Re: CV's Optical,PreCal, and Spyder for CRT/LCD
Re: CV's Optical,PreCal, and Spyder for CRT/LCD
- Subject: Re: CV's Optical,PreCal, and Spyder for CRT/LCD
- From: "Bruce J. Lindbloom" <email@hidden>
- Date: Thu, 23 Jan 2003 10:19:58 -0600
Regarding comparing Lab values from OptiCal to Lab values from Photoshop to
verify the accuracy of a Spyder:
Roger Breton wrote:
>
Did you test in OptiCal's colorimeter window whether the readings you take
>
with your Spyder match the Lab values reported by Photoshop when you display
>
corresponding RGB values with your monitor profile as the source? That could
>
give you an indication.
Jeff MacWright added:
>
That's an interesting thought. I have played around with the colorimeter but
>
haven't been able to establish a real test of the Spyder. Is anyone using
>
this method and have a process that they could share? Or any alternate
>
methods for qualifying a Spyder?
Roger Breton further added:
>
Is OptiCal's reported CIE xyY truly "D50" (I would assume so)?
>
If it is, then converting back to XYZ from xyY should be a straightforward
>
matter and from there it shouldn't be too difficult either to convert from
>
XYZ to CIE Lab, provided we know which reference white to use.
A colorimeter that measures an emissive source (like a monitor) does *not*
make use of a reference illuminant (D50 or anything else). The color sample
itself generates the energy, or colored light, so no reference illuminant is
needed or used. It's a different story with a reflective sample, because it
does not generate energy or colored light. Therefore, its color is
determined by the manner it which it modifies whatever light illuminates it.
Different illuminating light means different color and so a reference
illuminant is used to remove ambiguity.
That said, it is not straightforward to convert monitor xyY or XYZ to Lab
because in order to do so, an XYZ of a reference illuminant is needed and it
is not obvious which should be used or how it should be used. If the monitor
white point is used for this purpose, then the monitor XYZ must be adapted
to D50 before any comparison the Photoshop's Lab is meaningful. I don't know
how OptiCal converts from xyY or XYZ into Lab, but I would be interested in
learning from someone who *knows* (no speculators, please).
I also think that Roger's proposal has a certain "chicken and egg" nature to
it. When using the monitor profile as the source, Photoshop's Lab display is
affected by the (possibly faulty) monitor profile. So the test degenerates
to a situation where you are comparing something against itself. This
reveals nothing about the health of the instrument.
In assessing the quality of a monitor profile (and thereby the instrument
that made it), there's always the unscientific method of displaying an Lab
image of the ColorChecker chart on your monitor (I have one on my site) and
visually comparing it with a real ColorChecker chart. In most aspects of
color management, science takes a backseat to appearance. Customers pay for
what they see -- they do not pay for delta E. (Hey, that rhymes!)
--
Bruce J. Lindbloom
email@hidden
http://www.brucelindbloom.com
_______________________________________________
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.