On 6/20/2012 6:13 AM, Chuck Norcutt wrote:
> From Silverfast's site
> ------------------------
> How do I calibrate negatives?
> How do I select film profiles in positive mode?
> A:
>
> The orange mask of the emulsion surface of negatives makes it rather
> impossible to use a standard calibration process for negative images. To
> assure fine details, the contrast would necessarily be enforced to
> strongly.
Ignoring the last phrase as incomprehensible, this assertion is demonstrably
false. I've done it with VueScan, and it
works extremely well.
> To remedy this, LaserSoft Imaging has developed the Negafix
> tool which comes integrated into SilverFast. You simply select your film
> material and go ahead producing optimal results.
Sounds sort of like VS's presets for various films, where he corrected for
various film base orange colors. In VS's
case, they are terribly out of date. I think he stopped making them when he
added true ICC profiling.
> With positive film material it is just the other way round:
> You don't need to use a specific film material profile, because the IT8
> calibration standard allows to let the software know how a scanner sees
> colours. The main idea about calibration is linearisation. However, you
> do not calibrate a given film, but a given source of light.
> If the qualities of the scanner's lamp change (due to age, warmth, more
> than one lamp in the scanner), you'll need to recalibrate.
This makes no sense until the next pp is read.
> Although different films will show different colours for the same image,
> this isn't important to calibration, because what calibration is about
> that you get the colours on the film into your image file - regardless
> of the circumstances of the digitisation process, you shall ideally get
> the same colour data.
"... what calibration is about that you get the colours on the FILM into your
image file ... " This is about the
question CH raised. - On 6/20/2012 7:10 AM, C.H.Ling wrote:
> For scanning of slides I much prefer Nikonscan, the Kodakchrome settting
> gives rather accurate results.
>
> I don't know what results you expected, you want them look good to you or
> look close to the original slide? For me I choose the later. I use lightbox
> view to compare with the scanned image side by side.
SilverFast's creators clearly agree with CH, and many others. Those used to
choosing films for their particular looks
will tend to agree.
It is directly opposed to the intend of the ISO's ICC standards. The idea is to
allow any and all creators of images and
display media to adhere to absolute definitions of color, for complete
standardization.
Properly scanned and color corrected, images from Velvia, Kodak Gold, KR,
Ektachrome, Provia, Portra, all their variants
and all the other films, would all come out identical. As my goal is generally
to capture the scene with reasonable
accuracy, I never liked Velvia, I disliked differences between yellow/reddish
KR and Blue/greenish Ektachrome, and so
on. I didn't see why they couldn't make films that were simply neutral.
To me, to paraphrase the SF text, "... what calibration is about that you get
the colors on the SUBJECT into your image
file ... "
So I was the perfect customer for ICC profiling. And they make it so much more
easy, making lots of post scanning
adjustment unnecessary. For those relatively new to the list, here are some
examples of generic scanner output with
simple profiles for orange mask colors vs. full ICC profile scanner output.
<http://www.moosemystic.net/Gallery/tech/Scan/VuesProf/>
I assume SF's approach to scanning neg film is based on the same philosophy,
that the innate color/contrast inaccuracies
in those films should be preserved in the scanning process. Then it makes sense
that they would not do ICC profiling,
and that their Negafix process would only correct for film base color and
scanner light.
Although I personally prefer the ICC profile approach, I can certainly see why
others would prefer to retain the
idiosyncrasies of individual films, particularly when they chose films to use
those characteristics for their own
artistic vision. Or when they are so used to the colors of their favorite film
that anything else doesn't look right.
I'm aware that to pretend that even the fairly rigorous ICC profiling
specification and processes using it make images
true representations of the original subjects is naive and foolish. There are
colors that simply aren't captured by some
films and scanners. There are colors, like those of butterfly wings, that
aren't simple reflection or transmission, and
not accurately represented suing a process relying on reflective or
transmissive targets, there's metamerism, and so on.
Nevertheless, I prefer it to the available alternatives.
On a related note, ICC profiling may also be used to correct for light source.
So, for example, when shooting a target
for artificial light, mixed light, overcast, early light, and so on, it becomes
possible to make the results look like
they were shot in midday natural light. This may be highly desirable for many
professional uses, and I think it was
partially in response to needs of the commercial photography industry that the
ICC standards were developed. Change
studio lights, without changing the look of the images in an example. Accuracy
of product photos is another.
For photographers like me, that is seldom the point. I want late sunlight to
look warm, images in shadow to look at
least partially cool, so I use daylight profiles for all natural light images.
Even indoors, I want tungsten lighted
images to looks slightly warm. With scanned film, that balance is hard to
create directly with profiles. The Photo
Filters tool in PS is very useful there.
Fortunately, with Raw digital files, it's easy to adjust color temp in
conversion. In fact, most decent digital cameras
are remarkably color accurate, overall.*
Color Corrected** Moose
* There are some issues with Bayer array cameras, especially color accuracy at
the pixel level, and with some, overall,
for particular colors. Ag recently posted about this. There is also a lot of
color inaccuracy from clipping of
individual color channels when exposure is not careful with bright colors,
especially red and yellow, although it can
happen with the other channels too.
** But not necessarily Correct Color.
--
What if the Hokey Pokey *IS* what it's all about?
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|