to Nikonscan because of Vuescan's instability, poor
documentation and lack of true ICE implementation. To its credit, however,
performing a set of identical scans, with fixed exposure, base color
compensation, etc, is a trivial matter. Wish it were that simple with
Nikonscan. Or perhaps it is, and I just need to be pointed in the right
direction?