Film base color when profiling slides

[Markus Plail]

Hi there!

I got hold of Wolf Faust's IT8 targets for Fuji Astia/Sensia/Provia.
Now I am wondering whyt film base color I should use. As there are
different targets fpr different manufacturers I think I should lock it
to 1,1,1 and let the profiling software take care of it and hope that my
Fuji films all have approximately the same film base color.

Is this correct or should I calculate the film base color for every film
I use and do the same for the target and calibrate using this?

regards
Markus

 

[Bart van der Wolf]

Hi there!

I got hold of Wolf Faust's IT8 targets for Fuji Astia/Sensia/Provia.
Now I am wondering whyt film base color I should use.

Profiling doesn't require setting the film base color. The data file that
comes with the target refers to the color of the film base + dyes as seen
under a certain lightsource.

As there are
different targets fpr different manufacturers I think I should lock it
to 1,1,1 and let the profiling software take care of it and hope that my
Fuji films all have approximately the same film base color.

It'll vary somewhat from film to film, and from processing to processing,
but for the profile creation itself it isn't important.

Is this correct or should I calculate the film base color for every film
I use and do the same for the target and calibrate using this?

You can choose to determine the film base color for films you scan (not the
IT8 target) and then apply the generic profile you created to it, using
Neutral or None for colorbalance to start with.

Bart

 

[Markus Plail]

Profiling doesn't require setting the film base color. The data file
that comes with the target refers to the color of the film base + dyes
as seen under a certain lightsource.

So it actually automatically corrects for the film base color.

It'll vary somewhat from film to film, and from processing to
processing, but for the profile creation itself it isn't important.

But for using them it is, isn't it?

You can choose to determine the film base color for films you scan
(not the IT8 target) and then apply the generic profile you created to
it, using Neutral or None for colorbalance to start with.

Perhaps I should mention that I create the profile using lcms, not
vuescan.

So I think I should either determine the film base color for both:
target and slides or for none. I wonder which one is better.. or if I am
completely off the track.

regards
Markus

 

[Bart van der Wolf]

So it actually automatically corrects for the film base color.

Yes, well actually it doesn't correct. It matches the observed color with
the measured and documented color.

SNIP

But for using them it is, isn't it?
Correct.

SNIP
Perhaps I should mention that I create the profile using lcms, not
vuescan.

That is important, because for the best results it requires the actual scan
exposure to be approximately the same as when the profile was created.

So I think I should either determine the film base color for both:
target and slides or for none. I wonder which one is better.. or if I am
completely off the track.

Just see the profile creation stage as something different. It is more a
procedure to build a lookup table, than a scan. Don't worry about film base
color, just get the exposure right.
After you've got that lookup table, use it to convert the future scan
results for output. These actual scans, presumably with approx. the same
exposure as used to build the profile, will have colorimetrically correct
*film* colors, but may need tonescaling and colorbalancing to make the
*scene* look more pleasing to the eye. Part of that colorbalancing involves
the elimination of the film base color, so that is when it matters.

Bart

 

[Erik Krause]

Hello, Markus Plail
you wrote...

So it actually automatically corrects for the film base color.

No, it does not. You profile your scanner *not* the film. The target
data sheet contains values measured from the target.

Exactly. Or use media type 'image'. You should use 'generic slide'
anyway for creating and using profiles. 'Generic slide' with film base
locked to 1,1,1 should be exactly the same as 'image'.

But for using them it is, isn't it?

Yes. If you done as writte above, the profile simply does not affect
film base color (since you profiled your scanner!).

Perhaps I should mention that I create the profile using lcms, not
vuescan.

Please notice, that lcms profiles are (contrary to vuescan profiles)
exposure dependent. If you want to use them, you must use exactly the
same RGB exposure and any automatics in color balance is forbidden. So
you must use 'None' in all cases or you must lock image color to
certain values.