2018 LG OLED Calibration and User Settings (No price talk) - Page 121 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 2458Likes
Reply
 
Thread Tools
post #3601 of 3635 Old 11-05-2019, 10:12 AM
Advanced Member
 
janos666's Avatar
 
Join Date: Jan 2010
Location: Hungary
Posts: 760
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 108 Post(s)
Liked: 64
Quote:
Originally Posted by chros73 View Post
Nice!
Do you remember what was the max nit in HDR Cinema mode @ 100 OLED Light (without brightness drifting)? Which size do you have?
55C8PLA and I don't remember the peak of the HDR10 mode but the DolbyVision file recorded 810 nit and HDR10 should be the same (save for any random drift). So, the DV mode now offers ~15% higher peak brightness for highlights. It's a pity LG never made this configurable for the HDR10/HLG mode on these 2018 models even though they marketed them with "increased peak brightness in HDR" (which effectively makes them less accurate when it happens to be true, LOL).
Thank you @ConnecTEDDD for this OLED Light idea!
ConnecTEDDD and chros73 like this.

"DIY certified hobby-calibrator" (based on ChadB's "warning signs" list
janos666 is offline  
Sponsored Links
Advertisement
 
post #3602 of 3635 Old 11-05-2019, 12:50 PM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,688
Mentioned: 214 Post(s)
Tagged: 1 Thread(s)
Quoted: 3428 Post(s)
Liked: 4248
Quote:
Originally Posted by chros73 View Post
Can you upload an image of B7/C7? Thanks


This is from LG C7 with Technicolor firmware installed, which adds 17-Point Cube 3D LUT and 1024 entries 1D LUT
chros73 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 #3603 of 3635 Old 11-05-2019, 01:01 PM
AVS Forum Special Member
 
WiFi-Spy's Avatar
 
Join Date: Feb 2004
Location: Seattle, WA
Posts: 3,887
Mentioned: 234 Post(s)
Tagged: 0 Thread(s)
Quoted: 1608 Post(s)
Liked: 2194
Quote:
Originally Posted by janos666 View Post
55C8PLA and I don't remember the peak of the HDR10 mode but the DolbyVision file recorded 810 nit and HDR10 should be the same (save for any random drift). So, the DV mode now offers ~15% higher peak brightness for highlights. It's a pity LG never made this configurable for the HDR10/HLG mode on these 2018 models even though they marketed them with "increased peak brightness in HDR" (which effectively makes them less accurate when it happens to be true, LOL).
Thank you @ConnecTEDDD for this OLED Light idea!
DV uses different video pipeline configuration compared to HDR10/HLG. LG listened to calibrators/enthusiasts with regards to the peak brightness situation on 2018 TVs and that is why there is that control and also custom tone curves for the 2019 models.

Tyler Pruitt - Technical Evangelist - for CalMAN

10 Bit Gradient Test Patterns (HEVC) - Free Download
WiFi-Spy is online now  
Sponsored Links
Advertisement
 
post #3604 of 3635 Old 11-05-2019, 01:05 PM
AVS Forum Special Member
 
mrtickleuk's Avatar
 
Join Date: May 2016
Location: Birmingham, UK - you know, the original one!
Posts: 7,371
Mentioned: 182 Post(s)
Tagged: 0 Thread(s)
Quoted: 5097 Post(s)
Liked: 7431
Quote:
Originally Posted by ConnecTEDDD View Post
Spoiler!


This is from LG C7 with Technicolor firmware installed, which adds 17-Point Cube 3D LUT and 1024 entries 1D LUT
Quite incredible frame-rate. Thirty thousand frames per second! Wow!
chros73 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 offline  
post #3605 of 3635 Old 11-05-2019, 01:07 PM
AVS Forum Special Member
 
mrtickleuk's Avatar
 
Join Date: May 2016
Location: Birmingham, UK - you know, the original one!
Posts: 7,371
Mentioned: 182 Post(s)
Tagged: 0 Thread(s)
Quoted: 5097 Post(s)
Liked: 7431
Quote:
Originally Posted by WiFi-Spy View Post
DV uses different video pipeline configuration compared to HDR10/HLG. LG listened to calibrators/enthusiasts with regards to the peak brightness situation on 2018 TVs and that is why there is that control and also custom tone curves for the 2019 models.
This was good, but it's a crying shame that the group of people who own the C8 and made the suggestions don't get any benefit and LG gift the "fixes" to a completely different set of people - the C9 owners
chros73 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 offline  
post #3606 of 3635 Old 11-05-2019, 01:28 PM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,688
Mentioned: 214 Post(s)
Tagged: 1 Thread(s)
Quoted: 3428 Post(s)
Liked: 4248
Quote:
Originally Posted by WiFi-Spy View Post
DV uses different video pipeline configuration compared to HDR10/HLG. LG listened to calibrators/enthusiasts with regards to the peak brightness situation on 2018 TVs and that is why there is that control and also custom tone curves for the 2019 models.
When you perform DV/HDR10 AutoCAL, the whole additional processing for PQ or gamut mapping used for DV or HDR10 are totally bypassed, as the panel has native response and gamma tracking, even if you start DV AutoCAL or HDR10 AutoCAL, the panel response will be exact the same (and peak output), so makes no sense to calibrate 2 times the 1D LUT since the panel is super unstable per each time, while you waste a lot of your time also.

The video processing pipeline is different when the calibration of the panel has been disabled, when the PQ/gamut mapping are enabled, but the panel respond the same when you have these stuff disabled during AutoCAL.

CalMAN has to add the capability to copy the 1D LUT values of HDR10 to 1D LUT of DV, its not LG responsible for this ...and then ...from the 5 patch measurements to generate a 33-Point 3D LUT for HDR10 and a DV file for DoVi. (just to have workflow steps to say to the user.... now playback a DV file and press upload...now playback HDR10 content and press upload)

This procedure will help the whole calibration community (enthusiast/professionals) who spend double time now to calibrate the same native panel response.
L30Z3N 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 #3607 of 3635 Old 11-05-2019, 01:30 PM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,688
Mentioned: 214 Post(s)
Tagged: 1 Thread(s)
Quoted: 3428 Post(s)
Liked: 4248
Quote:
Originally Posted by janos666 View Post
After a few hours of fighting the quirks of the HDFury Integral I eventually managed to calibrate the DolbyVision Cinema mode on my C8. And I found an unexpected surprise: CalMAN left me a .CSV alongside the .TXT config file. This extra file looks suspiciously like a 1DLUT. So, it seems like it's possible to re-use a DolbyVision 1DLUT for HDR10 picture modes after all. Hurray! Well, at least on 2019 OLEDs and on a few random 2018 ones, because I opted to calibrate HDR10 with OLED Light set to 86 instead in order to hit the hard-coded constant 700 nit tone-mapping parameter of the 2018 models (and the 1DLUT shan't be re-used for picture modes with different OLED Light settings, similarly to SDR Day/Night or BFI/no-BFI picture modes). So, even though this is technically nice, it's practically still useless.
Can you upload these 2 files? (I want to see if that cvs file can be imported to LightSpace)

Probably the cvs file should be the last 1D LUT CalMAN uploaded to your TV, but there no upload button to send manually 1D LUT data to your TV using CalMAN.

This is possible with DeviceControl, if you playback an HDR10/DV content from your HDMI Input, when you will tick from DeviceControl to 'enable calibration', LG will enable the proper slot for 1D LUT (for DV or HDR10...based to the content you was playing when you enabled calibration) , so if the cvs file its a known 1D LUT format for LightSpace, you can import it and then export it as DeviceControl 1D LUT LG format.

Then you can use LG Template of DeviceControl to upload the 1D LUT to your TV.

The OLED Light in DV mode is not OLED panel gain control like is working in HDR10/SDR modes, its a slider to control a mid-level boost of PQ EOTF (so you control how much to deviate from PQ tracking to your mid-levels...it can be used to enhance picture if you watch DV during the day for example).

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 #3608 of 3635 Old 11-05-2019, 01:44 PM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,688
Mentioned: 214 Post(s)
Tagged: 1 Thread(s)
Quoted: 3428 Post(s)
Liked: 4248
Quote:
Originally Posted by WiFi-Spy View Post
The suggested way to calibrate for Full Range is To perform the calibration in limited range RGB, Then change LG menu setting: “black level=low to High” After calibration. This is the process that is used in Hollywood when using the TV in full range.
I have a question.. to all CalMAN 2018 LG guides and videos, the workflow has default 16-255 for 1D LUT and 16-255 for 3D LUT, but its not saying how to configure an external generator with more detail, its saying only to use 10% Windows, what colorspace output settings a Murideo Six-G or VideoForge PRO user has to use to perform 1D/3D LUT for 16-255?

The only notice I see is for users with generators not capable to generate super-white to use 16-235 patch generation.

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 #3609 of 3635 Old 11-05-2019, 01:44 PM
AVS Forum Special Member
 
WiFi-Spy's Avatar
 
Join Date: Feb 2004
Location: Seattle, WA
Posts: 3,887
Mentioned: 234 Post(s)
Tagged: 0 Thread(s)
Quoted: 1608 Post(s)
Liked: 2194
Quote:
Originally Posted by mrtickleuk View Post
This was good, but it's a crying shame that the group of people who own the C8 and made the suggestions don't get any benefit and LG gift the "fixes" to a completely different set of people - the C9 owners


There are certain things that cannot be added with a OTA or USB firmware update. One of those would be a video processing block. A firmware update can change the settings or values of a processing block, but it cannot add a processing block.

Tyler Pruitt - Technical Evangelist - for CalMAN

10 Bit Gradient Test Patterns (HEVC) - Free Download
WiFi-Spy is online now  
post #3610 of 3635 Old 11-05-2019, 04:17 PM
Advanced Member
 
janos666's Avatar
 
Join Date: Jan 2010
Location: Hungary
Posts: 760
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 108 Post(s)
Liked: 64
Quote:
Originally Posted by WiFi-Spy View Post
There are certain things that cannot be added with a OTA or USB firmware update. One of those would be a video processing block. A firmware update can change the settings or values of a processing block, but it cannot add a processing block.
If there are fix-function hardware units for certain tasks, then sure. But if the SoC is built up from very similar general purpose CPU and GPU cores in both years (as we suspect it to be the case) then theoretically anything should be possible to back-port as long as the raw processing power and RAM size, etc is enough for the new code to run. I think it's more of a question of will rather than possibility.
It's understandable if they don't wish to make huge changes to old models but the HDR10 tone-mapping parameter override doesn't seem like it would need fundamental changes.

Quote:
Originally Posted by ConnecTEDDD View Post
The video processing pipeline is different when the calibration of the panel has been disabled, when the PQ/gamut mapping are enabled, but the panel respond the same when you have these stuff disabled during AutoCAL.
Well, the 'pipeline' could technically be different. May be everything goes through the Dolby CMU chip in DV mode even if it's in passthrough/bypass (no processing) mode at the time (following the metadata). But I agree with the rest. The panel response is always the panel response and we should see that in both cases.

Quote:
Originally Posted by ConnecTEDDD View Post
from the 5 patch measurements to generate a 33-Point 3D LUT for HDR10
Wait, I thought HDR10 uses an actual 3x3 matrix similarly to DV and not a 33 cube generated from a matrix.

Quote:
Originally Posted by ConnecTEDDD View Post
The OLED Light in DV mode is not OLED panel gain control like is working in HDR10/SDR modes, its a slider to control a mid-level boost of PQ EOTF
Yes, I know that. I left OLED Light at 50 in DV (which matches 100 in HDR10) and set it to 86 in HDR10.

Quote:
Originally Posted by ConnecTEDDD View Post
Can you upload these 2 files? (I want to see if that cvs file can be imported to LightSpace)
Sure.
Attached Files
File Type: zip DolbyVision_UserDisplayConfiguration.zip (7.0 KB, 13 views)

"DIY certified hobby-calibrator" (based on ChadB's "warning signs" list

Last edited by janos666; 11-05-2019 at 05:42 PM.
janos666 is offline  
post #3611 of 3635 Old 11-05-2019, 05:24 PM
AVS Forum Special Member
 
mrtickleuk's Avatar
 
Join Date: May 2016
Location: Birmingham, UK - you know, the original one!
Posts: 7,371
Mentioned: 182 Post(s)
Tagged: 0 Thread(s)
Quoted: 5097 Post(s)
Liked: 7431
Quote:
Originally Posted by WiFi-Spy View Post
There are certain things that cannot be added with a OTA or USB firmware update. One of those would be a video processing block. A firmware update can change the settings or values of a processing block, but it cannot add a processing block.
Yeah. I appreciate the technical restrictions... frustrating though all the same.

_______________
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 offline  
post #3612 of 3635 Old 11-06-2019, 05:09 AM
Advanced Member
 
chros73's Avatar
 
Join Date: Jan 2015
Posts: 599
Mentioned: 12 Post(s)
Tagged: 0 Thread(s)
Quoted: 377 Post(s)
Liked: 174
Quote:
Originally Posted by janos666 View Post
I opted to calibrate HDR10 with OLED Light set to 86 instead in order to hit the hard-coded constant 700 nit tone-mapping parameter of the 2018 models
Have you measured the response of the panel around 700/800 nits? I'm not sure about that 700 nits at all.

Ryzen 5 2600,Asus Prime b450-Plus,16GB,MSI GTX 1060 Gaming X 6GB(v385.28),Win10 LTSB 1607,MPC-BEx64+LAV+MadVR,Yamaha RX-A870,LG OLED65B8(04.10.25+PC4:4:[email protected]/24/25/29/30/50/59/60Hz)
chros73 is online now  
post #3613 of 3635 Old 11-06-2019, 05:13 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,688
Mentioned: 214 Post(s)
Tagged: 1 Thread(s)
Quoted: 3428 Post(s)
Liked: 4248
Quote:
Originally Posted by mrtickleuk View Post
Yeah. I appreciate the technical restrictions... frustrating though all the same.
Since we talk for 1D LUT, the TV has 3x 1D LUT's to the exact same processing block per each picture mode.

Just the TV will select each slot based to your input signal.

So when you playback SDR content and enable calibration, it will be uploaded to SDR 1D LUT slot, if you playback HDR10 content and enable calibration it will be uploaded to HDR10 1D LUT slot, if you playback DoVi content and enable calibration it will be uploaded to DoVi 1D LUT slot.

1D LUT is the last table before the panel, so when you take measurements with native panel response, all processing in front of 1D LUT are bypassed.

Below you see Neil from LG explaining the processing blocks:



The 1D LUT I have highlight with red stroke is the 1D LUT its been used for uploading custom 1D LUT table. All previous processing blocks are bypassed when you measure for HDR/DV AutoCAL, SDR AutoCAL's, as you bypass all these processing and you calibrate the gamma 2.2
jrref and chros73 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 #3614 of 3635 Old 11-06-2019, 05:20 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,688
Mentioned: 214 Post(s)
Tagged: 1 Thread(s)
Quoted: 3428 Post(s)
Liked: 4248
Quote:
Originally Posted by janos666 View Post
Sure.
The LG_D65.cvs looks like it contain calibration data, but it has some 'steps' (at the panel are super unstable and you take multiple measurements to fix errors in real-time of each point).



Just for fun, I used excel to plot the first 30 values and there are some little kinks throughout the plot... when you have 'steps'...this can introduce banding.



I hope panels to be more stable in the future when its coming to HDR as for SDR are pretty stable.

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 #3615 of 3635 Old 11-06-2019, 05:22 PM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,688
Mentioned: 214 Post(s)
Tagged: 1 Thread(s)
Quoted: 3428 Post(s)
Liked: 4248
Quote:
Originally Posted by chros73 View Post
Amount of DRAM in different LG models from the diag screen:
1.5 GB : 55B7, 55C7
2GB : 65B8
3GB : 55C8 , 77C9
This is from 65E6:


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 #3616 of 3635 Old 11-07-2019, 02:41 AM
Advanced Member
 
chros73's Avatar
 
Join Date: Jan 2015
Posts: 599
Mentioned: 12 Post(s)
Tagged: 0 Thread(s)
Quoted: 377 Post(s)
Liked: 174
Quote:
Originally Posted by ConnecTEDDD View Post
This is from 65E6
Thanks, updated. What interesting is to see that 2017 models have smaller amount of DRAM.
We only need a B6 and B9 for a full list.

Ryzen 5 2600,Asus Prime b450-Plus,16GB,MSI GTX 1060 Gaming X 6GB(v385.28),Win10 LTSB 1607,MPC-BEx64+LAV+MadVR,Yamaha RX-A870,LG OLED65B8(04.10.25+PC4:4:[email protected]/24/25/29/30/50/59/60Hz)
chros73 is online now  
post #3617 of 3635 Old 11-07-2019, 11:26 AM
Advanced Member
 
Join Date: Jul 2017
Posts: 600
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 410 Post(s)
Liked: 204
Quote:
Originally Posted by chros73 View Post

Can you upload an image of B7/C7? Thanks
here my B7v
Attached Thumbnails
Click image for larger version

Name:	IMG_8259.jpg
Views:	25
Size:	1.21 MB
ID:	2637770  
chros73 likes this.

Video: LG OLED 55 B7v | Calibration: i1D3 OEM, Calman Enthusiast.
Audio: Yamaha RX-V585, Polk s15e, Polk s10e, Polk s35ce, Klipsch N12 SW.
MrRobotoPlus is offline  
post #3618 of 3635 Old 11-07-2019, 01:50 PM
Advanced Member
 
bryantc's Avatar
 
Join Date: Oct 2016
Location: Central
Posts: 750
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 513 Post(s)
Liked: 468
Quote:
Originally Posted by chros73 View Post
Amount of DRAM in different LG models from the diag screen:
1.5 GB : 55B7, 55C7
2GB : 65C6, 65E6, 65B8
3GB : 55C8 , 77C9


Here is the Z9:






And a 77 C8:


ConnecTEDDD and chros73 like this.

Media: LG 88" 8K OLED Z9 / Denon AVR-X8500H 7.1.4 / Oppo UDP-203 / Nvidia SHIELD (2017) / Apple TV 4K / Sony 4K FMP-X10 / Gaming PC (Threadripper 1950X / 2x GeForce GTX 1080 Ti SLI) / PlayStation 4 Pro with PS VR / PlayStation (original)
Family: Sony 85" 4K XBR-85X900F / Denon AVR-X3400H 5.1 / Nvidia SHIELD (2017) / Apple TV 4K / C64 Mini - Game: LG 77" 4K OLED C8 / Denon AVR-X3400H 5.1 / Nvidia SHIELD (2017) / Apple TV 4K
bryantc is offline  
post #3619 of 3635 Old 11-08-2019, 05:13 AM
Advanced Member
 
janos666's Avatar
 
Join Date: Jan 2010
Location: Hungary
Posts: 760
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 108 Post(s)
Liked: 64
Quote:
Originally Posted by ConnecTEDDD View Post
I hope panels to be more stable in the future when its coming to HDR as for SDR are pretty stable.
Did anyone try to apply direct airflow (with a regular floor standing fan) to the test patch area of the screen? Could it help or is it more about electric charge build-up (like it used to be with PDPs) than heath?
I remember Vincent (HDTVTest.co.uk) speculating about how the better IR resistance of the GZ2000 might be related to it's heath dispensing aluminum back-plate.

Another question: What's the real size of the HDR10 3DLUT in the Alpha9Gen1 hardware? It it 3x3 or 33^3?
I thought it's 3x3 (similar to the DV file) but I cleared all the .3dl files from CalMAN's folder, did a quick HDR10 cal (skipping the 1DLUT and going straight for the 3DLUT in Game mode) and found only 33^3 files (neutral and modified).
However, CalMAN seems to refuse uploading anything to the hardware slot (without giving any errors) when I select the "Lighting LUT" method (the red warning states "matrix method is recommended" but not that it's "mandatory" and it's also possible to create a 3x3 matrix from multi-point measurements, so...).
I was just curious to see if the "Lighting LUT" result alters the tone-response (to confirm if it's 3x3 or 33^3 in hardware) and if so, then if it could possibly be less "stepped" (the benefit of multiple samples per control points during the 1DLUT calibration might be diminished by the panel drift, so this could be a faster alternative with similar end results --- assuming it's 33^3 and it worked).

"DIY certified hobby-calibrator" (based on ChadB's "warning signs" list

Last edited by janos666; 11-08-2019 at 05:17 AM.
janos666 is offline  
post #3620 of 3635 Old 11-08-2019, 05:42 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,688
Mentioned: 214 Post(s)
Tagged: 1 Thread(s)
Quoted: 3428 Post(s)
Liked: 4248
Quote:
Originally Posted by janos666 View Post
Did anyone try to apply direct airflow (with a regular floor standing fan) to the test patch area of the screen? Could it help or is it more about electric charge build-up (like it used to be with PDPs) than heath?
I remember Vincent (HDTVTest.co.uk) speculating about how the better IR resistance of the GZ2000 might be related to it's heath dispensing aluminum back-plate.
Its been tested a Panasonic even with 2 fans to a 'syberia-like' air- conditioned room, even using black frame of 9 or 14 seconds before displaying each pattern, with Panasonic EZ1000 where you can upload an 1D LUT which will disable any active tone mapping, and profile took 48 hours to compete 21-Point Cube measurements with K-10A, but it was impossible to generate any valid 3D LUT correction due to that large drifting and due to WRGB problem in HDR.


This is drift plot with normal delay 1 sec (+-200 nits deviations)



This is drift plot using 9 sec black frame between displaying each patch:


...and the same TV in SDR:





Quote:
Originally Posted by janos666 View Post
Another question: What's the real size of the HDR10 3DLUT in the Alpha9Gen1 hardware? It it 3x3 or 33^3?
Its has 33-Point Cube for HDR10, just is using it as a container from the generated of 5 only patches (Black, 100% White/Red/Green/Blue) measurements required for the Matrix LUT and then its performed the matrix LUT to 33-Point 3D LUT conversion.

The 2019 models have additionally separate 3x3 matrix slot but its not being used by CalMAN.

Quote:
Originally Posted by janos666 View Post
I thought it's 3x3 (similar to the DV file) but I cleared all the .3dl files from CalMAN's folder, did a quick HDR10 cal (skipping the 1DLUT and going straight for the 3DLUT in Game mode) and found only 33^3 files (neutral and modified).
To the 2018 models the DV file had inside the Y number of Black/White and the xy of R,G,B + the 3x3 matrix from RGB to LMS colorspace.

To the 2019 models the DV file has inside the Y number of Black/White and the xy of R,G,B.

You feed that info the Dolby's CMU chip (inside LG) which takes totally control of the TV.
Rolls-Royce, Make73 and chros73 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 #3621 of 3635 Old 11-08-2019, 08:37 AM
Advanced Member
 
chros73's Avatar
 
Join Date: Jan 2015
Posts: 599
Mentioned: 12 Post(s)
Tagged: 0 Thread(s)
Quoted: 377 Post(s)
Liked: 174
Quote:
Originally Posted by ConnecTEDDD View Post
Its been tested a Panasonic even with 2 fans to a 'syberia-like' air- conditioned room, even using black frame of 9 or 14 seconds before displaying each pattern, with Panasonic EZ1000 ... and profile took 48 hours to compete 21-Point Cube measurements with K-10A, but it was impossible to generate any valid 3D LUT correction due to that large drifting and due to WRGB problem in HDR.
You guys are just crazy! Thanks for your work!

Ryzen 5 2600,Asus Prime b450-Plus,16GB,MSI GTX 1060 Gaming X 6GB(v385.28),Win10 LTSB 1607,MPC-BEx64+LAV+MadVR,Yamaha RX-A870,LG OLED65B8(04.10.25+PC4:4:[email protected]/24/25/29/30/50/59/60Hz)
chros73 is online now  
post #3622 of 3635 Old 11-08-2019, 09:21 AM
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 3,427
Mentioned: 56 Post(s)
Tagged: 0 Thread(s)
Quoted: 2754 Post(s)
Liked: 1896
Quote:
Originally Posted by ConnecTEDDD View Post
Its been tested a Panasonic even with 2 fans to a 'syberia-like' air- conditioned room, even using black frame of 9 or 14 seconds before displaying each pattern, with Panasonic EZ1000 where you can upload an 1D LUT which will disable any active tone mapping, and profile took 48 hours to compete 21-Point Cube measurements with K-10A, but it was impossible to generate any valid 3D LUT correction due to that large drifting and due to WRGB problem in HDR.


This is drift plot with normal delay 1 sec (+-200 nits deviations)
.
Were those drift charts with the original ansio sequence? It looks like they were judging by the strong pattern in the drift charts.
Has anyone since tried with either the SOFS7 sequences I did, or the revised LS sequence?
bobof is online now  
post #3623 of 3635 Old 11-08-2019, 09:23 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,688
Mentioned: 214 Post(s)
Tagged: 1 Thread(s)
Quoted: 3428 Post(s)
Liked: 4248
Quote:
Originally Posted by bobof View Post
Were those drift charts with the original ansio sequence? It looks like they were judging by the strong pattern in the drift charts.
Has anyone since tried with either the SOFS7 sequences I did, or the revised LS sequence?
They were from testing from ~2017/2018 period, so normal anisometric was used.

For HDR, whatever patch sequence you will use, it will not change the problem with stability, something is happening during movie playback also, as ABL is highly active also with HDR content.
bobof 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 #3624 of 3635 Old 11-08-2019, 12:18 PM
AVS Forum Special Member
 
mrtickleuk's Avatar
 
Join Date: May 2016
Location: Birmingham, UK - you know, the original one!
Posts: 7,371
Mentioned: 182 Post(s)
Tagged: 0 Thread(s)
Quoted: 5097 Post(s)
Liked: 7431
Quote:
Originally Posted by janos666 View Post
Ah! That explains it! (The grey text is now useless.)

That's the first officially released firmware with the "Macro Block Noise" fix (according to LG's change-log). I am pretty sure my calibrated picture modes got all messed up by the first fix. And I remember not being alone with that.
The picture modes calibrated with 05.10.20 show virtually zero "Macro Block Noise". The amount of dithering seem to be comparable between default-Cinema and AutoCAL-ISFDark. (Although the different tone responses make this comparison very hard. I am not looking at the exact same shades.)

Hmm... So...
May be this flag was only triggered when somebody calibrated a picture mode with >=04.10.31 but already calibrated picture modes were not flagged during the update to 04.10.31
And may be the firmware is smart enough to use the new dithering method if a picture mode gets calibrated with the current firmware. (By the way, I usually do a full factory reset between firmware updates and subsequent [re-]calibrations. May be that's another factor.)
Useless maybe, useful for people like me to keep for reference!

If I understand correctly, a summary would be something like:
  • firmware 4.10.31 (and later) behaves differently when a 1DLUT (either from an autocal session or a "unity 1DLUT") is uploaded to the TV. It adds a special flag.
  • firmware 4.10.31 (and later) therefore knows whether the 1DLUT was uploaded on firmware 4.10.31 (and later) or not, so it is able to behave in one of two ways:
    1. no flag; 1DLUT created on an older firmware => behave like the older firmware did, and use the white sub-pixel near black. Risk of the overshoot flashing, but this is required to match exactly the TV as it behaved on the day of calibration.
    2. flag; 1DLUT created when TV had firmware 4.10.31 (or later) => use the new dithering routines near black, as on the day of calibration.
jrref 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 offline  
post #3625 of 3635 Old 11-08-2019, 04:29 PM
Advanced Member
 
janos666's Avatar
 
Join Date: Jan 2010
Location: Hungary
Posts: 760
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 108 Post(s)
Liked: 64
Quote:
Originally Posted by ConnecTEDDD View Post
Its has 33-Point Cube for HDR10, just is using it as a container from the generated of 5 only patches (Black, 100% White/Red/Green/Blue) measurements required for the Matrix LUT and then its performed the matrix LUT to 33-Point 3D LUT conversion.
Ah. So, the upload is probably disabled in CalMAN because the Ligting LUT (as well as anything else but matrix) method is not ready to handle the gamut shaped by the unrestricted W. But it should throw an error instead of telling me it's done (or better yet, refuse to even begin the process). At first I thought nothing was uploaded because the TV needs an actual matrix and "Lighting LUT" didn't yield any.
However, "Module HDR" can be disabled in the Service Menu (for testing purposes, at least) and there is a user menu switch for this on the 2019 models.


This means we could try creating HDR10 .3dl with DisplayCAL (if the SDR and HDR10 3DLUT formats are the same then). But AlrgyllCMS probably needs the "W boost" disabled as well (or the initial ICM profile needs to be a matrix profile from ~100 nit RGB patches instead of 100% ones but may be that's not enough).
I don't wish to try using big patch sets for HDR, just something like 65 gray steps + RGB. I am simply curious if this might yields a little smoother end result.

Well, Yeah, I know, LightSpace can create a 1DLUT from static measurements. But I don't have a license for that software.
mrtickleuk likes this.

"DIY certified hobby-calibrator" (based on ChadB's "warning signs" list

Last edited by janos666; 11-08-2019 at 04:33 PM.
janos666 is offline  
post #3626 of 3635 Old 11-10-2019, 08:52 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 8,688
Mentioned: 214 Post(s)
Tagged: 1 Thread(s)
Quoted: 3428 Post(s)
Liked: 4248
Quote:
Originally Posted by janos666 View Post
Ah. So, the upload is probably disabled in CalMAN because the Ligting LUT (as well as anything else but matrix) method is not ready to handle the gamut shaped by the unrestricted W. But it should throw an error instead of telling me it's done (or better yet, refuse to even begin the process). At first I thought nothing was uploaded because the TV needs an actual matrix and "Lighting LUT" didn't yield any.
I have asked from Tyler to disable the options which are not working long time ago, to prevent users from spending their time doing incorrect stuff.

Even professional calibrator have done that mistake (using Lightning LUT for HDR10) and see the results: https://www.avsforum.com/forum/139-d...l#post56800464

Quote:
Originally Posted by janos666 View Post
However, "Module HDR" can be disabled in the Service Menu (for testing purposes, at least) and there is a user menu switch for this on the 2019 models.
The user menu of 2019 model is for enabling the maximum W-subpixel boost so to be able to get peak output levels similar with the ones you get at HDR10 mode but in SDR mode.

About how this control affect HDR, I have posted there.

The problem is that when you will feed the panel with SDR signal it will activate the SDR 1D/3D LUT tables, its not such easy the procedure you want to perform as I don't believe the poor internal processing of 1D LUT of LG will be able to provide a PQ-EOTF tracking without having any visible problems.

(in HDR10 1D LUT the correction is generated is gamma tracking based, not PQ)

[quote=janos666;58794450]This means we could try creating HDR10 .3dl with DisplayCAL (if the SDR and HDR10 3DLUT formats are the same then). But AlrgyllCMS probably needs the "W boost" disabled as well (or the initial ICM profile needs to be a matrix profile from ~100 nit RGB patches instead of 100% ones but may be that's not enough).
I don't wish to try using big patch sets for HDR, just something like 65 gray steps + RGB. I am simply curious if this might yields a little smoother end result.

So you will have to create an PQ 1D LUT you will load at SDR slot and then your player to send HDR10 signal to your LG (but the player has to be tricked that you have connected HDR10 capable display to able to output send full HDR10 signal, some players have that capability or you can use HD Fury) while the TV have to not receive any HDR10 metadata, to say at SDR mode.

Generally I don't think it will work, I have tested such stuff at 2016 with eeColor and HD Fury Linker and it didn't worked, using Linker where it was downscaling the signal to 1080p to pass eeColor etc., trying to create a 3D LUT for HDR while the TV was working in SDR mode with 350 nits peak output.

Quote:
Originally Posted by janos666 View Post
Well, Yeah, I know, LightSpace can create a 1DLUT from static measurements. But I don't have a license for that software.
LightSpace has tools from about 2015 to generate PQ 1D LUT's with any kind of clipping or by having control of the roll-off point/tone mapping, as Sony wanted some solutions for their X300 OLED, so LightSpace added such capabilities.


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 #3627 of 3635 Old 11-10-2019, 11:02 AM
Advanced Member
 
janos666's Avatar
 
Join Date: Jan 2010
Location: Hungary
Posts: 760
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 108 Post(s)
Liked: 64
Quote:
Originally Posted by ConnecTEDDD View Post
So you will have to create an PQ 1D LUT you will load at SDR slot and then your player to send HDR10 signal to your LG (but the player has to be tricked that you have connected HDR10 capable display to able to output send full HDR10 signal, some players have that capability or you can use HD Fury) while the TV have to not receive any HDR10 metadata, to say at SDR mode.
My imagined test case was resetting the 1DLUT and uploading a 3DLUT to the HDR10 slot which targets Rec2020, D65 and gamma 2.2 while LG's HDR mapping is kept enabled. So, basically I wanted to try moving the panel calibration (the white balance and gamma response) from the 1DLUT to the 3DLUT (just to see if the hardware handles the modified 3DLUT better than the modified 1DLUT and/or if the LUT itself is smoother when it's constructed from static measurements instead iterative calibration --- because I can't generate a 1DLUT from static data with the tools at my disposal).
So, in steps: Send an HDR signal, reset an HDR mode and put it into calibration/bypass mode, take some measurements (50-100 patches or so) with DisplayCAL (matrix+shaper profiling), create a 3DLUT with targets which fit between the native panel response and LG's tone-mapping functions, upload the 3DLUT, disable calibration mode. But do this either with the "W boost" disabled (and left disabled for good) or with a carefully chosen patch set (where R,G,B are below their respective peak brightness levels, so they aren't individually clipped at their arbitrary peaks but fit the same % level).

"DIY certified hobby-calibrator" (based on ChadB's "warning signs" list

Last edited by janos666; 11-10-2019 at 11:11 AM.
janos666 is offline  
post #3628 of 3635 Old 11-13-2019, 03:32 PM
AVS Forum Special Member
 
mrtickleuk's Avatar
 
Join Date: May 2016
Location: Birmingham, UK - you know, the original one!
Posts: 7,371
Mentioned: 182 Post(s)
Tagged: 0 Thread(s)
Quoted: 5097 Post(s)
Liked: 7431
Quote:
Originally Posted by mrtickleuk View Post
Ok, understood and thanks again. So I'll have the unavoidable known issue of "the panel behaves different when Enable Calibration is ticked", but that's ok.
Updated list of steps, with slight corrections. Discussion around end-September for a method to avoid using a user 1DLUT and keep the LG 1DLUT, performing a calibration using minimal processing (ie not the normal menu's 2pt correction or gamma correction).

Preamble
The best patch generation option is to use RGB 16-255 patches for the 3D LUT profiling.
Using RGB 16-255 patches will provide better results than
  • from using YCbCr (16-235)
  • or RGB (16-235).

Only relevant with the use-1DLUT process: there is a drop in peak brightness after using the 1DLUT process, of ~15%.
  • If using 16-235 patches, set OLED Light ~15% higher than target.
  • If using 16-255 patches, when measuring 235 white set OLED Light ~23% higher than target.

Relevant with the skip-1DLUT process: there is a drop in peak brightness, of ~5%-10%.
Skipping 1DLUT, if I want to use the 3DLUT with 16-255 patches:
  • The panel will be set to 2.2 Gamma mode so I cannot guess where 235 is going to land after calibration
  • I have to measure 255, and trace back from that knowing it will be the 109% white value.
  • If I wanted a peak (235 white) of 100 nits, 255=109.13% white = 123.34nits (at the end). So I add my 5-10% buffer to this, so I try to start with a patch of 255 reading 129.50-135.67 nits.
  • If I wanted a peak (235 white) of 120 nits, 255=109.13% white = 148.00nits (at the end). So I add my 5-10% buffer to this, so I try to start with a patch of 255 reading 155.40-162.80 nits.
-------------

Process overview
  1. In Calman, set the target Gamma to [desired target Gamma]. NB: Because we will be leaving the panel set at Gamma=2.2, this means that if the target Gamma is not 2.2, then any greyscale series that is read with the TV set to Gamma=2.2 will look wrong on the charts. Initially, using the TV's menus, set menu Gamma to [desired target Gamma].
  2. [New position in the order]: Take the "pre profiling" measurements in Calman [Gamma tracking, black level and peak output]. Optionally, take colour checker, saturation/luminance sweeps etc, and saturation sweeps at luminances other than 75%.
  3. Using the TV's menus, set menu Gamma to 2.2. This is the selection to which LG's processing will not add any gamma offset. Never touch it again. Nothing to do with actual target Gamma.
  4. Connect to the TV in Calman, tick "Enable Calibration"
    • Reset 3DLUT only (DDC window: 3DLUT controls: "Reset 3D Lut" button). This uploads a Unity 3DLUT to the TV. Expert Controls/Colour Gamut is now locked out until you do a full Factory Reset of the TV.
  5. Untick "Enable Calibration"
  6. Now we are ready to use Test Patterns to determine correct Brightness "B1" and Contrast "C1" for Gamma=2.2 state, leaving headroom to allow for the 235-254 values to be displayed (for all of CMYRGBW) without clipping. Use the available clipping test etc patterns for Contrast and PLUGE etc patterns for Brightness, as required. Note the values of B1/C1.
  7. Set Brightness B1 / Contrast C1 [the "panel correct" settings]
  8. SM White Balance gain (only) adjust, to fix RGB Balance [displaying 100% (White 235) 10% window]
  9. During this process adjust OLED light to be (Target nits + 10% buffer, see above about buffer)
  10. Re-check the correct Brightness "B1" and Contrast "C1" for Gamma=2.2 state, now that the White Balance is changed. Adjust if required.
  11. Old position in the order: Take the "pre profiling" measurements in Calman [Gamma tracking, black level and peak output]. Optionally, take colour checker, saturation/luminance sweeps etc; saturation sweeps at stimulus levels other than 75%.
  12. Tick "Enable Calibration"
    • Press the Autocal button on the 3DLUT page
    • Select the options, definitely choose levels 16-255, select file to save, and create 3DLUT (profiling then 3DLUT generation) which is automatically uploaded in Calman
    • The profile file can be re-used again later in a different Picture mode (eg to select a different Target Gamma)
  13. Untick "Enable Calibration", disconnect from TV in Calman
  14. Take "post profiling" measurements in Calman. Optionally, take colour checker, saturation/luminance sweeps etc; saturation sweeps at stimulus levels other than 75%.

Graphical summary of LUT process layers in >this post<.
SamuriHL 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; 11-13-2019 at 03:36 PM.
mrtickleuk is offline  
post #3629 of 3635 Old 11-13-2019, 04:04 PM
AVS Forum Special Member
 
SamuriHL's Avatar
 
Join Date: Jan 2005
Posts: 5,067
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 172 Post(s)
Liked: 233
I've got a LOT of reading to do. I'll be doing my first LG C8 calibration this week. It seems like there's a lot of nuance to it so I'll probably screw it up the first couple of times. Which is ok as it'll be a learning experience. Thanks for the awesome knowledge in this thread. I look forward to reading quite a bit of it to figure out what I'm doing.

LG C8 OLED | Pioneer VSX-LX303 | HTPC (9900k, RTX 2070) | nVidia SHIELD 2019 | Panasonic UB820
SamuriHL is online now  
post #3630 of 3635 Old 11-15-2019, 12:00 AM
Advanced Member
 
Anger.miki's Avatar
 
Join Date: Jan 2018
Posts: 686
Mentioned: 21 Post(s)
Tagged: 0 Thread(s)
Quoted: 413 Post(s)
Liked: 434
Quote:
Originally Posted by mrtickleuk View Post
Updated list of steps, with slight corrections. Discussion around end-September for a method to avoid using a user 1DLUT and keep the LG 1DLUT, performing a calibration using minimal processing (ie not the normal menu's 2pt correction or gamma correction).

Preamble
The best patch generation option is to use RGB 16-255 patches for the 3D LUT profiling.
Using RGB 16-255 patches will provide better results than
  • from using YCbCr (16-235)
  • or RGB (16-235).

Only relevant with the use-1DLUT process: there is a drop in peak brightness after using the 1DLUT process, of ~15%.
  • If using 16-235 patches, set OLED Light ~15% higher than target.
  • If using 16-255 patches, when measuring 235 white set OLED Light ~23% higher than target.

Relevant with the skip-1DLUT process: there is a drop in peak brightness, of ~5%-10%.
Skipping 1DLUT, if I want to use the 3DLUT with 16-255 patches:
  • The panel will be set to 2.2 Gamma mode so I cannot guess where 235 is going to land after calibration
  • I have to measure 255, and trace back from that knowing it will be the 109% white value.
  • If I wanted a peak (235 white) of 100 nits, 255=109.13% white = 123.34nits (at the end). So I add my 5-10% buffer to this, so I try to start with a patch of 255 reading 129.50-135.67 nits.
  • If I wanted a peak (235 white) of 120 nits, 255=109.13% white = 148.00nits (at the end). So I add my 5-10% buffer to this, so I try to start with a patch of 255 reading 155.40-162.80 nits.
-------------

Process overview
  1. In Calman, set the target Gamma to [desired target Gamma]. NB: Because we will be leaving the panel set at Gamma=2.2, this means that if the target Gamma is not 2.2, then any greyscale series that is read with the TV set to Gamma=2.2 will look wrong on the charts. Initially, using the TV's menus, set menu Gamma to [desired target Gamma].
  2. [New position in the order]: Take the "pre profiling" measurements in Calman [Gamma tracking, black level and peak output]. Optionally, take colour checker, saturation/luminance sweeps etc, and saturation sweeps at luminances other than 75%.
  3. Using the TV's menus, set menu Gamma to 2.2. This is the selection to which LG's processing will not add any gamma offset. Never touch it again. Nothing to do with actual target Gamma.
  4. Connect to the TV in Calman, tick "Enable Calibration"
    • Reset 3DLUT only (DDC window: 3DLUT controls: "Reset 3D Lut" button). This uploads a Unity 3DLUT to the TV. Expert Controls/Colour Gamut is now locked out until you do a full Factory Reset of the TV.
  5. Untick "Enable Calibration"
  6. Now we are ready to use Test Patterns to determine correct Brightness "B1" and Contrast "C1" for Gamma=2.2 state, leaving headroom to allow for the 235-254 values to be displayed (for all of CMYRGBW) without clipping. Use the available clipping test etc patterns for Contrast and PLUGE etc patterns for Brightness, as required. Note the values of B1/C1.
  7. Set Brightness B1 / Contrast C1 [the "panel correct" settings]
  8. SM White Balance gain (only) adjust, to fix RGB Balance [displaying 100% (White 235) 10% window]
  9. During this process adjust OLED light to be (Target nits + 10% buffer, see above about buffer)
  10. Re-check the correct Brightness "B1" and Contrast "C1" for Gamma=2.2 state, now that the White Balance is changed. Adjust if required.
  11. Old position in the order: Take the "pre profiling" measurements in Calman [Gamma tracking, black level and peak output]. Optionally, take colour checker, saturation/luminance sweeps etc; saturation sweeps at stimulus levels other than 75%.
  12. Tick "Enable Calibration"
    • Press the Autocal button on the 3DLUT page
    • Select the options, definitely choose levels 16-255, select file to save, and create 3DLUT (profiling then 3DLUT generation) which is automatically uploaded in Calman
    • The profile file can be re-used again later in a different Picture mode (eg to select a different Target Gamma)
  13. Untick "Enable Calibration", disconnect from TV in Calman
  14. Take "post profiling" measurements in Calman. Optionally, take colour checker, saturation/luminance sweeps etc; saturation sweeps at stimulus levels other than 75%.

Graphical summary of LUT process layers in >this post<.
Just wondering if you’ve found out all that by yourself or you should thank someone...

TVs: Pioneer PDP-LX5090H, LG OLED55C8PLA | SintoAmp: Pioneer VSX-921 | BD Player: Panasonic DMP-BDT260EG | External LUT box: Entertainment Experience eeColor | Softwares: Light Illusion Lightspace HTP, Portrait Displays CalMAN Home Enthusiast 2018 R3, HCFR, DisplayCAL | Probes: x-rite i1 Pro 2 - i1 Display Pro OEM B-02, basICColor DISCUS | Test Pattern Generator: DVDO AVLab TPG
Anger.miki is offline  
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