CRT Colorimetry woes
CRT Colorimetry woes
- Subject: CRT Colorimetry woes
- From: Roger Breton <email@hidden>
- Date: Thu, 29 May 2003 22:36:21 -0400
OSX 10.2.6 / G4 733
I just got back my DTP92 re-certified from X-Rite today. Anxious to put it
through its pace, I launched ColorShop 2.6.2 under Classic to take some
measurements off my faithful 19" Mitsubishi CRT (haven't found the $$ yet
to upgrade to ColorShop X). As a test area, I created a new 3 x 3 Photoshop
document filled with a white Background (R=255, G=255, B=255). The color
temperature I customarily calibrate my CRT to is 5000K.
With the Photoshop document window about positioned in the center of my
monitor, I took my first reading (see below). I am not quite sure how X-Rite
massages the raw colorimetry, in emissive mode, but I get a sense that a Y
value of 97.92 is too bright, soomehow, for what I know about my monitor
luminance, having measured it countless times over the years -- more like 85
Cd/m2. Not satistied with this result, I decided to launch MeasureTool and
make the the same reading. Naturally, the values returned my MeasureTool did
not compare with those from ColorShop. So, I repeated the process using
OptiCal, this time, and got another set of figures. Finally -- I wasn't
going to throw the towel that easy --, I took out my trusty EyeOnePro and
measured the same screen real-estate, and got a fourth set of numbers.
Here's a summary of the various tristimulus values I get for the same screen
are using two different instruments and three different software:
------------ DTP92 ------------ --- i1 ----
ColorShop OptiCal MeasureTool MeasureTool
2.6.2 3.7.5 4.1.5 4.1.5
Classic OSX OSX OSX
== ========= ======= =========== =========
X 94.00 77.79 75.7 70.0
Y 97.92 80.91 81.5 74.1
Z 83.92 68.78 57.0 61.6
All along, I made sure that the downloaded VCGT tag in my video card
remained the same.
Out of the four set of numbers, it is plain to see that X-Rite's data stick
out like a sore thumb. I suspect ColorShop must be going through some
normalization to the peak screen output because OptiCal's and MeasureTool's
are closer to what I believe is the true historical Luminance of my CRT.
I did not use basICColor because there is no way to make single measurements
that I know of. I wish I could have used my non-US version of my SpectroCam
but my PC is on repair and the Mac versions of the SpectroCam doesn't
support single Emissive mode readings... I will try to add these numbers to
my report at a later date.
Hope this is not a waste of the list's bandwidth. I'd appreciate any
comments and insights into the measurements.
Roger Breton
Laval, Canada
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.