[argyllcms] Re: patch 2561 of 3000: sample read failed

  • From: Graeme Gill <graeme@xxxxxxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Fri, 26 Nov 2010 09:10:03 +1100

Niccolò Belli wrote:
Sample read failed due to misread
Hit Esc or Q to give up, any other key to retry:
The instrument can be removed from the screen.
dispread: Error - test_crt returned error code 1

Hi, it's impossible to say for sure without some details of the problem -
ie. output of -D option (ie. "dispcal -D5 ..other options & filename .. 2> 
log.txt").

But note that if the instrument drifts too much from it's initial
black calibration, it's readings can then exceed the sensor saturation
point, and this triggers a misread. I made the margins pretty high though, so
it sounds like a heck of a lot of drift if that's the explanation! But
if it drifts that much, the readings may be useless anyway. So as well as
using -Ib (or -Ibw if your display doesn't have rock solid brightness over
the measurement period), you should cut the number of patches so that the
measurement is done over a shorter period of time (say 30 min max).

What's the problem? I wasted half a day to hardware calibrate and
profile the panel :-(

Sounds excessive. Displays are generally better behaved than printers, so
if you are using anything over 1000 test patches you need to determine if
it is adding anything, or simply adding problems due to instrument and/or
display drift.

Graeme Gill.

Other related posts: