2018 LG OLED Calibration and User Settings (No price talk) - Page 116 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 2475Likes
Reply
 
Thread Tools
post #3451 of 3665 Old 09-29-2019, 04:33 AM
Member
 
Join Date: Mar 2019
Posts: 67
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 48 Post(s)
Liked: 25
Quote:
Originally Posted by ConnecTEDDD View Post
Yes, it can improve the picture quality in SDR also, just in SDR the alignment of calibration controls with displayed patterns is much better.

The less processing normal UI will apply, the better the picture (not in dE, but visually).

So I could just calibrate the "warm" preset in SM for SDR (and "medium" for HDR/DV) and leave out 2p white balance in TV menu?
oledmod is offline  
Sponsored Links
Advertisement
 
post #3452 of 3665 Old 09-29-2019, 04:33 AM
AVS Forum Special Member
 
mrtickleuk's Avatar
 
Join Date: May 2016
Location: Birmingham, UK - you know, the original one!
Posts: 7,491
Mentioned: 184 Post(s)
Tagged: 0 Thread(s)
Quoted: 5190 Post(s)
Liked: 7508
Quote:
Originally Posted by ConnecTEDDD View Post
You can skip 1D LUT, then create a custom colorspace based to REC.709, enter the custom WP coordinates and save it.....then use that as target colorspace for your 3D LUT.

Later post your results, as we have to see 3D LUT volumetric calibration from someone with LG OLED a long time, upload your LG *.3dl file, and someone will import it to LightSpace to check how it looks.
Thanks, Tedd. In fact I have just finished one, so I will upload the file to this post.

I have switched over my SM WB settings, so "Cool" is now the custom white point setting, and the foundation for the SDR Cinema 3DLUT. I reset the 3DLUT in the Cinema mode first. Also SM "Warm" is back to the D65 white point, and the foundation for my other picture modes which use "Warm2". (I did not reset the 3DLUTs for the modes which will use this because I will continue with the factory 1DLUT/3DLUT in those modes and I just wanted to tweak "Warm2" to be more accurate for D65 for those modes.)

The only problems I had, which meant I had to re-do things, were to do with my "ending" positions. With this method of keeping the factory 1DLUT, as you explained we should use the Brightness/Contrast settings which are correct for the panel. For my panel, this is B=52 C=83.
When I did the 3DLUT with B=50 C=85 Gamma=2.2, I had terrible results with post-cal verification. It (the gamma curve) was clearly badly wrong.
When I did the 3DLUT with B=52 C=83 Gamma=2.4, I had very nice results with post-cal verification.
I tested this a few times to be sure.

If I understand correctly, my SDR Cinema mode is now using my 3DLUT (user menu shows "Colour Gamut: Wide" locked and greyed) running on top of the Service Menu WB "Cool" (when user menu is set to "Cool") and is correct when I set B=52 C=83 Gamma=2.4, because that is what I defined in the custom ColourSpace in Calman for the modified White Point. Also, the 3DLUT does some of the "lifting" to do the Greyscale and Gamma correction, in the 33pt greyscale in the 3DLUT. It looks good.

Anyway here's the file. NB I use dEITP in the charts. I did 8,000 pts and it took 7hrs 54mins.

Limitations I am aware of:
  • Capabilities of my C6HDR2000 meter near black.
  • Calman engines has problems with blue? Not sure of the details but I saw it meantioned
Attached Thumbnails
Click image for larger version

Name:	post.png
Views:	35
Size:	210.5 KB
ID:	2621218  
Attached Files
File Type: zip Cinema 8000pts 16-255.zip (165.8 KB, 9 views)

_______________
Denon AVR-X4200W, Arcam Alpha 8P; 5.1.4 setup: Mission 702e, M7C1i, 77DS, 731. Rel T5 Sub. Monitor Audio CT165 4 Tops | LG OLED55C8PLA TV | Samsung UBD-K8500 UHD Blu-Ray

HDMI 2.0 4K modes | Dolby & DTS core+outer audio tracks on (UHD) Blu-Rays | Hello to Jason Isaacs

Last edited by mrtickleuk; 09-29-2019 at 05:09 AM.
mrtickleuk is online now  
post #3453 of 3665 Old 09-29-2019, 04:43 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,773
Mentioned: 219 Post(s)
Tagged: 1 Thread(s)
Quoted: 3468 Post(s)
Liked: 4297
Quote:
Originally Posted by oledmod View Post
So I could just calibrate the "warm" preset in SM for SDR (and "medium" for HDR/DV) and leave out 2p white balance in TV menu?
Yes, but see what normal UI color temp will affect.

White Balance settings of 'Warm' in Service Menu will adjust the 'Warm 1, Warm 2 and Warm 3' preset of Normal TV Menu.

White Balance settings of 'Medium' in Service Menu will adjust the 'Medium and Warm 1' preset of Normal TV Menu.

White Balance settings of 'Cool' in Service Menu will adjust the 'Cool' preset of Normal TV Menu.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is online now  
Sponsored Links
Advertisement
 
post #3454 of 3665 Old 09-29-2019, 05:02 AM
AVS Forum Special Member
 
mrtickleuk's Avatar
 
Join Date: May 2016
Location: Birmingham, UK - you know, the original one!
Posts: 7,491
Mentioned: 184 Post(s)
Tagged: 0 Thread(s)
Quoted: 5190 Post(s)
Liked: 7508
Quote:
Originally Posted by ConnecTEDDD View Post
Yes, but see what normal UI color temp will affect.

White Balance settings of 'Warm' in Service Menu will adjust the 'Warm 1, Warm 2 and Warm 3' preset of Normal TV Menu.

White Balance settings of 'Medium' in Service Menu will adjust the 'Medium and Warm 1' preset of Normal TV Menu.

White Balance settings of 'Cool' in Service Menu will adjust the 'Cool' preset of Normal TV Menu.
Also, for the "dumb" modes which only have a slider from "Cool50 to Warm50", what you set for "Cool" and "Warm" affects these too. I've no idea if the "Medium" values are used in the middle part of the slider (and don't really care).
In my SM:
Cool=set for the OLED modified White Point (0.308, 0.313). Then I use "Cool" in my Picture Mode "SDR Cinema" which targetted that.
Medium=set for D65 WP, an older adjustment.
Warm=set for D65 WP, my latest adjustment. Then I use "Warm2" in all my Picture Modes which use LG's factory 1DLUT/3DLUTs.

The result is that the horrible modes which only have a slider from "Cool50 to Warm50", look slightly better because the "Cool" isn't terrible any more
bejoro likes this.

_______________
Denon AVR-X4200W, Arcam Alpha 8P; 5.1.4 setup: Mission 702e, M7C1i, 77DS, 731. Rel T5 Sub. Monitor Audio CT165 4 Tops | LG OLED55C8PLA TV | Samsung UBD-K8500 UHD Blu-Ray

HDMI 2.0 4K modes | Dolby & DTS core+outer audio tracks on (UHD) Blu-Rays | Hello to Jason Isaacs
mrtickleuk is online now  
post #3455 of 3665 Old 09-29-2019, 07:45 AM
Member
 
bejoro's Avatar
 
Join Date: Aug 2002
Posts: 186
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 70 Post(s)
Liked: 33
Quote:
Originally Posted by ConnecTEDDD View Post
Yes, it can improve the picture quality in SDR also, just in SDR the alignment of calibration controls with displayed patterns is much better.

The less processing normal UI will apply, the better the picture (not in dE, but visually).
In the service menu (SM) of my 65C8 for SDR and HDR only RGB high has adjustments implemented, max. value is 192 per color.
RGB low in SM shows always 64 per color. Is that the usual situation?

Should I ignore RGB low in SM?

Max. value of RGB high in SM is 192?

What is recommended and what has to be avoided in SM?

Warm regards,
bejoro
bejoro is offline  
post #3456 of 3665 Old 09-29-2019, 08:49 AM
AVS Forum Special Member
 
mrtickleuk's Avatar
 
Join Date: May 2016
Location: Birmingham, UK - you know, the original one!
Posts: 7,491
Mentioned: 184 Post(s)
Tagged: 0 Thread(s)
Quoted: 5190 Post(s)
Liked: 7508
Quote:
Originally Posted by bejoro View Post
In the service menu (SM) of my 65C8 for SDR and HDR only RGB high has adjustments implemented, max. value is 192 per color.
RGB low in SM shows always 64 per color. (1) Is that the usual situation?

(2) Should I ignore RGB low in SM?

(3) Max. value of RGB high in SM is 192?

What is recommended and what has to be avoided in SM?
(1) Yes, (2) yes, (3) yes, see also this post for the full set of advice.
bejoro likes this.

_______________
Denon AVR-X4200W, Arcam Alpha 8P; 5.1.4 setup: Mission 702e, M7C1i, 77DS, 731. Rel T5 Sub. Monitor Audio CT165 4 Tops | LG OLED55C8PLA TV | Samsung UBD-K8500 UHD Blu-Ray

HDMI 2.0 4K modes | Dolby & DTS core+outer audio tracks on (UHD) Blu-Rays | Hello to Jason Isaacs

Last edited by mrtickleuk; 09-29-2019 at 08:52 AM.
mrtickleuk is online now  
post #3457 of 3665 Old 09-29-2019, 11:04 AM
Member
 
bejoro's Avatar
 
Join Date: Aug 2002
Posts: 186
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 70 Post(s)
Liked: 33
Quote:
Originally Posted by mrtickleuk View Post
(1) Yes, (2) yes, (3) yes, see also this post for the full set of advice.
Thank you very much.

So, for SDR I use RGB GAIN of SM for general Whitepoint calibration and 20p of the "normal" user interface (UI) for lower IREs and gamma correction, as ConnecTEDDD suggested before? What about 20p of the SM, any advantages?

For HDR I have achieved a very good result (DeltaE below 0.6 from. 0% - 70%) by adjusting RGB High/Low in the normal UI.
If I do not use RGB low the DeltaE will be very high below 40%.
Is there a way to achieve that by using the SM?

Why is it not recommended to use RGB CUT of the SM for SDR/HDR calibration?

Warm regards,
bejoro
bejoro is offline  
post #3458 of 3665 Old 09-29-2019, 11:10 AM
Member
 
Join Date: Mar 2019
Posts: 67
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 48 Post(s)
Liked: 25
Quote:
Originally Posted by mrtickleuk View Post
(1) Yes, (2) yes, (3) yes, see also this post for the full set of advice.

I could significantly improve white balance by also adjusting SM RGB cut settings (and not only gain). Why not make use of it?

Last edited by oledmod; 09-29-2019 at 12:53 PM.
oledmod is offline  
post #3459 of 3665 Old 09-29-2019, 02:20 PM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,773
Mentioned: 219 Post(s)
Tagged: 1 Thread(s)
Quoted: 3468 Post(s)
Liked: 4297
Quote:
Originally Posted by bejoro View Post
In the service menu (SM) of my 65C8 for SDR and HDR only RGB high has adjustments implemented, max. value is 192 per color.
RGB low in SM shows always 64 per color. Is that the usual situation?

Should I ignore RGB low in SM?

Max. value of RGB high in SM is 192?

What is recommended and what has to be avoided in SM?
The panel default values for high dynamic range are RGB-Cut: 64.64.64 and RGB-Gain: 192.192.192.

The recommendation about to calibrate only RGB-Gain came initially as the best possible solution for 3D LUT profiling, instead of using internal 1D LUT or normal RGB-High controls to pre-cal 100% White, that method is the best for the best end result (in dE and visually also when you will evaluate with color reproduction patterns, ramps etc.)

Its been tested for 3D LUT pre-cal to be used the RGB-Cuts also, but there no any real advantage.

RGB-Cut adjustment can affect near black performance, so when you using them, verify with measurement and using patterns with near black to see that all are fine.

Just in factory they don't spend such time to calibrated RGB-Cuts also, and to all sets, only the RGB-gain values are adjusted from the factory typical calibration, more details are there: https://www.avsforum.com/forum/139-d...l#post58520702
bejoro, chros73 and mrtickleuk like this.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is online now  
post #3460 of 3665 Old 09-29-2019, 02:26 PM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,773
Mentioned: 219 Post(s)
Tagged: 1 Thread(s)
Quoted: 3468 Post(s)
Liked: 4297
Quote:
Originally Posted by bejoro View Post
Why is it not recommended to use RGB CUT of the SM for SDR/HDR calibration?
Quote:
Originally Posted by oledmod View Post
I could significantly improve white balance by also adjusting SM RGB cut settings (and not only gain). Why not make use of it?
You can use them if its helping with your manual calibrations, just for 3D LUT profiling (as instructions posted for 3D LUT''s) is not worth to spend such time to adjust RGB-Cuts, as 3D LUT will include many grayscale point calibration, for only adjusting near black with Brightness control and pre-cal 100% White is required, to get later results like these.
bejoro and oledmod like this.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is online now  
post #3461 of 3665 Old 09-30-2019, 12:45 AM
Member
 
bejoro's Avatar
 
Join Date: Aug 2002
Posts: 186
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 70 Post(s)
Liked: 33
ConnecTEDDD, thank you very much. Great clarification (as always) of SM for manual calibration.
I will use SM RGB GAIN/CUT in my next calibration session.

What about 20P in SM? Would there be the same advantages over the 20P of normal UI?

Are the SM RGB GAIN/CUT values stored in different slots for SDR, HDR and DV?
So different values for each mode?

Warm regards,
bejoro
bejoro is offline  
post #3462 of 3665 Old 09-30-2019, 12:50 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,773
Mentioned: 219 Post(s)
Tagged: 1 Thread(s)
Quoted: 3468 Post(s)
Liked: 4297
Quote:
Originally Posted by bejoro View Post
What about 20P in SM? Would there be the same advantages over the 20P of normal UI?
I haven't checked.

Are the SM RGB GAIN/CUT values stored in different slots for SDR, HDR and DV?
So different values for each mode?[/QUOTE]

It will affect the color temp selection of all modes (SDR/DV/HDR10/HLG).
bejoro likes this.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is online now  
post #3463 of 3665 Old 09-30-2019, 01:05 AM
Member
 
bejoro's Avatar
 
Join Date: Aug 2002
Posts: 186
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 70 Post(s)
Liked: 33
Quote:
Originally Posted by ConnecTEDDD View Post
I haven't checked.

Are the SM RGB GAIN/CUT values stored in different slots for SDR, HDR and DV?
So different values for each mode?
It will affect the color temp selection of all modes (SDR/DV/HDR10/HLG).[/QUOTE]

Faboulos results of Anger.miki's 3DLUT. Unfortunately I can not afford an expensive colorimeter like a basICColor Discus, so 3DLUT is out of the question for me.

Because the values of the SM affect all modes, the SM-20P makes no sense because it is only needed for SDR for correction of gamma tracking and these values would destroy a correct EOTF.

Warm regards,
bejoro
bejoro is offline  
post #3464 of 3665 Old 09-30-2019, 01:16 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,773
Mentioned: 219 Post(s)
Tagged: 1 Thread(s)
Quoted: 3468 Post(s)
Liked: 4297
Quote:
Originally Posted by bejoro View Post
Faboulos results of Anger.miki's 3DLUT. Unfortunately I can not afford an expensive colorimeter like a basICColor Discus, so 3DLUT is out of the question for me.
For these posted results, its been used the i1Display PRO for the measurements: https://www.avsforum.com/forum/139-d...l#post58522806

https://www.avsforum.com/forum/139-d...l#post58523134

...and the dE average from 1000 color points measured of verification (0.25dE2000) was lower from Anger.Miki's, which has 0.26dE2000.

The meter is not a limitation, the posted results came from users with i1Display PRO/Discus and Klein K-10A also.

Another example with 1000p verification with 0.3dE2000 average with i1Display PRO: https://www.avsforum.com/forum/139-d...l#post58609714

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is online now  
post #3465 of 3665 Old 09-30-2019, 02:52 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,773
Mentioned: 219 Post(s)
Tagged: 1 Thread(s)
Quoted: 3468 Post(s)
Liked: 4297
Quote:
Originally Posted by mrtickleuk View Post
Thanks, Tedd. In fact I have just finished one, so I will upload the file to this post.

I have switched over my SM WB settings, so "Cool" is now the custom white point setting, and the foundation for the SDR Cinema 3DLUT. I reset the 3DLUT in the Cinema mode first. Also SM "Warm" is back to the D65 white point, and the foundation for my other picture modes which use "Warm2". (I did not reset the 3DLUTs for the modes which will use this because I will continue with the factory 1DLUT/3DLUT in those modes and I just wanted to tweak "Warm2" to be more accurate for D65 for those modes.)

The only problems I had, which meant I had to re-do things, were to do with my "ending" positions. With this method of keeping the factory 1DLUT, as you explained we should use the Brightness/Contrast settings which are correct for the panel. For my panel, this is B=52 C=83.
When I did the 3DLUT with B=50 C=85 Gamma=2.2, I had terrible results with post-cal verification. It (the gamma curve) was clearly badly wrong.
When I did the 3DLUT with B=52 C=83 Gamma=2.4, I had very nice results with post-cal verification.
I tested this a few times to be sure.

If I understand correctly, my SDR Cinema mode is now using my 3DLUT (user menu shows "Colour Gamut: Wide" locked and greyed) running on top of the Service Menu WB "Cool" (when user menu is set to "Cool") and is correct when I set B=52 C=83 Gamma=2.4, because that is what I defined in the custom ColourSpace in Calman for the modified White Point. Also, the 3DLUT does some of the "lifting" to do the Greyscale and Gamma correction, in the 33pt greyscale in the 3DLUT. It looks good.

Anyway here's the file. NB I use dEITP in the charts. I did 8,000 pts and it took 7hrs 54mins.

Limitations I am aware of:
  • Capabilities of my C6HDR2000 meter near black.
  • Calman engines has problems with blue? Not sure of the details but I saw it meantioned
Have you used the latest CalMAN 2019 R2 release correct?

Your results confirm that skipping 1D LUT provides better results, as we have found with LightSpace also.

BTW why you have 10 dEIPT to your black?

What is your black level measurement in nits (pre-post)?

Use a workflow where we can see your gamma tracking and the numbers of black/white (pre-post).

It will be better idea to use dE2000, to be able to compare with results of other people who used dE2000 for verification also.

Your meter is fine, as Tyler has posted, the whole LG testing based to C6 measurements, so i1Display PRO/C6 is performing good, nothing to worry.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is online now  
post #3466 of 3665 Old 09-30-2019, 07:35 AM
Member
 
bejoro's Avatar
 
Join Date: Aug 2002
Posts: 186
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 70 Post(s)
Liked: 33
Quote:
Originally Posted by ConnecTEDDD View Post
The panel default values for high dynamic range are RGB-Cut: 64.64.64 and RGB-Gain: 192.192.192.

The recommendation about to calibrate only RGB-Gain came initially as the best possible solution for 3D LUT profiling, instead of using internal 1D LUT or normal RGB-High controls to pre-cal 100% White, that method is the best for the best end result (in dE and visually also when you will evaluate with color reproduction patterns, ramps etc.)

Its been tested for 3D LUT pre-cal to be used the RGB-Cuts also, but there no any real advantage.

RGB-Cut adjustment can affect near black performance, so when you using them, verify with measurement and using patterns with near black to see that all are fine.

Just in factory they don't spend such time to calibrated RGB-Cuts also, and to all sets, only the RGB-gain values are adjusted from the factory typical calibration, more details are there: https://www.avsforum.com/forum/139-d...l#post58520702
I performed a new calibration and used the service menu RGB GAIN/CUT settings and achvieved very good results (deltaE below 0.7, mostly below 0.4) for SDR and HDR (except EOTF 70% and above where I got a deltaE of below 1.1, which is still very good).

Maybe an interesteing observation (or not):

The delta values for RGB GAIN/CUT of the service menu are identical to the 2P high/low settings of the normal UI.
In the normal user interface I had RGB high of 0/-4/-5 and RGB low of +1/0/+2.

If I add these values to RGB GAIN/CUT I get the calibrated settings as I have it now.

All modes for SDR and HDR show now almost identical results, very nice.

I have still a slight falloff in gamma tracking with SDR at IRE10 and below with gamma 2.4 because I did not use 20P yet (2.45 at IRE10 and 2.55 at IRE5).

With gamma 2.2 the gamma tracking is very good (2.28 at IRE5).

I will watch some content to see if I really need a 20P correction for SDR for the very low IREs.

Thank you for encouraging me to use the service menu.
mrtickleuk likes this.

Warm regards,
bejoro
bejoro is offline  
post #3467 of 3665 Old 09-30-2019, 07:41 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,773
Mentioned: 219 Post(s)
Tagged: 1 Thread(s)
Quoted: 3468 Post(s)
Liked: 4297
Quote:
Originally Posted by bejoro View Post
I performed a new calibration and used the service menu RGB GAIN/CUT settings and achvieved very good results (deltaE below 0.7, mostly below 0.4) for SDR and HDR (except EOTF 70% and above where I got a deltaE of below 1.1, which is still very good).

The delta values for RGB GAIN/CUT of the service menu are identical to the 2P high/low settings of the normal UI.
In the normal user interface I had RGB high of 0/-4/-5 and RGB low of +1/0/+2.

If I add these values to RGB GAIN/CUT I get the calibrated settings as I have it now.

Thank you for encouraging me to use the service menu.
When you will use SM WB 2-Point, use normal UI 2-Point RGB to default (untouched) positions.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is online now  
post #3468 of 3665 Old 09-30-2019, 07:58 AM
Member
 
bejoro's Avatar
 
Join Date: Aug 2002
Posts: 186
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 70 Post(s)
Liked: 33
Quote:
Originally Posted by ConnecTEDDD View Post
When you will use SM WB 2-Point, use normal UI 2-Point RGB to default (untouched) positions.
Yes, of course, thank you. I performed a reset of the white balance settings in the normal UI before changing anything in the SM.

Maybe I explained it misleadingly. I could use the identical 2P values of the normal UI that I had before to adjust the SM settings RGB GAIN/CUT and achieved identical results.

I checked the slight falloff of the gamma tracking with your Calibration Disk (Black Level, Black Tunnel etc.) and I can distinguish 17 and 18, easily with gamma 2.2. It takes a bit longer for the eyes to adapt with gamma 2.4, 17 is very dark at 100 nits white point and gamma 2.4. So I think I can live without a 20P (normal UI) adjustment for SDR. I prefer to use as less adjustments as possible.

Warm regards,
bejoro
bejoro is offline  
post #3469 of 3665 Old 09-30-2019, 02:06 PM
AVS Forum Special Member
 
mrtickleuk's Avatar
 
Join Date: May 2016
Location: Birmingham, UK - you know, the original one!
Posts: 7,491
Mentioned: 184 Post(s)
Tagged: 0 Thread(s)
Quoted: 5190 Post(s)
Liked: 7508
Quote:
Originally Posted by ConnecTEDDD View Post
Have you used the latest CalMAN 2019 R2 release correct?

Your results confirm that skipping 1D LUT provides better results, as we have found with LightSpace also.

BTW why you have 10 dEIPT to your black?

What is your black level measurement in nits (pre-post)?
Correct, yes this was with the new version from last week. Ah, I think I set the option off from "use measured black level" and onto "use X for black" and entered 0. This is probably why it's giving large errors. When I look at old calibrations I did months ago, I normally measured 0.0019 for black.

Quote:
Use a workflow where we can see your gamma tracking and the numbers of black/white (pre-post).
ok sir . But what do you want to see for "pre"? If I (for example) re-did the mode I have done here, "pre" could either be the state it is now, or the state after resetting the 3DLUT, or the state after using the SM to get the Gain value as good as I can?

Quote:
It will be better idea to use dE2000, to be able to compare with results of other people who used dE2000 for verification also.

Your meter is fine, as Tyler has posted, the whole LG testing based to C6 measurements, so i1Display PRO/C6 is performing good, nothing to worry.
Thanks good to hear about the meter. I'm happy with it, I just wanted to list is as a possible "weakness" but I'm only a hobbyist and I can't imagine buying any other meter for many years.

Does anyone have an opinion on the problem of what settings to use during the 3DLUT profiling? I outlined the problems I had above. I was very confused by this issue (because the method you guys used for LS was to do the profile with the panel set to Gamma2.2)?
  • When I did the 3DLUT with B=50 C=85 Gamma=2.2, I had terrible results with post-cal verification. It (the gamma curve) was clearly badly wrong.
  • When I did the 3DLUT with B=52 C=83 Gamma=2.4, I had very nice results with post-cal verification.

Mine was set to Gamma2.4 (in the menus) with "Enable calibration" ticked in Calman and the target Gamma set to 2.4 in Calman.

The sequence I had been following, was:
  1. Using the TV's menus, set Gamma to 2.4, Brightness 52, Contrast 83 [the "destination" settings for this calibration]
  2. Take "pre" measurements in Calman
  3. Connect to the TV in Calman, tick "Enable Calibration"
  4. Using the TV's menus, set Gamma to 2.2 ["bypass" setting]
  5. Using Calman's DDC, Set OLED Light to Target / Brightness 50 / Contrast 85 [the "neutral native" settings]
  6. Do not touch 1DLUT. Do not "full DDC reset". Do not press Autocal button on Greyscale page. Do not pass go. Do not collect £200.
  7. Reset 3DLUT only (DDC window: 3DLUT controls: "Reset 3D Lut" button)
  8. Using the TV's menus, set Gamma to 2.4
  9. SM White Balance gain (only) adjust, to fix RGB Balance [displaying 100% grey 10% window +OLED light at target]
  10. Using the TV's menus, set Gamma to 2.2 ["required for 3DLUT"]
  11. Using Calman's DDC, set Brightness 52, Contrast 83 [the "destination" settings for this calibration]
  12. Create 3DLUT and upload in Calman, untick "Enable Calibration", disconnect from TV in Calman
  13. Using the TV's menus, set Gamma to 2.4, Brightness 52, Contrast 83 [the "destination" settings for this calibration]
  14. Take "post" measurements in Calman

(Step 11 might have been B=52, C=83, actually I can't remember). But it was definitely Gamma2.2

This was the result afterwards, clearly wrong, when I changed the TV gamma back to 2.4 which is what I was targeting:


Then, when I changed the TV gamma to 2.2 (used during the 3DLUT creation), I got this which looked right but is not what I wanted!


So, I changed to this procedure:
  1. Using the TV's menus, set Gamma to 2.4, Brightness 52, Contrast 83 [the "destination" settings for this calibration]
  2. Take "pre" measurements in Calman
  3. Connect to the TV in Calman, tick "Enable Calibration"
  4. Using the TV's menus, set Gamma to 2.2 ["bypass" setting]
  5. Using Calman's DDC, Set OLED Light to Target / Brightness 50 / Contrast 85 [the "neutral native" settings]
  6. Do not touch 1DLUT. Do not "full DDC reset". Do not press Autocal button on Greyscale page. Do not pass go. Do not collect £200.
  7. Reset 3DLUT only (DDC window: 3DLUT controls: "Reset 3D Lut" button)
  8. Using the TV's menus, set Gamma to 2.4
  9. SM White Balance gain (only) adjust, to fix RGB Balance [displaying 100% grey 10% window +OLED light at target]
  10. Using the TV's menus, make sure Gamma is still at 2.4
  11. Using Calman's DDC, set Brightness 52, Contrast 83 [the "destination" settings for this calibration]
  12. Create 3DLUT and upload in Calman, untick "Enable Calibration", disconnect from TV in Calman
  13. Using the TV's menus, set Gamma to 2.4, Brightness 52, Contrast 83 [the "destination" settings for this calibration]
  14. Take "post" measurements in Calman

This was the result afterwards, better (I didn't keep the greyscale for this one, but this shows the flat/correct gamma):


So it would really help me if someone else would verify that yes, you do have to set the TV Gamma to the destination gamma that you will want to use, before you start the 3DLUT points measurement. Ie if you want to end up with Gamma2.4, you must set Gamma2.4 not Gamma2.2 Thanks!
Attached Thumbnails
Click image for larger version

Name:	post 2nd (2kpt) grey (tv gamma 2.4).png
Views:	140
Size:	181.1 KB
ID:	2621840   Click image for larger version

Name:	post 2nd (2kpt) grey (tv gamma 2.2).png
Views:	141
Size:	170.3 KB
ID:	2621846   Click image for larger version

Name:	post 3rd (Lightning) (tv gamma 2.4)-MatchScale.png
Views:	139
Size:	217.9 KB
ID:	2621850  

_______________
Denon AVR-X4200W, Arcam Alpha 8P; 5.1.4 setup: Mission 702e, M7C1i, 77DS, 731. Rel T5 Sub. Monitor Audio CT165 4 Tops | LG OLED55C8PLA TV | Samsung UBD-K8500 UHD Blu-Ray

HDMI 2.0 4K modes | Dolby & DTS core+outer audio tracks on (UHD) Blu-Rays | Hello to Jason Isaacs

Last edited by mrtickleuk; 09-30-2019 at 02:26 PM.
mrtickleuk is online now  
post #3470 of 3665 Old 10-03-2019, 03:36 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,773
Mentioned: 219 Post(s)
Tagged: 1 Thread(s)
Quoted: 3468 Post(s)
Liked: 4297
Quote:
Originally Posted by mrtickleuk View Post
ok sir . But what do you want to see for "pre"? If I (for example) re-did the mode I have done here, "pre" could either be the state it is now, or the state after resetting the 3DLUT, or the state after using the SM to get the Gain value as good as I can?
Gamma tracking, black level and peak output, to see how was your black/white level before and after the profiling.

Quote:
Originally Posted by mrtickleuk View Post
Thanks good to hear about the meter. I'm happy with it, I just wanted to list is as a possible "weakness" but I'm only a hobbyist and I can't imagine buying any other meter for many years.
Most people which are using 3D LUT with LG's are using i1DisplayPRO's (branded or OEM, retail).

Quote:
Originally Posted by mrtickleuk View Post

Does anyone have an opinion on the problem of what settings to use during the 3DLUT profiling? I outlined the problems I had above. I was very confused by this issue (because the method you guys used for LS was to do the profile with the panel set to Gamma2.2)?
  • When I did the 3DLUT with B=50 C=85 Gamma=2.2, I had terrible results with post-cal verification. It (the gamma curve) was clearly badly wrong.
  • When I did the 3DLUT with B=52 C=83 Gamma=2.4, I had very nice results with post-cal verification.
Mine was set to Gamma2.4 (in the menus) with "Enable calibration" ticked in Calman and the target Gamma set to 2.4 in Calman.
The reason the guys using LightSpace (and my recommendation also) to use gamma 2.2 as a selection when you will not do 1D LUT upload, is because the 2.2 selection will not force LG processing to alter the values it will get from SM white balance.

The plan we do SM WB and not normal menu WB is because normal menu processing of adjuststents will degrade the picture, so for that reason we don't apply any WB or gamma adjustment from normal TV UI menu.

When you select other setting, like 2.4, the will use its processing of normal menu to alter the panel response, we don't need any processing from LG side, for that reason people use 2.2, as its the bypass setting where LG will not add any offset to gamma tracking.

Also when you set 2.4, this means that the whole color have reduced luminance (due to gamma change) so the meter will take more time to read, as the meter is faster with brighter patches. So its better idea when you take measurements for profiling your gamma to be a bit brighter from your target gamma, for the meter to read better the patches.

The issue you see is not normal, as with display characterization you can have at target any gamma values you want, and this will not affect the end results.

Quote:
Originally Posted by mrtickleuk View Post

The sequence I had been following, was:
  1. Using the TV's menus, set Gamma to 2.4, Brightness 52, Contrast 83 [the "destination" settings for this calibration]
  2. Take "pre" measurements in Calman
  3. Connect to the TV in Calman, tick "Enable Calibration"
  4. Using the TV's menus, set Gamma to 2.2 ["bypass" setting]
  5. Using Calman's DDC, Set OLED Light to Target / Brightness 50 / Contrast 85 [the "neutral native" settings]
  6. Do not touch 1DLUT. Do not "full DDC reset". Do not press Autocal button on Greyscale page. Do not pass go. Do not collect £200.
  7. Reset 3DLUT only (DDC window: 3DLUT controls: "Reset 3D Lut" button)
  8. Using the TV's menus, set Gamma to 2.4
  9. SM White Balance gain (only) adjust, to fix RGB Balance [displaying 100% grey 10% window +OLED light at target]
  10. Using the TV's menus, set Gamma to 2.2 ["required for 3DLUT"]
  11. Using Calman's DDC, set Brightness 52, Contrast 83 [the "destination" settings for this calibration]
  12. Create 3DLUT and upload in Calman, untick "Enable Calibration", disconnect from TV in Calman
  13. Using the TV's menus, set Gamma to 2.4, Brightness 52, Contrast 83 [the "destination" settings for this calibration]
  14. Take "post" measurements in Calman
(Step 11 might have been B=52, C=83, actually I can't remember). But it was definitely Gamma2.2
1. As your have find these settings are the correct your your specific TV, you use these during profiling, when you will not use 1D LUT Upload, you can set whatever Contrast/Brightness setting is better for your TV.

4+5) Bypass settings required with you are using 1D LUT upload.

As for your profiling you will not use 1D LUT, your have to set the BR/CR to values where with the Gamma 2.2 selection (the bypass processing) you will have correct CR/BR for your setup.

7) You reset the 3D LUT before starting the 1) step. Also have calibration disabled when you do 1) step.

After the initial setting of BR/CR and OLED Light (when you will have calibrated SM WB), you will not touch again all of there controls after the 3D LUT profiling.

Quote:
Originally Posted by mrtickleuk View Post
So it would really help me if someone else would verify that yes, you do have to set the TV Gamma to the destination gamma that you will want to use, before you start the 3DLUT points measurement. Ie if you want to end up with Gamma2.4, you must set Gamma2.4 not Gamma2.2 Thanks!
We don't want any LG processing, for that reason we do WB in SM and not in normal WB menu.
LeRoyK and mrtickleuk like this.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is online now  
post #3471 of 3665 Old 10-03-2019, 08:47 AM
Senior Member
 
LeRoyK's Avatar
 
Join Date: Dec 2004
Location: Woodstock Georgia
Posts: 426
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 88 Post(s)
Liked: 104
@ConnecTEDDD does the above recommendation apply to HDR modes as well?

Reset 3D LUT, set Brightness, Contrast and SM whitepoint (high only). Skip 1D LUT and calibrate 3D LUT.

Calman won't allow a full 3D LUT calculation in HDR, it only uses a few points.

LG OLED65C8PUA, Onkyo TX-RZ920, Panasonic DP-UB820, Apple TV 4K, Amazon Fire TV Stick 4K, AT&T Gigabit Internet & U-verse
LeRoyK is online now  
post #3472 of 3665 Old 10-03-2019, 08:59 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,773
Mentioned: 219 Post(s)
Tagged: 1 Thread(s)
Quoted: 3468 Post(s)
Liked: 4297
Quote:
Originally Posted by LeRoyK View Post
@ConnecTEDDD does the above recommendation apply to HDR modes as well?

Reset 3D LUT, set Brightness, Contrast and SM whitepoint (high only). Skip 1D LUT and calibrate 3D LUT.

Calman won't allow a full 3D LUT calculation in HDR, it only uses a few points.
3D LUT with volumetric measurements is possible for SDR only.

For HDR10, its impossible to perform 3D LUT with any WRGB OLED T, for that reason CalMAN is measuring only 5 patches.

100% White, Black, 100% Red/Green/Blue with 100% Stimulus level.

Based to these 5 patches, it will create a matrix LUT, and then use the 33-Point 3D LUT slot as a container.

Because of WRGB design of the panel, the large instability it has when its working in HDR mode, its impossible to do something more.

https://www.avsforum.com/forum/139-d...l#post56234908

https://www.avsforum.com/forum/139-d...l#post56201380

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is online now  
post #3473 of 3665 Old 10-03-2019, 09:17 AM
Senior Member
 
ebr9999's Avatar
 
Join Date: Jun 2013
Posts: 400
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 234 Post(s)
Liked: 100
Quote:
Originally Posted by ConnecTEDDD View Post
Gamma tracking, black level and peak output, to see how was your black/white level before and after the profiling.



Most people which are using 3D LUT with LG's are using i1DisplayPRO's (branded or OEM, retail).



The reason the guys using LightSpace (and my recommendation also) to use gamma 2.2 as a selection when you will not do 1D LUT upload, is because the 2.2 selection will not force LG processing to alter the values it will get from SM white balance.

The plan we do SM WB and not normal menu WB is because normal menu processing of adjuststents will degrade the picture, so for that reason we don't apply any WB or gamma adjustment from normal TV UI menu.

When you select other setting, like 2.4, the will use its processing of normal menu to alter the panel response, we don't need any processing from LG side, for that reason people use 2.2, as its the bypass setting where LG will not add any offset to gamma tracking.

Also when you set 2.4, this means that the whole color have reduced luminance (due to gamma change) so the meter will take more time to read, as the meter is faster with brighter patches. So its better idea when you take measurements for profiling your gamma to be a bit brighter from your target gamma, for the meter to read better the patches.

The issue you see is not normal, as with display characterization you can have at target any gamma values you want, and this will not affect the end results.



1. As your have find these settings are the correct your your specific TV, you use these during profiling, when you will not use 1D LUT Upload, you can set whatever Contrast/Brightness setting is better for your TV.

4+5) Bypass settings required with you are using 1D LUT upload.

As for your profiling you will not use 1D LUT, your have to set the BR/CR to values where with the Gamma 2.2 selection (the bypass processing) you will have correct CR/BR for your setup.

7) You reset the 3D LUT before starting the 1) step. Also have calibration disabled when you do 1) step.

After the initial setting of BR/CR and OLED Light (when you will have calibrated SM WB), you will not touch again all of there controls after the 3D LUT profiling.



We don't want any LG processing, for that reason we do WB in SM and not in normal WB menu.
Unluckly, using that method, I get something like gamma=2.8 at IR5 ... Something I cannot fix. This I think comes from my pannel (I have to set BR at 52). In such a case with gamma=2.2 I have excess of luminance. So the LS engine (correctly) kills luminance converting to 2.4. Not the case for friends that can keep BR=50. Starting from gamma=2.4 I still have a gamma=2.6 I fix with a parametric gamma forcing 2.1, 2.2, 2.3 at IR3, IR6, IR9. I have not followed the alternative way to work on UI, as I should have to set L=15 at IR5 ... somehow an extreme value .... surely better fixing using a 3DLUT

TV: LG 55E8PLA, Pana TX-L37ET5E BD/DVR: Panasonic BMR-BWT735 Media player: Teufel Connector 2, HT Rec: ANTHEM MRX 510,
Front: Woofer&Tweeter, Center: 2Woofer&1Tweeter, Surround: Bose acoustimass 5 series II, Sub: SVS SB2000
ebr9999 is offline  
post #3474 of 3665 Old 10-03-2019, 09:54 AM
Senior Member
 
LeRoyK's Avatar
 
Join Date: Dec 2004
Location: Woodstock Georgia
Posts: 426
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 88 Post(s)
Liked: 104
Quote:
Originally Posted by ConnecTEDDD View Post
3D LUT with volumetric measurements is possible for SDR only.

For HDR10, its impossible to perform 3D LUT with any WRGB OLED T, for that reason CalMAN is measuring only 5 patches.

100% White, Black, 100% Red/Green/Blue with 100% Stimulus level.

Based to these 5 patches, it will create a matrix LUT, and then use the 33-Point 3D LUT slot as a container.

Because of WRGB design of the panel, the large instability it has when its working in HDR mode, its impossible to do something more.

https://www.avsforum.com/forum/139-d...l#post56234908

https://www.avsforum.com/forum/139-d...l#post56201380
So what method do you recommend for Calman calibration of HDR10/HLG? Use 1D LUT autocal, or user menu 20pt or service menu 20pt, or leave the gamma alone after the matrix LUT?

Thanks

LeRoy

LG OLED65C8PUA, Onkyo TX-RZ920, Panasonic DP-UB820, Apple TV 4K, Amazon Fire TV Stick 4K, AT&T Gigabit Internet & U-verse
LeRoyK is online now  
post #3475 of 3665 Old 10-03-2019, 10:04 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,773
Mentioned: 219 Post(s)
Tagged: 1 Thread(s)
Quoted: 3468 Post(s)
Liked: 4297
Quote:
Originally Posted by LeRoyK View Post
So what method do you recommend for Calman calibration of HDR10/HLG? Use 1D LUT autocal, or user menu 20pt or service menu 20pt, or leave the gamma alone after the matrix LUT?
I don't have a method to recommend you for HDR10, because there no proven method which will work for all.

Try and compare:

1) 1D LUT AutoCAL + Matrix LUT

2) 1D LUT AutoCAL + No Matrix LUT, just use factory CMS.

3) SM WB with HDR10 pattern + some modification of some 20-Point Controls.

One of those 3 will work better.

But except verification measurements, evaluate with gradation patterns and a lot of movies evaluation.
mrtickleuk likes this.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is online now  
post #3476 of 3665 Old 10-03-2019, 10:36 AM
Senior Member
 
LeRoyK's Avatar
 
Join Date: Dec 2004
Location: Woodstock Georgia
Posts: 426
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 88 Post(s)
Liked: 104
Quote:
Originally Posted by ConnecTEDDD View Post
I don't have a method to recommend you for HDR10, because there no proven method which will work for all.

Try and compare:

1) 1D LUT AutoCAL + Matrix LUT

2) 1D LUT AutoCAL + No Matrix LUT, just use factory CMS.

3) SM WB with HDR10 pattern + some modification of some 20-Point Controls.

One of those 3 will work better.

But except verification measurements, evaluate with gradation patterns and a lot of movies evaluation.
Thanks Ted, I was afraid of that answer. I have tried #1 resulting in some obvious flaws with the Spears & Munsil 4K 4000 nit HDR demo material. For instance the background of the horses in the snow scene is not coming together like the Dolby Video clip.

I will try #3 next using the Spears and Munsil patterns with lots of black screen rest between readings. The panel gets brighter the longer it shows a bright image.

Should I use OLED light to hit the 700 nit 100% white before using these methods? Or will 1D LUT fail with any OLED light under 100?

LG OLED65C8PUA, Onkyo TX-RZ920, Panasonic DP-UB820, Apple TV 4K, Amazon Fire TV Stick 4K, AT&T Gigabit Internet & U-verse
LeRoyK is online now  
post #3477 of 3665 Old 10-03-2019, 10:40 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,773
Mentioned: 219 Post(s)
Tagged: 1 Thread(s)
Quoted: 3468 Post(s)
Liked: 4297
Quote:
Originally Posted by LeRoyK View Post
Thanks Ted, I was afraid of that answer. I have tried #1 resulting in some obvious flaws with the Spears & Munsil 4K 4000 nit HDR demo material. For instance the background of the horses in the snow scene is not coming together like the Dolby Video clip.

I will try #3 next using the Spears and Munsil patterns with lots of black screen rest between readings. The panel gets brighter the longer it shows a bright image.

Should I use OLED light to hit the 700 nit 100% white before using these methods? Or will 1D LUT fail with any OLED light under 100?
According how 2018 models work, the tone mapping engine calculations performed expecting the 700 nits as panel peak brightness.

A different test will be to see if by reducing OLED Light to reach 700 nits calibrated it will be better from leaving OLED Light @ 100.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is online now  
post #3478 of 3665 Old 10-03-2019, 10:48 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,773
Mentioned: 219 Post(s)
Tagged: 1 Thread(s)
Quoted: 3468 Post(s)
Liked: 4297
Quote:
Originally Posted by mrtickleuk View Post
Thanks, Tedd. In fact I have just finished one, so I will upload the file to this post.

Anyway here's the file. NB I use dEITP in the charts. I did 8,000 pts and it took 7hrs 54mins.

Limitations I am aware of:
  • Calman engines has problems with blue? Not sure of the details but I saw it meantioned
Quote:
Originally Posted by mrtickleuk View Post
Correct, yes this was with the new version from last week.
About your results your posted using only the 3D LUT profiling of CalMAN by skipping to perform 1D LUT AutoCAL, using the latest CalMAN 2019 R2.

Beyond measurements, how picture looks? Have you display any reference pattern with gradation to take a look if you can spot any issue?

The more problems come from blue/yellow area, and to other colors at smaller scale.

I have import your file to LightSpace to see what is happening looking the Cube Preview / 1D LUT viewer, also I have opened a 4500 point measurement file of another user with C8 who posted to LightSpace 3D LUT for LG OLED to compare.

The generated 3D LUT file has the typical noise of CalMAN LUT's we see the last years, the over-saturation yellow issue don't appear to your LUT, but there is another problem to the yellow area.

Lets start with Cube Preview:










If you want to see how some better generated cube's look like see:

https://www.avsforum.com/forum/139-d...l#post58469276

https://www.avsforum.com/forum/139-d...l#post58522806

https://www.avsforum.com/forum/139-d...l#post58546994

Now see how the C8 3D LUT (using the same pre-calibration with SM WB and skipping 1D LUT) compared between LightSpace (Anger.miki) and CalMAN (yours):



You can see the problem to your blue area.

Also your cube is smaller, means that its de-saturated and reduce luminance more than from LightSpace.

This can be visible looking your 1D LUT graph.

Your total luminance has been reduced by CalMAN, for the unknown reason this is happening with every LUT I have checked, which is not normal, as you have previously calibrated your 100% White and the display has native gamut, so enough luminance for colors, any software has no reason to reduce your data for so specific reason, I have posted about this from 2018 to other examples: https://www.avsforum.com/forum/139-d...l#post56559306

You can see that details to that 1D LUT below, where I have focused to top end only (60-100%), and you see the UNITY (Black) line also.

When RGB lines are below black (UNITY Line) means that data reduced from your signal, as you can see, both displays has been pre-calibrated perfectly their 100% White before starting the profiling (for that reason the RGB lines are close, as its not required changes to channels to fix RGB balance), in LS LUT there no any reduction, while with CM LUT you see a reduction.



Loading some reference pictures to LUT Preview, you can see the problem to Blue, Yellow, and to the edge of other colors.






mrtickleuk likes this.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is online now  
post #3479 of 3665 Old 10-03-2019, 12:14 PM
AVS Forum Special Member
 
mrtickleuk's Avatar
 
Join Date: May 2016
Location: Birmingham, UK - you know, the original one!
Posts: 7,491
Mentioned: 184 Post(s)
Tagged: 0 Thread(s)
Quoted: 5190 Post(s)
Liked: 7508
Hi @ConnecTEDDD , thanks for your detailed reply.

Quote:
Originally Posted by ConnecTEDDD View Post
Gamma tracking, black level and peak output, to see how was your black/white level before and after the profiling.
ok thanks; I meant which time should I take these measurements for the "pre" charts?
  • now,
  • after resetting the 3DLUT,
  • after using the SM to get the RGB Balance as good as I can?
(doesn't really matter but that's what I was asking)

Quote:
The reason the guys using LightSpace (and my recommendation also) to use gamma 2.2 as a selection when you will not do 1D LUT upload, is because the 2.2 selection will not force LG processing to alter the values it will get from SM white balance.

The plan we do SM WB and not normal menu WB is because normal menu processing of adjuststents will degrade the picture, so for that reason we don't apply any WB or gamma adjustment from normal TV UI menu.

When you select other setting, like 2.4, the will use its processing of normal menu to alter the panel response, we don't need any processing from LG side, for that reason people use 2.2, as its the bypass setting where LG will not add any offset to gamma tracking.
Yes understood about less LG processing, but are you only talking about the state of the Gamma setting during profiling and calibration, or are you talking about the state of the Gamma setting when I watch content which I prefer to watch with Gamma=2.4?

I don't think you mean this:
Leave the TV's Gamma menu set to "2.2", but do the calibration with Calman set to 2.4 light levels (targets), so that in the end, the TV's UI isn't doing any processing, but the end result will be Gamma 2.4 and less processing. You just have to remember it and it's fine, but the "2.2" entry in the TV's menu will be a lie, it will really give you 2.4, and then "2.4" menu entry will give you something like 2.6, etc!)
I do think you mean this:
Leave the TV's Gamma menu set to "2.2", and do the calibration with Calman set to 2.2 light levels (targets), so that the end result will be Gamma 2.2. (The "2.2" in the TV's menu will be true). Then, when watching content, use the TV's menu setting (and LG processing) to get Gamma 2.4.
The reason I don't think you mean the first one is, because of what you said next:
Quote:
Also when you set 2.4, this means that the whole color have reduced luminance (due to gamma change) so the meter will take more time to read, as the meter is faster with brighter patches. So its better idea when you take measurements for profiling your gamma to be a bit brighter from your target gamma, for the meter to read better the patches.
(profiling gamma=2.2, brighter than target viewing gamma=2.4?).

Also, I had not though of doing that, especially that it would be both quicker and more accurate. Double win!

Quote:
The issue you see is not normal, as with display characterization you can have at target any gamma values you want, and this will not affect the end results.
Maybe the "Enable Calibration" setting sets Gamma 2.2 for profiling, but then I was accidentally over-riding it when I set Gamma=2.4 in the menus and causing that problem?

Anyway, this is what I think we are now saying I should try next time:
  1. Connect to the TV in Calman, but do not tick "Enable Calibration"
  2. Using the TV's menus, set actual Gamma to 2.2. In Calman, set the target Gamma to 2.2 also.
  3. Reset 3DLUT only (DDC window: 3DLUT controls: "Reset 3D Lut" button)
  4. Now we are ready to use Test Patterns to determine correct Brightness "B1" and Contrast "C1" for Gamma=2.2 state. [NB: Might be different to the correct Brightness and Contrast for Gamma=2.4 state, worry later]
  5. Take "pre" measurements in Calman [Gamma tracking, black level and peak output]
  6. Connect to the TV in Calman, tick "Enable Calibration"
  7. Using Calman's DDC, Set OLED Light to Target / Brightness 50 / Contrast 85 [the "neutral native" settings]
  8. Do not touch 1DLUT. Do not "full DDC reset". Do not press Autocal button on Greyscale page. Do not pass go. Do not collect £200.
  9. SM White Balance gain (only) adjust, to fix RGB Balance [displaying 100% grey 10% window +OLED light at target]
  10. Using Calman's DDC, set Brightness 52, Contrast 83 [the "correct for Gamma=2.2" settings]
  11. Create 3DLUT and upload in Calman, untick "Enable Calibration", disconnect from TV in Calman
  12. Take "post" measurements in Calman, still with Calman's target Gamma as 2.2

Then after calibration:
  • Now we are ready to use Test Patterns to determine correct Brightness "B2" and Contrast "C2" for Calibrated Gamma=2.4 state. [NB: Might be different to the correct Brightness and Contrast for Gamma=2.2 state]
  • We then have two sets of values: for Gamma=2.2, Brightness B1/Contrast C1; Gamma=2.4, Brightness B2/Contrast C2.
  • When watching content, use the TV's menus to switch from B1/C1 to B2/C2 depending on which Gamma is used. The TV will use the "pure" SM+3DLUT state for Gamma 2.2, and it will use SM+3DLUT + a little processing for Gamma 2.4.

Quote:
We don't want any LG processing, for that reason we do WB in SM and not in normal WB menu.
Yes understood but please see my question above? TIA

_______________
Denon AVR-X4200W, Arcam Alpha 8P; 5.1.4 setup: Mission 702e, M7C1i, 77DS, 731. Rel T5 Sub. Monitor Audio CT165 4 Tops | LG OLED55C8PLA TV | Samsung UBD-K8500 UHD Blu-Ray

HDMI 2.0 4K modes | Dolby & DTS core+outer audio tracks on (UHD) Blu-Rays | Hello to Jason Isaacs

Last edited by mrtickleuk; 10-03-2019 at 01:26 PM.
mrtickleuk is online now  
post #3480 of 3665 Old 10-03-2019, 12:39 PM
AVS Forum Special Member
 
mrtickleuk's Avatar
 
Join Date: May 2016
Location: Birmingham, UK - you know, the original one!
Posts: 7,491
Mentioned: 184 Post(s)
Tagged: 0 Thread(s)
Quoted: 5190 Post(s)
Liked: 7508
Quote:
Originally Posted by ConnecTEDDD View Post
About your results your posted using only the 3D LUT profiling of CalMAN by skipping to perform 1D LUT AutoCAL, using the latest CalMAN 2019 R2.

Beyond measurements, how picture looks? Have you display any reference pattern with gradation to take a look if you can spot any issue?

The more problems come from blue/yellow area, and to other colors at smaller scale.
It looks fantastic, the best I've seen on the TV with real content.

With patterns, I have checked more today than I did before and I think it's very interesting. I've used your RGB Ramps 1, RGB Ramps 2, Colour Ramps horizontal+vertical, and the Linearisation patterns.

My other Picture Modes (ISF Dark/Bright and Technicolour) are all using the SM-corrected "Warm2" which is D65, better than from the factory, and LG's default 1DLUT and 3DLUT.
My Cinema Picture Mode is my calibration with different White Point, LG factory 1DLUT, and my 3DLUT.

Results:
All picture modes, including the LG ones, had some banding visible in all patterns.
With some patterns, my Cinema mode had less banding, with others the LG modes had less banding. Mostly, the banding just moved within the pattern.
For each one it's normally difficult for me to decide which is best. The only exception was in the case of the Blue Linearisation pattern, where it's easy to see the banding but looking at the cubes you posted Blue was always going to be the most "evil" test for it.
But I haven't noticed any problems at all in content.

Quote:
I have import your file to LightSpace to see what is happening looking the Cube Preview / 1D LUT viewer, also I have opened a 4500 point measurement file of another user with C8 who posted to LightSpace 3D LUT for LG OLED to compare.

The generated 3D LUT file has the typical noise of CalMAN LUT's we see the last years, the over-saturation yellow issue don't appear to your LUT, but there is another problem to the yellow area.

You can see the problem to your blue area.
Thanks for your analysis. Other than the blue being a little bit ragged, is that it? I take it this is due to the particular sequence of patches, and I know that the LS one was from a "hardcore" LS user with a very special sequence designed to minimise drift and equalise voltage. Without that, mine will never be as good, I have decided to cope with how CM works now and be pleased if/when it improves.

Quote:
Loading some reference pictures to LUT Preview, you can see the problem to Blue, Yellow, and to the edge of other colors.

I do see, yes. Thanks for all that.

After I establish which gamma settings to use for 3DLUT generation vs viewing, I can easily run it again to see if I can improve it (my previous post).

_______________
Denon AVR-X4200W, Arcam Alpha 8P; 5.1.4 setup: Mission 702e, M7C1i, 77DS, 731. Rel T5 Sub. Monitor Audio CT165 4 Tops | LG OLED55C8PLA TV | Samsung UBD-K8500 UHD Blu-Ray

HDMI 2.0 4K modes | Dolby & DTS core+outer audio tracks on (UHD) Blu-Rays | Hello to Jason Isaacs
mrtickleuk is online now  
Sponsored Links
Advertisement
 
Reply Display Calibration

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off