- This topic has 4 replies, 2 voices, and was last updated 7 years, 1 month ago by
.
-
Posts
-
I was wondering if there is a way to split the profiling measurements in two sessions: one session using the spectro for high luminance, and another session using the colorimeter for low luminance.
The reason I would like to do this is because a correction matrix done with i1Pro2 for a i1D3 on a WOLED display does not seem to provide good enough results. I have profiled the display using a i1D3 with the correction matrix, then I made two verification measurement reports, using the extended test patch (for video): one using the i1D3, and another one using the i1Pro2. As can be seen from the first two attached files, for some patches the dE is quite high when measured by the spectro, while it’s very low when measured with the i1D3.
I have created another correction matrix based on new measurements on the spectro and i1D3, one with the default settings, and one with the minimize xy option, and then did new measurement reports with these two correction matrices. The last two files attached are these measurement reports. The results are very similar to the original one done with the i1D3.
This makes me believe the four colors matrix correction is not a good tool for calibrating these WOLEDs, maybe because these displays have four type of pixels, instead of being a 3 colorant display. This is the reason I would like to use the spectro for the high luminance range, as I trust it more when measuring this kind of display. As can be seen from its verification report, the spectro measures badly, and is extremely unstable when measuring low light. There are wide swings in the measurements from one read to the next for the same patch at the low luminance end. A colorimeter must be used for these patches.
Calibrite Display Pro HL on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.I forgot to say this is a brand new i1Pro2, which passes all the tests in i1Diagnostics, and the measurements with it were done using the HighRes Adaptive option.
I have attached to this post the three correction matrices used as well.
-You assume tha a i1Pro2 is a reference and it is not
-It is true that WOLED lack of RGB additivity may be a problem for colorimeter corrections
-Try a CCSS created from your WOLED TV native gamut for your i1d3 trusting data stored in firmware, then validate again. You only need ONE ccss for all modes in your TV if it is taken at native gamut. You can use ArgyllCMS tool specplot to ensure that this CCSS you compute is at native gamut.Hi thanks for your reply, but my question was not how to calibrate, it was whether it’s possible to create a profile using data from two measurement sessions: one for higher luminance levels using the spectro, and another one for lower luminance levels using the colorimeter.
My measurement files were there only to prove that there are differences between the spectro and the colorimeter measurements for some patches, and for increasing luminance, and that such a two-sessions method would prove valuable.
I do not assume the i1Pro2 is a reference spectro (or close to the measurements from a reference spectro). The FCCM method should work for non-reference spectros just as it does for reference ones though: the measurements done with the colorimeter using the correction matrix (a “transform” matrix would be better to say, as that would not imply the results are magically “correct” after applying it) should be close to the spectro measurements.
Using a generic CCSS is not a better method than creating a correction matrix using a spectro. In the case of i1D3, it’s still a correction matrix that is computed based on the CCSS and the internal calibration of the i1D3. The i1D3 internal calibration is not guaranteed to be valid for this unit, as there is considerable inter-unit variations, as Tom from Chromapure found out after measuring lots of them. And a generic CCSS is not guaranteed to be quite correct for my panel either.
Anyway, my i1D3 shows pretty much the same variations from the i1Pro2 that Tom encountered: red and luminance have the greatest dE from the spectro measurements. This makes me believe both instruments are somewhat typical.
I forgot to say: all the measurements were indeed taken using the panel at native gamut (the “Wide” setting). I first used the 2-point white balance controls to match the D65 white point at 120 nits, using the spectro. Then the measurements for the correction matrix were carried out.
And using a CCSS made by me instead of a generic one for this type of panel would still rely on the colorimeter internal calibration for the correction matrix Argyll will compute, which is what I’m trying to avoid by using a spectro in the first place. 🙂
