Hey Ruslan, I have a newish Xrite i1 Display Pro Plus handy so I fired up
spotread and tried three orientations on Dell UP2516D:
(Word on street is i1dp is the most solid colorimeter, and the design
hasn't changed much in years, for whatever this lore is worth)
Measured at center of display where I last calibrated a couple months ago,
with three orientations
(CCT — VDT)
Normal: 6539K (Duv 0.0035) — 6529K (DE2K 0.3)
90 deg: 6517K (Duv 0.0034) — 6511K (DE2K 0.2)
45 ded: 6510K (Duv 0.0040) — 6475K (DE2K 1.1)
I was holding the puck on the screen by hand and I can cause 0.3 dE
deviations by putting more or less pressure on the face of the display.
Firm pressure, normal orientation
6533K (Duv 0.0032) — 6537K (DE2K -0.0)
Some displacement on face of display face produces much diff readings:
3 inch vertical displacement, normal
6704K (Duv 0.0041) — 6655K (DE2K 1.3)
(Purely as a matter of taste I like CCT measure, as I can more easily
visualize the numbers re offset from D65 target—blue/yellow, purple/green.
But clearly VDT is more qualitative as to degree of missing target.
Something I enjoy about this subject so much is that the better things get,
the more refined at being wrong you can be!)
At rest, my readings have a tolerance of about 2–3K and Duv 0.0005, or a
dE2K of maybe 0.1? Maybe drifting with puck temp a bit?
Display color stability with temp (power-on-cold to operating temp) pretty
good at about 1 dE. However, brightness lowest to highest is about 5 dE.
Do any of these other factors help inform about the significance of the
discrepancy you are seeing?
WRT your readings, the 3 dE spread doesn't seem too bad compared to mine,
given a broader range of factors. However, if everything is being kept
fairly constant except for orientation...
Next question is do your results have say dE 0.1 repeatability given puck
position, orientation, temp, pressure, etc? If so, then seems like
instrument gets a pass as long as you use consistently.
As to knowing if instrument is correct. That question remains if you
replaced it... No?
There's this funny zone, for me, between getting readings I like, versus
knowing how correct they are. You can really go around in circles
/wire
On Sat, May 23, 2020 at 1:54 PM Ruslan Kabatsayev <b7.10110111@xxxxxxxxx>
wrote:
Hi all!
I'm trying to calibrate my monitor (Samsung S22E391H), and already at
the step of finding white point I'm stuck with inconsistent readings
of SpyderX colorimeter. In particular, if I hang it as it's supposed
to be used, I get color temperature of 6233K from dispcal. But if I
turn it by 90°, the result is 6137K, and when turned diagonally it
shows 6066K. Here's the full dispcal's line for each case:
Upright:
Current Br 209.41, x 0.3167+, y 0.3342- VDT 6233K DE 2K 0.9 R+ G-- B+
Horizontal:
Current Br 207.46, x 0.3163+, y 0.3381- VDT 6137K DE 2K 3.8 R+ G-- B+
Diagonal:
Current Br 209.60, x 0.3202+, y 0.3369- VDT 6066K DE 2K 0.4 R+ G-- B+
I suppose the device should still be useful somehow. But given this
discrepancy, I'm not sure how I should e.g. manipulate my monitor RGB
max value controls to calibrate it to D65. When I do it naïvely,
simply hanging the instrument in upright position, the resulting white
differs drastically (for my human eyes) from corresponding result on
another monitor (AOC 2269WM, which has opposite polarization).
How do you cope with such problems? Do you simply ditch such a
colorimeter and get a better one? Or is there some (easy enough) way
of averaging the measurement results from different orientations of
the instrument to set the RGB controls, and then create the actual
profile?
Thanks,
Ruslan
