Originally Posted by dbillen
The new meter helped a lot but there's still a lack of color resolution in saturated blues that looks vulgar when they're involved in gradients. That ConnecTED guy saw a question I had about it and asked me to send him my calib files and he's going to tell me what's wrong with them.
Hi, I received your 3D LUT correction file generated from CalMAN, to investigate it, since CalMAN don't provide any tool for the user to be able to see what the correction is doing, which is a very important detail, the only way to locate problems or evaluate the performance is the take measurements with a meter, verify color reproduction patterns (this is why I have added about 150 Color Reproduction patterns to my calibration disk for that evaluation) or watch real content to see if there problems introduced that pretty dE reports are not showing. All that kind of verification takes some time, the measurements, the verification etc... if you have CalMAN only.
For me who has CalMAN but LightSpace also, since LightSpace has tools and you can import any 3D LUT correction table, I can locate the errors in cube generation in a few seconds, not only for those generated for my system but form any other user used any calibration solution and it's experiencing problems (like you).
There 3 ways to locate the problems, looking the: 1) 1D LUT Viewer of LightSpace, 2) Looking and rotating the 3D Cube Viewer, 3) using LUT Preview.
Looking the 1D LUT Viewer, I see that you have re reduction of green channel at your low end, this will impact your near black chromatity and reduction of near black luminance. Looking your RGB channels seems that you have pre-calibrated your display 20-point grayscale before running the 3D LUT, so since you have pre-calibrated 5% Gray also, this extra correction @ near black will provide more dE at your low end. You can very this by taking a 21-Point Grayscale run.
Looking the 1D LUT Viewer also, I see that the cLUT has dropped your peak output, so if had 130nits pre-calibrated, I estimate that you have about 110nits post-calibrated.
Your correction is not looking so smooth, I can spot points where calman has calibrated, so seems that it's focusing more to specific measured points only and not looking a lot about the linearity of the correction from point to point...but propably your pre-calibration of grayscale is responsible also for this non-linearity from point-to-point.
Looking the 3D LUT viewer I see that your blue has desaturated a lot (towards to inside the cube), it's obviously an error, because LG OLED never had so much oversaturated blue's by default to force calman to do such high desaturation. Also your Cyan top end (high luminance levels) have shifted a lot towards to blue.
Finally using the LUT Preview, you can load your own pictures and you can see how applying your correction is affecting the image. Here you have see the problems, if you have my calibration disk and you upload that correction to your eeColor and see the same patterns you will see the same results. Just that LUT Preview of LightSpace saves a lot of time to upload and verify manual from your player.
This is my help (using LightSpace tools) for you locate where you have problems, send the file that CalMAN generated to SpectraCAL also, among with the other files CalMAN creates during LUT generation to help them find locate what causing these issue.
Also take 10-Point Luminance run with eeColor at UNITY to see if there all good to your Cyan/Blue before running a new AutoCAL.
Also check with a 7-color clipping pattern, your problem is not visible with contrast pattern.
Using the Advanced Contrast Flashing Bars Pattern, you are checking:
Using the 7-Color Clipping Bars Pattern to prevent clipping of each color channel (WRGBCMY), this sometimes can be fixed be removing some additionally clicks from the contrast control also, you are checking these areas: