AVS Forum banner

2301 - 2320 of 2655 Posts

·
Registered
Joined
·
2,200 Posts
It's in the post-LS-cal color checker Fabio ran in CalMAN and pictured in his post.
Yes, I can see that - but 'jrref' is suggesting he's seen it too.
I am asking where he has seen it, as he is not a LightSpace user.
Would be interesting to know, as this is not a LightSpace issue, so I guess it must be a limitation of the display?
For display owners, that would be good to know.

Steve
 

·
Registered
Joined
·
8,905 Posts
What I find interesting is even with this new method you still have that elevated brown to deltaE of around 2 in the CMS scan. I've noticed this and elevated orange on both the Sony and the LGs.
We are using that specific thread to talk about 3D LUT results using LightSpace.

The stuff you have noticed are not related with results you have seen from LightSpace.

So next time be more specific about how you saw that 'elevated' errors in colors, for not confusing people who don't know the fact that you haven't used ever LightSpace for LUT generation.

The color patch that BlackJoker has measured and has error is a dark brown with target Y about 2.5 nits, to the picture below which is the RGB cube space presentation, it's displaying the 21-Point Grayscale, the 3 primary colors (with 100% Saturation and 100% Luminance) and the brown dot near to 15-20% Gray, that one was the measured color.



I'm posting this without having any info from BlackJoker or his complete verification run result file to check.
 

·
Registered
Joined
·
4,903 Posts
Yes, I can see that - but 'jrref' is suggesting he's seen it too.
I am asking where he has seen it, as he is not a LightSpace user.
Would be interesting to know, as this is not a LightSpace issue, so I guess it must be a limitation of the display?
For display owners, that would be good to know.

Steve
John can speak for himself, but I know he uses CalMAN for his professional calibrations, so that has to be where he has seen it.
 

·
Registered
Joined
·
2,200 Posts
John can speak for himself, but I know he uses CalMAN for his professional calibrations, so that has to be where he has seen it.
That's what I'm interested in.
Would be interesting to know, as we can than flag this as a display issue.
I will be assessing the data from Blackjoker shortly.

It's likely is an artefact from the volumetric non-linearity of the WOLED.

Steve
 

·
Registered
Joined
·
6,590 Posts
John can speak for himself, but I know he uses CalMAN for his professional calibrations, so that has to be where he has seen it.
Please don't misunderstand my post, sorry for the confusion. I'm not a LS user but may be in the future which is why I like reading this thread. I was referring to BlackJoker's Calman scan, in that I see similar elevated colors when using Calman on the OLEDs. I was just wanted to point this out since it would be interesting to see what might be causing it and where it is to determine how important the error actually is.
 
  • Like
Reactions: Rolls-Royce

·
Registered
Joined
·
8,905 Posts
John can speak for himself, but I know he uses CalMAN for his professional calibrations, so that has to be where he has seen it.


John is talking for LG 2018 results he has seen using CalMAN and 3D LUT, the problem of Yellow (it has the same brown problem like BlackJoker's) as I have explained to him there: https://www.avsforum.com/forum/139-display-calibration/2962814-2018-lg-oled-calibration-user-settings-no-price-talk-44.html#post56558088

More info about these calibrations are available there: https://www.avsforum.com/forum/139-display-calibration/2962814-2018-lg-oled-calibration-user-settings-no-price-talk-47.html#post56575484

I don't think BlackJoker's cLUT cube shape has so many issues like John's results. If BlackJoker will post a cube viewer picture of LightSpace we can confirm that is 'clean'.

About the Sony, John has to let us know more details.

All Sony OLED's are problematic to many areas, especially to lower end, since they don't map correctly the REC.709 color transformation from native panel gamut, so lower end will be over-saturated to all colors, so any measurement at these areas of luminance will return higher dE errors.



That problem affects all Sony OLED's, from A1 until AF9, but I haven't measured 2019 models.

For more details about Sony problem:

https://www.avsforum.com/forum/139-display-calibration/3001138-sony-master-series-a9f-calibration-settings-5.html?fbclid=IwAR0e19uoDgWIOFzOu2RjtJLCcJVxaK8b-TpWycjvWaR6EThq0cPYsXm0L80#post57100374

https://www.avsforum.com/forum/139-display-calibration/3001138-sony-master-series-a9f-calibration-settings-5.html?fbclid=IwAR3HilS88Ykeqf1ByKk01Q3_zqsLhnKVrUy3B_TDXvL_dPYnISZ2sBHbCwU#post57103058

...and its voted as best TV the Sony AF9 lol.
 

·
Registered
Joined
·
8,905 Posts
LightSpace CMS 10.0.0.2969 (26 July 2019) has been released.

Release Notes

Added Hybrid LUT Generation mode to Convert Colour Space menu

Add access to the Hint capability for older licenses that match XPT, PRO, HTP

Release Info

The latest LightSpace release includes two new additions that greatly enhance LUT generation.

Hint (for LightSpace XPT, PRO, and HTP users)

The first is a new Hint mode, which is a very powerful process that enhances calibration accuracy attainable with smaller patch sets.

Hint can also be used as a multi-point probe matching process, by inverting the normal application (using a large Tristimulus profile as the Hint, with a smaller Spectro profile as the main profile).

The concept behind Hint is to enable the use of a complex full volumetric profile as a 'guide' or 'master' for a smaller profile set. For example, a Primary Only, or Grey Only Quick Profile can have a display's in-built non-linear volumetric distortions mapped into the data set prior to LUT generation. This means that sparse profiles (Quick Profiles for example) can generate more accurate LUTs, based on the more complex profile data being used as a Hint.

Ideally, the Hint profile would be generated from the display in question, but can also be used for displays within the same family that have exactly the same underlying non-linear response.

Hybrid

Additionally, a new Hybrid mode has been added to the LUT Generation options.

Hybrid mode is based on Peak Chroma, but does two things that can potentially help with the accuracy of the LUT's grey scale and low-light performance, by isolating those areas of the profile, and processing them independently from the rest of the profile's volumetric data.

This maximizes the accuracy of the grey scale, and helps remove low-light errors introduced by inaccurate probe readings.

Download Link: https://www.lightillusion.com/downloads.html

Note: All LightSpace related guides which are available online has been updated to reflect changes the latest released version of LightSpace introduced.
 

·
Registered
Joined
·
4,903 Posts
LightSpace CMS 10.0.0.2969 (26 July 2019) has been released.

Release Notes

Added Hybrid LUT Generation mode to Convert Colour Space menu

Add access to the Hint capability for older licenses that match XPT, PRO, HTP

Release Info

The latest LightSpace release includes two new additions that greatly enhance LUT generation.

Hint (for LightSpace XPT, PRO, and HTP users)

The first is a new Hint mode, which is a very powerful process that enhances calibration accuracy attainable with smaller patch sets.

Hint can also be used as a multi-point probe matching process, by inverting the normal application (using a large Tristimulus profile as the Hint, with a smaller Spectro profile as the main profile).

The concept behind Hint is to enable the use of a complex full volumetric profile as a 'guide' or 'master' for a smaller profile set. For example, a Primary Only, or Grey Only Quick Profile can have a display's in-built non-linear volumetric distortions mapped into the data set prior to LUT generation. This means that sparse profiles (Quick Profiles for example) can generate more accurate LUTs, based on the more complex profile data being used as a Hint.

Ideally, the Hint profile would be generated from the display in question, but can also be used for displays within the same family that have exactly the same underlying non-linear response.

Hybrid

Additionally, a new Hybrid mode has been added to the LUT Generation options.

Hybrid mode is based on Peak Chroma, but does two things that can potentially help with the accuracy of the LUT's grey scale and low-light performance, by isolating those areas of the profile, and processing them independently from the rest of the profile's volumetric data.

This maximizes the accuracy of the grey scale, and helps remove low-light errors introduced by inaccurate probe readings.

Download Link: https://www.lightillusion.com/downloads.html

Note: All LightSpace related guides which are available online has been updated to reflect changes the latest released version of LightSpace introduced.
Are there any differences between this and the 10.0.0.2969 beta I installed yesterday? In other words, do I need to install this one?
 

·
Registered
Joined
·
8,905 Posts
Are there any differences between this and the 10.0.0.2969 beta I installed yesterday? In other words, do I need to install this one?
No need to install the public version since its exact the same as the beta you had.
 

·
Registered
Joined
·
6,590 Posts


John is talking for LG 2018 results he has seen using CalMAN and 3D LUT, the problem of Yellow (it has the same brown problem like BlackJoker's) as I have explained to him there: https://www.avsforum.com/forum/139-display-calibration/2962814-2018-lg-oled-calibration-user-settings-no-price-talk-44.html#post56558088

More info about these calibrations are available there: https://www.avsforum.com/forum/139-display-calibration/2962814-2018-lg-oled-calibration-user-settings-no-price-talk-47.html#post56575484

I don't think BlackJoker's cLUT cube shape has so many issues like John's results. If BlackJoker will post a cube viewer picture of LightSpace we can confirm that is 'clean'.

About the Sony, John has to let us know more details.

All Sony OLED's are problematic to many areas, especially to lower end, since they don't map correctly the REC.709 color transformation from native panel gamut, so lower end will be over-saturated to all colors, so any measurement at these areas of luminance will return higher dE errors.



That problem affects all Sony OLED's, from A1 until AF9, but I haven't measured 2019 models.

For more details about Sony problem:

https://www.avsforum.com/forum/139-display-calibration/3001138-sony-master-series-a9f-calibration-settings-5.html?fbclid=IwAR0e19uoDgWIOFzOu2RjtJLCcJVxaK8b-TpWycjvWaR6EThq0cPYsXm0L80#post57100374

https://www.avsforum.com/forum/139-display-calibration/3001138-sony-master-series-a9f-calibration-settings-5.html?fbclid=IwAR3HilS88Ykeqf1ByKk01Q3_zqsLhnKVrUy3B_TDXvL_dPYnISZ2sBHbCwU#post57103058

...and its voted as best TV the Sony AF9 lol.
Ted, thanks for the summary. Just to clarify, i did see the elevated brown and orange/yellow with Calman on the 2018 LG OLEDs and still see it with Calman on both the 2019 Sony's and LG OLEDs. I know the Calman 3D LUT could be causing the problem but even just using the user controls to eliminate the LUT process, also you only have user controls on the Sony, i still see these errors which may be a display problem as Steve says. As you know i see a lot of the same sets vs most calibrators so i was just trying to share given all of the advancements in LS.
 

·
Registered
Joined
·
8,905 Posts
Yes Master, but if you read more accurately my post, I suggested to leave a unity LUT as 1DLUT.
Performing 1D LUT first, and then 3D LUT, it can provide better image at the end, since the adjustments from 3D LUT will be less, so the video signal will less be manipulated.

But there specific panels (or specific installed firmware versions) where you have to ignore the 1D LUT and perform only 3D LUT, panels with problem as described below:

Near Black Issues from poor factory Quality Control of LG 2018 OLED's

Some LG OLED's have passed the factory QC (Quality Control) with near black clipping issues and shipped to stores, for these panels it will be required to adjust the Brightness to 55-56 or up to 60-61 (in worst case scenario) from the default 50 value for the near black clipping issue to be resolved.



Also it will not work any service menu sub-adjustment related to Brightness when you will use internal 1D LUT capability.

By adjusting the sub-brightness in the service menu it will cause the 1D LUT points to be misaligned at near black, set Sub-brightness to default 128 value.

When you have such issues with near black clipping and you see that 1D LUT profiling can't resolve completely the problem (with Brightness @ 50 setting as required for internal 1D LUT to work) then you can skip using the internal 1D LUT completely and resolve the problem by adjusting Brightness control to whatever values you see that is not providing any clipping (while it will not raise your native panel black level) and then start the 3D LUT profiling measurements.

If you will skip performing 1D LUT profiling, select Warm 2 color temperature, Gamma 2.2 preset, Wide Color Space and the adjust Brightness/Contrast to the value you see that is correct for your TV. Perform also only 100% White (or 109% White) pre-calibration using RGB-High Controls only (keep one color channel at zero and reduce values from other 2 channels) while adjusting OLED Light you will control the luminance output (nits). Don't use RGB balance controls to adjust the Luminance output levels. After these pre-calibration adjustments you can proceed to 3D LUT profiling.

For expert users, its recommended instead of adjusting normal TV menu White Balance settings, to adjust only the White Balance values of color temperature from inside the TV's Service Menu only. The reason for that is because the normal menu White Balance settings combined with White Balance of Service Menu settings to work as offset from the native panel, and when you have active two controls which are performing the same job, this can make the internal processing algorithm more complex, so its better idea one White Balance menu to be used only to adjust from native panel response and not from both menus (normal TV + service) the same time.

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

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

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

LG's factory pre-calibrate these service menu White Balance color temperature modes using a 80% Gray pattern with 6500K for 'Warm', 9300K for 'Medium' and 11000K for 'Cool'.



Use the EZ-ADJ button of Service Remote to access the LG OLED TV Service Menu, it will ask for a password, type '0413', navigate to '12. White Balance', and select the color temperature you want to adjust.

If you want to adjust for example the 'Medium' of Service Menu, start by setting the default 192 value to R-Gain, G-Gain and B-Gain and 64 value to R-Cut, G-Cut and B-Cut controls.



When R/G/B gain in the OSD is at 192 value, it means that the panel works at its Full Dynamic Range. In order to prevent saturation of Full Dynamic range and data, one of R/G/B channels need to be fixed at 192 value, for pre-calibrating the White Balance, lower only the other two channels.

Be careful with IN-STOP button, don't press it, because it will perform a complete factory reset, it will reset the TV's UTT (number of panel usage hours) counter which will affect the compensation cycle operation.
 

·
Registered
Joined
·
8,905 Posts
LG released the final firmware 5.10.03 for the 2018 models so I decided to give the hybrid mode a chance and I can say that im pretty amazed. Here is the setup I used:

LG 65 C8 with 5.10.03 FW
Klein K10-A
DVDO TPG @ 4:2:2 12bit
LightSpace 10.0.0.2969 BETA
21^3 Cube with Finseq Patchset
1080p24 YCbCr 4:2:2 12 bit as BlackJoker used for patch generation is the best patch generation setting for these TV's.

When RGB-Video patch generation will be used (16-235) or RGB-Video Extended (16-255) then there no difference if the bit-depth output of the patch generator will be 8 or 10 or 12-bits.

But when YCbCr 4:2:2 or 4:4:4 patch generation will be used, its recommended to set the bit-depth output of the patch generator to 10 or 12-bit (not 8-bit).

Calibration software will send request of 8-bit RGB triplet values with colorspace output flag also to the generator processor and from there the RGB triplet values will be converted to YCbCr colorspace, the 3x3 colorspace conversion matrix (RGB-Video -> YCbCr) calculation will introduce some rounding errors when the output will be 8-bit, so having pre-configured the output as 10 or 12-bit, this will compensate for rounding errors.

There no 1:1 mapping between RGB and YCbCr colorspaces. There YCbCr values with no corresponding RGB values and RBG values with no corresponding YCbCr values.

Only about 60% of all RGB values can be represented in YCbCr space when you will use the same amount of bits for both triplets.

This means the most damage happens in RGB -> YCbCr when you take a 3 * 8-bit RGB triplet, convert and round it back to 3* 8-bits of precision.

For that reason is recommended to set YCbCr colorspace bit-depth output of the pattern generator to 10 or 12-bit.
 

·
Registered
Joined
·
8,905 Posts
LG 65 C8 with 5.10.03 FW
Klein K10-A
DVDO TPG @ 4:2:2 12bit
LightSpace 10.0.0.2969 BETA

21^3 Cube with Finseq Patchset

...here is the result of the new hybrid mode nearly perfect 6500k and the almost perfect RGB balance the colorcheker performance improves also. This is a huge step for more accurate grayscale performance well done!
and here is BlackJoker's 3D LUT generated file as it looks at evaluation using the 3D Cube Viewer of LightSpace:






 

·
Registered
Joined
·
122 Posts
Hi fellas,

After weeks of testing here are my findings:

LG OLED65C8PLA (5.10.03 FW)
i1D3 profiled against an i1PRO 2
madTPG @ 16-235 8bit
LightSpace 10.0.0.2969
17^3 Cube with Finseq Patchset

The 1st normal approach:
1D-LUT (GS only Large) with augemented data followed by 3D-LUT made using Hybrid mode.





The 2nd approach:
- 2P high @100% or 109%
- Gamma 2.2 (Bypass setting)
- Reset 3D-LUT with DCC
- Peak Chroma colorspace conversion (You might try hybrid mode with this approach but for me peak chroma give better results)
- Apply videoscale
- Upload the 3D-LUT to the TV
- Make another Gray large QP for augemented data and add it the the first 3D-LUT
- Peak Chroma colorspace conversion (Or use hybrid if it works good for you)
- Apply videoscale
- Upload the new file








According to those charts, the 1st method look better, but visually when i skip the 1dlut i get smoother gradiation and much much more clean near black details.
I have no explanation for that but i think it's something to do with LG's internal processing.


looking forward to see more testing from members here.


cheers
 

·
Registered
Joined
·
8,905 Posts

·
Registered
Joined
·
972 Posts
Hi fellas,

After weeks of testing here are my findings:

LG OLED65C8PLA (5.10.03 FW)
i1D3 profiled against an i1PRO 2
madTPG @ 16-235 8bit
LightSpace 10.0.0.2969
17^3 Cube with Finseq Patchset

The 1st normal approach:
1D-LUT (GS only Large) with augemented data followed by 3D-LUT made using Hybrid mode.





The 2nd approach:
- 2P high @100% or 109%
- Gamma 2.2 (Bypass setting)
- Reset 3D-LUT with DCC
- Peak Chroma colorspace conversion (You might try hybrid mode with this approach but for me peak chroma give better results)
- Apply videoscale
- Upload the 3D-LUT to the TV
- Make another Gray large QP for augemented data and add it the the first 3D-LUT
- Peak Chroma colorspace conversion (Or use hybrid if it works good for you)
- Apply videoscale
- Upload the new file








According to those charts, the 1st method look better, but visually when i skip the 1dlut i get smoother gradiation and much much more clean near black details.
I have no explanation for that but i think it's something to do with LG's internal processing.


looking forward to see more testing from members here.


cheers
Probably because 3DLUT is before 1DLUT in the internal video pipeline.
 
  • Like
Reactions: sat4all

·
Registered
Joined
·
8,905 Posts
LightSpace CMS 10.0.0.2972 (06 August 2019) has been released.

Release Notes

Corrected application of Drift with Hybrid mode.


Changed default HGL Surround Luminance to 5 nits.


Corrected profile data shown on Grey Only profiles with invalid black level readings.

Download Link: https://www.lightillusion.com/downloads.html
 
2301 - 2320 of 2655 Posts
Top