Pink white

Home Forums Help and Support Pink white

Viewing 15 posts - 1 through 15 (of 26 total)
  • Author
    Posts
  • #2474

    Victor Wolansky
    Participant
    • Offline

    Why do you think I could be getting a pink white? it is like all is good until 250,250,250 and then it goes really pink…. I’m measuing with a Spyder 4 and selected the white gammut RGB LED correction.

    Attachments:
    You must be logged in to view attached files.
    #2476

    Victor Wolansky
    Participant
    • Offline

    FYI, is pink until 254.254.254, 255.255.255 is white.

    #2483

    Florian Höch
    Administrator
    • Offline

    Is this the same problem from here:

    https://hub.displaycal.net/forums/topic/very-weird-problem-with-new-monitor/
    https://hub.displaycal.net/forums/topic/display-instrument-tab-question/

    Or a different issue? Have you tried decreasing calibration speed (e.g. to medium)? The report actually looks good as far as grayscale accuracy is concerned, but the steps in the CCT graph seem a liitle odd, although that could be the effect of a less than 8 bit panel with FRC dithering (quantization?).

    #2492

    Victor Wolansky
    Participant
    • Offline

    sorry… I might duplicated the post, I know I had problems with the levels going abruptly from zero and ending abruptly too, but I didn’t remember if I posted about becoming pink….  according to the spec sheet of the screen is not a panel with FRC. It is for sure an 8 bits wide gammut according to the reports like 158% sRGB over 100% AdobeRGB and 90 something DCI ….

    The weird thing is that I notice this only when I use this software, if I use the Colormunki software or the Spyder Software I do not get this issue.

    Could it be the XYZ having way too many points of correction for an 8 bits LUT? May be I need to use the smoothing option? The calibration I’m doing right now takes like one hour, you mean to make one that take less time or more time?

    #2498

    Florian Höch
    Administrator
    • Offline

    Is it “pink” only in color managed apps (e.g. Photoshop) or also elsewhere (e.g. Windows desktop wallpaper)?

    Can you attach both the “pink” and “non-pink” profiles?

    #2503

    Victor Wolansky
    Participant
    • Offline

    sure, you mean the ICM file made with DisplayCal and the one made with the ColorMunki software?

    #2506

    Florian Höch
    Administrator
    • Offline

    Yes.

    #2509

    Victor Wolansky
    Participant
    • Offline

    here is

    Thanks!

    Attachments:
    You must be logged in to view attached files.
    #2511

    Victor Wolansky
    Participant
    • Offline

    the white, grays and blacks of the profile that gives me pink white before full white are nore pleasing that the one done with the Colorminki software

    #2513

    Victor Wolansky
    Participant
    • Offline

    can you see it?

    Attachments:
    You must be logged in to view attached files.
    #2515

    Florian Höch
    Administrator
    • Offline

    The ColorMunki software profile gets around the problem by scaling down the output values via the calibration curves so that the maximum digital driving level is around 250. You can do the same in DisplayCAL by setting a white level target that is lower than the current panel brightness (i.e. start a calibration with “Interactive display adjustment” enabled, white level target “As measured”, look at the current cd/m2 readout during interactive adjustment, cancel, set white level target to 95% of the reported value, disable interactive display adjustment, start calibration again).

    #2519

    Victor Wolansky
    Participant
    • Offline

    Thanks, is there a way to manually adjust those 5 last values before full white to get rid of the pink? Either on a text editor or on your software? It would be very cool if one could choose a specific color and work on it interactively.

    Now the question is why this is happening? Obviously the panel can reproduce full white correctly since the last value is OK, is this because DisplayCal is talking way much more measures that the Colormunky software and is trying to fix every single tone of white? And may be the ColorMunki is sampling white every 10 % or less?

    Shouldnt the the option to smooth the result on yiur software take care of this problem? Or can I make it to skip measuring that last 5%? Obviously by compressing the range that 5% I would be creating fixing one issue but creating another that would be less dynamic range

    I noticed when I have black compensation ON that it selects a compensation of like .4 cd/m2 which is a lot, it takes the lowest black I get of .15 cd/m2 to .55cd/m2 and things looks a bit washed out. And of course it compress even more the dynamic range, the uncalibrated monitor is about 950:1 and ends at 680:1

    as usual , tanks a lot for yiur help.

    #2525

    Victor Wolansky
    Participant
    • Offline

    FYI, my friend with the same panel has not seen this issue and he used both black as measures and white as measured on his calibration so he should had the same issue.

    #2528

    Victor Wolansky
    Participant
    • Offline

    Which files I have to send to a person with the same hardware to do the exact same calibration and profiling and verification as I’m doing?

    Thanks again.

    #2534

    Florian Höch
    Administrator
    • Offline

    Thanks, is there a way to manually adjust those 5 last values before full white to get rid of the pink?

    The correct way to deal with these is to clip them as I outlined above.

    Shouldnt the the option to smooth the result on yiur software take care of this problem?

    No, it’s not meant to.

    Or can I make it to skip measuring that last 5%?

    Yes, like I said above.

    I noticed when I have black compensation ON that it selects a compensation of like .4 cd/m2 which is a lot, it takes the lowest black I get of .15 cd/m2 to .55cd/m2 and things looks a bit washed out.

    That’s to be expected: The only way to correct the black point hue is to add some % of R, G or B (you can’t subtract from zero), so naturally this will lighten the black to some extent.

    FYI, my friend with the same panel has not seen this issue

    The effect you’re seeing may be limited to your particular panel – it’s not an indication that every panel of the same model will exhibit similar problems.

Viewing 15 posts - 1 through 15 (of 26 total)

You must be logged in to reply to this topic.

Log in or Register

Display Calibration and Characterization powered by ArgyllCMS