Re: 2 degree vs. 10 degree
Re: 2 degree vs. 10 degree
- Subject: Re: 2 degree vs. 10 degree
- From: Peter MacLeod <email@hidden>
- Date: Sun, 9 Feb 2003 10:59:40 -0800
At 1:25 AM -0700 2/9/03, Chris Murphy wrote:
The LAB debate has caused me to think of yet another possibly "more
interesting than useful" topic. Everything in ICC-based color
management is 2-degree, D50 LAB (for output profiles) and I've
always wondered why we use 2-degree and not the newer 10-degree
Standard Observer data. Annex A of the ICC spec says the PCS is to
be based on the 2-degree Standard Observer.
What little I've been able to find on 2-degree vs 10-degree
generally says that 10-degree is considered to be better (better for
what is somewhat unclear), and some sources say it's more common.
But not in ICC-based color management.
Just curious. About five years ago I set a DTP41 to report back
10-degree, D50 LAB and built a profile. I have no data on this
anymore I'm sure, but what I recall was that the result was slightly
better in some areas, and no change in others. I'd have to repeat
the test in order to provide something other than vague supposition
though.
As you probably know, the difference between the two is how
large the color field was (in the observer's field of view)
in the color matching experiments used to derive the color
matching functions.
The 10 degree observer was based on some newer, better data, but
AFAIK the CIE believed that the original data
was still quite good, and good enough that a new 2 degree standard
observer was not necessary. (I'm amazed by an empirically-derived
standard that has lasted 70+ years!)
The bigger difference between the two is just what it says--the size
of color field. The visual system's response is a little different for
larger fields of view. If you think about how wide a 10 degree field
is, you'll realize that it would be the right thing to use when
matching large areas like wall and auto paint, but for graphic
arts usage, the number of times you have a critical color that's
going to cover 10 degrees of your field of view is pretty low,
which is why the 2 degree observer is the standard.
Note that the observer data doesn't tell you anything about the
aperture of the instrument you use to measure the data. Obviously
a 4 or 8mm spectro isn't going to cover anywhere near what you'd
see looking at a 10 degree field at normal reading distance (or even
a 2 degree field). The CMF is just used to compute CIEXYZ from the given
spectrum, such that if XYZs match under similar viewing conditions,
the standard observer would see a match at 10 degrees (but not 2,
or vice versa.)
--Peter
_______________________________________________
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.