- This topic has 4 replies, 3 voices, and was last updated 4 years, 2 months ago by
.
-
Posts
-
Hello everybody!
I have encountered an anomaly while testing the same 15,6″ IGZO display (Sharp 43N80-LQ156D1) in the same mobile workstation running Ubuntu 18.04.2 LTS operating system and using DisplayCAL 3.5 calibration software with the same settings. The graphic card is integrated Intel UHD Graphics 630. The ambient temperature and light were almost identical and the measurement were made in the centre of the display.
I performed 8 calibrations with the following settings: white point D65 and Gamma 2.2. I used three different colorimeters from X-rite:
1. ColorMunki Display (serial number CM-15.A-02.170897.09 made in 09/2015),
2. i1 Display Pro (serial number I1-17.B-02.264199.01 made in 01/2017),
3. i1 Display Pro (serial number I1-18.B-02.316261.10 made in 10/2018).All of the measurements were almost identical between the tests and devices, particularly good for a mobile workstation display. One of the result sets was: 99.4% sRGB coverage, 99.2% AdobeRGB coverage, 88.6% DCI P3 coverage, average dE*76 0.10 and maximum dE*76 1.26.
Strangely, despite the white balance set to D65 the actual colour temperature in the profiles varied significantly between the colorimeters, but the results were consistent within one device. All of the profiles from the colorimeter #1 give the warmest colour temperature, all from the colorimeter #2 give cooler colour temperature and all from the colorimeter #3 give the coolest colour temperature. The profiles are indistinguishable within a device and easily distinguishable between devices.
I called X-rite, but the could not explain the anomaly. One of X-rite engineers suggested to run calibrations with the native display white point (“As measured” option in DisplayCAL). As a result, I have got three visually indistinguishable profiles, but every colorimeter measured different temperature. Colorimeter #1 measured the native display white point at 6900K, the colorimeter #2 measured 6700K and the colorimeter #3 measured D65.
I wonder if colorimeters deteriorate over time and tend to give warmer profiles as they get older. I also would like to ask if know that X-rite has changed firmware over the last few years in a way which could effect white point measurements only. Any ideas?
I would like to highlight that the newest i1 Display Pro has slightly different sticker on the counterweight, but is still the “Rev. B-02”.
Calibrite Display Pro HL on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.-No information about which spectral correction did you use for colorimeters, they need it.
Also I’m not sure about which one for that panel, you’ll have to do a little research. ” 99.4% sRGB coverage, 99.2% AdobeRGB coverage, 88.6% DCI P3 coverage” looks like old GB-LED but it will be weird for a laptop with ~1yo or newer.
If that laptop has some vendor calibration on windows like some dells or lenovos, try uing RG_phosphor or RGBLED, because they are the one that such software may use.
There is no way that such laptop has to use bundled “WLED”… if in your research you found “W-LED” it is NOT WLED (like sRGB WLED) but widegamut W-LED like W-LEE PFS phosphor corrections (and they ae bundled with DisplayCAL)-Color correlated temperature IS NOT A VALID WAY TO COMPARE WHITES, so your comparisons are wrong. Compare color distance between them of maybe against calibration target.
Hello Vincent, thank you for the comprehensive answer, but I need some explanation as I am not a pro in display calibration.
I used the “Auto (None)” correction setting as I was not sure which one was the right for my display. The display is a 15.6″ IGZO display in Dell Precision 7530. Dell calls it “15.6” UltraSharp UHD IGZO, 3840×2160 AG, NT, No WWAN, Cam/Mic, w/Prem Panel Guar 100% color gamut” and the actual model is Sharp 43N80-LQ156D1. I think a very similar display is used by Dell since 2015, so it might be something a bit old-school. Which correction should I use?
Correct me if I am wrong, but I think that three colorimeters (if not defective) should give the same results on the same display with the same settings.
Which one:
If that laptop has some vendor calibration on windows like some dells or lenovos, try uing RG_phosphor or RGBLED, because they are the one that such software may use.
Regarding:
Correct me if I am wrong, but I think that three colorimeters (if not defective) should give the same results on the same display with the same settings.
No, uncorrected they may or may not agree… that why we use corrections. Corrected should be between some “valid margin”… lets put 2-3dE maximum.
Spectral corrections (CCSS) are applied reading EACH colorimeter unique (supossed to be unique…or at least by batch) spectral sensivities (colorimeter “observer”) stored in firmware, then evaluate such spectral sample (CCSS) against colorimeter observer and some reference standard observer (usually CIE 1931 2degree). Colorimeter observer and standard observer may not agree, a matrix correction is computed on the fly from evaluation of RGBW samples in CCSS.
1 common CCSS for n colorimeters and a display (type) , n corrections computed on the fly from firmware data,
