Calibrating HDR10 on LG OLED - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 12Likes
Reply
 
Thread Tools
post #1 of 76 Old 04-14-2017, 04:10 PM - Thread Starter
AVS Forum Special Member
 
BigCoolJesus's Avatar
 
Join Date: Feb 2009
Posts: 4,389
Mentioned: 16 Post(s)
Tagged: 0 Thread(s)
Quoted: 1674 Post(s)
Liked: 1386
Calibrating HDR10 on LG OLED

Hey all,

Finally decided to calibrate my LG OLED C6 for HDR10 playback using the CalMAN workflow and Masciola's patterns on an USB stick. Not going to jump into UHD/HDR too much, but my Xbox One S is finally connected to an HDR capable receiver and a few games I have support HDR (like Mass Effect Andromeda), so figured I should at least get the settings in line.

From my understanding, pretty much the only thing I can/should measure and change, using the HDR Standard picture mode, is the the Brightness, OLED Light, Contrast, Color (Tint) and the Gamma adjustments. Correct?

Also, is the ST 2084 HDR (PQ) Gamma curve the only curve one should use for HDR10 calibration or can the traditional power values (such as 2.2, 2.3, 2.4, etc) be used? I'm a little confused by the EOTF curve graph in place of the traditional Gamma curve graph for CalMAN...

Thanks!
mo949 likes this.

Last edited by BigCoolJesus; 04-14-2017 at 07:08 PM.
BigCoolJesus is online now  
Sponsored Links
Advertisement
 
post #2 of 76 Old 04-15-2017, 02:16 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,539
Mentioned: 64 Post(s)
Tagged: 1 Thread(s)
Quoted: 1733 Post(s)
Liked: 2138
Quote:
Originally Posted by BigCoolJesus View Post
Hey all,

Finally decided to calibrate my LG OLED C6 for HDR10 playback using the CalMAN workflow and Masciola's patterns on an USB stick. Not going to jump into UHD/HDR too much, but my Xbox One S is finally connected to an HDR capable receiver and a few games I have support HDR (like Mass Effect Andromeda), so figured I should at least get the settings in line.

From my understanding, pretty much the only thing I can/should measure and change, using the HDR Standard picture mode, is the the Brightness, OLED Light, Contrast, Color (Tint) and the Gamma adjustments. Correct?

Also, is the ST 2084 HDR (PQ) Gamma curve the only curve one should use for HDR10 calibration or can the traditional power values (such as 2.2, 2.3, 2.4, etc) be used? I'm a little confused by the EOTF curve graph in place of the traditional Gamma curve graph for CalMAN...

Thanks!
Hi, Contrast/Brightness settings in HDR are not working like we know in SDR, Brightness should left untouched, because the display gamut/tone mapping is based to these default settings.

HDR is using ST.2084 which is an absolute curve, the display has to follow specific luminance levels per digital level, according to the Dolby's golden reference numbers, so you are following to up to the luminance levels the display is capable and you clipp or roll-off (hard/soft)...(it's up to each display internal gamut/tone mapping programming)...the higher from display's peak output levels the display can't follow.

So when you have a 650nits capable display you see about up to 70.5% of the signal, the other info will be clipped or roll-off (hard/soft), it's up to each display internal gamut/tone mapping programming.

Other examples:

75.2% is 1000 nits
90.2% is 4014 nits
97.7% is 8047 nits
100% is 10000 nits

The PQ curve charts have different curve shape from the familiar ones of SDR calibration because 100% is 10.000 nits.

If you reduce Contrast/OLED from LG, you will reduce the peak output, OLED/Contrast @ 100 provides better color tracking at these default values.

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 level luminance levels, but in HDR the PQ Curve transfer function has fixed numbers from 0-10.000nits, you follow or you clip, you can't modify or use other values.

Ignore CMS adjustments and do RGB Balance adjustments only.

Each of the White Balance control in HDR has a specific +- value working range which not introduce problems to real content. The range of values are different from point-to-point. At low end controls the adjustments has to be very minimal. Play some different content and play with each control until you find up to which values you don't see problems. Use different movies/scenes to have plenty of different palettes to evaluate.

Using LG instructions for HDR calibration it will increase your color errors and provide lower peak output but better Grayscale and gamma tracking, but the color errors are more important from grayscale errors in that case.

After some testing performed using a 65E6 @ HDR mode, lowering the OLED Light, can improve the Grayscale but increase the color errors. To compare what is happening you have to take 5-Point Saturations (or 10-Point)

If you keep the Contrast/OLED @ 100, and calibrated the RGB levels for HDR, then you get an average Grayscale of about 1.6dE2000 and the errors are coming from gamma (brigher) at mid range while the RGB balance is near perfect....you get about 670nits calibrated.

If you run a 5-Point Saturation with targets DCI P3 inside a REC2020 (look CalMAN HDR workflow) then you get about ~2.0dE2000 average.

If you keep the Contrast @ 100 and reduce OLED about 80? (I don't remember the exact value...to get 540nits) and calibrated the RGB levels for HDR, then you get an average Grayscale of about ~1.0dE2000 (or lower)and the errors of gamma are reduced at mid range while the RGB balance is near perfect....you get about 540nits calibrated.

If you run a 5-Point Saturation with targets DCI P3 inside a REC2020 (look CalMAN HDR workflow) then you get about ~4.0dE2000 average.
mascior, Miguel Ronda and donkrx 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 offline  
post #3 of 76 Old 04-15-2017, 04:38 AM - Thread Starter
AVS Forum Special Member
 
BigCoolJesus's Avatar
 
Join Date: Feb 2009
Posts: 4,389
Mentioned: 16 Post(s)
Tagged: 0 Thread(s)
Quoted: 1674 Post(s)
Liked: 1386
THANK YOU!

Excellent response. Covers everything I need to know.

Sent from my Nexus 6P using Tapatalk
ConnecTEDDD likes this.
BigCoolJesus is online now  
 
post #4 of 76 Old 04-15-2017, 06:12 AM - Thread Starter
AVS Forum Special Member
 
BigCoolJesus's Avatar
 
Join Date: Feb 2009
Posts: 4,389
Mentioned: 16 Post(s)
Tagged: 0 Thread(s)
Quoted: 1674 Post(s)
Liked: 1386
Quote:
Originally Posted by ConnecTEDDD View Post
HDR is using ST.2084 which is an absolute curve, the display has to follow specific luminance levels per digital level, according to the Dolby's golden reference numbers, so you are following to up to the luminance levels the display is capable and you clipp or roll-off (hard/soft)...(it's up to each display internal gamut/tone mapping programming)...the higher from display's peak output levels the display can't follow.

So when you have a 650nits capable display you see about up to 70.5% of the signal, the other info will be clipped or roll-off (hard/soft), it's up to each display internal gamut/tone mapping programming.
Ted,

So for ST.2084 Gamma using the CalMAN workflow, when I make adjustments, the goal is simply to try and get RGB tracking near 0 for all values without introducing real world content issues (the lower end adjustments are the most sensitive to change) and the EOTF curve will fall into line accordingly?

I ask because as you know, for SDR, one might use both the Gamma curve and the RGB tracking to tailor their results. But it sounds like for HDR, the only goal is to get RGB tracking in line and wherever the curve falls is where it falls without really being able to change it (since the curve corresponds to the displays overall nit ability).

Thanks!

Last edited by BigCoolJesus; 04-15-2017 at 08:53 AM.
BigCoolJesus is online now  
post #5 of 76 Old 04-15-2017, 11:09 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,539
Mentioned: 64 Post(s)
Tagged: 1 Thread(s)
Quoted: 1733 Post(s)
Liked: 2138
Quote:
Originally Posted by BigCoolJesus View Post
Ted,

So for ST.2084 Gamma using the CalMAN workflow, when I make adjustments, the goal is simply to try and get RGB tracking near 0 for all values without introducing real world content issues (the lower end adjustments are the most sensitive to change) and the EOTF curve will fall into line accordingly?

I ask because as you know, for SDR, one might use both the Gamma curve and the RGB tracking to tailor their results. But it sounds like for HDR, the only goal is to get RGB tracking in line and wherever the curve falls is where it falls without really being able to change it (since the curve corresponds to the displays overall nit ability).

Thanks!
Fixing only RGB Balance errors will require less control adjustments, to these displays calibration is more like art than science (as D-Nice said), this means that you have to sacrifice high adjustments (so leave errors) to be able to get watchable picture. dE Charts don't tell the whole story, controls are not aligh with patterns also, you have to test some stuff and them watch some movies scenes to confirm that there no added problems to shades/distortions/pixel-blocking etc..

Tracking of PQ transfer function is not possible because it will require high adjustments which will introduce serious problems to real content.

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 #6 of 76 Old 04-15-2017, 11:17 AM
Member
 
T( )( )L's Avatar
 
Join Date: May 2008
Location: Sweden
Posts: 184
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 81 Post(s)
Liked: 41
A nice thing to do is keep watching a greyramp while balancing rgb,its easy to get it look like a rainbow : )
David Mathews likes this.
T( )( )L is offline  
post #7 of 76 Old 04-15-2017, 11:56 AM - Thread Starter
AVS Forum Special Member
 
BigCoolJesus's Avatar
 
Join Date: Feb 2009
Posts: 4,389
Mentioned: 16 Post(s)
Tagged: 0 Thread(s)
Quoted: 1674 Post(s)
Liked: 1386
Thanks for the input!

Looks like for HDR, baby steps is the name of the game (small adjustments, watch, small adjustments, watch)
BigCoolJesus is online now  
post #8 of 76 Old 04-18-2017, 06:42 AM
M-V
Member
 
Join Date: Feb 2017
Posts: 31
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 25 Post(s)
Liked: 1
Hi all!

Self-calibrating HDR on an LG OLED and would like to join the discussion and ask for opinions and advices. Any input would be much appreciated!

I'm aware that HDR near black leves have very low luminance (0.00X nits) but starting from what levels (Black(64) - 68 - 73 - 77 - 81 - 86 - 90 - etc. ) you can see bars blinking in the Black Clipping HDR pattern in a dark room with a naked eye after some adaptation period?
All bars are distinguishable from a full black field or some are crushed?


When referring to a real world content what movie scenes (dark, bright, colorful?) and test patterns are you using for tests? (Grayscale ramp, gradient, ...?)
Any 'known issues' to check in the first place?
M-V is offline  
post #9 of 76 Old 04-18-2017, 07:27 AM - Thread Starter
AVS Forum Special Member
 
BigCoolJesus's Avatar
 
Join Date: Feb 2009
Posts: 4,389
Mentioned: 16 Post(s)
Tagged: 0 Thread(s)
Quoted: 1674 Post(s)
Liked: 1386
Quote:
Originally Posted by M-V View Post
Hi all!

Self-calibrating HDR on an LG OLED and would like to join the discussion and ask for opinions and advices. Any input would be much appreciated!

I'm aware that HDR near black leves have very low luminance (0.00X nits) but starting from what levels (Black(64) - 68 - 73 - 77 - 81 - 86 - 90 - etc. ) you can see bars blinking in the Black Clipping HDR pattern in a dark room with a naked eye after some adaptation period?
All bars are distinguishable from a full black field or some are crushed?


When referring to a real world content what movie scenes (dark, bright, colorful?) and test patterns are you using for tests? (Grayscale ramp, gradient, ...?)
Any 'known issues' to check in the first place?
From my recent delving into the topic, I would highly recommend you purchase Masciola's HDR test pattern download (Google Masciola). All you need to do is put his test patterns on a USB stick and plug it into the LG and voila, the patterns kick the LG into HDR mode and you are free to adjust away.

Regarding Black Clipping, using HDR Standard picture mode with default Brightness 50, near black value 73 can be seen flashing from a few inches away and, if you focus really hard and block out the white light from the numbers on the pattern, value 68 can just barely be seen flashing...at least on my display. It is advised to NOT TOUCH the Brightness setting regardless of what the pattern shows on your TV as it throws other HDR values and pre-mapped points out of whack. I have found that what you see with default values is about what you are going to get aside from using a meter to calibrate the grayscale.

Regarding real world content, the Masciola download has five excellent HDR test scenes that will show obvious issues if you change settings too much.

My advice, based on playing around yesterday, is to use HDR Standard with Contrast and OLED at 100, Brightness at 50, other settings left on default (such as Tint and Color), Dynamic Contrast Off, Color Gamut Normal, and all other enhancements turned off as usual.
BigCoolJesus is online now  
post #10 of 76 Old 04-18-2017, 08:52 AM
M-V
Member
 
Join Date: Feb 2017
Posts: 31
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 25 Post(s)
Liked: 1
Quote:
Originally Posted by BigCoolJesus View Post
From my recent delving into the topic, I would highly recommend you purchase Masciola's HDR test pattern download (Google Masciola). All you need to do is put his test patterns on a USB stick and plug it into the LG and voila, the patterns kick the LG into HDR mode and you are free to adjust away.

Regarding Black Clipping, using HDR Standard picture mode with default Brightness 50, near black value 73 can be seen flashing from a few inches away and, if you focus really hard and block out the white light from the numbers on the pattern, value 68 can just barely be seen flashing...at least on my display. It is advised to NOT TOUCH the Brightness setting regardless of what the pattern shows on your TV as it throws other HDR values and pre-mapped points out of whack. I have found that what you see with default values is about what you are going to get aside from using a meter to calibrate the grayscale.

Regarding real world content, the Masciola download has five excellent HDR test scenes that will show obvious issues if you change settings too much.

My advice, based on playing around yesterday, is to use HDR Standard with Contrast and OLED at 100, Brightness at 50, other settings left on default (such as Tint and Color), Dynamic Contrast Off, Color Gamut Normal, and all other enhancements turned off as usual.
Thanks for the info!

I'm already using Ryan's patterns and they are doing the job and are great overall.
But when it comes to 'Test footage' and I start looking at this photo (girl in a studio with a color target behind) ->


I personally have a very strong feeling that exactly the same photo in SDR is looking way much better than a HDR version. (More natural skintone, shadow, etc.)
I'm certainly not trying to get full HDR/SDR match but such results are a bit confusing.

With black level -
Whenever I get 68 barely visible 64 is not zero black...
When I try to get WB point (code value 127) luminance around what the standard is saying (PQ 127 = 0,1410 nits, LG doc says code 127 = 0.14 nits)
I can barely (almost invisible) see 72 flashing and ~68-70 are clipped.

So here is another question
What is more important: seeing all near black levels distinctly but loosing zero-black or clipping some of the near black but keeping 0-black and following PQ more accurately on near black levels?

Real picture looks "nicer" and "as if there is more contrast" when some of the near black is clipped.
M-V is offline  
post #11 of 76 Old 04-18-2017, 03:27 PM - Thread Starter
AVS Forum Special Member
 
BigCoolJesus's Avatar
 
Join Date: Feb 2009
Posts: 4,389
Mentioned: 16 Post(s)
Tagged: 0 Thread(s)
Quoted: 1674 Post(s)
Liked: 1386
Quote:
Originally Posted by M-V View Post
Thanks for the info!

I'm already using Ryan's patterns and they are doing the job and are great overall.
But when it comes to 'Test footage' and I start looking at this photo (girl in a studio with a color target behind) ->

I personally have a very strong feeling that exactly the same photo in SDR is looking way much better than a HDR version. (More natural skintone, shadow, etc.)
I'm certainly not trying to get full HDR/SDR match but such results are a bit confusing.

With black level -
Whenever I get 68 barely visible 64 is not zero black...
When I try to get WB point (code value 127) luminance around what the standard is saying (PQ 127 = 0,1410 nits, LG doc says code 127 = 0.14 nits)
I can barely (almost invisible) see 72 flashing and ~68-70 are clipped.

So here is another question
What is more important: seeing all near black levels distinctly but loosing zero-black or clipping some of the near black but keeping 0-black and following PQ more accurately on near black levels?

Real picture looks "nicer" and "as if there is more contrast" when some of the near black is clipped.
I cannot give an in-depth "professional" reply as I am merely a high level amateur when it comes to understanding calibrating (at least for SDR...HDR is an ongoing lesson for me right now).

But, from what I have gathered so far is two things are fairly certain for HDR on the 2016 LG OLED:
1) Do not adjust the 127 value for WB...leave it alone, forget about it, ignore it
2) You shouldn't touch the Brightness value, either
ConnecTEDDD likes this.
BigCoolJesus is online now  
post #12 of 76 Old 04-26-2017, 11:42 PM
Member
 
Join Date: Jun 2015
Posts: 37
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 28 Post(s)
Liked: 4
Hi guys,

i am about to "calibrate" my LG E6 for HDR10 as well using Mascior`s patterns and HCFR. Could you please let me know which of his patterns you have used?

And Ted... thanks for your outstanding support once again!

Greetings

Felix
ConnecTEDDD likes this.
RoninF is offline  
post #13 of 76 Old 09-20-2017, 06:19 PM
Member
 
jamesb23's Avatar
 
Join Date: Nov 2002
Posts: 113
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 38 Post(s)
Liked: 13
Quote:
Originally Posted by BigCoolJesus View Post
I cannot give an in-depth "professional" reply as I am merely a high level amateur when it comes to understanding calibrating (at least for SDR...HDR is an ongoing lesson for me right now).

But, from what I have gathered so far is two things are fairly certain for HDR on the 2016 LG OLED:
1) Do not adjust the 127 value for WB...leave it alone, forget about it, ignore it
2) You shouldn't touch the Brightness value, either
Posted in the HCFR thread but will post here in case someone has some guidance.

So I've moved on from SDR to HDR calibration but I'm a little bit confused. I purchased the Masciola patterns for my USB to use on my LG c6 OLED. In HCFR preferences I'm assuming I select 2084 for gamma, but which standard is it in the drop down menu - there's Rec2020, Rec2020/P3, DCI-P3? Also, given that I have to use the LG code patterns rather than standard 5-100 IRE grayscale, how does HCFR track it? It's not going to show something by LG's code values right? So how do I know which grayscale pattern aligns with what?
jamesb23 is offline  
post #14 of 76 Old 09-20-2017, 07:35 PM
A/V Addicted Member
 
mascior's Avatar
 
Join Date: Apr 2008
Location: Currituck, NC
Posts: 666
Mentioned: 27 Post(s)
Tagged: 0 Thread(s)
Quoted: 278 Post(s)
Liked: 406
Quote:
Originally Posted by jamesb23 View Post
Posted in the HCFR thread but will post here in case someone has some guidance.

So I've moved on from SDR to HDR calibration but I'm a little bit confused. I purchased the Masciola patterns for my USB to use on my LG c6 OLED. In HCFR preferences I'm assuming I select 2084 for gamma, but which standard is it in the drop down menu - there's Rec2020, Rec2020/P3, DCI-P3? Also, given that I have to use the LG code patterns rather than standard 5-100 IRE grayscale, how does HCFR track it? It's not going to show something by LG's code values right? So how do I know which grayscale pattern aligns with what?
Hi Jamesb23,

Select SMPTE 2084 HDR for "Gamma calculation" and UHDTV-Rec2020 for the "Color Space" standard option. Regarding the LG's code specific grayscale patterns, you can enter the 8-bit values(shown below) within the user.csv file in the HCFR program directory. Reference HERE and HERE. Thanks!

Here is the 10-bit to 8-bit reference of the LG OLED code specific patterns:

1. 127/32.....7.2%
2. 254/64.....21.7%
3. 320/80.....29.2%
4. 386/97.....36.8%
5. 419/105.....40.5%
6. 451/113.....44.2%
7. 467/117.....46.0%
8. 482/121.....47.7%
9. 498/125.....49.5%
10. 513/128.....51.3%
11. 529/132.....53.1%
12. 544/136.....54.8%
13. 560/140.....56.6%
14. 575/144.....58.3%
15. 591/148.....60.2%
16. 606/152.....61.9%
17. 622/156.....63.7%
18. 637/159.....65.4%
19. 653/163.....67.2%
20. 668/167.....68.9%

- Ryan M.

Last edited by mascior; 09-20-2017 at 07:41 PM.
mascior is online now  
post #15 of 76 Old 09-20-2017, 08:01 PM
Member
 
jamesb23's Avatar
 
Join Date: Nov 2002
Posts: 113
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 38 Post(s)
Liked: 13
Quote:
Originally Posted by mascior View Post
Hi Jamesb23,

Select SMPTE 2084 HDR for "Gamma calculation" and UHDTV-Rec2020 for the "Color Space" standard option. Regarding the LG's code specific grayscale patterns, you can enter the 8-bit values(shown below) within the user.csv file in the HCFR program directory. Reference HERE and HERE. Thanks!

Here is the 10-bit to 8-bit reference of the LG OLED code specific patterns:

1. 127/32.....7.2%
2. 254/64.....21.7%
3. 320/80.....29.2%
4. 386/97.....36.8%
5. 419/105.....40.5%
6. 451/113.....44.2%
7. 467/117.....46.0%
8. 482/121.....47.7%
9. 498/125.....49.5%
10. 513/128.....51.3%
11. 529/132.....53.1%
12. 544/136.....54.8%
13. 560/140.....56.6%
14. 575/144.....58.3%
15. 591/148.....60.2%
16. 606/152.....61.9%
17. 622/156.....63.7%
18. 637/159.....65.4%
19. 653/163.....67.2%
20. 668/167.....68.9%

- Ryan M.
Ah great, thank you!
jamesb23 is offline  
post #16 of 76 Old 09-22-2017, 10:17 PM
Member
 
jamesb23's Avatar
 
Join Date: Nov 2002
Posts: 113
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 38 Post(s)
Liked: 13
Quote:
Originally Posted by BigCoolJesus View Post
From my recent delving into the topic, I would highly recommend you purchase Masciola's HDR test pattern download (Google Masciola). All you need to do is put his test patterns on a USB stick and plug it into the LG and voila, the patterns kick the LG into HDR mode and you are free to adjust away.

Regarding Black Clipping, using HDR Standard picture mode with default Brightness 50, near black value 73 can be seen flashing from a few inches away and, if you focus really hard and block out the white light from the numbers on the pattern, value 68 can just barely be seen flashing...at least on my display. It is advised to NOT TOUCH the Brightness setting regardless of what the pattern shows on your TV as it throws other HDR values and pre-mapped points out of whack. I have found that what you see with default values is about what you are going to get aside from using a meter to calibrate the grayscale.

Regarding real world content, the Masciola download has five excellent HDR test scenes that will show obvious issues if you change settings too much.

My advice, based on playing around yesterday, is to use HDR Standard with Contrast and OLED at 100, Brightness at 50, other settings left on default (such as Tint and Color), Dynamic Contrast Off, Color Gamut Normal, and all other enhancements turned off as usual.
So to help get my black level dialed in for SDR, I did the sub127 service menu trick, which enabled me to raise my brightness to 52 on the ISF Expert Dark mode. For HDR with the masciola pattern, leaving brightness at 50 badly clips black, and in order to see 73 flash (68 barely, maybe) I have to raise brightness to 52. This undoubtedly is because of the sub127 service menu thing. But is it still bad to touch brightness in this case?

Some other weirdness that's confusing me is I'm seeing flashing well above 1000 nits on the white clipping pattern, and when measuring nits on pattern 668 in HCFR I can't get the 670 nits that's been mentioned as max when OLED and Contrast are both set to 100. Any advice?
jamesb23 is offline  
post #17 of 76 Old 09-23-2017, 01:48 AM
Senior Member
 
256k's Avatar
 
Join Date: Oct 2016
Posts: 282
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 235 Post(s)
Liked: 35
Quote:
Originally Posted by jamesb23 View Post
So to help get my black level dialed in for SDR, I did the sub127 service menu trick, which enabled me to raise my brightness to 52 on the ISF Expert Dark mode. For HDR with the masciola pattern, leaving brightness at 50 badly clips black, and in order to see 73 flash (68 barely, maybe) I have to raise brightness to 52. This undoubtedly is because of the sub127 service menu thing. But is it still bad to touch brightness in this case?

Some other weirdness that's confusing me is I'm seeing flashing well above 1000 nits on the white clipping pattern, and when measuring nits on pattern 668 in HCFR I can't get the 670 nits that's been mentioned as max when OLED and Contrast are both set to 100. Any advice?
Don't waste your time, HDR calibration for 2016 is absolutely broken, if you try to calibrate it with the codes LG gives you, and then you take look at a grayscale ramp, there's tint everywhere, and not to say that the one that really matters, 127 IRE is impossible to touch since it lead to +100% tint on that color you raise

Last edited by 256k; 09-23-2017 at 07:32 AM.
256k is offline  
post #18 of 76 Old 09-25-2017, 08:03 PM
Member
 
jamesb23's Avatar
 
Join Date: Nov 2002
Posts: 113
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 38 Post(s)
Liked: 13
Quote:
Originally Posted by ConnecTEDDD View Post
Hi, Contrast/Brightness settings in HDR are not working like we know in SDR, Brightness should left untouched, because the display gamut/tone mapping is based to these default settings.

HDR is using ST.2084 which is an absolute curve, the display has to follow specific luminance levels per digital level, according to the Dolby's golden reference numbers, so you are following to up to the luminance levels the display is capable and you clipp or roll-off (hard/soft)...(it's up to each display internal gamut/tone mapping programming)...the higher from display's peak output levels the display can't follow.

So when you have a 650nits capable display you see about up to 70.5% of the signal, the other info will be clipped or roll-off (hard/soft), it's up to each display internal gamut/tone mapping programming.

Other examples:

75.2% is 1000 nits
90.2% is 4014 nits
97.7% is 8047 nits
100% is 10000 nits

The PQ curve charts have different curve shape from the familiar ones of SDR calibration because 100% is 10.000 nits.

If you reduce Contrast/OLED from LG, you will reduce the peak output, OLED/Contrast @ 100 provides better color tracking at these default values.

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 level luminance levels, but in HDR the PQ Curve transfer function has fixed numbers from 0-10.000nits, you follow or you clip, you can't modify or use other values.

Ignore CMS adjustments and do RGB Balance adjustments only.

Each of the White Balance control in HDR has a specific +- value working range which not introduce problems to real content. The range of values are different from point-to-point. At low end controls the adjustments has to be very minimal. Play some different content and play with each control until you find up to which values you don't see problems. Use different movies/scenes to have plenty of different palettes to evaluate.

Using LG instructions for HDR calibration it will increase your color errors and provide lower peak output but better Grayscale and gamma tracking, but the color errors are more important from grayscale errors in that case.

After some testing performed using a 65E6 @ HDR mode, lowering the OLED Light, can improve the Grayscale but increase the color errors. To compare what is happening you have to take 5-Point Saturations (or 10-Point)

If you keep the Contrast/OLED @ 100, and calibrated the RGB levels for HDR, then you get an average Grayscale of about 1.6dE2000 and the errors are coming from gamma (brigher) at mid range while the RGB balance is near perfect....you get about 670nits calibrated.

If you run a 5-Point Saturation with targets DCI P3 inside a REC2020 (look CalMAN HDR workflow) then you get about ~2.0dE2000 average.

If you keep the Contrast @ 100 and reduce OLED about 80? (I don't remember the exact value...to get 540nits) and calibrated the RGB levels for HDR, then you get an average Grayscale of about ~1.0dE2000 (or lower)and the errors of gamma are reduced at mid range while the RGB balance is near perfect....you get about 540nits calibrated.

If you run a 5-Point Saturation with targets DCI P3 inside a REC2020 (look CalMAN HDR workflow) then you get about ~4.0dE2000 average.
So if I keep my contrast and OLED at 100 to minimize the dE on colors and I don't follow the LG HDR calibration values, how do I know what my luminance targets should be (aside from 670 nits at the max 668 code level for the set)? Thanks.
jamesb23 is offline  
post #19 of 76 Old 09-28-2017, 01:18 PM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,539
Mentioned: 64 Post(s)
Tagged: 1 Thread(s)
Quoted: 1733 Post(s)
Liked: 2138
Quote:
Originally Posted by jamesb23 View Post
So if I keep my contrast and OLED at 100 to minimize the dE on colors and I don't follow the LG HDR calibration values, how do I know what my luminance targets should be (aside from 670 nits at the max 668 code level for the set)? Thanks.
Hi, to these sets when you will try to fix luminance errors, this will require for you to do large adjustments of the internal controls, but when you do this, there added problems to grayscale/color ramps/shades-discoloration and other issues. For that reason you have to carefully use as less as possible adjustments, fixing only RGB balance errors, some times it will require to leave some errors because trying to fix them it will introduce problems to real content. It need a lot of practice until to find out what is the safe-working-range of each internal calibration control, which is different per each step.

If you see the latest shootouts (HDTVtest.co.uk/ Value Electronics) the calibrators used only 2-Point RGB balanced controls to calibrate the 2017 LG OLED's. (for the above reasons). Since 2016 LG's don't have 2-Point, it require more time to spent to find out how you have have good numbers at charts and no issues in real content re-production.

The display's don't have fixed response, the EOTF tracking it's changing when the display will see the content metadata. When you calibrate HDR10, you calibrate the display response to the specific metadata the patterns are sending (lets say 1000nits). If you change the metadata of the patterns to 4000 nits for example, then the response of the display to EOTF tracking will be different. Some displays are changing their response based to the MaxCLL or MaxFALL also, for that reason I have suggested a calibration using a fixed metadata, bypassing each movie metadata, so the display will have a known and fixed calibration status with fixed tracking to EOTF (and RGB balance also): LG 2017 OLED Calibration Thread and Settings

For example if in the future it will be possible 3D LUT for HDR10, you will have to bypass each movie metadata also.

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 #20 of 76 Old 09-28-2017, 01:37 PM
Member
 
jamesb23's Avatar
 
Join Date: Nov 2002
Posts: 113
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 38 Post(s)
Liked: 13
Quote:
Originally Posted by ConnecTEDDD View Post
Hi, to these sets when you will try to fix luminance errors, this will require for you to do large adjustments of the internal controls, but when you do this, there added problems to grayscale/color ramps/shades-discoloration and other issues. For that reason you have to carefully use as less as possible adjustments, fixing only RGB balance errors, some times it will require to leave some errors because trying to fix them it will introduce problems to real content. It need a lot of practice until to find out what is the safe-working-range of each internal calibration control, which is different per each step.

If you see the latest shootouts (HDTVtest.co.uk/ Value Electronics) the calibrators used only 2-Point RGB balanced controls to calibrate the 2017 LG OLED's. (for the above reasons). Since 2016 LG's don't have 2-Point, it require more time to spent to find out how you have have good numbers at charts and no issues in real content re-production.

The display's don't have fixed response, the EOTF tracking it's changing when the display will see the content metadata. When you calibrate HDR10, you calibrate the display response to the specific metadata the patterns are sending (lets say 1000nits). If you change the metadata of the patterns to 4000 nits for example, then the response of the display to EOTF tracking will be different. Some displays are changing their response based to the MaxCLL or MaxFALL also, for that reason I have suggested a calibration using a fixed metadata, bypassing each movie metadata, so the display will have a known and fixed calibration status with fixed tracking to EOTF (and RGB balance also): LG 2017 OLED Calibration Thread and Settings

For example if in the future it will be possible 3D LUT for HDR10, you will have to bypass each movie metadata also.
Sigh... HDR, especially on these LG OLEDs is a real clusterf**k. Makes me almost wish I had waited a bit longer for things to stabilize before spending all that money on my C6. Appreciate your response Ted! For now I'll follow your advice and just try to get RGB balanced while leaving OLED and Contrast at 100.

One other quick question (which I posted above but you may not have seen it) - regarding HDR brightness, I know you've said the correct point is always 50, but does that still hold true if I've lowered sub-brightness to 127 in the service menu? Using the Masciola patterns I discovered that I need a brightness of 52 in HDR to see level 73. At 50 it severely crushed, which I suspect is because of my sub127 service menu setting.
jamesb23 is offline  
post #21 of 76 Old 09-28-2017, 01:55 PM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,539
Mentioned: 64 Post(s)
Tagged: 1 Thread(s)
Quoted: 1733 Post(s)
Liked: 2138
Quote:
Originally Posted by jamesb23 View Post
Sigh... HDR, especially on these LG OLEDs is a real clusterf**k. Makes me almost wish I had waited a bit longer for things to stabilize before spending all that money on my C6. Appreciate your response Ted! For now I'll follow your advice and just try to get RGB balanced while leaving OLED and Contrast at 100.

One other quick question (which I posted above but you may not have seen it) - regarding HDR brightness, I know you've said the correct point is always 50, but does that still hold true if I've lowered sub-brightness to 127 in the service menu? Using the Masciola patterns I discovered that I need a brightness of 52 in HDR to see level 73. At 50 it severely crushed, which I suspect is because of my sub127 service menu setting.
Hi,

Contrast/Brightness settings in HDR are not working like we know in SDR, Brightness/Contrast should left untouched, because the display gamut/tone mapping is based to these default settings.

All the contrast/brightness patterns for HDR10 (from a calibration disk or from reference pattern generator) should be displayed only to see what is happening, testing different metadata for example, not for performing any adjustments. Display manufactures have to disable these 3 controls to future models (OLED Light/Contrast/Brightness) to the TV's HDR modes.

The reality from current HDR calibration status is that having some correctly encoded patterns (or a reference external HDR pattern generator) is not capable to provide you the level of color fidelity you can have with SDR calibration. So even if you calibrate that HDR Grayscale, if you use different Metadata for the patterns (Mastering Metadata/Content medata settings also) it will affect final results, also calibrating perfectly that Grayscale; it doesn't guarantee to you an acceptable final result; if you start watching HDR movies the results will not reflect to your calibration accuracy adjustment. If you calibrate 5 different models the image will look so different, there many issues about calibration, the industry has started slowly to understand these problems but they don't know how to solve that.

There limited operations you can do, only Grayscale/RGB balance adjustments, but even doing this, there un-alighment of calibration controls with patterns.

Generally once you will find out what adjustment you have to do which will not introduce problems to real content, you will be fine, but it will take some additional time....and you need to verify looking various movies and different scenes.

This a calibration report from HDR10 calibration from E6:



There no issues in shades looking real content playback (beyond charts), but if you try to make even one value adjustment to 288 step to improve the RGB balance for example, problem will be introduced to real content (but the chart will be better).

I don't suggest any adjustment in service menu. (except disabling ASBL)

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 #22 of 76 Old 09-28-2017, 02:20 PM
Member
 
jamesb23's Avatar
 
Join Date: Nov 2002
Posts: 113
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 38 Post(s)
Liked: 13
Quote:
Originally Posted by ConnecTEDDD View Post
Hi,

Contrast/Brightness settings in HDR are not working like we know in SDR, Brightness/Contrast should left untouched, because the display gamut/tone mapping is based to these default settings.

All the contrast/brightness patterns for HDR10 (from a calibration disk or from reference pattern generator) should be displayed only to see what is happening, testing different metadata for example, not for performing any adjustments. Display manufactures have to disable these 3 controls to future models (OLED Light/Contrast/Brightness) to the TV's HDR modes.

The reality from current HDR calibration status is that having some correctly encoded patterns (or a reference external HDR pattern generator) is not capable to provide you the level of color fidelity you can have with SDR calibration. So even if you calibrate that HDR Grayscale, if you use different Metadata for the patterns (Mastering Metadata/Content medata settings also) it will affect final results, also calibrating perfectly that Grayscale; it doesn't guarantee to you an acceptable final result; if you start watching HDR movies the results will not reflect to your calibration accuracy adjustment. If you calibrate 5 different models the image will look so different, there many issues about calibration, the industry has started slowly to understand these problems but they don't know how to solve that.

There limited operations you can do, only Grayscale/RGB balance adjustments, but even doing this, there un-alighment of calibration controls with patterns.

Generally once you will find out what adjustment you have to do which will not introduce problems to real content, you will be fine, but it will take some additional time....and you need to verify looking various movies and different scenes.

This a calibration report from HDR10 calibration from E6:



There no issues in shades looking real content playback (beyond charts), but if you try to make even one value adjustment to 288 step to improve the RGB balance for example, problem will be introduced to real content (but the chart will be better).

I don't suggest any adjustment in service menu. (except disabling ASBL)
Hmmm. So I've made things worse in HDR by adjusting sub brightness then? It seemed to work very well to help with my SDR calibration. I guess my other option would be to go back into service menu each time to put the sub-B back at 128 before I watch any HDR content, although that's a bit of a pain. This HDR stuff is giving me a massive headache. At least other TVs use regular grayscale, not this LG code nonsense.
jamesb23 is offline  
post #23 of 76 Old 09-28-2017, 02:30 PM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,539
Mentioned: 64 Post(s)
Tagged: 1 Thread(s)
Quoted: 1733 Post(s)
Liked: 2138
Quote:
Originally Posted by jamesb23 View Post
Hmmm. So I've made things worse in HDR by adjusting sub brightness then? It seemed to work very well to help with my SDR calibration. I guess my other option would be to go back into service menu each time to put the sub-B back at 128 before I watch any HDR content, although that's a bit of a pain. This HDR stuff is giving me a massive headache. At least other TVs use regular grayscale, not this LG code nonsense.
Any brightness adjustment in HDR10 mode it will provide you de-saturation in dark area of colors, except the major shifting of PQ transfer function tracking (you will see stuff supposed not to be seen, lifted near black...added noise)

In SDR mode, at your own risk, you can use Service Mode for the adjustments you are talking. (which I don't recommend).

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 #24 of 76 Old 09-28-2017, 02:57 PM
AVS Forum Special Member
 
WiFi-Spy's Avatar
 
Join Date: Feb 2004
Location: Seattle, WA
Posts: 2,809
Mentioned: 58 Post(s)
Tagged: 0 Thread(s)
Quoted: 521 Post(s)
Liked: 966
You can use the medium color temp for HDR and then adjust the medium 2pt white balance in the service menu for HDR.

Tyler Pruitt - Technical Liaison - CalMAN

10 Bit Gradient Test Patterns (HEVC) - Free Download
WiFi-Spy is offline  
post #25 of 76 Old 09-29-2017, 07:31 AM
Member
 
jamesb23's Avatar
 
Join Date: Nov 2002
Posts: 113
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 38 Post(s)
Liked: 13
Quote:
Originally Posted by WiFi-Spy View Post
You can use the medium color temp for HDR and then adjust the medium 2pt white balance in the service menu for HDR.
Interesting, I may try that, just have to find where that is in service menu - is it clearly labeled as medium 2pt white balance? Will it affect my SDR calibration, which I've already completed? Also, if I do this, which code patterns should use? Typically in 2pt you look at 80 and 30 IRE or something like that, so I'm guessing codes 451 and 606 or around there?

EDIT: I think I found the 2pt white balance in the SM now. But, I'm still wondering if the adjustments would affect my SDR settings? Or only if I use medium on my SDR content? Also, what's the advantage of using medium color temp to begin with? Isn't that too cool? I've always used Warm2...

@ConnecTEDDD any thoughts on the above?

Last edited by jamesb23; 09-29-2017 at 08:58 AM.
jamesb23 is offline  
post #26 of 76 Old 10-03-2017, 09:16 AM
Member
 
toddrain1's Avatar
 
Join Date: Nov 2006
Posts: 155
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 62 Post(s)
Liked: 53
Quote:
Originally Posted by WiFi-Spy View Post
You can use the medium color temp for HDR and then adjust the medium 2pt white balance in the service menu for HDR.
Thanks for this advice Tyler
I used that method and it worked extremely well. I used the medium temperature like you said, so it doesn't mess Warm 2 for Dolby vision and SDR. I was able to get almost flat 6500k grayscale. I used the High point controls to get the upper end of the grayscale flat, upper end had excessive blue in my case. Then used the low point controls to get the lower grayscale reasonable, too much adjustment with that really effects the very bottom of the grayscale and 127 control does not work properly. So after that I used the 20 point adjustment to get code values 254 to 529 on target. I'm really glad I tried it because up until now HDR never had the color accuracy of SDR with just the HDR 20 point control.
WiFi-Spy likes this.
toddrain1 is offline  
post #27 of 76 Old 10-03-2017, 12:00 PM
Member
 
T( )( )L's Avatar
 
Join Date: May 2008
Location: Sweden
Posts: 184
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 81 Post(s)
Liked: 41
This goes for the 2017 models right?

Calman Enthusiast,Lightspace HTP,Klein K-10A,I1pro2,Murideo 6G.
T( )( )L is offline  
post #28 of 76 Old 10-03-2017, 12:05 PM
Member
 
toddrain1's Avatar
 
Join Date: Nov 2006
Posts: 155
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 62 Post(s)
Liked: 53
Quote:
Originally Posted by T( )( )L View Post
This goes for the 2017 models right?
I'm not sure, I have a 2016 65B6.
toddrain1 is offline  
post #29 of 76 Old 10-03-2017, 12:09 PM
Member
 
T( )( )L's Avatar
 
Join Date: May 2008
Location: Sweden
Posts: 184
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 81 Post(s)
Liked: 41
Same as me then,will have a look at it tomorrow

Calman Enthusiast,Lightspace HTP,Klein K-10A,I1pro2,Murideo 6G.
T( )( )L is offline  
post #30 of 76 Old 10-03-2017, 12:53 PM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 5,539
Mentioned: 64 Post(s)
Tagged: 1 Thread(s)
Quoted: 1733 Post(s)
Liked: 2138
Quote:
Originally Posted by toddrain1 View Post
Thanks for this advice Tyler
I used that method and it worked extremely well. I used the medium temperature like you said, so it doesn't mess Warm 2 for Dolby vision and SDR. I was able to get almost flat 6500k grayscale. I used the High point controls to get the upper end of the grayscale flat, upper end had excessive blue in my case. Then used the low point controls to get the lower grayscale reasonable, too much adjustment with that really effects the very bottom of the grayscale and 127 control does not work properly. So after that I used the 20 point adjustment to get code values 254 to 529 on target. I'm really glad I tried it because up until now HDR never had the color accuracy of SDR with just the HDR 20 point control.
Hi, do you have measurement to show that using 2-Point HDR10 RGB balance from service menu provide better color accurancy from 20-Point HDR10 RGB balance with saturation and colorchecker?

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  
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