[argyllcms] Re: Camera calibration: LUT only as good as matrix?
- From: Graeme Gill <graeme@xxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Wed, 03 Jul 2013 18:41:36 +1000
Torsten Bronger wrote:
> indeed reduced. This is a good thing. But I would have thought
> that the *much* bigger flexibility of a LUT would decrease the maximal
> residual error drastically. But this is not the case. Some patches
> are still way off in a noticable way. Why is this? Is my workflow
> flawed, or is this as good as it gets?
Hi,
hard to say without a detailed investigation. But you get irreducible errors
when two patches are in conflict - ie. they have close to the same device
values, but have different CIE values. This can happen for a variety of
reasons, but one cause specific to input devices is if their spectral
sensitivity is different enough from the standard observer that they
"perceive" metameric matches quite differently.
Graeme Gill.
Other related posts:
