[argyllcms] spotread Question

  • From: "Brian W. Gortney, II" <bwg@xxxxxxxxxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Wed, 9 Aug 2017 13:36:08 -0400

Good day,

Really love Argyll, generally.  Thanks for all of your hard work, Graeme (and others) !

A quick question for the list and/or Graeme:  I'm using spotread with an i1Pro2 (rev E) to read XYZ values to be later processed, but I'm seeing readings (in a fixed spot) that are initially pretty inconsistent.

Here's an Excel screenshot showing the differences in readings over time (reading number) in XYZ_Y (luminance, which is what I'm mostly interested in) when taking readings with ~1s between samples (hitting the space bar to sample so as not to move the i1 itself):


As you can see, after ~20 samples the Y values level off and are more-or-less stable (range-bound), but the difference between the initial sample and stable samples is roughly 0.2%, which is a little more error than I'd like to see for this application, generally.

So a few questions, I guess:

 * I assume this is likely a heating-related effect, in that the
   illuminant (bulb) outputs more or less light depending on its
   temperature; is that reasonable from others' perspectives?  Have
   others seen this same effect?  Is it possible anything else
   (firmware or software-related) is involved?
 * Assuming that temperature-variable light output is the issue, I
   assume the only way to eliminate these effects is to 'warm up' the
   bulb by taking multiple readings before beginning to sample.  I do
   generally plan to oversample and then average, but I'd like to
   eliminate the variations/error caused by whatever is going on as
   much as possible/practical.  Does simply warming the bulb by taking
   ~20 readings seem reasonable, reliable, sustainable?  Or are there
   any other thoughts or ideas?

To be fair, I think this is probably less an Argyll question and more an X-Rite question, but I don't necessarily see the same issue with i1Profiler (which I can't use for my application and also rounds away precision) and they're less helpful than I'd like, generally - and especially when it comes to questions about output from software that isn't theirs - so I'm hoping someone here can shed some light on whatever might be going on.

Very grateful for any thoughts on the above.  Thanks in advance!


Other related posts: