How to find constant luminance? Photoshop LAB disagrees with DigitalColor Meter, and looks wrong
How to find constant luminance? Photoshop LAB disagrees with DigitalColor Meter, and looks wrong
- Subject: How to find constant luminance? Photoshop LAB disagrees with DigitalColor Meter, and looks wrong
- From: John <email@hidden>
- Date: Thu, 26 Oct 2006 12:07:30 -0700
Goal: To display, on a calibrated iMac monitor (OSX 10.4), all hues
and saturations available at a given Luminance value.
Problem: Photoshop LAB color-picker does not appear to show correct
constant L*, and disagrees with Apple DigitalColor Meter.
In LAB mode, Photoshop 7 color picker displays a range of hues at a
*supposedly* constant Luminance value.
A simple visual inspection shows this to not be true. For example,
let "L = 25%" in photoshop LAB color-picker. Notice the bright purple
(bottom right) is much brighter to the eye than the dark yellow (top
center). Photoshop reports both of these as L=25.
Apple's DigitalColor Meter, on the other hand, shows a large
variation in L* across the photoshop "constant" L display. In fact,
the DigitalColor Meter shows there is only a narrow band of colors in
the center of the color-picker display that have 25% luminance
(including dark yellow). Here are the LAB values for Purple, at the
photoshop L=25 level, as reported by the two tools:
Photoshop picker (L=25, a=127, b=-127);
DigitalColor Meter (L*=48, a*=55, b*=-81)
Clearly a* and b* are scaled differently, but L* in DigitalColor
Meter has the same range [0-100] as L in photoshop LAB. The point is,
the Luminance of purple is almost twice what photoshop reports.
The question is: Is there any way to display all colors for a given
Luminance, so that the colors actually appear with the same
Luminance? So that if the display were in grayscale, there would be,
theoretically, only one gray level? (I know conversion to gray levels
introduces other concerns).
Thanks for your response.
I am basing all of this on the following assumption:
The L* of CIELAB (and photoshop LAB), represents the cone-response
Luminance value for color sensation. Because L* models the
physiological cone-response (via the Luminance Efficiency Function),
all colors having the same L* should appear to have the same
luminance, should seem equally "bright". (Referring to Fairchild's
"Color Appearance Models", and leaving aside issues of simultaneous
contrast, surround, background, etc. ).
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden