[argyllcms] Re: [Possible bug in Spyd2.c] Inconsistant readings from Spyder3

  • From: howdy555@xxxxxxxxx
  • To: Graeme Gill <argyllcms@xxxxxxxxxxxxx>
  • Date: Sat, 7 Mar 2009 13:41:37 +0100

> Does this occur if you remove the instrument from the screen for
> the first 50 measurements, and then place it on the screen ?
> (ie. is it a thermal effect as the instrument warms up ?)
I think you hit the bullseye here! :)

Take a look at the enclosed ResultsFull.txt - the result of leaving
the Spyder3 on screen for about 9 hours :)... Open it in any
spreadsheet and plot the X data - you will notice a "dog's tail" at
the beginning, then, after about 2000 readings, it stabilises.

Question to any person who knows statistics - is it possible to write
a software for warming up the spyder before commencing read?

> If I choose and R=G=B on my display to give a Y of about 0.2, this is what
> I get:

> Reading X   Y   Z
> 1   0.000000    0.175008    0.124630
> 2   0.000000    0.176625    0.126857
> 3   0.000000    0.177723    0.127206
> 4   0.106032    0.187275    0.159078
> 5   0.106109    0.187766    0.162659

Now this is interesting!!! Have you used the original integration time
or did you incorporate my "hack" that was aimed at exactly this
problem... Note that Y and Z values are OK whereas X is often a "trash
read"... I do not think X is set to zero when it is less than 0.100 -
I have managed to see values of about X=0.05 when covering the spyder
by hand (I do not have a monitor dark enough to see blacks of such a
low level). It was "flapping (switching between zero and 0.05) all the
time also the results were very inaccurate (it does not trouble me as
the error on X seems to be more than 0.04... All the results were made
using the increased integration time.

This night I will try to test the response of the Spyder3 to my
monitor's black using my hack. The question I want to answer myself is
whether I will get any flapping (=hack is useless) or not(=hack is
working).

Other related posts: