[argyllcms] Re: dispcal - blackpoint still corrected despite -k 0.0

  • From: Graeme Gill <graeme@xxxxxxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Wed, 27 Oct 2010 23:40:08 +1100

Anders Torger wrote:
With that I expected that for RGB 0,0,0 the calibration curves would
also be at 0,0,0, but looking at the calibration curves in the .cal file
I see:

I      R        G        B
0.0000 0.023793 0.020968 0.020392

That is non-zero for the black point, and also slightly different values
for RGB, so desipte that -k is 0.0 there seems to be black point
correction. Is this the expected behaviour, or am I doing something

Notice that these are device values. The device response could be
almost anything, so how do you know that the above produces something
other than black ?

It seems like the -A parameter can control when calibration is reduced,
but I expected that with -k 0.0 the curves will always end at 0,0,0.
Could it be that the gamma setting affects the process?

No, it is expected if the display response has a flat spot at zero.

Because the calibration curve is composed of finite steps, it
can't instantaneously transition from zero to what it needs to be to start
displaying something (ie. the inverse of flat is an infinite slope
step). So to make the transition smooth and the response progressive,
the zero is a device level that is as large as possible and has a color
that (hopefully) is the same as black.

Graeme Gill.

Other related posts: