[argyllcms] Re: Strange contrast clipping during calibration

  • From: Graeme Gill <graeme@xxxxxxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Mon, 20 Aug 2012 12:11:34 +1000

János, Tóth F. wrote:
> Possible bug: contrast is notably clipped on the higher end of the
> calibration LUT
> 1.0000000 0.9105900 0.8348000 0.9275700

        you're asking for a specific brightness that exceeds what the
display is capable of (200 cd/m^2) and a non native white point, so some
sort of device model is needed to clip the white aim point
before calibration can commence. An additive model is used, so there
are two possible explanation of why one of the device is not 100%:
Either the display is not very additive causing an inaccuracy in
the clip calculation, or perhaps it's brightness changed during the calibration.

For instance, if this is a fluorescent backlit LCD display, maybe the light
has warmed up and increased in brightness. If it's the latter, then letting
the display brightness stabilise for longer before calibrating, or perhaps
using -Iw could reduce this effect. Doing an uncalibrated display report
after the calibration as well as before could give a clue if it's a display
brightness change.

Graeme Gill.

Other related posts: