3D Lut Verification

Home Forums Help and Support 3D Lut Verification

Viewing 15 posts - 16 through 30 (of 61 total)
  • Author
    Posts
  • #540

    anonymous SourceForge
    Member
    • Offline

    The monitor is internally calibrated to Rec.709, with Windows color management turned off for that monitor

    #541

    Florian Höch
    Administrator
    • Offline

    There’s no way to turn Windows color management off as far as displays are concerned. If no display profile is associated, the internal fallback is sRGB. In your case, you need to enable “Use own settings for this device” in Windows color management settings and assign sRGB, because that is the only way DCG will be able to pick it up.

    #542

    anonymous SourceForge
    Member
    • Offline

    I see. But what is the effect of having sRGB enabled in Windows and Rec 709 inside the Eizo? Are you saying that when I first calibrated, ColorNavigator took its measurements from the default sRBG profile in Windows and adjusted for Rec 709?

    #543

    Florian Höch
    Administrator
    • Offline

    Whoops, my bad. I meant assign a Rec709 profile, not sRGB. Depending on the tone curve used for the internal calibration of the Eizo, you may have to create a custom Rec709 profile with that tone curve (I’m assuming the Eizo does not use the Rec709 TRC because that is a pure capture/’camera’ TRC that’s usually not used for displays. Common for Rec709 when used for a display is BT1886 or gamma 2.x). You can use the synthetic ICC profile creator to make this custom Rec709 profile.

    #544

    anonymous SourceForge
    Member
    • Offline

    Oh ok 🙂 I need Gamma 2.2 so that’s when I chose when I calibrated the Eizo. Okay will try that. So far I’m not getting good results with my calibration tests, so I may be doing something wrong. I’m testing with ICC for now.

    I first did a test using Calman colorchecker and my delta results were pretty good (with my current Eizo calibration). After applying the ICC created by DCG, deltas were over the limit. Could it have anything to do with what you’re explaining above?

    #545

    Florian Höch
    Administrator
    • Offline

    After applying the ICC created by DCG, deltas were over the limit. Could it have anything to do with what you’re explaining above?

    Possibly. Assigning a gamma 2.2 Rec709 profile, and then following my hints for the verification should work.

    #546

    anonymous SourceForge
    Member
    • Offline

    I checked and actually the Rec 709 profile from the Eizo is already enabled (default) in the Windows color management settings.

    I did a few more calibrations, created ICCs, verifying them each time, and this is what DCG says:

    Measured vs. assumed target whitepoint 2.63 NOT OK
    Average 0.53 OK ✔✔
    Maximum 1.76 OK ✔✔

    I also tested in Calman colorchecker before and after applying the DCG ICC in Windows. Here are the results:

    Before (with Eizo Rec 709):

    White Luminance: 99 cd/m2
    Black Luminance: 0,123 cd/m2
    DeltaE Avg: 1,86
    DeltaE Max: 4,43
    (GREEN – Test ok)

    After (with DCG Rec 709 applied):

    White Luminance: 89,3 cd/m2
    Black Luminance: 0,123 cd/m2
    DeltaE Avg: 1,67
    DeltaE Max: 5,01
    (RED – Test not ok)

    The main color differences are in the bright whites and grays; quite warmer with the ICC profile applied. Also the luminance reading should be at 100 and is down at 89 after DCG calibration.

    Why is that? I’m using an X-Rite iDisplay Pro, and I’m feeding my monitor via a GeForce Quadro NVidia K1000M using DisplayPort.

    Thanks for your help
    Jeremy

    #547

    anonymous SourceForge
    Member
    • Offline

    Okay, changed a few parameters (mainly chose a different spectral correction “LCD CCFL Wide Gamut IPS) and got much better results. Now getting 0.9/2.4.

    There might be a better spectral correction yet. Calman recommends “LCD (LED Blue Green)” for the Eizo CS240, I wonder what would be the equivalent in DCG. Will continue testing.

    #552

    Florian Höch
    Administrator
    • Offline

    Now getting 0.9/2.4.

    Looks good.

    Calman recommends “LCD (LED Blue Green)” for the Eizo CS240, I wonder what would be the equivalent in DCG.

    It might be something Calman specific.

    #553

    anonymous SourceForge
    Member
    • Offline

    Hello again,

    I now have a DCG ICC profile I’m pretty happy with (based on the numbers). It does look quite warmer than the Eizo profile so I need to be certain before I move on with grading.

    I tested the ICC in Calman colorchecker with a result of 0.8/2.3. Then I launched a verification in DCG with “simulation profile” enabled, Rec709 ITU-R BT.709, set my Gamma to 2.2 (same as my ICC profile) and my DCG summary results are [0.26/0.54/1.9].

    Good to go? Or is it possible to get such results in two different apps and still have a wrong calibration?

    Thanks

    #554

    Florian Höch
    Administrator
    • Offline

    Looks reasonable. Can you attach/link the measurement report?

    #555

    anonymous SourceForge
    Member
    • Offline

    Sure, here it is.

    Note that the DCG test is successful only when I select “LCD CCFL Wide Gamut” for Spectral correction. It also works with White LED IPS and OLED, but fails when I select regular CCFL, or White LED IPS.

    Thanks

    Attachments:
    You must be logged in to view attached files.
    #557

    Florian Höch
    Administrator
    • Offline

    Ok, I think you’re good to go. Whitepoint is somewhat off from the 6500K standard, but may be acceptable.

    Note that the DCG test is successful only when I select “LCD CCFL Wide Gamut” for Spectral correction. It also works with White LED IPS and OLED, but fails when I select regular CCFL, or White LED IPS.

    Makes sense, CCFL or white LED (standard gamut) corrections are not a good match for your display. Unfortunately there’s no white LED wide gamut spectral correction available.

    #558

    anonymous SourceForge
    Member
    • Offline

    Awesome. Thanks for your help all along! I think I’ll live without a 3D Lut for now 🙂

    Yeah, I’ve seen that. Is that something I could fix on the monitor settings? Cause I’ve disabled interactive calibration.

    #559

    Florian Höch
    Administrator
    • Offline

    Is that something I could fix on the monitor settings? Cause I’ve disabled interactive calibration.

    I think so.

Viewing 15 posts - 16 through 30 (of 61 total)

You must be logged in to reply to this topic.

Log in or Register

Display Calibration and Characterization powered by ArgyllCMS