DaVinci Resolve reference monitor calibrating problems

Home Forums Help and Support DaVinci Resolve reference monitor calibrating problems

Viewing 5 posts - 1 through 5 (of 5 total)
  • Author
    Posts
  • #27707

    Marek Frankiewicz
    Participant
    • Offline

    Hi All!

    I am using a wide gamut Benq SW271 monitor connected to Davinci Resolve Studio 17 via Decklink card on PC. I have this monitor calibrated with producer software (Palette Master Element) and X-Rite I1 Display Pro calibrator to Adobe RGB, SRGB & Rec709 with brightness 120 cand (3d LUTS are stored in monitor and color profiles in windows). Then I have calibrated it using DisplayCal and its DaVinci Resolve basic profile with a LUT, which is kept and used inside Resolve (Video Monitoring).  These are my observations:

    1. Using Deckling output (and even with Cleen feed) with monitor calibrated by Palette Master Element to Rec709 with LUT inside monitor and no LUT for Video Monitoring in Resolve seems to be quite accurate (checked rendered video on few other monitors and TV),
    2. whereas calibration with DisplayCal (of course in manual mode on monitor with RGB & brightnes twicking) with LUT inside DaVinci looks a little less contrasty and less saturated.

    In theory I hoped, that calibration with DisplayCal (+Decklink) will give me at least the same if not better results, whereas so far it is worse. I’ve been tweaking different settings in DisplayCal with brightness 120 cand  and still no better results. I  was choosing the number of patches for ca. 37 minutes, probably longer would be more accurate but not that much…

    What are your experiences maybe also with other hardware calibrated monitors? Is it just that I still need to try other options and it shall be fine, or  there is something wrong with DisplayCal? I know that it is free software (but was not updated since over a year?) but payable ones like Calman costs quite a lot… Please share your opinions, thanks!

    Calibrite Display Pro HL on Amazon  
    Disclosure: As an Amazon Associate I earn from qualifying purchases.

    #27711

    Vincent
    Participant
    • Offline
    1. Using Deckling output (and even with Cleen feed) with monitor calibrated by Palette Master Element to Rec709 with LUT inside monitor and no LUT for Video Monitoring in Resolve seems to be quite accurate (checked rendered video on few other monitors and TV),
    2. whereas calibration with DisplayCal (of course in manual mode on monitor with RGB & brightnes twicking) with LUT inside DaVinci looks a little less contrasty and less saturated.

    Do not guess, measure: Video samples with Rec709 R,G,B,C,M,Y 100% saturation patches, measure both setups (ArgyllCMS spotread with proper colorimeter correction if you use one). Check numerically against references.

    Also compare apples vs apples. If PME aims for a let say… Rec709 primaries and g2.4 (DisplayCAL “measurement report” to check this), do not use Rec1886 in DisplayCAL.

    • This reply was modified 3 years, 2 months ago by Vincent.
    #27726

    Marek Frankiewicz
    Participant
    • Offline

    OK, so maybe in other words here are my questions/doubts for calibration Wide gamut monitor  for Resolve via Decklink:

    Monitor itself:

    • in manual mode shall I set it up for AdobeRGB mode or Rec 709? Does it make a difference?
    • I also chose gamma 2.4.

    DisplayCal settings:

    Settings Tab:

    • I chose Video 3D Lut for Resolve, Display Resolve, I1 Display Pro, LCD Generic, Correction Matrix for I1 Display Pro, Override display delay: 1000 ms and Output levels: Full

    Calibration Tab:

    • Whitepoint:  6500 K
    • White level: 100 cd (otherwise it will take native of the panel – 350 cd)
    • Black level:  as mesured
    • Tone curve: as measured

    Profiling Tab:

    • all default settings

    3D Lut Tab:

    • all default; Input/Output Full range, Tone Curve Rec 1886 gamma 2.4 Absolute

    Are these settings correct? Where  shall I not use Rec 1886 – in 3D Lut?  What shall I change it for – Custom with gamma 2.4?

    #27727

    Vincent
    Participant
    • Offline

    OK, so maybe in other words here are my questions/doubts for calibration Wide gamut monitor  for Resolve via Decklink:

    Monitor itself:

    • in manual mode shall I set it up for AdobeRGB mode or Rec 709? Does it make a difference?
    • I also chose gamma 2.4.

    Native or P3 if available. Limiting native gamut to AdobeRGB and use that HW cal as a basis for a LUT3D meant for editing or playing video makes no sense.

    Using Rec709 vendor software preset has the potential flaw that PME works wrong, and we know that it DOES work in the rong way since it cannor measure accurately an arbitrary SW display with an arbitrary i1DisplayPro.
    So unless double checked that it works as intented better the bigger native gamut.

    DisplayCal settings:

    Settings Tab:

    • I chose Video 3D Lut for Resolve, Display Resolve, I1 Display Pro, LCD Generic, Correction Matrix for I1 Display Pro, Override display delay: 1000 ms and Output levels: Full

    Calibration Tab:

    • Whitepoint:  6500 K
    • White level: 100 cd (otherwise it will take native of the panel – 350 cd)
    • Black level:  as mesured
    • Tone curve: as measured

    Profiling Tab:

    • all default settings

    3D Lut Tab:

    • all default; Input/Output Full range, Tone Curve Rec 1886 gamma 2.4 Absolute

    Are these settings correct? Where  shall I not use Rec 1886 – in 3D Lut?  What shall I change it for – Custom with gamma 2.4?

    Google Rec1886. It is not 2.4 on a finite & very limited contrast display like an IPS. Your dispaly is not really meant to use that to see content.

    Equivalent to PME (no abs) is likely to be 2.4 relative or 2.4 abs with 100% balck output offset. Anyway as said previosuly run a verification report on PME calibration and look for “gamma vs input” 2D plot. It should look +- 2.4 flat untill it’s close to black/darkest grays where “it bends”. Using that other suggested settings in DisplayCAL LUT3D should render same results.
    Measure them, compare then.

    #27759

    Marek Frankiewicz
    Participant
    • Offline

    Thanks. So I’ve set up 2.4 with black output offset 100% (also with relative) instead and it worked! The verification went fine! I tested also with initial monitor settings AdobeRGB vs DCI-P3 and a little better results I received with AdobeRGB –  both in terms of  DeltaE and contrast.

Viewing 5 posts - 1 through 5 (of 5 total)

You must be logged in to reply to this topic.

Log in or Register

Display Calibration and Characterization powered by ArgyllCMS