This requirement for accurate calibration of displays with poor linearity (which is most displays as stated above) requires the use of 3D LUTs generated from full 3D cube based profiles.
But, not all 3D LUT based calibration is made equal - because of the overriding desire of many calibration systems to focus, incorrectly, on Delta-E, grey scale and primary colours as the definition for accurate calibration.
This is not to say that Delta-E reports are useless, and should be ignored, or that the values they report are untrustworthy (ignoring the fact that the actual values reported can be deceptive), but that good Delta-E values alone are no guarantee of accurate calibration. All the colours that Delta-E vales do not report on are equally important, and must be equally as accurate for good final calibration.
Every Colour Point MUST Be Considered Equal
From the above description of calibration issues it can be seen that every colour point has to be given equal importance during profiling and calibration, not just grey scale and primary colours.
That above statement is so important for accurate calibration it is worth stating again!
Every colour point has to be given equal importance during profiling and calibration, not just grey scale and primary colours
And the only way to do that is to profile all points within a 3D cube equally, and we recommend the use of full 21^3 profiling for any critical calibration, as this covers the entire colour space very accurately, with a good level of granularity.
The cube here shows a full 21^3 point profile, with all colours measured equally, and so all colours calibrated to a higher level of accuracy.
Within LightSpace CMS there are options for a Quick profile and Hybrid profile modes, which can produce excellent results, but they will never be as 'accurate' as a full 17^3, or better still a 21^3 profile.
(But, depending on the display the variation may be next to invisible - but never as accurate).
The image here shows a 5^3 cube profile combined with 17 point grey scale and RGB colour. It is easy to see that none of the points align with the skin tone colour locations, and there is still a lot of 'black' un-calibrated space within the cube.
The use of skin tones as an example of 'real accuracy' is a perfect one, as they are critical memory colours and you can only get them accurate by accurate profiling - no interpolation will ever be accurate as is it - well - interpolated information, not accurate profiling.
The problem with interpolated data is lack of accuracy, as can be seen in the following images, where you can see the difference between calibrations generated by the above two approaches on the same display.
The edges of the LUT (primary colours) and the grey scale are identical, while the interior of the cube is very different, showing the additional accuracy afforded by treating all colours equally.
What all the above shows is that relying Delta-E as a guide to overall calibration accuracy can be fraught with issues, and should be used as a 'guide' to accuracy only.