Probably one for the geeks among you
My IMatch 3 software has a validation feature which gives the dE2000 errors for the current monitor icc profile.
These dE2000 errors, in simple terms, are a measure of out-of-gamut amount of the monitor in relation to the test colour patches. A value is given for each patch and an
average.
I always use the advanced method of calibration which allows you to set the monitor colour temperature with the RGB gain controls which leaves less for the graphics card to adjust via the LUT.
I have been aware for some time that the individual RGB gain settings and the brightness setting combination can affect the validation values, both on my old cheapo monitor and my current Eizo wide gamut unit.
Recently I played with the RGB gain settings on the Eizo monitor to see what could be achieved.
By reducing the RGB (individually) values when the colour temperature is set, and correspondingly increasing the brightness to still get 120CD/Msq, the average dE2000 error went from 0.74 to 0.59. Not much overall but, significantly, the error on some individual colours improved considerably, but a few got worse.
There is a limit as to how much you can alter (reduce in this case) the gain settings as they go out of range and the optimum setting indicated by the software.
A table from a spreadsheet is attached below: the green numbers show improvement and the red are where the values got worse.
I suppose this shows it's worth using the advanced method of calibration and setting the gain values individually to achieve the colour temperature and to try different RGB values to get the best possible overall result. Once done and the trend is understood, it's then easy to maintain the improvement on subsequent calibrations.
Or is it all rather academic?
Terry.