Re: Weird Color Behavior?
Re: Weird Color Behavior?
- Subject: Re: Weird Color Behavior?
- From: Roger Breton <email@hidden>
- Date: Tue, 02 Mar 2004 21:57:28 -0500
I have two monitors side by side, two 19" monitors, that is. One is a BenQ
LCD (I wouldn't use it for color critical work but for the purpose of this
test it's OK) and the other is my faithfull Mitsu900u aging CRT. The 900u is
calibrated to D50 while I've experimented with calibrating the BenQ at D50
and D65 while finally settling on D65. I'm using the excellent MonacoOptix
XR Pro package so all my measurements are done on the same absolute basis.
First experiment, how does the Lab version of the ColorChecker II chart
display across the two monitors -- the eternally revisited D50 vs D65
question. I'm using Photoshop CS. Well, it turns out that, when I calibrate
the BenQ's white luminance to match that of my 900u's luminance, about 80
Cd/m2, the comparison is fair. That is, I get two rendition of the
ColorChecker reasonably looking similar in appearance and luminance. On
closer inspection, there are some disturbing differences between both
monitors's colorimetric accuracy, as expected. And one does not need a
senstivei instrument to spot those... But the BenQ grayscale capabilities is
surprising, not at all over the place as I would have expected. Probaly due
to Monaco's excellent colorimeter and good algorithms, and probably due to
BenQ LCD technology. That's good news for LCDs, I figure. For the record,
the DeltaE of the grayscale on both monitors is within 1 to 2 units of
expected. Now, there is substantial difference between the two ColorCheckers
color appearance, in this side by side comparison: the one at D65 (BenQ)
clearly looks bluish while to the one at D50 (Mitsu) does not. The reference
physical ColorChecker chart sits between the two monitors and is being lit
by a 4700K Solux lamp.
Second experiment, when I convert the Lab ColorChecker chart to sRGB, using
AbsCol -- yes AbsCol --, the resulting colors do take on a global yellowish
cast because Photoshop tries to approsimate a D50 white point on a
D65-calibrated monitor. Fair and that's as expected. I've observed that in
the process, some of the 24 patches do appear exagerately tainted with
yellow but, overall, it's not as bad as I thought it would. For sure it does
not come near the color appearance of the chart as displayed on the
physically D50-calibrated monitor but it's not bad. Nothing phenomenal but
certainly not the disaster I expected. Interesting and encouraging...
Roger Breton | Laval, Canada | email@hidden
http://pages.infinit.net/graxx
_______________________________________________
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.