Finally I got the change to play around with LTE a while. I used Resolve as TPG and mostly used the grey ramp RBG. I didn't have time to do much except to take some pre-cal measuments, but they look okay, don't they? I worry most about clipping since the previous calibration software I used handeled that automatically so didn't really have to think about it, but I think it looks fine? I also adjusted the RBG balance after thanking that screencap but the readings in the image look pretty okay too. The monitor was set to native gamut, didn't have to try out the other modes. Covers 709 really well, expect some parts in blue.

Another thing is that I used full patches, my calibrated computer monitor was set to legal and so was Resolve's video monitoring. Is that okay? Using legal patches gave me black level of 0.20 but full gave 0.08 which is what I got from my previous calibration software.

I'm planning to do a full 3D LUT calibration later this week when I can, hopefully using the 17-point cube mode. How long does that roughly take when using i1D3?

As you have only used a Quick Profile, and not a volumetric profile, it is really impossible to say how good the gamut coverage really is.
And without knowing the monitor technology is it not possible to say if a 0.08 black level is good, but it seems high for most modern displays.
And it is impossible for any system to automatically define settings for clipping issues, as they are more often caused by incorrect settings outside of the calibration software.
That is why actually learning calibration workflows, and not relying on next-next-next workflows is so critical for accurate end results.
It is also why you need to know the correct Resolve and monitor settings, as well as ColourSpace settings, to have correct levels throughout the signal path.
And a full volumetric profile duration will vary depending on probe settings, as well as monitor near black levels, etc.

Steve

Steve
I think in the world of budget friendly monitors, 0.08 is slightly lower than average.

My question, though, was if anything in the attached image showed signs of clipping in the low end? I don't think it does, but I would appreciate your input. Also, I used full patches, the computer monitor I measured was set to legal and so was Resolve's video monitoring. Is that okay?

Sorry - without knowing the monitor make/model it is still impossible to say if 0.08 nits a as expected for black.
And as you posted the Clip graph I did assume you could see from that there is no clipping.
But that doesn't mean the levels are set correctly as brigntness could be set too high, and contrast set too low.
Clipping should really be verified with the BrightnessCal and ContrastCal images, as per the User Guide workflow instructions.
The Resolve, monitor and ColourSpace level settings are not on their own enough information.
Are you using a dedicate video card?
If not, what was the graphics card set to, and was EDID defining values?
Is there range scaling set on the Resolve timeline for input to output?
There are a lot of places range levels can be set incorrectly.

Steve

Steve
I did plan to use the BrightnessCal and ContrastCal patterns but I couldn't locate those in time, I was a bit lost with all the options. I did find them later, and will be using them tomorrow

I'm using Decklink 4K. Resolve's Video Monitoring settings are 10 bit and set for video, SDI 4:4:4 is ticked. I don't think there's anything else affecting the levels in the Resolve project, no active nodes or any active LUTs in the Color Management. I'm little confused about the Resolve settings since in the Resolve guide it's said that I should use 8 bit since Resolve is limited to that. It also say I should set Resolve to Full Range, and use ColourSpace Settings to define the actual required Patch Scale, however in my post Ted replied that I should use Video levels and 10 bit, so that's what I did.

Also, when Resolve was playing the test patterns, it did show 10-bit values in the small calibration window. You may not know what I'm referring to, so I'll take a photo of it tomorrow.

As per the User Guide, if Resolve is set to 10 bit it will bit-shift the input from 8 bit to 10 bit.
That is why it is suggested 8 bit it use.
The issue is a Resolve issue that BMD have not bothered to update, even though they have been provided all the necessary info a number of times.
Same as BG colour and patch position.

Steve

Jonahan
Name the brand and model of your monitor!

Well, my first 709 LUT (peak chroma) doesn't look so great after pretty good profiling results with the 17^3 cube on BenQ PD2725. Probe was i1D3, White LED (Konica Minolta CS-1000 5nm) CVS was used.

No active nodes or LUTs in Resolve, video monitoring settings were:
Levels: video
8 bit
SDI 4:4:4 ticked
Output via Decklink 4K

BenQ PD2725 was set to Video, ColourSpace patches were Full.

Definitely not sure what happened with EOTF during the active LUT verification (i don't recall the cube size, maybe around 2800 patches or so) but now the blue channel seems to be clipping in the dark end, could I have lowered the blue channel too much while setting the grey balance? The monitor only has RGB balance settings that affect the bright end of the image, but that doesn't look too great either. The EOTF graphs look fine when the monitor is in the pre-cal, native gamut state.

What also troubles me are those drift readings, I used 100 patches of 240/240/240 white, no idea what happened there or what that's really suppose to look like, this is my first time looking at that graph. Also, there were some issues with the gamut coverage, especially with Blue, but I don't really know how to interpret those any better.

These anything elso jump out to you? I'm really not sure what to try next or what could be causing these issues. Bad probe, bad/wrong CVS file, display instability come to mind.

Also, please note that the active LUT verification image shows the activeLUT selection being empty but it was on when I did the verification, it must have turned off afterwards, before I took the screencap.

Sorry, but it is basically impossible assess anything from the screen grabs.
Please post the before profile, the LUT, and the verification profile.
Also please stated your full settings.
Foe example, I can see you do not have the correct Luma Min/Max values in the Target vs. the Profile.
That is well explained in the User Guides.
(Easiest to tick the 'Auto' box, or press the upload '^' arrow button.)
And what Extra Delay, etc., do you have set?

Steve

Steve
How do I post the before profile, the LUT, and the verification profile?

Stick them into a .zip file.

Steve

Added the BCS files. The LUT I had not saved, like a little dum dumb...

No extray delay added, didn't think about it at the time.

Attached file: BenQ calibration.zip

Initial view is there is likely errors due to incorrect Extra Delay, as well as the incorrect Min/Max settings, etc.
I would start again.

Steve

Steve
Thank you, I will.

My second attempt with 0.500 sec extra delay. The results look pretty good to me at least, but does anything jump out? The LUT was uploaded to Resolve, but due to time constraints I had to do use the 12^3 cube for verification, for active LUT checking I used 17^3. There is a bit of a dip in the brighter end of EOTF but I'm not sure how much of an issue that really is.

Attached file: BenQ calibration second atte.zip