- This topic has 4 replies, 2 voices, and was last updated 1 week, 6 days ago by
.
-
Posts
-
So in fact I try to calibrate an old 5:4 CCFL monitor. I calibrated this monitor WP against a White LED monitor with smoothed CCSS (from 3.33…. to 1 nm). I have the same white point but my I1Dplus says me I have too much green. I attach the computed 1 nm CCSS of my CCFL monitor and the 2012 10deg observer reading with matched white point. So what could be done ?
First of all those CCSS may not be a proper description for your displays. If you made them by yourself try to match numerically @3nm both displays with the spectrophotometer measuremensts , then check if i1d3 CCSS corrected difts too much from the spectro 3nm reference.
A) If CCSS correced i1d3 colorimeter drifts from 3nm spectro reading, it could be that firmware data in i1d3 does not match spectral sensivity on those wavelegths where de spikes arise. A custom matrix correction (CCMX) would solve the issue, but it will only be valid for your i1d3 and your display.
B) If spectro @3nm and CCSS corrected i1d3 match between some expected ~3dE tolerance radius but it does not look white to you it may be that “you” deviate from the STD observer you used for computations (std observers are just statistical means/medians). Choose the whitest as reference, visual white point matching to the other. If you make LUT3D for the 2nd one remember to use relative colorimetric, not absolute or whitepoint visually matched will be undone.
C) if you do not have an spectrophotometer of your own, so you cannot try to appraise i1d3 firmware data accuracy, then use B) approach
Calibrite Display Pro HL on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.CCMX didn’t helped and changing observer too. Is it possible that my eyes are making observer metameric failure ? I tried to measure with my EFI ES-2000 and I got similar results. What should I think ?
Edit : I used the perceptual appearance and I created a custom REC709 profile with 2012 10 deg observer RGBW coordinates. But I can use perceptual instead
If using 3nm readings you got the same visual mismatch than with corrected i1d3, no matter if yiutried 1931 or 2012… visual whitepoint match (and using relative whitepoint LUT3Ds)
The main problem I have too is the fact I converted from 3.3 nm to 1 nm by extracting missing information from the CCFL IPS familly CCSS. It should increase the FWHM of SPD so I could get better results but I was wrong. I gonna have to use visual whitepoint as you said. I bought this spectro for a tiny improvement. Can the PCS be bothered by the visual whitepoint match ?
