LG 2017 OLED Calibration Thread and Settings - Page 9 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 3286Likes
Reply
 
Thread Tools
post #241 of 3127 Old 08-21-2017, 11:06 PM
AVS Forum Addicted Member
 
Join Date: Sep 2002
Location: Wetumpka, AL
Posts: 17,407
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 1708 Post(s)
Liked: 738
Quote:
Originally Posted by bmccoupe View Post
...snip...

I ended up with a contrast of 76.... Reading the old 2016 OLED cal thread, most people had about 85. Is my setting too low?
On the 2016s, using 85 had to do with improving bit rates. It was stated as using a minimum of 85.

Is there some sort of test where you can see if that also applies to the 2017s?
JimP is offline  
Sponsored Links
Advertisement
 
post #242 of 3127 Old 08-22-2017, 05:18 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 7,752
Mentioned: 179 Post(s)
Tagged: 1 Thread(s)
Quoted: 2921 Post(s)
Liked: 3569
Quote:
Originally Posted by PeteNeat View Post
D-Nice, thank you for the enlightenment, I also have PE2 BlueRay and the scene changes are perfect black. Not only that the black of space from the Earth shots are also perfect black.

Naturally this makes my brain fart somewhat....
Hi, Planet Earth II HDR10 static metadata infoframe has 'zeros' to all it's Mastering Display Color Volume or Content Metadata. It's a conversion of HLG -> HDR10 and its peak output doesn't exceed 1000nits.

Mastering has been performed from [email protected] (UK) which are using LightSpace for Calibration (it's at the LightSpace's customer list) using a Dolby PRM-4220 Professional Reference Monitor (Dolby talking that one of their monitor features is the capability to use LightSpace for its calibration), which is an SDR 1080p but high-brightness panel, real 12-bit panel with 100% DCI coverage, it can output about 600nits calibrated, but because it's RGB-LED it's not having the blacks of OLED's.

Dolby Monitor has 2x 65-Point Cube 3D LUT table uploading capability (same 3D LUT Cube size as eeColor 3D LUT Box we have in consumer market), so they used an HLG 3D LUT with FilmLight's Baselight color granding suite.

When you use a 600nit monitor for grading HDR content you can still grade for higher nits (for 1000 nits) if you not clip above 600 nits, leaving the highlights unclipped and evaluate using the software histograms.

For make metadata analysis, to find the Content Metadata (MaxCLL & MaxFALL) you need to have the RAW 16-TIFF master of the Planet Earth II and the calculation should be performed before encoding to UHD-BD HEVC.

Except Planet Earth II, there a lot of movies which has MaxCLL & MaxxFALL to 0 nits....(usually the 20th Century Fox releases), others are using a typical default which BDA is recommending to be 1000 MaxCLL /400 MaxFALL.

There displays, like the Panasonic EZ1000 which counting the MaxCLL number to their internal calculation, for example if you pause one HDR movie using Panasonic and send a different MaxCLL using HD Fury, you will see that the still picture will be change.

Basically consumer display don't follow specs completely or correctly this time, this is why the black level (and highlights also) performance of released UHD titles is varying, there movies which have lifted blacked, other with clipped near blacks etc.

About Black levels, in SDR, with 100nits as Reference White, the Target Luminance for Power Law Gamma 2.2 (with a display with 0 nits black, like OLED) is:

0.5% Gray (17 level) is 0.0007 nits
1% Gray (18 level) is 0.0033 nits

In SDR, with 100nits as Reference White, the Target Luminance for Power Law Gamma 2.4/BT.1886 (with a display with 0 nits black, like OLED) is:

0.5% Gray (17 level) is 0.0002 nits
1% Gray (18 level) is 0.0013 nits

In SDR we use Gamma curve as transfer function (where you can set it to 2.2/2.4/BT1886 or whatever value you like....having in calculation the display Black/Peak White level to generate each digital code luminance levels, but in HDR with PQ Curve transfer function, it has fixed numbers for each digital code luminance level (0-10.000nits) when you have a display; lets say an LCD-LED which has about 0.039 nits black level (Samsung 55KS9000 for example), all the near black information below that, from the 10-bit HDR signal will not be displayed, so it's lost.

So when you have a movie mastered with min 0.005 nits and you watch it with OLED which has 0 nits, it's up to the display internal programming what to do....

BT.2390 (which is not a published standard, it's a report) it has a minLum parameter to do a normalize counting to the display real black measurements but even if a display will follow that minLum parameter (which is baked inside to a display data) .....if BY.2390 was a spec and not just a report.../, how does a display know it's current black level measurement? ... There a lot of users talking for lifted blacks and glowing blacks even at using default setting to OLED's, also if a calibration adjustment from the user using the 127 digital point calibration setting adjustment (5% lets say) is high, this can affect the black level measurement. (of it the user have wrongly adjusted Contrast/Brightness,OLED Light to values other from default values.)



See also for example CalMAN 5 calculated 64-Point Grayscale luminance targets for ST.2084, based as example to Samsung UE55KS9000 measurement data (Black Level 0.039 nits - Peak White: 1460 nits), some near black will clip.

Adding Dynamic frame per frame features will just undo your calibration, since it will be changing the panel response, so at the end you don't see a movie how supposed to look but how LG want, preserving all details that you monitor can't display which will reduce the levels of your display...those who can correctly display and follow ST.2084 PQ, so the calibrated part of your display (0-750 nits) from the user menu adjustments it will be ignored and levels will be re-adjusted in real-time....if the mastering is from 4000 nits, the whole picture will be a lot darker compared to 1000 nit mastered movie.

This is the reason why calibrated displays in HDR mode will reproduce the most movies at average levels of color performance.

One of the solutions to this can be using HD Fury to calibrate using patterns with metadata... for example 1000nits max, 1000 MaxFALL, 400MaxCLL, and after the end of calibration enable from HD Fury to send the same metadata you used to calibrated to the display, so the display will always be receiving the same fixed metadata (to it will have fixed and calbrated response) to any movies you will watch...it will not be receiving each movie metadata.

If someone has HD Fury, he can try this and post some feedback, from watching 4000nits movies.

When we say ''4000 nits monitor used'' doesn't mean that the movie will have up to 4000 nits levels, it can have much lower or much higher, look examples below:



So when you have that titles which their max luminance levels is below the max output of an OLED, it means that it will not have to use any roll-off since all data from the movie will be below the max panel output capability, but the most dislpays will see the 4000 max only and start the roll of a lot nits earlier so the total picture will be very dimmed.


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 offline  
post #243 of 3127 Old 08-22-2017, 05:54 AM
Advanced Member
 
randal_r's Avatar
 
Join Date: Mar 2010
Posts: 522
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 232 Post(s)
Liked: 131
Quote:
Originally Posted by ConnecTEDDD View Post
Hi, Planet Earth II HDR10 static metadata infoframe has 'zeros' to all it's Mastering Display Color Volume or Content Metadata. It's a conversion of HLG -> HDR10 and its peak output doesn't exceed 1000nits.

Mastering has been performed from [email protected] (UK) which are using LightSpace for Calibration (it's at the LightSpace's customer list) using a Dolby PRM-4220 Professional Reference Monitor (Dolby talking that one of their monitor features is the capability to use LightSpace for its calibration), which is an SDR 1080p but high-brightness panel, real 12-bit panel with 100% DCI coverage, it can output about 600nits calibrated, but because it's RGB-LED it's not having the blacks of OLED's.

Dolby Monitor has 2x 65-Point Cube 3D LUT table uploading capability (same 3D LUT Cube size as eeColor 3D LUT Box we have in consumer market), so they used an HLG 3D LUT with FilmLight's Baselight color granding suite.

When you use a 600nit monitor for grading HDR content you can still grade for higher nits (for 1000 nits) if you not clip above 600 nits, leaving the highlights unclipped and evaluate using the software histograms.

For make metadata analysis, to find the Content Metadata (MaxCLL & MaxFALL) you need to have the RAW 16-TIFF master of the Planet Earth II and the calculation should be performed before encoding to UHD-BD HEVC.

Except Planet Earth II, there a lot of movies which has MaxCLL & MaxxFALL to 0 nits....(usually the 20th Century Fox releases), others are using a typical default which BDA is recommending to be 1000 MaxCLL /400 MaxFALL.

There displays, like the Panasonic EZ1000 which counting the MaxCLL number to their internal calculation, for example if you pause one HDR movie using Panasonic and send a different MaxCLL using HD Fury, you will see that the still picture will be change.

Basically consumer display don't follow specs completely or correctly this time, this is why the black level (and highlights also) performance of released UHD titles is varying, there movies which have lifted blacked, other with clipped near blacks etc.

About Black levels, in SDR, with 100nits as Reference White, the Target Luminance for Power Law Gamma 2.2 (with a display with 0 nits black, like OLED) is:

0.5% Gray (17 level) is 0.0007 nits
1% Gray (18 level) is 0.0033 nits

In SDR, with 100nits as Reference White, the Target Luminance for Power Law Gamma 2.4/BT.1886 (with a display with 0 nits black, like OLED) is:

0.5% Gray (17 level) is 0.0002 nits
1% Gray (18 level) is 0.0013 nits

In SDR we use Gamma curve as transfer function (where you can set it to 2.2/2.4/BT1886 or whatever value you like....having in calculation the display Black/Peak White level to generate each digital code luminance levels, but in HDR with PQ Curve transfer function, it has fixed numbers for each digital code luminance level (0-10.000nits) when you have a display; lets say an LCD-LED which has about 0.039 nits black level (Samsung 55KS9000 for example), all the near black information below that, from the 10-bit HDR signal will not be displayed, so it's lost.

So when you have a movie mastered with min 0.005 nits and you watch it with OLED which has 0 nits, it's up to the display internal programming what to do....

BT.2390 (which is not a published standard, it's a report) it has a minLum parameter to do a normalize counting to the display real black measurements but even if a display will follow that minLum parameter (which is baked inside to a display data) .....if BY.2390 was a spec and not just a report.../, how does a display know it's current black level measurement? ... There a lot of users talking for lifted blacks and glowing blacks even at using default setting to OLED's, also if a calibration adjustment from the user using the 127 digital point calibration setting adjustment (5% lets say) is high, this can affect the black level measurement. (of it the user have wrongly adjusted Contrast/Brightness,OLED Light to values other from default values.)



See also for example CalMAN 5 calculated 64-Point Grayscale luminance targets for ST.2084, based as example to Samsung UE55KS9000 measurement data (Black Level 0.039 nits - Peak White: 1460 nits), some near black will clip.

Adding Dynamic frame per frame features will just undo your calibration, since it will be changing the panel response, so at the end you don't see a movie how supposed to look but how LG want, preserving all details that you monitor can't display which will reduce the levels of your display...those who can correctly display and follow ST.2084 PQ, so the calibrated part of your display (0-750 nits) from the user menu adjustments it will be ignored and levels will be re-adjusted in real-time....if the mastering is from 4000 nits, the whole picture will be a lot darker compared to 1000 nit mastered movie.

This is the reason why calibrated displays in HDR mode will reproduce the most movies at average levels of color performance.

One of the solutions to this can be using HD Fury to calibrate using patterns with metadata... for example 1000nits max, 1000 MaxFALL, 400MaxCLL, and after the end of calibration enable from HD Fury to send the same metadata you used to calibrated to the display, so the display will always be receiving the same fixed metadata (to it will have fixed and calbrated response) to any movies you will watch...it will not be receiving each movie metadata.

If someone has HD Fury, he can try this and post some feedback, from watching 4000nits movies.

When we say ''4000 nits monitor used'' doesn't mean that the movie will have up to 4000 nits levels, it can have much lower or much higher, look examples below:



So when you have that titles which their max luminance levels is below the max output of an OLED, it means that it will not have to use any roll-off since all data from the movie will be below the max panel output capability, but the most dislpays will see the 4000 max only and start the roll of a lot nits earlier so the total picture will be very dimmed.

Hi Ted,

No display will ever have an absolute 0 for black, as this is reserved for the horizontal and vertical blanking process. Within this process is the only time a display will have absolute 0 . This allows for the various syncing for color, scanning position and others attributes to occur in order for the image to appear properly on the display.
randal_r is offline  
Sponsored Links
Advertisement
 
post #244 of 3127 Old 08-22-2017, 05:55 AM
AVS Forum Addicted Member
 
D-Nice's Avatar
 
Join Date: Dec 2004
Location: Columbia, SC
Posts: 16,780
Mentioned: 120 Post(s)
Tagged: 1 Thread(s)
Quoted: 1832 Post(s)
Liked: 3596
Quote:
Originally Posted by ConnecTEDDD View Post
Hi, Planet Earth II HDR10 static metadata infoframe has 'zeros' to all it's Mastering Display Color Volume or Content Metadata. It's a conversion of HLG -> HDR10 and its peak output doesn't exceed 1000nits.

Mastering has been performed from [email protected] (UK) which are using LightSpace for Calibration (it's at the LightSpace's customer list) using a Dolby PRM-4220 Professional Reference Monitor (Dolby talking that one of their monitor features is the capability to use LightSpace for its calibration), which is an SDR 1080p but high-brightness panel, real 12-bit panel with 100% DCI coverage, it can output about 600nits calibrated, but because it's RGB-LED it's not having the blacks of OLED's.

Dolby Monitor has 2x 65-Point Cube 3D LUT table uploading capability (same 3D LUT Cube size as eeColor 3D LUT Box we have in consumer market), so they used an HLG 3D LUT with FilmLight's Baselight color granding suite.

When you use a 600nit monitor for grading HDR content you can still grade for higher nits (for 1000 nits) if you not clip above 600 nits, leaving the highlights unclipped and evaluate using the software histograms.

For make metadata analysis, to find the Content Metadata (MaxCLL & MaxFALL) you need to have the RAW 16-TIFF master of the Planet Earth II and the calculation should be performed before encoding to UHD-BD HEVC.

Except Planet Earth II, there a lot of movies which has MaxCLL & MaxxFALL to 0 nits....(usually the 20th Century Fox releases), others are using a typical default which BDA is recommending to be 1000 MaxCLL /400 MaxFALL.

There displays, like the Panasonic EZ1000 which counting the MaxCLL number to their internal calculation, for example if you pause one HDR movie using Panasonic and send a different MaxCLL using HD Fury, you will see that the still picture will be change.

Basically consumer display don't follow specs completely or correctly this time, this is why the black level (and highlights also) performance of released UHD titles is varying, there movies which have lifted blacked, other with clipped near blacks etc.

About Black levels, in SDR, with 100nits as Reference White, the Target Luminance for Power Law Gamma 2.2 (with a display with 0 nits black, like OLED) is:

0.5% Gray (17 level) is 0.0007 nits
1% Gray (18 level) is 0.0033 nits

In SDR, with 100nits as Reference White, the Target Luminance for Power Law Gamma 2.4/BT.1886 (with a display with 0 nits black, like OLED) is:

0.5% Gray (17 level) is 0.0002 nits
1% Gray (18 level) is 0.0013 nits

In SDR we use Gamma curve as transfer function (where you can set it to 2.2/2.4/BT1886 or whatever value you like....having in calculation the display Black/Peak White level to generate each digital code luminance levels, but in HDR with PQ Curve transfer function, it has fixed numbers for each digital code luminance level (0-10.000nits) when you have a display; lets say an LCD-LED which has about 0.039 nits black level (Samsung 55KS9000 for example), all the near black information below that, from the 10-bit HDR signal will not be displayed, so it's lost.

So when you have a movie mastered with min 0.005 nits and you watch it with OLED which has 0 nits, it's up to the display internal programming what to do....

BT.2390 (which is not a published standard, it's a report) it has a minLum parameter to do a normalize counting to the display real black measurements but even if a display will follow that minLum parameter (which is baked inside to a display data) .....if BY.2390 was a spec and not just a report.../, how does a display know it's current black level measurement? ... There a lot of users talking for lifted blacks and glowing blacks even at using default setting to OLED's, also if a calibration adjustment from the user using the 127 digital point calibration setting adjustment (5% lets say) is high, this can affect the black level measurement. (of it the user have wrongly adjusted Contrast/Brightness,OLED Light to values other from default values.)



See also for example CalMAN 5 calculated 64-Point Grayscale luminance targets for ST.2084, based as example to Samsung UE55KS9000 measurement data (Black Level 0.039 nits - Peak White: 1460 nits), some near black will clip.

Adding Dynamic frame per frame features will just undo your calibration, since it will be changing the panel response, so at the end you don't see a movie how supposed to look but how LG want, preserving all details that you monitor can't display which will reduce the levels of your display...those who can correctly display and follow ST.2084 PQ, so the calibrated part of your display (0-750 nits) from the user menu adjustments it will be ignored and levels will be re-adjusted in real-time....if the mastering is from 4000 nits, the whole picture will be a lot darker compared to 1000 nit mastered movie.

This is the reason why calibrated displays in HDR mode will reproduce the most movies at average levels of color performance.

One of the solutions to this can be using HD Fury to calibrate using patterns with metadata... for example 1000nits max, 1000 MaxFALL, 400MaxCLL, and after the end of calibration enable from HD Fury to send the same metadata you used to calibrated to the display, so the display will always be receiving the same fixed metadata (to it will have fixed and calbrated response) to any movies you will watch...it will not be receiving each movie metadata.

If someone has HD Fury, he can try this and post some feedback, from watching 4000nits movies.

When we say ''4000 nits monitor used'' doesn't mean that the movie will have up to 4000 nits levels, it can have much lower or much higher, look examples below:



So when you have that titles which their max luminance levels is below the max output of an OLED, it means that it will not have to use any roll-off since all data from the movie will be below the max panel output capability, but the most dislpays will see the 4000 max only and start the roll of a lot nits earlier so the total picture will be very dimmed.

I would try this but my Furys are attached to my A1Es which ignore the MaxCLL & MaxxFALL.
D-Nice is offline  
post #245 of 3127 Old 08-22-2017, 06:15 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 7,752
Mentioned: 179 Post(s)
Tagged: 1 Thread(s)
Quoted: 2921 Post(s)
Liked: 3569
Quote:
Originally Posted by ERuiz View Post
Ok, so when playing a movie mastered to 1000 nits, do I need to adjust my brightness to adjust blacks back to normal?

If so, then how do I figure out to what nits was a movie mastered to?

Whatever happened to just popping in a movie and sit back to enjoy it. Seems like we have to jump through hoops from movie to movie in order to watch them correctly.

Sent from my iPhone using Tapatalk
Hi, as D-Nice said, you can use OPPO to see the mastering display peak output and black.

Or you can wait some days because I will upload a special page to my website with 220 UHD titles checked, with full metadata details.

Below there is a preview to see how it will look, there a lot of titles being checked from different markets (US/UK/DE/IT); I say region to the list; UHD don't have region codes like BD/DVD, they play everywhere. There some differences between metadata of MaxCLL and MaxFALL of the same title but from other region market.


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 offline  
post #246 of 3127 Old 08-22-2017, 06:33 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 7,752
Mentioned: 179 Post(s)
Tagged: 1 Thread(s)
Quoted: 2921 Post(s)
Liked: 3569
Quote:
Originally Posted by randal_r View Post
Hi Ted,

No display will ever have an absolute 0 for black, as this is reserved for the horizontal and vertical blanking process. Within this process is the only time a display will have absolute 0 . This allows for the various syncing for color, scanning position and others attributes to occur in order for the image to appear properly on the display.
Hi Randal,

Your post is valid only if we were talking for CRT's and analog interlaced video signal before 20 years. From signal point of view, signal waveform devices you are studding, they can only examining an input signal, you can't examine what the display is doing, this is why we use meter to measure black levels and do measurement, so for LG OLED's Klein will report 0 nits. The info you are posting is related with raster of a signal, which is not visible to the active resolution we see.

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 offline  
post #247 of 3127 Old 08-22-2017, 06:41 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 7,752
Mentioned: 179 Post(s)
Tagged: 1 Thread(s)
Quoted: 2921 Post(s)
Liked: 3569
Quote:
Originally Posted by D-Nice View Post
I would try this but my Furys are attached to my A1Es which ignore the MaxCLL & MaxxFALL.
Hi D-Nice,

Try 540 nits peak also, maybe doing this you will have better PQ tracking when you calibrating and use it as fixed metadata for all the titles you will check.

Maybe this solution is the best one for current HDR calibration status improvement, sending a fixed metadata for all movies and by disabling any active HDR additional processing.

Just need some test what metadata exactly will able to display proper different movies for each specific display brand/model, you have to test 1000 - 000.5, 1000-000.0, 1100-0.005, and 4.000-000.5 (nits max- nits min) mastered movies...this is why the list I will post it will be useful

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 offline  
post #248 of 3127 Old 08-22-2017, 06:45 AM
Advanced Member
 
randal_r's Avatar
 
Join Date: Mar 2010
Posts: 522
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 232 Post(s)
Liked: 131
Hi Ted,

I bed to differ, just because the system has gone from analog to digital does not mean that the methodology and procedural attributes have completely been changed. If you note that black has been recorded as low as 0.0005 nits. Could you please forward technical based information on any display that actually displays black at absolute zero (0) nits as appose to marketing data which is highly flawed.
randal_r is offline  
post #249 of 3127 Old 08-22-2017, 07:10 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 7,752
Mentioned: 179 Post(s)
Tagged: 1 Thread(s)
Quoted: 2921 Post(s)
Liked: 3569
Quote:
Originally Posted by randal_r View Post
Hi Ted,

I bed to differ, just because the system has gone from analog to digital does not mean that the methodology and procedural attributes have completely been changed. If you note that black has been recorded as low as 0.0005 nits. Could you please forward technical based information on any display that actually displays black at absolute zero (0) nits as appose to marketing data which is highly flawed.
Hi Randal,

The waveform monitors you are checking they visualize the digital signal to analog YPrCr waveform, so they use analog charts/graphs signal screens (component analog emulation) which are useful for visual identify some analog patterns (specific lines/steps form they display (to locate digital errors using analog classic methods, nothing more.

For digital signal analysis, we talk about HDR here, you need Accupel 6000 Plus Ultra (with Engineer Option) pixel checker feature if you want to check the pixel levels or more advanced analyzers, to check CEA-VIC, AVI, GCP, VSIF, TMDS Clock Rate, EDID, HDCP etc.

To see how the signal you send to display is actually displayed, you need the best colorimeter in global market (for low end/speed) which is the Klein K-10A to take a fast results, you can use Minolta CS-2000A if you have time to wait about 10-20 minutes while it's internal peltier thermo-electric cooling system cooling down the sensors to get a single measurement (in an ideal world were displaying 10 minutes the same pattern is not enabling any protection mode, while with Klein K-10A you will need 0.05 sec if you use 1 single read with CalMAN.

Do you want me to send you a document that says that OLED's have zero nits black? are you disagree with 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

Last edited by ConnecTEDDD; 08-22-2017 at 07:15 AM.
ConnecTEDDD is offline  
post #250 of 3127 Old 08-22-2017, 07:38 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 7,752
Mentioned: 179 Post(s)
Tagged: 1 Thread(s)
Quoted: 2921 Post(s)
Liked: 3569
Quote:
Originally Posted by D-Nice View Post
If you have an Oppo 203, holding down the info button will provide that info. I don't know of any other UHD player that has that option.

In general Warner and their subsideraries will be 4000nit. Everything else is 1000. I'm speaking of North American released titles.
Hi D-Nice, that feature is very useful of OPPO, I hope they can add more features in the future, to display MaxCLL and MaxFall...but the Mastering Primaries and White Point used it will be useful also.

While all current titles are using DCI-P3 Primaries with REC.2020 D65 White Point, there one only HDR title which has REC.2020 Primaries/White Point.

I'm talking about the Wild Africa / Tiny Giants from BBC UK, it's masted @ 4000 - 0.005.

There 2 ways to calibrated a Studio Mastering Monitor, one is to calibrated it for DCI-P3 with REC.2020 D65 and at the end to put these data to a REC.2020 container and the other way is to calibrated the monitor to REC.2020 directly; to take advance a monitor which has higher coverage from DCI-P3. So seems they used the second way, probably using Dolby Pulsar or Dolby Eclipse monitor.

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 offline  
post #251 of 3127 Old 08-22-2017, 07:49 AM
AVS Forum Addicted Member
 
D-Nice's Avatar
 
Join Date: Dec 2004
Location: Columbia, SC
Posts: 16,780
Mentioned: 120 Post(s)
Tagged: 1 Thread(s)
Quoted: 1832 Post(s)
Liked: 3596
Quote:
Originally Posted by ConnecTEDDD View Post
Hi D-Nice,

Try 540 nits peak also, maybe doing this you will have better PQ tracking when you calibrating and use it as fixed metadata for all the titles you will check.

Maybe this solution is the best one for current HDR calibration status improvement, sending a fixed metadata for all movies and by disabling any active HDR additional processing.

Just need some test what metadata exactly will able to display proper different movies for each specific display brand/model, you have to test 1000 - 000.5, 1000-000.0, 1100-0.005, and 4.000-000.5 (nits max- nits min) mastered movies...this is why the list I will post it will be useful
The A1E automatically ignores this info. It's dynamic HDR all the time.
D-Nice is offline  
post #252 of 3127 Old 08-22-2017, 09:19 AM
Senior Member
 
Join Date: Jul 2017
Posts: 300
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 161 Post(s)
Liked: 96
Wow, a lot to digest.... thank you Ted for the large amount of info!
I really hope things settle down soon and we can all start to enjoy a similar standard.

Head ---> SPLODE !
ConnecTEDDD likes this.

LGOLED55C7P - Calman Enthusiast - I1Display / I1Pro
PeteNeat is offline  
post #253 of 3127 Old 08-22-2017, 09:23 AM - Thread Starter
Oppo Beta Group
 
RichB's Avatar
 
Join Date: Nov 1999
Location: Massachusetts
Posts: 11,021
Mentioned: 29 Post(s)
Tagged: 0 Thread(s)
Quoted: 1834 Post(s)
Liked: 1318
Quote:
Originally Posted by D-Nice View Post
The A1E automatically ignores this info. It's dynamic HDR all the time.

If only a little birdie could whisper into LG's ear


- Rich

Oppo UPD-205 x 2 | UPD-203 | Sonica DAC | Emotiva XMC-1 (v3) | Revel Salon2s, Voice2, Studio2s | Benchmark AHB2 x 4 | ATI AT522NC | Velodyne HGS-15 | LG 77C9 | Lumagen 2020 | HDFury Vertex x 2
RichB is offline  
post #254 of 3127 Old 08-22-2017, 09:24 AM - Thread Starter
Oppo Beta Group
 
RichB's Avatar
 
Join Date: Nov 1999
Location: Massachusetts
Posts: 11,021
Mentioned: 29 Post(s)
Tagged: 0 Thread(s)
Quoted: 1834 Post(s)
Liked: 1318
Quote:
Originally Posted by ConnecTEDDD View Post
Hi D-Nice, that feature is very useful of OPPO, I hope they can add more features in the future, to display MaxCLL and MaxFall...but the Mastering Primaries and White Point used it will be useful also.

While all current titles are using DCI-P3 Primaries with REC.2020 D65 White Point, there one only HDR title which has REC.2020 Primaries/White Point.

I'm talking about the Wild Africa / Tiny Giants from BBC UK, it's masted @ 4000 - 0.005.

There 2 ways to calibrated a Studio Mastering Monitor, one is to calibrated it for DCI-P3 with REC.2020 D65 and at the end to put these data to a REC.2020 container and the other way is to calibrated the monitor to REC.2020 directly; to take advance a monitor which has higher coverage from DCI-P3. So seems they used the second way, probably using Dolby Pulsar or Dolby Eclipse monitor.

Are you suggesting that a providing a fixed override of the MaxCLL and MaxFall could be used to ignore and defeat tone-mapping in the LG?


- Rich

Oppo UPD-205 x 2 | UPD-203 | Sonica DAC | Emotiva XMC-1 (v3) | Revel Salon2s, Voice2, Studio2s | Benchmark AHB2 x 4 | ATI AT522NC | Velodyne HGS-15 | LG 77C9 | Lumagen 2020 | HDFury Vertex x 2
RichB is offline  
post #255 of 3127 Old 08-22-2017, 09:29 AM
Senior Member
 
Join Date: Jul 2017
Posts: 300
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 161 Post(s)
Liked: 96
For the average Joe in the street (me), what I would like all UHD and DV material to do is start at level 0.
No matter what formulas and math these things comprise of, the starting point for all gamma and black level is zero.

No light output. Nadda. Does this sound impossible?

LGOLED55C7P - Calman Enthusiast - I1Display / I1Pro
PeteNeat is offline  
post #256 of 3127 Old 08-22-2017, 09:51 AM
AVS Forum Addicted Member
 
D-Nice's Avatar
 
Join Date: Dec 2004
Location: Columbia, SC
Posts: 16,780
Mentioned: 120 Post(s)
Tagged: 1 Thread(s)
Quoted: 1832 Post(s)
Liked: 3596
Quote:
Originally Posted by RichB View Post
Are you suggesting that a providing a fixed override of the MaxCLL and MaxFall could be used to ignore and defeat tone-mapping in the LG?


- Rich
Not ignore but make the HDR input consistent instead of variable.
RichB and ConnecTEDDD like this.
D-Nice is offline  
post #257 of 3127 Old 08-22-2017, 09:52 AM
AVS Forum Addicted Member
 
D-Nice's Avatar
 
Join Date: Dec 2004
Location: Columbia, SC
Posts: 16,780
Mentioned: 120 Post(s)
Tagged: 1 Thread(s)
Quoted: 1832 Post(s)
Liked: 3596
Quote:
Originally Posted by PeteNeat View Post
For the average Joe in the street (me), what I would like all UHD and DV material to do is start at level 0.
No matter what formulas and math these things comprise of, the starting point for all gamma and black level is zero.

No light output. Nadda. Does this sound impossible?
Improbable not impossible.
PeteNeat likes this.
D-Nice is offline  
post #258 of 3127 Old 08-22-2017, 10:50 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 7,752
Mentioned: 179 Post(s)
Tagged: 1 Thread(s)
Quoted: 2921 Post(s)
Liked: 3569
Quote:
Originally Posted by RichB View Post
Are you suggesting that a providing a fixed override of the MaxCLL and MaxFall could be used to ignore and defeat tone-mapping in the LG?


- Rich
Hi Rich,

Sending a fixed metadata which will be the same for calibration and any movie player, you have enable a tone mapping LG but this tone mapping is fix, and it's not changing for every movie, so you calibration will be valid for that specific display response.

When it will be possible in the future (I hope) the capability to do 3D LUT for HDR, then it will work with same logic as I'm suggesting, the metadata of the movie playback has to be the same with the one you used to generate the patterns for that 3D LUT HDR display characterization.

The ideal is for any display to have a switch to it's normal menu where you will be able to enable the HDR mode without requiring to send any metadata....so no internal tone-mapping......when this will be possible you will be able to do 3D LUT for HDR using Lumagen PRO for example, but you have to block with Lumagen or HD Fury any movie metadata to go to the display.

The correct approach to 3D LUT a display is to just clip to the nits the display can't reproduce. This correction will be the reference one, it's exactly what Sony BVM-X300 OLED is doing, hard clip. Or you can make a small roll-off before reaching the max display nits....for not being so hard clip.

But since you will have more 3D LUT slots, you can create another custom roll-off (as a secondary) with earlier nits roll-off, but when you do this, you sacrifice the correct reproduction of nits levels the display can reproduce to be able to see some more details in the clouds or to lights etc....

For me is important to watch correct exactly what your display can display, this means 750nits for 2017 LG's and when in the future you get a brighter display, to give you the capability to see more detail in highlights......than trying to see from now all detail but getting a picture different from that the movie supposed to look.

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

Last edited by ConnecTEDDD; 08-22-2017 at 05:21 PM.
ConnecTEDDD is offline  
post #259 of 3127 Old 08-22-2017, 11:46 AM
Senior Member
 
pazman2000's Avatar
 
Join Date: Feb 2011
Posts: 330
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 216 Post(s)
Liked: 72
Quote:
Originally Posted by D-Nice View Post
Quote:
Originally Posted by pazman2000 View Post
I have run 100 hours up on my c7 and couldn't wait to give it a Calibration.
Greyscale was very accurate and needed very little 2pt and 20pt to get dE under 0.5
But colour on was not good at all ,very low on luminance for most colours .
I was very disappointed at this as I had read the 2017s had more accurate colours then 2016s .
I have tried CMS But just like the 2016 using that is cause visible problems in content , adjustments to green luminance of +5 is needed but one to two clicks creates false contouring and polarisation.
Any ideas guys ?
Do you have any examples for the false contouring? What Color Gamut setting are you using?
What you call this I'm not sure but this is one the effects of using CMS in the UK
The same as last year on my e6
Green lum +20 looks like mud on the field
Attached Thumbnails
Click image for larger version

Name:	IMG_1455_1503427571159.JPG
Views:	92
Size:	6.16 MB
ID:	2271396   Click image for larger version

Name:	IMG_1456_1503427608003.JPG
Views:	87
Size:	5.18 MB
ID:	2271398  

Last edited by pazman2000; 08-22-2017 at 11:52 AM.
pazman2000 is offline  
post #260 of 3127 Old 08-22-2017, 12:27 PM
AVS Forum Addicted Member
 
D-Nice's Avatar
 
Join Date: Dec 2004
Location: Columbia, SC
Posts: 16,780
Mentioned: 120 Post(s)
Tagged: 1 Thread(s)
Quoted: 1832 Post(s)
Liked: 3596
Quote:
Originally Posted by pazman2000 View Post
What you call this I'm not sure but this is one the effects of using CMS in the UK
The same as last year on my e6
Green lum +20 looks like mud on the field
Your attachment did not come through. I've only seen the need to do +-5 on anyone of the controls and see zero issues with the CMS (verified with Ted's disc). Perhaps it is your meter? Also, I use the Extended Color gamut with a main color setting of 46.
ConnecTEDDD and PeteNeat like this.
D-Nice is offline  
post #261 of 3127 Old 08-22-2017, 12:35 PM
Senior Member
 
pazman2000's Avatar
 
Join Date: Feb 2011
Posts: 330
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 216 Post(s)
Liked: 72
Quote:
Originally Posted by D-Nice View Post
Quote:
Originally Posted by pazman2000 View Post
What you call this I'm not sure but this is one the effects of using CMS in the UK
The same as last year on my e6
Green lum +20 looks like mud on the field
Your attachment did not come through. I've only seen the need to do +-5 on anyone of the controls and see zero issues with the CMS (verified with Ted's disc). Perhaps it is your meter? Also, I use the Extended Color gamut with a main color setting of 46.
I used + 20 just to make the problem more visible in a photo
I referenced my i1d3 with an i1 pro 2
I also only need + green lum but this still causes issues .
This may well be only Uk sets for some strange reason again .
Thanks again D nice
Attached Thumbnails
Click image for larger version

Name:	IMG_1455_1503430338865.JPG
Views:	96
Size:	6.16 MB
ID:	2271432   Click image for larger version

Name:	IMG_1456_1503430362378.JPG
Views:	86
Size:	5.18 MB
ID:	2271434  
pazman2000 is offline  
post #262 of 3127 Old 08-22-2017, 12:40 PM
AVS Forum Addicted Member
 
D-Nice's Avatar
 
Join Date: Dec 2004
Location: Columbia, SC
Posts: 16,780
Mentioned: 120 Post(s)
Tagged: 1 Thread(s)
Quoted: 1832 Post(s)
Liked: 3596
Quote:
Originally Posted by pazman2000 View Post
I used + 20 just to make the problem more visible in a photo
I referenced my i1d3 with an i1 pro 2
I also only need + green lum but this still causes issues .
This may well be only Uk sets for some strange reason again .
Thanks again D nice
Still can't see your attachment
D-Nice is offline  
post #263 of 3127 Old 08-22-2017, 01:03 PM
Senior Member
 
pazman2000's Avatar
 
Join Date: Feb 2011
Posts: 330
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 216 Post(s)
Liked: 72
Quote:
Originally Posted by D-Nice View Post
Still can't see your attachment
Cheers D nice ill give extended go now
Attached Images
File Type: jpg image1.JPG (91.6 KB, 246 views)
File Type: jpg image2.JPG (78.4 KB, 182 views)
pazman2000 is offline  
post #264 of 3127 Old 08-22-2017, 01:32 PM
Senior Member
 
Join Date: Jul 2017
Posts: 300
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 161 Post(s)
Liked: 96
Quote:
Originally Posted by D-Nice View Post
...I've only seen the need to do +-5 on anyone of the controls and see zero issues with the CMS (verified with Ted's disc). Perhaps it is your meter? Also, I use the Extended Color gamut with a main color setting of 46.
Snippets of information seeping out

LGOLED55C7P - Calman Enthusiast - I1Display / I1Pro
PeteNeat is offline  
post #265 of 3127 Old 08-22-2017, 03:14 PM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 7,752
Mentioned: 179 Post(s)
Tagged: 1 Thread(s)
Quoted: 2921 Post(s)
Liked: 3569
Quote:
Originally Posted by ERuiz View Post
So what you are saying is, the director's intent with Girl On The Train was to show us a washed out image? We sure are suckers for spending all this money on the latest technology to watch a movie rendered horribly because that's the way the director wanted it.
Hi, the problem is the same movie will look so different because of differences in tone mapping, peak output and gamut coverage of displays used for mastering vs. the ones consumers can buy. So it's impossible to see what director wanted to show you for the HDR release of a movie, unless you watch the same movie some years later when the displays will have 1000/4000 nits and DCI-P3 100% coverage (but that time the mastering will have moved to larger gamut coverage from DCI-P3, so we always we will try to follow the mastering specs of mastering studios (which will always be higher from consumer models capabilities)

With current displays which are wonderful, if you want to watch the movies as supposed to look, you have to invest in 3D LUT for SDR. All modern TV's can cover 100% calibrated the REC.709 colorspace, and the luminance levels movies are mastered (100-120 nits), so there 2 ways:

1) If you source is a HTPC and you have a decent VGA card, you can use a software player which supports madVR renderer (MadVR has software patch generation also from inside the player you will use), madVR is supporting 65-Point 3D Cube files (same as eeColor 3D LUT Box). You can use DisplayCAL/ArgyllCMS (free) or LightSpace/CalMAN (paid) solutions to create the 3D LUT generation for madVR.

The ideal, if you have a HTPC is to periodically check by doing a quick verification of the active 3D LUT correction you are using, because VGA drivers, window drivers, madvr updates, software player updates, it's unknown if all these after an new installed update if it will affect you calibration. The safest way is to have a HTPC only for movies without applying any updates when you have a valid 3D LUT loaded.....and to apply all the updates at once next time (after months) before you will run a new 3D LUT calibration.

2) If you source is a BD player or Media Player or UHD Player*, then you can add a SDR 1080p 3D LUT Box, there many solutions out, the lowest priced but the same time with highest 3D LUT table size is the eeColor 3D LUT Box. It's the only 3D LUT Box which can be used with DisplayCAL/ArgyllCMS (free) to keep the total cost low or you can use CalMAN Enthousiast or LightSpace HTL (from paid solutions)

For patch generation required for the thousand colors you will measure; you can use Amazon FireStick (for CalMAN/LightSpace or Google Chromecast (for DisplayCAL/ArgyllCMS) and each software will connect thru wireless network to command these patch generators.

When you will use 3D LUT display characterization, you will have calibrated multiple levels of hue/saturation/brightness, the mastering monitors at post-production are using for ages 3D LUT's only (about 15 years already).

Because the Gamma SDR movies are mastered, it varies from years-to-years, it's unknown what gamma each studio is using, currently most of the studios are using 2.4, before some years it was 2.35 (per EBU...now EBU has changed it to 2.4), older movies were 2.2. With 3D LUT you can have different calibrated gamma's to 3 different slots, to swap real-time to see how it's affecting you picture and select the one that you satisfied most...since we are don't know what gamma each studio is using for SDR.

There a lot of solutions from Lumagen also to check (smaller 3D LUT size, 5 or 9 or 17-Point Cube), they will require a paid software, but they are including patch generation internally.

The good news with HDR is that all are using the same transfer function to studios but it's existing a problem in home reproduction of PQ due to different approaches each display is using, different peaks/gamut coverage etc..

If all these sound complicated, you can hire a calibrator which performs 3D LUT, so he will use ever better meters for this job.

From the feedback I have from my disk users, D-Nice is performing 3D LUT calibration for eeColor (maybe he is doing for Lumagen's also), AFIK he is using CalMAN or LightSpace.

I use only 3D LUT's from 2012, there no way back to normal calibration.

* If you have UHD player and you need to have the same time 2 outputs (one for DV/HDR10 and another one for 1080p), then you can add to your UHD player output the HD Fury Integral which can work as 1 Input -> 2 Output Spliter.

The top output of HD Intergral is HDCP2.2, so from there you will use an HDMI cable to go to your TV for DV/HDR10 signaling, and to that input of your TV you will have adjust the settings for HDR calibration.

The bottom port of HD Intergral is HDCP 1.4, there you can connect the 3D LUT Box and then the TV...so to that input you will have the settings for 3D LUT. (native gamut colorspace and pre-calibration only of 100% White required for 3D LUT pre-calibration, not parametric adjustments)

You will configure your UHD player to have play at native resolution, so when you will load a BD Movie it will output 1080p24/50/60 (according to each title framing) and TV will do the upscale to 2160p.

Alternative solution if you have OPPO 20x UHD player and a secondary BD-Player is to use the secondary player as a source -> from there 1080p to go to 3D LUT Box -> from there to go to OPPO 20x HDMI Input, and OPPO 20x will do the upscale to the TV directly. In that case you don't need HD Integral (in case the TV stores different settings for SDR and HDR...so for LG it will be fine, you will not need HD Fury Integral)

AVSForums (or others) users are contacting me from such questions so it was a good opportunity to have there a post with a more complete answer that may helpful for all
cardoski, PeteNeat and Anger.miki 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

Last edited by ConnecTEDDD; 01-22-2018 at 01:53 PM.
ConnecTEDDD is offline  
post #266 of 3127 Old 08-22-2017, 05:34 PM
Senior Member
 
Join Date: Jul 2017
Posts: 300
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 161 Post(s)
Liked: 96
Quote:
Originally Posted by ConnecTEDDD View Post
...it was a good opportunity to have there a post with a more complete answer that may helpful for all
Thanks very much for the detailed info.
Please someone release a 4k LUT box for a grand

LGOLED55C7P - Calman Enthusiast - I1Display / I1Pro
PeteNeat is offline  
post #267 of 3127 Old 08-22-2017, 06:04 PM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 7,752
Mentioned: 179 Post(s)
Tagged: 1 Thread(s)
Quoted: 2921 Post(s)
Liked: 3569
Quote:
Originally Posted by PeteNeat View Post
Thanks very much for the detailed info.
Please someone release a 4k LUT box for a grand
Even for 1$ 2160p 3D LUT Box, will not work for HDR 3D LUT unless companies and an option to disable the internal display tone/gamut mapping. So you are limited for SDR 4K if you buy any 3D LUT Box with 2160p now (only Lumagen PRO is available as an option)

Since HDR for displays is impossible (because internal processing can't be bypassed), the only market for 3D LUT Box which supports HDR is Lumagen PRO, the only one choice in consumer world, because it has HDCP 2.2 support and HDMI 2.0a input/output.

There is a much better and cheaper 3D LUT Box (without scalling and enhancements) that can be used for HDR (HDR 3D LUT in post production don't require metadata, so no need to support HDMI 2.0a) in professional world, the FSI LUT Box which has 4096-Point Input 1D LUT that can be used as a Shaper LUT to create amazing smoothly ST.2084 curves where you can clip to exact nits you want, 33-Point 3D LUT Cube and Output 1D LUT than can be used to calibrated it first...(LightSpace has helped FSI a lot to design that box) but it doesn't have HDMI input-output and not support HDCP in case someone try to use SDI to HDMI converters....so we can't use in HT world.....

Lumagen PRO has to increase from 21-Point 1D LUT now to at least a 1024 adjustable (10-bit) Input 1D LUT and increase the 3D LUT size in the future.....since it will require a lot of new space probably not currently available, there can be added an option to select between 6x 17-Point slots or 2x 33-Point, so this will cover all setups.

17-Point is a good size for SDR, to cover 0-100 nits but very small to cover 0-10.000 nits, it will be required an increase of 3D LUT and 1D LUT for more accurate final results.

So when you have only 21-Points to follow PQ transfer function; the adjustment points have very large distance between them, so even a very smooth generated roll-off from LightSpace for example, it will be not possible to be reproduced when it will be uploaded to Lumagen PRO memory. An exact clip (or roll-off starting) to specific nits it will possible.

See how the 21-Points 1D LUT matching the absolute values of ST-2048 for HDR:

64 -> 0 nits
108 -> 0.61 nits
152 -> 0.33 nits
196 -> 1.01 nits
240 -> 2.47 nits
284 -> 5.24 nits
328 -> 10.21 nits
372 -> 18.78 nits
416 -> 33.1 nits
460 -> 56.55 nits
504 -> 94.38 nits
544 -> 148.03 nits
588 -> 239.81 nits
632 -> 384.71 nits
676 -> 613.08 nits
720 -> 973.13 nits
764 -> 1542.21 nits
808 -> 2445.67 nits
852 -> 3848.17 nits
896 -> 6214.61 nits
940 -> 10000 nits

When there will be a way to enable HDR mode of the display without sending metadata, and you will calibrated with normal patterns (without sending metadata), when you will playback movies you have to block any movie to send any metadata to the display.

To be able to generate a valid 3D LUT correction you need a display at it's HDR mode, as you see below LG OLED 2016 is very bad example:

Here you see the Drift Plot of a 10-Point Cube, it took 17 minutes to be completed using my Klein K-10A, the peak output of the LG is drifting +-15 nits, the graph is displaying +-2.5 nits range (this test is from November 2016 using 65E6 @ HDR Mode):



During the profiling LightSpace was taking 1 White measurement patch per 20 other color patches, for it's Drift Compensation algorithm. 30 nits difference is too much for proper 3D LUT generation.

Just for comparison, see how an Epson LS-10000 projector is drifting during the profiling (very stable device):



While the Luminance output is stable, the blue channel is changing and this affecting a bit the chromaticity, but this will be fixed by LightSpace during the colorspace correction generation.

So even with 3D LUT Box for HDR with $1 cost, it's impossible to calibrate with 3D LUT for HDR, while you will have no problem for SDR, post 3D LUT measurements of LG 65E6 using LightSpace and eeColor 3D LUT Box I have posted here: LG OLEDs - how to enhance near black detail
PeteNeat 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 offline  
post #268 of 3127 Old 08-22-2017, 06:15 PM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 7,752
Mentioned: 179 Post(s)
Tagged: 1 Thread(s)
Quoted: 2921 Post(s)
Liked: 3569
An additional solution if we found a stable display in HDR mode which will have an option to disable internal tone gamut mapping so enable it's HDR mode without metadata is the BlackMagic Design Teranex Mini.

Teranex Mini don't have model with HDMI 2.0 input - output....it has one model with HDMI 2.0 Input and SDI 12G output and one model with SDI 12G input and HDMI 2.0 output.

You need to get these 2 boxes (1K cost both) and connect them ''one after the other'' so you will have HDMI 2.0 input and HDMI 2.0 output....and the boxes will that SDI 12G connection between them.

Before the first box you have to connect HD Fury, to remove any metadata and downgrade to from HDCP 2.2 -> HDCP 1.4.....then you need a device to remove any HDCP (after the HD Fury) and then you connect the first Teranex there.

Terranex don't have 1D LUT but it has 33-Point Cube 3D LUT, so the first box with be used as 1D Input LUT 33-Point (Shaper LUT) and the seconds unit will do the 3D LUT.

Very simple stuff, I have solutions for all loooooooooool.
PeteNeat 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 offline  
post #269 of 3127 Old 08-22-2017, 09:48 PM
Senior Member
 
Join Date: Jun 2017
Posts: 289
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 200 Post(s)
Liked: 66
which all this means that HDR is still in its infancy and require a lot of work to play nicely; if at all......
PeteNeat likes this.
rotarydude is offline  
post #270 of 3127 Old 08-22-2017, 10:37 PM
Newbie
 
Join Date: Aug 2017
Posts: 1
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 0
So, after reading how precarious it is to calibrate HDR, I'm now looking at local calibration experts with a more critical eye. This is a long shot but do any of you happen to know of someone with HDR calibration experience in the Seattle area? Also, I've somewhat vaguely put together that I should wait after using the TV for a certain amount of hours before having it calibrated. Is this correct?
optimusprime8062 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