[argyllcms] Re: ColorMunki measurement drift

  • From: Juergen Lilien <jp.lilien@xxxxxxxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Tue, 28 Sep 2010 06:08:21 +0200

Graeme Gill wrote:

> I did a simple test using my ColorMunki and a CRT based monitor.
> I've had the instrument taped to the one spot, and the display
> and instrument powered up for some hours. There was a noticeable
> amount of drift over the first hour or two, but after that the
> readings were fairly constant.

Thanks for testing!
I would consider the temperature of a CRT tube as very cool (isolating
vacuum), compared to a LCD screen with backlight, so I'm really
astonished that even with a CRT there is such a long lasting drift.

> Using spotread, the error to a reading taken as a reference once the drifting
> had finished remains under 1 delta E, and averages about 0.5 delta E.
> The smallest error is less than 0.1 delta E.

The smaller temperature difference (room temperature<->CRT) should
result in a lower drift, so this does not signify that your
ColorMunki is better/less prone to drift, right?

> Recalibrating does not seem to affect this.

This means that your ColorMunki is also not back to its first readings
after a self-calibration?

> [I'd guess that the variation in this case is due to the lack of
> synchronization between the CRT refresh rate and the instrument integration 
> time.]

This reminds me of the question, if it is possible that the temporal
dithering used in panels to gain 10 Bit resolution is jointly
responsible for measurement variance?

Regards, Juergen

Other related posts: