[argyllcms] Re: dispcal VDT DE 2K measurement discrepancy

  • From: Vallo Kallaste <kalts@xxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Fri, 19 Dec 2014 16:30:12 +0200

On Tue, Dec 16, 2014 at 09:44:57AM +1100, Graeme Gill <graeme@xxxxxxxxxxxxx> 
wrote:

> Vallo Kallaste wrote:
> 
> > 5) Check all
> >   Current white = x 0.3263, y 0.3431, VDT 5750K DE 2K  0.8
> 
> > 2) White point (Color temperature, R,G,B, Gain/Contrast)
> 
> > / Current Br 105.84, x 0.3263+, y 0.3421-  VDT 5770K DE 2K  0.1  R=  G=  B= 
> >  q
> 
> It's not unexpected that they have different DE's because the white point has 
> changed
> between the measurements - i.e. y .3431 vs. .3421.

Out of curiosity I looked at JETI specbos 1201 specification. There is CCT
reproducibility data, given as ± 20 K. Of course LCD backlight is not
illuminant A, but can I relate dispcal shown VDT drift to specbos CCT
reproducibility? I mean that when near lab-grade instrument is able to give
reproducibility in ± 20 K range then my assumptions about consumer grade
instruments are plainly wrong. It seems logical to expect that reproducibility
for CM Photo is much worse than ± 20 K. This in turn means that VDT drift (and
DE 2K) numbers shown by dispcal should be taken with a lot of salt so to say.
The numbers are what they are, limited by instrument absolute error and
reproducibility.
Of course I could be wrong and those CCT reproducibility numbers mean something
else and we're blessed with near lab-grade CM Photo, but I doubt it. That's why
I am thinking out loud. Anybody knowledgeable please set me on right track.

thanks,
-- 
Vallo

Other related posts: