[argyllcms] Re: Awful result with Argyll 1.3.5 + dispcalGUI 0.8 + i1Display Pro & Dell U2711 :-(((

  • From: "Philip Reed" <philipreed@xxxxxxxxxxxxx>
  • To: <argyllcms@xxxxxxxxxxxxx>
  • Date: Thu, 12 Jan 2012 19:58:28 -0500


Regards - Phil Reed
Phil's Appliance Science
49 Francis St N
Kitchener ON N2H 5B4
Canada
 
http://www.mixerfixer.com
519-747-7220
 

-----Original Message-----
From: argyllcms-bounce@xxxxxxxxxxxxx [mailto:argyllcms-bounce@xxxxxxxxxxxxx]
On Behalf Of Paolo Avezzano
Sent: Thursday, January 12, 2012 6:42 PM
To: argyllcms@xxxxxxxxxxxxx
Subject: [argyllcms] Re: Awful result with Argyll 1.3.5 + dispcalGUI 0.8 +
i1Display Pro & Dell U2711 :-(((


On 13/gen/2012, at 00:28, Graeme Gill wrote:

> Paolo Avezzano wrote:
>>   do you have any link with a test on this?
> 
> There was some discussion a while ago on this list. Adjusting the
secondaries
> (CMY) to be an additive combination of the primaries (RGB) proved to
> be fiddly, and not very sucessful (matrix profiles fitted poorly).
> There were hints that the RGBCMY transform the display was applying
> was clipping the gamut.

OK, I'll try standard as well and test for gamut volume.


> 
>> I just made a XYZLUT+MTX profile and everything is quite good, except
for:
>> - global greenish hue
> 
> Compared to what ?
> 
> A purely subjective impression isn't much value I'm afraid,
> since our visual system is so quick to adapt to various
> influences outside the display itself.
> 
> If the white point hue doesn't match 50% gray, then that's a problem
> that bears investigating.
> 
> If the whole response is far from daylight/black body, then this
> indicates a problem with the instrument absolute accuracy.

Compared to the i1Profiler and, for what is worth, for my taste.
I used the same 7200K for both test. I'm not saying is way off, just a bit
greenier.


> 
>> - far from neutral darkest grey patches
> 
> Note that the default for LCD is not to adjust the black end for
> neutrality, so as to preserve the best possible contrast range.
> You can fiddle with the crossover rate between the corrected
> neutral and black using "dispcal -A rate". Try something like 8.0
> to push the correction closer to black.
> You can also have correction go all the way to black by using
> "dispcal -k 0.0" (or some in-between value), but your contrast
> ratio will suffer. In theory you could also change your white point
> to better match your black point, but the visual result may not
> be desirable.

I suspect I can't set these advanced parameters under dispcalGUI: I'll have
to test Argyll under command line again tomorrow.
Thanks for the info, appreciated.

Paolo

The advanced parameters are available in dispcalGUI.  You need to enable
them under the Options menu.  Then adjust Black output offset, Black point
correction and Rate.

Phil


Other related posts: