Testing a spectrophotometer


(Public Enemy Nº 2) #21

Hi @Claes,
Thanks! I already solved it. It was everything on the docs, shame on me.
Long story short: Under windows, instruments that are not auto-detected need an extra driver. Launch DisplayCAL and select “Install ArgyllCMS instrument drivers…” from the “Tools” menu.
Did that and the colormunki design was detected properly.
Regarding the other question I asked to @gwgill, it’s also documented and it’s quite straight-forward.
I mean producing a correction matrix for colorimeters.
It just takes a quick reading with the spectro, then switch instruments and take a second reading with the colorimeter. DisplayCal will do the rest.
I’ve already uploaded a couple of CCMX for the Colorhug1 from some of my monitors. So far one for the solid LG 23MP55 that completely fixes the red tint I always had trying to set a custom white point, and other one for the LG Flatron E2340, probably the worst monitor I have ever seen. I hope nobody has one of those, it belongs to the garbage, but I’m still curious about how close to usable I can get it.
I have a few more that I’m going to calibrate so I might try to produce CCMX for them too.
So far it’s being fun, so I’ll keep playing.

Oh, btw. I’m also playing with my aged i1 display2, trying to figure it out whether a correction matrix can get it back to work. I don’t know if this is a good or a bad idea (probably bad) but at least I can gather some information comparing the results (gamut, gamut volume, curves) with my reference which is now the colormunki design.

(Public Enemy Nº 2) #22

As I expected, the correction matrix improved the situation of the aged i1 display2 but the results aren’t good enough. I guess that aged filters affect the sensitivity of the device for certain wavelengths, so although it was better than nothing, the result was still off to be considered seriously.
I should say goodbye for once and thank it for its (rather short) service.

The Colorhug1 went from zero to decent with the CCMX, so I call it a win. It took years, but it’s solved. I think I can conclude that the factory calibration was crap, because now that I loaded the CCMX for one screen I can get the rest of my screens reasonably close with it, when it was impossible with the factory settings or the contributed CCMX (which in turn were probably created using the factory calibration as baseline, but that’s just speculation, I really don’t know).

I’m bringing this up here because I know there are dozens of people who bought their colorhugs and they are resting now in a drawer because they could never fix the damn red shift. For them, it is possible that the LG 22MP55 CCMX I uploaded improves things, although a custom matrix from a spectro and the target screen is definitely the proper way to go.