Home › Forums › Help and Support › Create profile from measurement data…
- This topic has 1 reply, 2 voices, and was last updated 8 years, 3 months ago by
Florian Höch.
-
AuthorPosts
-
2018-03-02 at 17:45 #10912
Hello,
is it normal that, creating a profile starting from an existing profile, previously created using Displaycal, even without modifying a single parameter of the latter, the new obtained profile gives different visual results compared to the existing one? Shouldn’t both give exactly the same visual results?
I’m asking just because I was trying, just as a learning exercise without to much pretencies, to obtain a decent calibration and profiling for a “standard” laptop screen (HP 250 G6) and, in the tentative of fighting the nasty crushed blacks that I had obtained with this first tentative profiling, I was playing with some of the options to see if this would have led to better results (example, activating black point compensation). In this learning experience, I’ve tried that also: leaving all the options of an existing profile untouched and generate a new one from the data measurements of the former. And I discovered this strange (to me, at least) behaviour….
Attached you’ll find the two profiles, the original and the “cloned one”.
Who can enlight me?
P.S. To see clearly the differences, Ive used an underexposed image (included in the attachment), and I’ve checked the behaviour of the two profiles in Rawtherapee 5.3, Lightrooom 5.7, Darktable 2.4.1: all the 3 softwares show marked differences between the two profiles.
Moreover, in Rawtherapee, I’ve also compared the differences of the two profiles varying the rendering intent (percettive, relative colorimetric, absolute colorimetric), still, there are major differences between the two profiles, regardless of the rendering intent.Thank you!
-
This topic was modified 8 years, 3 months ago by
Gaetano Cirillo.
Attachments:
You must be logged in to view attached files.2018-03-03 at 22:46 #10931Hi,
your display seems to suffer from a nonlinearity near black. The DisplayCAL version you used previously (3.3.3) hard clipped those near black values (making the problem worse visually), while the recent version tries to maintain tonality.
In your case, a “Curves + matrix” profile may give you a smoother result (be sure to enable black point compensation as well).
-
This topic was modified 8 years, 3 months ago by
-
AuthorPosts