- This topic has 10 replies, 2 voices, and was last updated 4 years, 6 months ago by
.
-
Posts
-
So I tried to calibrate a Panasonic TH-42PF20E, a plasma display, with i1Display Pro, targeting to 100 nits, 2.2 gamma and wb 6500K, with white level drift compensation turned on.
The results however were abysmal after the first attempt which weren’t super great either, as you can see from the report. I tried to turn off every power saving feature with no success to counter act the ABL. The contrast got worse with every attempt and at the same time the black levels rose.
With every attempt it took almoust an hour to finish the calibration, but I also tried to calibrate the display via Resolve and created a 3D LUT. I used only 175 patches and the results were almoust okey that time. It took less then 15 minutes.
Can this all be attributed to the ABL? What else go could make the results so bad?
Calibrite Display Pro HL on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.Black level seems high for a Pana plasma. Are you driving the TV with correct output levels (e.g. nVidia control panel)?
Can this all be attributed to the ABL?
ABL can have very detrimental effects. Are you using the smallest possible patch size + white level drift compensation? The Resolve result looks ok (apart from high black level), possibly due to reduced patch count which helps display stability.
Black level seems high for a Pana plasma. Are you driving the TV with correct output levels (e.g. nVidia control panel)?
I’m not really sure how to answer. The TV was connected to my Macbook Pro via HDMI-DVI cable, and I always leave Displaycal’s output levels setting to auto.
Are you using the smallest possible patch size + white level drift compensation
Yes to both, and I also used the black background. Colorimeter correction was set to plasma.
The TV was connected to my Macbook Pro via HDMI-DVI cable
If you view the attached picture, can you see all numbers from 1-8 on each side?
I always leave Displaycal’s output levels setting to auto.
Good.
Also, attach the profile you are currently using please (“create compressed archive…” in DisplayCAL).
If you view the attached picture, can you see all numbers from 1-8 on each side?
Not on my laptop, but I assume you mean do I see them on the TV? I don’t have access to it until tomorrow.
The profile is the from the last attempt.
The TV seems to crush the first 10 or so levels. You can probably fix this with the TV’s settings (brightness and/or gamma controls in OSD). Output levels from your mac to the TV seem to be ok.
You are limiting the peak luminance via the 1D LUT calibration. This is not recommended because it will reduce contrast. Do not set a white level target, you can still adjust the TV to a desired luminance during interactive adjustment.
The gamma was set to 2.2 but during the interactive adjustments I lowered brightness by couple of points to lower the the black point, do you think I should raise them back to the point they were after the factory reset?
Also, maybe off topic, but gamma menus had an s-curve option, something i havent seen in any other display, can you explain what it means?
I lowered brightness by couple of points to lower the the black point, do you think I should raise them back to the point they were after the factory reset?
Possibly, that may get rid of the black crush.
gamma menus had an s-curve option, something i havent seen in any other display, can you explain what it means?
An s-curve alters contrast. while it may look pleasing, it is not recommended for calibration.
Okey, so the todays results weren’t much better. The contrast and black level were even worse. For the third attempt I switched from the dvi input to hdmi, if that makes any difference. I also switched the tone curve from 2.2 to as measured to speed up the process.
After I gave up and normalised the TV’s image back to the starting point, I remembered to check the black and white picture on the TV and it look pretty good. Black 1 was barely visible as was the white one.
Your black level has risen from 0.07 to 0.2 cd/m2 (problems with rising black levels are actually documented for the 2009/2010 Panasonic models like this one, e.g. see https://www.avforums.com/review/panasonic-pf20-th-42pf20-review.128). Near-black tone response is much better (no more crush), but overall the fit to the measurement data is poor, meaning the monitor is not stable. It may be the best you can do, given that it is already about 10 years old. I would recommend if you are using it mainly for Resolve to create a LUT profile instead of matrix though.
