Nikon D1x
Nikon D1x
- Subject: Nikon D1x
- From: "tlianza" <email@hidden>
- Date: Sat, 5 Apr 2003 09:39:13 -0500
Roger Breton wrote:
/***********************************************************************/
"I would be surprised if the camera would internally convert from a 'native'
space to AdobeRGB. I don't think Nikon's engineering is that advanced.
Kodak's engineers, on the other hand, are doing just that.
I believe that the camera raw pixels are just slapped with AdobeRGB on the
way out to the storage media."
/***************************************************************************
/
I can't speak to the D1x but I routinely use a Nikon D100. Roger, Nikon
engineering IS that advanced. The cameras contain matricies that do map the
colors nominally to their respective spaces. As a matter of fact, nearly
all of the cheapest cameras actually contain and RGB->sRGB matrix mechanism.
When I first got my D100 I ran a Color Checker through the system and tested
that, when properly configured, and color managed, the results were the same
through each of the color spaces, after color management. The D100 has two
sRGB modes, each with it's own bias towards color errors in the final space.
Nikon recommends that a photographer select the WORKFLOW that they feel most
comfortable with.
As far as camera manufacturers are concerned, profiling and the ICC in
general, have been obstacles to electronic imaging, not enabling factors.
The choice of a fixed color space RGB space, such as sRGB or AdobeRGB,
allows them to measure and derive a working transformation that is
measurable and quantifieable. Many camera manufacturers "precondition" the
data with some viewing assumptions as well. If they balanced cameras to
reproduce colors exactly, no one would buy them. If they had to tag every
image with a profile, it would take up a lot of useful space on the memory
card.
The prime goal of a color managed system is to reproduce the colors that are
presented to it. The goal of most photographic systems is to generate color
images that look "good". A great deal of photographic research has
demonstrated that 1:1 tone reproduction and accurate color reproduction are
not only NOT required, but actually NOT desireable when those images are
judged for quality. If you contact me off list, i'll give you some
references that date back nominally to the year of my birth, that are the
basis for much of the thinking in color imaging today. If you can find a
copy Hunt's "The reproduction of colour", there is some great insight and
discussion on the goals and priorities color reproduction on input. Note
that these are very different than the objective requirements of the ICC
based methods.
I believe that the great camera manufacturers like Canon and Nikon, were
presented with a host of problems that they never had to deal with. The
didn't make the film, they weren't responsible for the color and tone
reproduction, that was the film guys responsibility. They bring out a
digital camera and now they get complaints that the digital images from the
camera don't look "as good" as the digital images from scanned
transparencies. That is not a color management problem per se. If you
actually measure the tone reproduction characteristics of these cameras,
they utilize a non-linear transfer function (semi-gamma related). They tend
to reproduce colors with greater saturation than the orignal, and they tend
to exhibit a somewhat higher contrast than the original scene. They make
the image look great, not accurate. Our goal as color scientists, is to
preserve and manage the image that the artist presents to us.
Take care.
Tom Lianza
Technical Director
Sequel Imaging Inc.
25 Nashua Rd.
Londonderry, NH 03053
email@hidden
_______________________________________________
colorsync-users mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/colorsync-users
Do not post admin requests to the list. They will be ignored.