[argyllcms] Re: DE to locus = 17.3
- From: Graeme Gill <graeme@xxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Sat, 10 Jan 2009 02:26:35 +1100
Lars Tore Gustavsen wrote:
So instead of using native white point I have this concluded that at
5000K would be a naturally choice. I have the brightness at my monitor
at 100, the rest of the settings are at factory default. I have tried
with the daylight locus and the planckian locus:
dispcal -v -c1 -yl -m -t 5002.78 d50
White Correlated Color Temperature = 5025K, DE to locus = 5.5
White Correlated Daylight Temperature = 5026K, DE to locus = 1.0
White Visual Color Temperature = 4909K, DE to locus = 5.3
White Visual Daylight Temperature = 5009K, DE to locus = 0.9
And here with planckian locus:
dispcal -v -m -yl -c1 -T 5000 T5000
White Correlated Color Temperature = 5013K, DE to locus = 0.0
White Correlated Daylight Temperature = 5011K, DE to locus = 4.8
White Visual Color Temperature = 5014K, DE to locus = 0.0
White Visual Daylight Temperature = 5114K, DE to locus = 4.6
At lest the error is much better than with native white point. But I
think there is some more banding and maybe some color cast compared to
native white point.
Yes, that is to be expected when largish adjustments are made using
8 bit LUTs.
I also really wonder why the native white point is
so far away from locus?
Hard to say. If it were a CRT, the answer would be that the
guns need a tweak. Since it is an LCD, it is mainly down to
the backlight color, unless there is some sort of adjustment
still being made by the display electronics.
I have also tried higher temperatures but I don't think my monitor
have enough brightness to do that. Another nice feature of my monitor
are, If I do a recheck after some time the white point may be off with
0.5 DE or even more to locus. :-(
That's pretty good given that CRTs and Fluorescent backlit
LCDs can drift quite a bit with time (warm-up) and temperature.
Even LED backlit monitors can shift somewhat.
cheers,
Graeme Gill.
Other related posts:
