We have recently received a number of questions regarding the use of Colour Engines for LUT generation within calibration systems.

The questions seem to stem from recent PR from an alternative calibration system manufacturer, who has, for years, had issues with their LUTs causing posterization, banding, colour issues, and even what became known as 'Shark Fins', with their LUTs.
The recent PR has been pushing a new colour engine that proports to reduce or eliminate posterization and banding...
But as is often the case with PR, there is a lot of marketing in use, and not a lot of factual information.
(Or put another way, the PR is Bollox...)

So, the question is, how do you know how good a calibration system's colour engine actually is?

In reality, it is actually very to spot poor colour engines, as any good colour engine will have no operational restrictions, and will effectively do its job virtually invisibly.

To explain further, the following points highlight areas that should be of concern when assessing any given calibration system, and the colour engine used.

1) Any patch set can be used - not limited to fixed sets
Any good colour engine will be able to work with any patch set, without restriction.
That includes patch sets that are non-grid based, with what are effectively randomly spaced patches.
Any restriction to grid-based patch sets immediately points to underlying issues with any colour engine

2) No restriction on compatible TVs/displays/LUT boxes
Any calibration system that generated calibration LUTs must be able to use the generated LUTs in any TV, display, or LUT box, using the required LUT forma/file type.
If there are restrictions on what TV, display, or LUT box can be used, that again flags serious issue with the underlying colour engine.

3) Doesn't use dE Formulas in the LUT generation
And probably the biggest giveaway that there are issues with the specific colour engine is when the colour engine relies on a defined dE formula to generate the calibration LUT.
Any good colour engine will not use dE formulas in any way, as they are just for verification of a given calibration, and should never be used in the actual calibration itself.
Using dE formulas as part of the calibration means the calibration result will vary depending on the dE formula used - and that is fundamentally wrong...

If a calibration system has any of the above issues, there are underlying colour engine issues which will always compromise the final calibration accuracy.
Any good colour engine will have no operational restrictions, and will effectively do its job virtually invisibly.

Very obviously, the colour engine within ColourSpace has no such issues/restrictions.

Steve

Just bouncing this back to the top, due to resent queries we have received from users of Calman, and confusion with dE being used as a method for calculating calibration LUTs... dE should NEVER be used for LUT calculation - ONLY as a method for calibration asessment.

Also see: dE ITP

Steve