AVS Forum banner

421 - 440 of 3759 Posts

·
Registered
Joined
·
588 Posts
Last night I did my first run for my 65C8 in SDR mode. Scans for HDR mode will follow. I discovered some important things during my first run I would like to share. I set up OLED Light via DDC and aimed ~100 nits and after the Grayscale AutoCAL of 42p I ended up with ~85 nits so the peak light output drop is heavy so keep that in mind and set the OLED light before AutoCal at ~115-120 nits if you aim 100 nits. I don't recommend to change the OLED light after the Grayscale run like mentioned in the workflow. I discovered also some high peaks like other users especially at the 90 ish IRE.

Also I have some questions:

- Do I need to do a factory Reset or is it enough to reset DDC for the pictures models?
- Before hitting AutoCal for 3D LUT what do I have to choose for SMTP 16-255 Or 16-235? My TPG is sending RGB Full mode

I used to sensitive meter settings which ended up in a long calibration run but on the other hand I got pretty stable and reliable results especially in the low end. But I'm going to tweak the meter settings next time to improve the duration time without a big loss in accuracy.

I did 42p Grayscale and 101p lightning LUT and I used SMTP 16-255 and I don't know exactly of this was correct after all the Rec.709 coverage was to high with 105% so pleaseet me know if have to change that value.

I used 18% Window and 0% Pattern Inseration every 15s for 5s.

















 

·
Registered
Joined
·
4,901 Posts
Last night I did my first run for my 65C8 in SDR mode. Scans for HDR mode will follow. I discovered some important things during my first run I would like to share. I set up OLED Light via DDC and aimed ~100 nits and after the Grayscale AutoCAL of 42p I ended up with ~85 nits so the peak light output drop is heavy so keep that in mind and set the OLED light before AutoCal at ~115-120 nits if you aim 100 nits. I don't recommend to change the OLED light after the Grayscale run like mentioned in the workflow. I discovered also some high peaks like other users especially at the 90 ish IRE.
I'm curious as to what your actual readings were at the 100% and above points. In particular, the difference between actual Y vs target Y. In my experience, they usually aren't very large at all. As you move up the gray scale from 0 any differences between actual and target luminance have an increasing effect on the gamma plot. However, you do have some considerable spikes, so those differences could be higher than normal.
 

·
Registered
Joined
·
4,291 Posts
I'm curious as to what your actual readings were at the 100% and above points. In particular, the difference between actual Y vs target Y. In my experience, they usually aren't very large at all. As you move up the gray scale from 0 any differences between actual and target luminance have an increasing effect on the gamma plot. However, you do have some considerable spikes, so those differences could be higher than normal.


The log/log gamma chart amplifies small luminance differences as you reach 90-100%. This is just a side effect of the math itself that powers the chart. A .2 nit Difference from target Luminance versus measured, can make the graph swing wildly.

Gamma number is not a good accuracy metric to use at the top end.
 

·
Registered
Joined
·
6,587 Posts
Discussion Starter #424
@BlackJoker:Your results look typical and similar to mine.
@tyler: Thanks for the explanation on the gamma graph. I wonder if calman can be modified in a future release to "fix" the representation of the high end gamma since it makes the user think there is something seriously wrong. When i went in with the DDC controls and made a 1 click adjustment, i could see the results make a large swing as you are indicating.
 
  • Like
Reactions: chunon

·
Registered
Joined
·
588 Posts
Also I have some questions:

- Do I need to do a factory Reset or is it enough to reset DDC for the pictures models?
- Before hitting AutoCal for 3D LUT what do I have to choose for SMTP 16-255 Or 16-235? My TPG is sending RGB Full mode

Can someone tell me something about my questions?
 

·
Registered
Joined
·
972 Posts
DDC reset should be ok

Choose 16-235

If your tpg sends full rgb then Expand video to pc levels should be selected on calman souce
 
  • Like
Reactions: BlackJoker

·
Registered
Joined
·
98 Posts
I apologize for the lack of documentation. We are currently working on it and working on tutorial videos for the LG integration with CalMAN. We tried to be as explicit as possible in the LG workflows themselves, however we understand everyone doesn't have the same calibration equipment, which can cause the configuration needed to be different.

Most of the issues seem to be ABSL related. in SDR the pattern insertion is used to defeat the ABSL algorithm, in HDR it needs to keep the panel cool, and defeat the ABSL algorithm. It seems like the recommend configuration isn't working for everyone, so that will need to be revised, but for the time being, the easiest thing for people to eliminate that as the cause of their issues, would be to temporarily disable ABSL in the service menu during calibration (something that most pro calibrators have been doing for years on LG OLEDs). Then to re-enable it after calibration.

I will try to replicate your exact setup - VF HDMI + HDFURY Integral (I only have an Integral and Vertex) + i1 Display Pro - and see if I can figure out what is going wrong.

Tyler
News from @HDfury over here, he claims the AVI frame provided by calman to enable HDR10/BT2020 might simply be wrong. He provided another IF which puts my TV in the right mode, and I can see the pattern but the colors are wrong due to RGB output of my PatternGen which does not match the IF. Can you confirm we must use RGB output for HDR10 or does YCbCr work?.
 

·
Registered
Joined
·
588 Posts
There must be another problem with your gear and settings. I'm using a RPI as TPG and the HDFury linker for HDR10 calibration and BT2020 colorspace like mentioned in the spectracal PDF without any issues.
 

·
Registered
Joined
·
98 Posts
There must be another problem with your gear and settings. I'm using a RPI as TPG and the HDFury linker for HDR10 calibration and BT2020 colorspace like mentioned in the spectracal PDF without any issues.
Could you go a bit more into detail about your PI-PatternGen? Is it something you wrote/configured by yourself or is there an ˋofficialˋ project where one can download? I have a PI here in my flat.
 

·
Registered
Joined
·
98 Posts

·
Registered
Joined
·
4,901 Posts
News from @HDfury over here, he claims the AVI frame provided by calman to enable HDR10/BT2020 might simply be wrong. He provided another IF which puts my TV in the right mode, and I can see the pattern but the colors are wrong due to RGB output of my PatternGen which does not match the IF. Can you confirm we must use RGB output for HDR10 or does YCbCr work?.
Currently you can't use YCbCr for HDR if you are using a pattern generator unless it's an Accupel 6000. It is the only pattern generator right now that can do the conversion math needed to provide the YCbCr used in HDR (it's not the same as for SDR). Also after some research, I found that the Custom Info Frame you had in the Integral GUI pictured in your initial posts was for DV, not HDR10.
 

·
Registered
Joined
·
98 Posts
Currently you can't use YCbCr for HDR if you are using a pattern generator unless it's an Accupel 6000. It is the only pattern generator right now that can do the conversion math needed to provide the YCbCr used in HDR (it's not the same as for SDR). Also after some research, I found that the Custom Info Frame you had in the Integral GUI pictured in your initial posts was for DV, not HDR10.
Thanks. The IF shown in the first posts was a relict from my DV attempts. I definitly use the ones Calman provides for HDR10 and they donˋt work.

 

·
Registered
Joined
·
588 Posts
Thanks. The IF shown in the first posts was a relict from my DV attempts. I definitly use the ones Calman provides for HDR10 and they donˋt work.





Same here I have exactly the same issue.
 

·
Registered
Joined
·
4,901 Posts
Thanks. The IF shown in the first posts was a relict from my DV attempts. I definitly use the ones Calman provides for HDR10 and they donˋt work.

That CalMAN AVI IF (00:E8:64:5D:00) works fine on my Samsung KS8000. I'm looking at the color bar pattern from my DVDO AVLab TPG running through my Integral, both controlled by CalMAN, as we speak. No color issues, and the display is in HDR mode. That IF string should work for all HDR displays, unless LG has borked something and created a need for an LG-specific IF string.
 

·
Registered
Joined
·
588 Posts
The string worked perfectly on my LG 2016 model but now I only get a black screen
 

·
Registered
Joined
·
588 Posts

·
Registered
Joined
·
98 Posts
So, several questions arise:

- Could someone else with a 2018 LG model and a HDFury Integral test the situation?
- Could someone with a 2017 LG model and a HDFury Integral test the situation?
- Could someone with a dedicated HDR10 Pattern Generator (i.e. Videoforge Pro or Murideo) and a C8 (or 2018 model) test the situation? Additionally, if he is successful, could he extraxt the IF his PatternGen is sending?
- What about the announcment from Tyler, that it should work? Has anybody at Spectracal confirmed this? For me, i based by decision to purchase the Fury as well as the VideoForge HDMI on this announcement.


For SDR any CalMAN controllable pattern generator that supports RGB triplets will work.

For HDR10/HLG any CalMAN controllable HDR10 compatible pattern source will work (including HDfury setups to inject metadata)

For Dolby Vision you can use the VideoForge Pro, Murideo six-G, Or CalMAN’s pop out pattern window with a PC HDMI output (w/ certain GPUs) + HDFury to inject the Dolby Vision VSIF.


Even if you don’t want to use our autoCAL algorithms for grayscale, you can still use the DDC controls to tweak the grayscale manually yourself.

For SDR and HDR10/HLG The color gamut is controlled with 3D look up tables.

Dolby Vision still uses the custom DV config file to tell the DV engine what the actual performance of the panel is, then their color volume mapping is adjusted to that specific panel’s peak brightness/native primaries.

We will be posting tutorial videos to our YouTube page very soon.

Tyler
 

·
Registered
Joined
·
4,901 Posts
When I did my test, I used the HDR Analysis workflow, because there is no Autocal for HDR on the KS series. Would one of you be willing to switch to that workflow and try it again? The Source settings should be independent of the workflows, but you never know...
 
421 - 440 of 3759 Posts
Top