AVS Forum banner

1921 - 1940 of 1990 Posts

·
Registered
Joined
·
11,956 Posts
I assume calibrating at 100nits and then increasing OLED light after the fact is not a correct thing to do?
I have a ISF day cal done with Oled light at 65 no issues that I can see . I know that doesn’t help you at the moment but it is possible to get good results at that luminance level.


77CX 55E8 YTTV AppleTv 4K Sony X800M2 PS4 Denon X3700H
 
  • Like
Reactions: Abizzlenyc

·
Registered
Joined
·
193 Posts
I have a ISF day cal done with Oled light at 65 no issues that I can see . I know that doesn’t help you at the moment but it is possible to get good results at that luminance level.


77CX 55E8 YTTV AppleTv 4K Sony X800M2 PS4 Denon X3700H
Thanks, did you also address this type of panel stability issue somehow or did you not encounter anything like that?
 

·
Registered
Joined
·
11,956 Posts
Thanks, did you also address this type of panel stability issue somehow or did you not encounter anything like that?
I just used the Calman defaults for pattern insertion .


77CX 55E8 YTTV AppleTv 4K Sony X800M2 PS4 Denon X3700H
 
  • Like
Reactions: Abizzlenyc

·
Registered
Joined
·
106 Posts
I have a ISF day cal done with Oled light at 65 no issues that I can see . I know that doesn’t help you at the moment but it is possible to get good results at that luminance level.
How many nits do you end up with OLED light at 65?

I found that for targeting ~200nits after calibration (for daytime viewing), I need to set OLED light to at least 80. In my case, the panel drifts more during calibration and hence has more measured errors. The suggestions from @jrref should help with this.
 
  • Like
Reactions: jrref and Andy112

·
Registered
Joined
·
11,956 Posts
How many nits do you end up with OLED light at 65?

I found that for targeting ~200nits after calibration (for daytime viewing), I need to set OLED light to at least 80. In my case, the panel drifts more during calibration and hence has more measured errors. The suggestions from @jrref should help with this.
Not sure I would have to look at my calibration . It’s definitely over 180 though if I recall correctly .


77CX 55E8 YTTV AppleTv 4K Sony X800M2 PS4 Denon X3700H
 

·
Registered
Joined
·
193 Posts
@jrref and @chunon

All good advice. I did fiddle with my service menu white point before I did my stuff. May have improved the pre-cal. Anyway, think these results are pretty decent considering my nits.

There was no doubt a difference when I switched the pattern frequency. Was good advice for sure.

Edit: FYI I did do lightning lut at first but it didn't look great. So I hit back, enable calib again and redid it with fixed grid. I'm not sure of the mechanics of doing something like that but it seems to have took.

I actually think the pre-cal showed here did not match what I originally measured. Might be because of what I did going backwards.

3114997
 

·
Registered
Joined
·
63 Posts
Edit: FYI I did do lightning lut at first but it didn't look great. So I hit back, enable calib again and redid it with fixed grid. I'm not sure of the mechanics of doing something like that but it seems to have took.
Eeeeeeek! This is giving me PTSD...
Oh CalMAN :)
 

·
Registered
Joined
·
106 Posts
It is known that Lightning or Fixed Grid are the only two methods in Calman so far that produce acceptable results for 3D LUTs. At least for LG OLEDs, the issues with other methods have been discussed at length. Even with these two methods, users have reported artifacts when inspecting calibration results visually, despite the post-cal numbers being good (although one could argue that using just a handful colors for verification isn‘t enough anyway...)

There‘s hope that the upcoming beta release of Calman Home will have fixes for at least some of these long standing issues.
 
  • Like
Reactions: jrref

·
Registered
Joined
·
11,956 Posts
@jrref and @chunon

All good advice. I did fiddle with my service menu white point before I did my stuff. May have improved the pre-cal. Anyway, think these results are pretty decent considering my nits.

There was no doubt a difference when I switched the pattern frequency. Was good advice for sure.

Edit: FYI I did do lightning lut at first but it didn't look great. So I hit back, enable calib again and redid it with fixed grid. I'm not sure of the mechanics of doing something like that but it seems to have took.

I actually think the pre-cal showed here did not match what I originally measured. Might be because of what I did going backwards.

View attachment 3114997
Looks good . Could I ask what you changed the pattern insertion to ? May redo my day mode and bump it up some .


77CX 55E8 YTTV AppleTv 4K Sony X800M2 PS4 Denon X3700H
 

·
Registered
Joined
·
193 Posts
Looks good . Could I ask what you changed the pattern insertion to ? May redo my day mode and bump it up some .


77CX 55E8 YTTV AppleTv 4K Sony X800M2 PS4 Denon X3700H
Course. Changed it from 45 to 15. I went with jrrefs initial suggestion of 15/5/10 for insertion frequency, duration and level. I'm not sure what level does because other than the timing, I didn't notice anything different. But the default for level was 25 I think?
Even with these two methods, users have reported artifacts when inspecting calibration results visually,
I really shouldn't ask... But can you define artifact in this context?
 

·
Registered
Joined
·
106 Posts
I really shouldn't ask... But can you define artifact in this context?
For example: banding and posterization. Visually inspecting with test patterns after calibration reveals these for some users. Also, if you look at the 3D LUTs generated by Calman with tools that can visualize them, the sources of some of these issues become apparent. You can find references to these in the various calibration threads.
 

·
Registered
Joined
·
11,956 Posts
Course. Changed it from 45 to 15. I went with jrrefs initial suggestion of 15/5/10 for insertion frequency, duration and level. I'm not sure what level does because other than the timing, I didn't notice anything different. But the default for level was 25 I think?

I really shouldn't ask... But can you define artifact in this context?
Thanks


77CX 55E8 YTTV AppleTv 4K Sony X800M2 PS4 Denon X3700H
 

·
Registered
Joined
·
7,836 Posts
Course. Changed it from 45 to 15. I went with jrrefs initial suggestion of 15/5/10 for insertion frequency, duration and level. I'm not sure what level does because other than the timing, I didn't notice anything different. But the default for level was 25 I think?

I really shouldn't ask... But can you define artifact in this context?
Right, shortened the interval and lowered the level. The level becomes an issue when calibrating at high nits. The default is 45/5/25
 

·
Registered
Joined
·
12 Posts
It is known that Lightning or Fixed Grid are the only two methods in Calman so far that produce acceptable results for 3D LUTs. At least for LG OLEDs, the issues with other methods have been discussed at length. Even with these two methods, users have reported artifacts when inspecting calibration results visually, despite the post-cal numbers being good (although one could argue that using just a handful colors for verification isn‘t enough anyway...)

There‘s hope that the upcoming beta release of Calman Home will have fixes for at least some of these long standing issues.
Is this applicable only for SDR calibration I assume? Calman recommend Matrix LUT as the only option for HDR calibrations on their guide
 

·
Registered
Joined
·
106 Posts
Calman recommend Matrix LUT as the only option for HDR calibrations on their guide
Yes, you can only use Matrix LUT for HDR calibration in the latest Calman version (which I think is R2).
 

·
Registered
Joined
·
12 Posts
I'm having trouble calibrating the Dolby Vision mode - can someone help? edit: i might have answered my own question...

I'm using an LG CX OLED and the built-in pattern generator. I've put the screen in to DV mod using the Calman DV mp4 video and confirmed it's in that mode. Followed the steps in their guide and this is the result from the greyscale autocal step:

3115826

All seems to look good so far.

I completed the next step to create the DV config file and then uploaded it - or as far as i can tell it was uploaded. you select it from where it's saved but you never really seem to get a confirmation it's been completed.

when you get in to the end of the calibration you're asked to change the metadata for the DV source to "absolute" before completing the validation. but because i'm using the built in pattern generator there doesn't seem to be any option for this?

3115828


doing the post cal capture though produces poor results:

3115829


what am i doing wrong? is this something to do with the metadata needing to be "absolute"? why is it that in the greyscale autocal step the curves on the right are different to the target curves in the post cal capture?

is this because you can't use the internal pattern generator to validate DV as it says here? although that is for 2019 LG TV's but there's no such message for the 2020 models here
 

·
Registered
Joined
·
106 Posts
is this because you can't use the internal pattern generator to validate DV as it says here? although that is for 2019 LG TV's but there's no such message for the 2020 models here
Exactly, you cannot use the internal pattern generator for DV validation because of where the iTPG is in the video pipeline.
 
1921 - 1940 of 1990 Posts
Top