- This topic has 4 replies, 2 voices, and was last updated 6 years, 1 month ago by
.
-
Posts
-
Florian,
First of all, let me thank you profusely for your work on DisplayCal! I really appreciate your dedication to color science, your hard work creating and maintaining DisplayCal, and in your sharing of arcane knowledge!
I am attempting to set up a DaVinci Resolve Color Correction system and calibrate the monitors for Rec. 709 BT1886 (I will have follow-up questions after the issue below is cleared-up!).
The system is as follows:
HP Z840 Workstation with 64GB Ram
Windows 7 64bit SP 1
Nvidia Quadro M4000 and M5000 video cards
BlackMagic Decklink 4K Extreme 12G
Dell UP2715K Ultrasharp monitor (GUI)
Ezio ColorEdge CG318-4K monitor (DaVinci Resolve via Decklink HDMI)
In doing some reading online, I ran across this:
http://www.root6.com/support-2/tips/whats-the-best-setting-for-gamma-with-rec-709-video/
The author’s summation that: “So, it seems the choice is this – 2.4 gamma in well-controlled grading/mastering conditions (particularly if your monitor has good dynamic range; OLED or Dolby reference monitor) and 2.2 in brighter editing rooms with lower-end LCDs.”
On the assumption that I have good quality monitors, I have tried profiling this system in Rec. 709 BT1886 @ 2.40 gamma with white level set at 100 cd/m2, for both the Resolve system and GUI monitors and the results appear to be too high in contrast.
This might be a result of my being confused as to principals of “iterative gray balance” and the proper use of the “Apply Calibration (vcgt)” checkbox in the 3D LUT tab, or it could be that I should be using the 2.20 gamma setting for these monitors; I am unsure…
After repeated readings of the Resolve LUT generation document https://hub.displaycal.net/wiki/3d-lut-creation-workflow-for-resolve/ , iterative gray balance appears to be implemented via the calibration tone curve setting under the “Calibration” tab.
Am I right in assuming that the gray balance will be non-iterative when at default setting, but become iterative at any other setting? What determines if iterative gray balance is enabled?
Not a criticism, but I have no point of reference when you refer to “iterative gray balance” and how to determine if I need to disable the “Apply Calibration (vcgt)” checkbox under the 3D LUT tab.
So I really don’t know if I need to enable or disable the “Apply Calibration (vcgt)” checkbox in either monitor profile…
In any event, I profiled both monitors in BT 1866 and did NOT apply the (vcgt) calibration to the LUTs. Was this incorrect?
Hi,
On the assumption that I have good quality monitors, I have tried profiling this system in Rec. 709 BT1886 @ 2.40 gamma with white level set at 100 cd/m2, for both the Resolve system and GUI monitors and the results appear to be too high in contrast.
The UP2715K should have a native actual contrast of around 850:1, the Eizo is advertised as 1500:1. If you want equal contrast on both, this means black levels need to be equalized (e.g. by setting the black level target in DisplayCAL to a value that both monitors can achieve, and setting tone curve to anything else than “As measured” – e.g. Rec. 1886) as well as white level. Rec. 1886 leads to a less contrasty picture for displays with non-zero black level (this is intended by Rec. 1886).
Am I right in assuming that the gray balance will be non-iterative when at default setting
When it’s enabled (i.e. calibration tone curve set to anything else but “As measured”), it’s always iterative. There is no “non-iterative” setting for calibration.
Not a criticism, but I have no point of reference when you refer to “iterative gray balance” and how to determine if I need to disable the “Apply Calibration (vcgt)” checkbox under the 3D LUT tab.
Normally “Apply Calibration (vcgt)” should always be enabled, unless you create a 3D LUT for a desktop/GUI monitor (not driven through DeckLink or similar) and also install the profile it was created from. I think this is explained in the respective section of the Resolve 3D LUT Wiki entry.
In any event, I profiled both monitors in BT 1866 and did NOT apply the (vcgt) calibration to the LUTs. Was this incorrect?
Taking into account the above, for the DeckLink monitor the calibration needs to be applied. You can just re-create the 3D LUT from the existing profile, no new measurements are necessary.
Florian,
Thank you very much!
Sorry I missed or misunderstood the apply calibration (vcgt) section in the Wiki entry!
I am an old film timer/grader venturing into this young man’s land of bits and bytes and it’s all a bit too much for me sometimes…
Thank you for your time,
Frank
Florian,
So I just enable the “Apply Calibration (vcgt)” checkbox under the 3D LUT tab and generate a new LUT for the Ezio, correct?
Also, do you have any opinions on 120cd/m2 Vs 100cd/m2 brightness for the Rec. 709 profile?
I see conflicting opinions online.
Thanks again.
So I just enable the “Apply Calibration (vcgt)” checkbox under the 3D LUT tab and generate a new LUT for the Ezio, correct?
Yes.
Also, do you have any opinions on 120cd/m2 Vs 100cd/m2 brightness for the Rec. 709 profile?
You should use what you (respectively your eyes) feel comfortable with. Personally, in a dim room, I prefer 100/cdm2, anything higher causes me undue eye strain. In a bright room (daylight streaming in), considerably higher.
