[argyllcms] Re: Color profile shadow issues
- From: Graeme Gill <graeme@xxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Tue, 26 Sep 2017 17:14:16 +1000
Brandon Bates wrote:
Hi,
It appears that
the video lookup curves have a power curve on the gamma (straight gamma)
while the profile has a roll off in the blacks (follows the BT1886 from what
HCFR says) First I'd like to know which is correct (power curve or BT1886
for the profile),
it's not clear which profile you are referring to here, the source profile
or the destination profile. I'll assume source profile.
Calibration curves are there to improve the display response (make it better
behaved/more controllable as the [destination] profile sees it), and to set
the response for non-color managed applications.
Because the destination profile is created on top of the calibrated response,
it will make the calibration have no effect on the overall response when
color managed, apart from what I've already mentioned - setting the white point.
When color managed, it's the source profile that determines the overall
response, not calibration curves. So the source profile response and
calibration curves are largely independent, meaning that there is no "correct"
or "incorrect".
and second if the BT1886 roll off is correct, I'd to know
how to set my display calibration video LUT to match it (one so that images
viewed in non color managed programs will be a bit closer to correct, and
two so that "bits" arent used up doing level translations twice causing more
banding than necessary)
Using dispcal you can determine the calibration curve response using
the -g parameter, combined with the -f parameter and -a parameter.
<
http://www.argyllcms.com/doc/dispcal.html>
You can certainly achieve BT1886 like responses with those controls.
Cheers,
Graeme Gill.
Other related posts:
