Spears & Munsil UHD HDR Benchmark Disc - Discussion - Page 11 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 150Likes
Reply
 
Thread Tools
post #301 of 376 Old 08-12-2019, 01:33 PM
AVS Forum Special Member
 
sspears's Avatar
 
Join Date: Feb 1999
Location: Sammamish, WA, USA
Posts: 5,250
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 272 Post(s)
Liked: 460
Quote:
Originally Posted by Dominic Chan View Post
Shouldn’t tone mapping affect only the highlights, without changing the overall brightness?
Short answer is no.

First there is the bottom end tone mapping. e.g. you grade on an OLED with 0.0001 black level and then playback on an LCD with 0.05 black level. This may impact your shadow detail.

There are many schools of thought on tone mapping. One is to only tone map the highlights. You track PQ as high as possible and then roll-off only the highlights. This can result in what looks like clipped highlights as you end up compressing just the highlights and given the limited range to compress them, it looks like clipping. To avoid this, you start tone mapping much sooner, e.g. you my actually start rolling off around 100 nits.

Another school of thought is you also pull down the mid tones, which results in an overall darker image, but more highlight detail is preserved smoothly.

Which one is correct? Since there is no standard, there is no correct. Just different approaches.

The lower the brightness capability of the display, the more important dynamic tone mapping becomes.

We are really at the beginning of HDR. Much will change over the coming years as more experience is gained and more content is created.

Stacey Spears
Co-Creator, Spears & Munsil UHD HDR Benchmark

Last edited by sspears; 08-12-2019 at 01:39 PM.
sspears is offline  
Sponsored Links
Advertisement
 
post #302 of 376 Old 08-12-2019, 01:37 PM
Advanced Member
 
zeonstar's Avatar
 
Join Date: Sep 2005
Posts: 703
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 487 Post(s)
Liked: 115
Quote:
Originally Posted by sspears View Post
Depends on how it was calibrated. If it uses the CalMAN approach by replacing the 1D and 3D LUT in the display, then DTM will work just fine. In the CalMAN calibration of the LG, it turns off tone mapping, so it does not get in the way of calibration. No other HDR display lets you disable tone mapping during calibration that I am aware of.

With the C8 (8 series) you measure the peak brightness of the panel and that is used for Dolby Vision. For HDR10, the panel brightness is a fixed value regardless of panel capability.

For the 9 series, you measure the panel peak for HDR10 just like Dolby Vision. So you get to use the full brightness of the panel for HDR10, just like Dolby Vision.

Here is a really old xls with HDR metadata. This must have been pre Disney support of UHD BD since I don't see any Disney titles in it.

Here are the tone mapping curves in the first gen Panasonic OLED based on different metadata settings. Same thing for an older Samsung display.
My TV was Calibrated with CalMAN by my calibrator but I couldn't tell you if he did that 1D and 3D LUT thing. The technical aspects of Calibration are well beyond me at this point. But like you said, it was most definitely off during Calibration.


***

I've gotten a bit of conflicting info on this topic so I'd like to summarize:


LG C8 Tone Mapping does not really affect a calibrated TV negatively. But...

In order to preserve a creator's intent, DTM should be OFF.

Using the HDR Optimizer from a Panasonic Player + The TV's standard Tone Mapping (Not DTM) should preserve the creator's intent more closely.

Last edited by zeonstar; 08-12-2019 at 01:54 PM.
zeonstar is offline  
post #303 of 376 Old 08-12-2019, 01:48 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,647
Mentioned: 126 Post(s)
Tagged: 0 Thread(s)
Quoted: 4654 Post(s)
Liked: 1728
Quote:
Originally Posted by sspears View Post
There are many schools of thought on tone mapping. One is to only tone map the highlights. You track PQ as high as possible and then roll-off only the highlights. This can result in what looks like clipped highlights as you end up compressing just the highlights and given the limited range to compress them, it looks like clipping. To avoid this, you start tone mapping much sooner, e.g. you my actually start rolling off around 100 nits.

Another school of thought is you also pull down the mid tones, which results in an overall darker image, but more highlight detail is preserved smoothly.

Which one is correct? Since there is no standard, there is no correct. Just different approaches.
BT.2390 seems to be recommending the first approach.
Dominic Chan is online now  
Sponsored Links
Advertisement
 
post #304 of 376 Old 08-12-2019, 01:51 PM
AVS Forum Special Member
 
sspears's Avatar
 
Join Date: Feb 1999
Location: Sammamish, WA, USA
Posts: 5,250
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 272 Post(s)
Liked: 460
Quote:
Originally Posted by Dominic Chan View Post
BT.2390 seems to be recommending the first approach.
BT.2390 is simplistic. That is not a comment on which approach is better. Just pointing out how basic BT.2390 is.

Stacey Spears
Co-Creator, Spears & Munsil UHD HDR Benchmark
sspears is offline  
post #305 of 376 Old 08-12-2019, 01:54 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,647
Mentioned: 126 Post(s)
Tagged: 0 Thread(s)
Quoted: 4654 Post(s)
Liked: 1728
Quote:
Originally Posted by sspears View Post
BT.2390 is simplistic. That is not a comment on which approach is better. Just pointing out how basic BT.2390 is.
I believe the Panasonic players also tone-map the highlights only, although they also allow the mid-tone to be adjusted manually.
Dominic Chan is online now  
post #306 of 376 Old 08-12-2019, 03:55 PM
AVS Forum Addicted Member
 
Kris Deering's Avatar
 
Join Date: Oct 2002
Location: The Pacific Northwet
Posts: 10,701
Mentioned: 189 Post(s)
Tagged: 0 Thread(s)
Quoted: 3838 Post(s)
Liked: 6540
Quote:
Originally Posted by Dominic Chan View Post
I believe the Panasonic players also tone-map the highlights only, although they also allow the mid-tone to be adjusted manually.
No, they balance the tone map based on not only the MaxCLL but also the dynamic range specified. So it will change depending on where you set it. You can't do a single curve as you don't want to waste dynamic range. And you can't just do highlights only because APL greatly effects the whole range. A single tone map for a movie wastes dynamic range to appease a set MaxCLL that may apply to literally a single frame of a movie. With a flat panel this may not mean much, but with a projector it certainly does.
monkaquinas likes this.

My Home Theater UPDATED DEC 2017
Technical Editor/Writer Sound and Vision Magazine
Deep Dive AV - Calibration, Consulting and Education
Kris Deering is offline  
post #307 of 376 Old 08-13-2019, 05:41 AM
Advanced Member
 
Join Date: Jul 2017
Posts: 584
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 402 Post(s)
Liked: 197
Quote:
Originally Posted by sspears View Post
Yes, green only mode on HD color bars will show it.

This problem was common years ago on everyone's displays, but its usually correct on displays these days.
Hi @sspears where may I find more infos about this topic about LG 2017 wrong color conversion?

I never heard about this in both owner and calibration threads.

Does it affect ycbcr on HDMI only or any kind of input or streaming app?

LG OLED 55 B7v | LG SK10Y Soundbar | Samsung 55KS8000 | LG Plasma 50PZ550 | i1D3 OEM | Calman Enthusiast
MrRobotoPlus is offline  
post #308 of 376 Old 08-13-2019, 06:19 AM
AVS Forum Special Member
 
Join Date: Nov 2014
Location: New York
Posts: 5,349
Mentioned: 185 Post(s)
Tagged: 0 Thread(s)
Quoted: 4381 Post(s)
Liked: 5373
Quote:
Originally Posted by zeonstar View Post
My TV was Calibrated with CalMAN by my calibrator but I couldn't tell you if he did that 1D and 3D LUT thing. The technical aspects of Calibration are well beyond me at this point. But like you said, it was most definitely off during Calibration.


***

I've gotten a bit of conflicting info on this topic so I'd like to summarize:


LG C8 Tone Mapping does not really affect a calibrated TV negatively. But...

In order to preserve a creator's intent, DTM should be OFF.

Using the HDR Optimizer from a Panasonic Player + The TV's standard Tone Mapping (Not DTM) should preserve the creator's intent more closely.
For a flat panel display at a high level, you are correct. You need to separate the calibration of white point, gray scale, CMS from tone mapping. On the LG when you calibrate it with CalMAN autocal, the tone mapping is turned off. Once done, it's turned back on either DTM or not. Because there is not a fully developed tone mapping standard every manufacturer does what it feels is the best way to tone map. Vincent's video goes over this in detail. As Kris points out, tone mapping considerations are different with flat panel displays and projectors.

At the end of the day, stop worrying which setting on your OLED is correct and just set DTM on or off based on how you personally perceive the picture is best overall. My personal experience is if your set is not calibrated, use DTM On. If it's calibrated, I personally like DTM Off. Also, as D-Nice pointed out when we were initially evaluating the 2018 and 2019 LG OLEDs, LG initially wasn't following the EOTF curve 100% to make the picture look brighter. Samsung TVs are really a lot off. Although this can be corrected with calibration, I've seen LG improve their ootb HDR with subsequent firmware updates.

John
Sony 55A1E, A9F / LG 55OLEDC8
Marantz 7012, Ohm Walsh Speakers
Klein K10-A, Jeti 1501, Murideo Six-G Gen2
Calman Ultimate, ISF Level III Certified
jrref is offline  
post #309 of 376 Old 08-13-2019, 06:42 AM
AVS Forum Special Member
 
sspears's Avatar
 
Join Date: Feb 1999
Location: Sammamish, WA, USA
Posts: 5,250
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 272 Post(s)
Liked: 460
Quote:
Originally Posted by MrRobotoPlus View Post
Hi @sspears where may I find more infos about this topic about LG 2017 wrong color conversion?

I never heard about this in both owner and calibration threads.

Does it affect ycbcr on HDMI only or any kind of input or streaming app?
You can fix the color decoding on the 2017, but not sure if gamut is wrong after. You may need to use CMS to fix the gamut. Go to Expert Controls -> Colour Gamut and change it to Wide and then color decoding for SDR is correct.

To verify this change, put up SMPTE color bars. In green only mode, you will see four solid bars of green. Change it to extended or auto and you now have four bars of green.

On the 8 and 9 series, you don't need to change colour gamut.
venkatesh_m and MrRobotoPlus like this.

Stacey Spears
Co-Creator, Spears & Munsil UHD HDR Benchmark
sspears is offline  
post #310 of 376 Old 08-13-2019, 06:51 AM
Advanced Member
 
Join Date: Jul 2017
Posts: 584
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 402 Post(s)
Liked: 197
Quote:
Originally Posted by sspears View Post
You can fix the color decoding on the 2017, but not sure if gamut is wrong after. You may need to use CMS to fix the gamut. Go to Expert Controls -> Colour Gamut and change it to Wide and then color decoding for SDR is correct.

To verify this change, put up SMPTE color bars. In green only mode, you will see four solid bars of green. Change it to extended or auto and you now have four bars of green.

On the 8 and 9 series, you don't need to change colour gamut.
OK, last year I did calibrate the set myself following D-Nice advises. So, my color gamut is set to Extended and color 43 (or 46 don't remember). I never noticed anything wrong with this settings within Calman. BTW this is one of the most interesting thread since months, thank you

LG OLED 55 B7v | LG SK10Y Soundbar | Samsung 55KS8000 | LG Plasma 50PZ550 | i1D3 OEM | Calman Enthusiast
MrRobotoPlus is offline  
post #311 of 376 Old 08-13-2019, 07:13 AM
AVS Forum Special Member
 
sspears's Avatar
 
Join Date: Feb 1999
Location: Sammamish, WA, USA
Posts: 5,250
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 272 Post(s)
Liked: 460
Quote:
Originally Posted by MrRobotoPlus View Post
OK, last year I did calibrate the set myself following D-Nice advises. So, my color gamut is set to Extended and color 43 (or 46 don't remember). I never noticed anything wrong with this settings within Calman. BTW this is one of the most interesting thread since months, thank you
I just ran some tests. Just to be clear, this should only apply to a 6 or 7 series. 8 and 9 I recall being correct, but it is easy for anyone to verify my results. This is also for SDR, not HDR. HDR is fine on the 7 series.

I am currently in technicolor expert mode and have gamma set to BT.1886. In other modes, the bars will appear even more different (error). Dynamic contrast is off and edge enhancer is on (on is the off position on the 7 series)

1. Put up some SMPTE bars. I am using the SMPTE bars from the 2nd edition disc since they are HD resolution and 709. Wanted to eliminate UHD being an issue. (EDIT: I did go back to the UHD HDR version and verify things match between the 2nd edition and UHD HDR edition when looking at the SDR BT.709 bars. Just wanted to confirm resolution or disc format was not the cause.)
2. Go to Expert Controls -> Colour Filter and set to green.

If SDR BT.709 color decoding is correct, then you should see what looks like one big massive green bar. Four of the bars turn green and should all match in level.

At default settings, you see one green, followed by a lighter green, followed by two greens. So you can see three different levels of green.

If you go to Expert Controls -> Colour Gamut and change it to Wide, the three green levels become one, so all four bars match. More importantly, if you go to our Color and Tint Green Only bars on the 2nd edition, the right half is green and the left half is black.

If you don't have the 2nd edition and do have the UHD HDR edition, then go to color and tint bars in SDR BT.709. In blue only mode, the left has colors leaking through instead of being black in default. When you change to Wide, the left half becomes black. HD color bars under SDR BT.709 will let you verify the four bars are green or not.

SDR BT.709 -> Setup -> Color and Tint for the blue only version of our color and tint bars
SDR BT.709 -> Evaluation -> HD Color Bars for the SMPTE style HD color bars

Just in case you have no idea what I am talking about by left and right half or left half turning black vs. showing color, here is our article that explains it. It is how you know the color filter or blue only mode is working correctly.

Just for the fun of it, I tried using extended mode and setting color to 43 and 46 and they are way off. I had to actually set tint to G19 for tint to be correct, but then there was one green bar that stands WAY out when looking at the SMPTE / HD color bars in green only mode.
gomo657 and MrRobotoPlus like this.

Stacey Spears
Co-Creator, Spears & Munsil UHD HDR Benchmark
sspears is offline  
post #312 of 376 Old 08-13-2019, 07:24 AM
AVS Forum Special Member
 
sspears's Avatar
 
Join Date: Feb 1999
Location: Sammamish, WA, USA
Posts: 5,250
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 272 Post(s)
Liked: 460
Quote:
Originally Posted by Dominic Chan View Post
I believe the Panasonic players also tone-map the highlights only, although they also allow the mid-tone to be adjusted manually.
What I might do, though not sure I will have time this week, is to go through the various HDR Optimizer modes and plot the tone map curve. This will at least tell us when the roll-off starts and how steep it is.

If Kris is correct, then I would also need to measure the various nit level versions too. That is a lot of data.

Probably should also measure some luminescence loading to see if the numbers change based on APL.
chunon and monkaquinas like this.

Stacey Spears
Co-Creator, Spears & Munsil UHD HDR Benchmark

Last edited by sspears; 08-13-2019 at 07:27 AM.
sspears is offline  
post #313 of 376 Old 08-13-2019, 07:31 AM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,647
Mentioned: 126 Post(s)
Tagged: 0 Thread(s)
Quoted: 4654 Post(s)
Liked: 1728
Quote:
Originally Posted by sspears View Post
What I might do, though not sure I will have time this week, is to go through the various HDR Optimizer modes and plot the tone map curve. This will at least tell us when the roll-off starts and how steep it is.
That will be really useful. I have not seen any Panasonic tone map curves plotted.

Presumably the mid-tone will depend heavily on the Dynamic Range setting, while the roll-off will depend on the Display Type.

Looking forward to your results, and thank you in advance!
Dominic Chan is online now  
post #314 of 376 Old 08-13-2019, 08:24 AM
AVS Forum Special Member
 
sspears's Avatar
 
Join Date: Feb 1999
Location: Sammamish, WA, USA
Posts: 5,250
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 272 Post(s)
Liked: 460
Here is the curve with display type set to OLED.

I ran through all nit versions of the patterns with HDR Optimizer on and set to default settings. The curve did not change when the APL changed. (luminance loading)

0, 5, 10, ..., 100 is the window level. The numbers are the code values. The baseline numbers are the default code values for 0-100%.

I used the Accupel DGA-6000 HDMI Analyzer to grab the pixel values coming out of the Panasonic DP-UB9000. The cd/m² is the HDR version of the pattern. This is the max mastering display luminance as well as the MaxCLL value.

I really don't want to measure every possible combination of settings. If someone wants some specific settings measured, let me know. Give me all the settings to change and I can re-measure all of the windows.
chunon and Dominic Chan like this.

Stacey Spears
Co-Creator, Spears & Munsil UHD HDR Benchmark

Last edited by sspears; 08-13-2019 at 08:34 AM.
sspears is offline  
post #315 of 376 Old 08-13-2019, 08:49 AM
Advanced Member
 
zeonstar's Avatar
 
Join Date: Sep 2005
Posts: 703
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 487 Post(s)
Liked: 115
Would it be too much to ask to have the result explained in laymans terms? It's great you did those tests, so thank you.

What does the results say about Panasonic's HDR Optimizer?
zeonstar is offline  
post #316 of 376 Old 08-13-2019, 08:55 AM
AVS Forum Special Member
 
sspears's Avatar
 
Join Date: Feb 1999
Location: Sammamish, WA, USA
Posts: 5,250
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 272 Post(s)
Liked: 460
Quote:
Originally Posted by zeonstar View Post
Would it be too much to ask to have the result explained in laymans terms? It's great you did those tests, so thank you.

What does the results say about Panasonic's HDR Optimizer?
It just tells you when the roll-off starts and how it changes its shape based on the sources HDR metadata. The roll-off starts above 100 nits. (code value 509 would be ~100 nits).

Stacey Spears
Co-Creator, Spears & Munsil UHD HDR Benchmark
sspears is offline  
post #317 of 376 Old 08-13-2019, 08:57 AM
Advanced Member
 
Join Date: Jul 2017
Posts: 584
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 402 Post(s)
Liked: 197
Quote:
Originally Posted by sspears View Post
I just ran some tests. Just to be clear, this should only apply to a 6 or 7 series. 8 and 9 I recall being correct, but it is easy for anyone to verify my results. This is also for SDR, not HDR. HDR is fine on the 7 series.

I am currently in technicolor expert mode and have gamma set to BT.1886. In other modes, the bars will appear even more different (error). Dynamic contrast is off and edge enhancer is on (on is the off position on the 7 series)

1. Put up some SMPTE bars. I am using the SMPTE bars from the 2nd edition disc since they are HD resolution and 709. Wanted to eliminate UHD being an issue. (EDIT: I did go back to the UHD HDR version and verify things match between the 2nd edition and UHD HDR edition when looking at the SDR BT.709 bars. Just wanted to confirm resolution or disc format was not the cause.)
2. Go to Expert Controls -> Colour Filter and set to green.

If SDR BT.709 color decoding is correct, then you should see what looks like one big massive green bar. Four of the bars turn green and should all match in level.

At default settings, you see one green, followed by a lighter green, followed by two greens. So you can see three different levels of green.

If you go to Expert Controls -> Colour Gamut and change it to Wide, the three green levels become one, so all four bars match. More importantly, if you go to our Color and Tint Green Only bars on the 2nd edition, the right half is green and the left half is black.

If you don't have the 2nd edition and do have the UHD HDR edition, then go to color and tint bars in SDR BT.709. In blue only mode, the left has colors leaking through instead of being black in default. When you change to Wide, the left half becomes black. HD color bars under SDR BT.709 will let you verify the four bars are green or not.

SDR BT.709 -> Setup -> Color and Tint for the blue only version of our color and tint bars
SDR BT.709 -> Evaluation -> HD Color Bars for the SMPTE style HD color bars

Just in case you have no idea what I am talking about by left and right half or left half turning black vs. showing color, here is our article that explains it. It is how you know the color filter or blue only mode is working correctly.

Just for the fun of it, I tried using extended mode and setting color to 43 and 46 and they are way off. I had to actually set tint to G19 for tint to be correct, but then there was one green bar that stands WAY out when looking at the SMPTE / HD color bars in green only mode.
This was great, thank you for your time.

In the meanwhile I've found an old Chadb post where he was talking about this problem. Wide fixes things but the only way to fix skin tones and low saturation colors (the most used ones in real content) for a correct calibration is Auto and bumping color (chad method) or Extended and lowering Color (D-nice method). There's no way to get the right saturation with Wide.

I did forget about this. It was discussed 2 years ago

LG OLED 55 B7v | LG SK10Y Soundbar | Samsung 55KS8000 | LG Plasma 50PZ550 | i1D3 OEM | Calman Enthusiast
MrRobotoPlus is offline  
post #318 of 376 Old 08-13-2019, 09:01 AM
AVS Forum Special Member
 
sspears's Avatar
 
Join Date: Feb 1999
Location: Sammamish, WA, USA
Posts: 5,250
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 272 Post(s)
Liked: 460
Quote:
Originally Posted by MrRobotoPlus View Post
This was great, thank you for your time.

In the meanwhile I've found an old Chadb post where he was talking about this problem. Wide fixes things but the only way to fix skin tones and low saturation colors (the most used ones in real content) for a correct calibration is Auto and bumping color (chad method) or Extended and lowering Color (D-nice method). There's no way to get the right saturation with Wide.

I did forget about this. It was discussed 2 years ago
The real fix, of course, is a new display. LOL Not usually the answer people want to hear. Sadly the 9 series removed the colour filter option to verify color decoding, so it was the C8 where I verified it was correct for SDR.

Wide puts the panel into native mode, so the gamut is larger than 709. Fix one thing and break another.
MrRobotoPlus likes this.

Stacey Spears
Co-Creator, Spears & Munsil UHD HDR Benchmark

Last edited by sspears; 08-13-2019 at 09:16 AM.
sspears is offline  
post #319 of 376 Old 08-13-2019, 09:50 AM
Advanced Member
 
zeonstar's Avatar
 
Join Date: Sep 2005
Posts: 703
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 487 Post(s)
Liked: 115
Quote:
Originally Posted by sspears View Post
It just tells you when the roll-off starts and how it changes its shape based on the sources HDR metadata. The roll-off starts above 100 nits. (code value 509 would be ~100 nits).
Thanks. Graph made more sense to me that I expected it to with my limited understand of all this. But it basically looks like the HDR Optimizer messes with the curve as little as possible. Which is great.
zeonstar is offline  
post #320 of 376 Old 08-13-2019, 10:23 AM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,647
Mentioned: 126 Post(s)
Tagged: 0 Thread(s)
Quoted: 4654 Post(s)
Liked: 1728
Quote:
Originally Posted by sspears View Post
I really don't want to measure every possible combination of settings. If someone wants some specific settings measured, let me know. Give me all the settings to change and I can re-measure all of the windows.
Thank you so much Stacey. It will take me some time to digest this, but if it’s not too much to ask, can you please also measure the following two separate sets:
- same settings as what was already measured, but with Display Type set to Basic Luminance (500 nits)
- as above, but with the Dynamic Range adjustment boosted to some suitable level.

I believe the three sets together will give people a very good idea of the behavior of the HDR Optimizer.

Thanks again.
shs1234 likes this.
Dominic Chan is online now  
post #321 of 376 Old 08-13-2019, 01:55 PM
AVS Forum Addicted Member
 
Kris Deering's Avatar
 
Join Date: Oct 2002
Location: The Pacific Northwet
Posts: 10,701
Mentioned: 189 Post(s)
Tagged: 0 Thread(s)
Quoted: 3838 Post(s)
Liked: 6540
Quote:
Originally Posted by sspears View Post
Here is the curve with display type set to OLED.

I ran through all nit versions of the patterns with HDR Optimizer on and set to default settings. The curve did not change when the APL changed. (luminance loading)

0, 5, 10, ..., 100 is the window level. The numbers are the code values. The baseline numbers are the default code values for 0-100%.

I used the Accupel DGA-6000 HDMI Analyzer to grab the pixel values coming out of the Panasonic DP-UB9000. The cd/m² is the HDR version of the pattern. This is the max mastering display luminance as well as the MaxCLL value.

I really don't want to measure every possible combination of settings. If someone wants some specific settings measured, let me know. Give me all the settings to change and I can re-measure all of the windows.
That is what I would expect. The curve is based on the MaxCLL being reported, and since your disc is actually changing that data, it works well. The default assumed brightness used for that tone map is 800 nits (OLED). You are right that it does not change at all based on APL, that is where the slider comes into play. That assigns how much dynamic range you actually have to work with. So with the slider at default, you are working with 800 nits. If you raise it you are lowering that number (and the image gets brighter due to compression). If you lower it you are telling the player you have MORE nits to work with and the image gets darker (less dynamic range compression and more saturation). The key to a static tone map like this is to find the best balance of dynamic range to overall brightness. There will always be a trade-off, the key is finding the best balance of the two. But that also means it applies on a title by title basis because content is all over the place in this regard.

Again, I don't see this as much of an issue for OLED owners because you already have a decent amount of range to work with by default. I would probably raise the number slightly because most OLEDs don't really have 800 nits to work with, but even in default it shouldn't be too bad. When you start doing this with projectors it is another beast entirely!
jrref likes this.

My Home Theater UPDATED DEC 2017
Technical Editor/Writer Sound and Vision Magazine
Deep Dive AV - Calibration, Consulting and Education
Kris Deering is offline  
post #322 of 376 Old 08-14-2019, 07:29 AM
AVS Forum Special Member
 
RickD_99's Avatar
 
Join Date: Nov 1999
Location: Birmingham, AL USA
Posts: 1,540
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 292 Post(s)
Liked: 200
Quote:
Originally Posted by sspears View Post
I just ran some tests. Just to be clear, this should only apply to a 6 or 7 series. 8 and 9 I recall being correct, but it is easy for anyone to verify my results. This is also for SDR, not HDR. HDR is fine on the 7 series.

I am currently in technicolor expert mode and have gamma set to BT.1886. In other modes, the bars will appear even more different (error). Dynamic contrast is off and edge enhancer is on (on is the off position on the 7 series)

1. Put up some SMPTE bars. I am using the SMPTE bars from the 2nd edition disc since they are HD resolution and 709. Wanted to eliminate UHD being an issue. (EDIT: I did go back to the UHD HDR version and verify things match between the 2nd edition and UHD HDR edition when looking at the SDR BT.709 bars. Just wanted to confirm resolution or disc format was not the cause.)
2. Go to Expert Controls -> Colour Filter and set to green.

If SDR BT.709 color decoding is correct, then you should see what looks like one big massive green bar. Four of the bars turn green and should all match in level.

At default settings, you see one green, followed by a lighter green, followed by two greens. So you can see three different levels of green.
Stacey I just performed the test you referenced above using your 2nd edition SMPTE bars and the green filter on my 2016 C6 OLED and I see only one big green bar...so maybe the errors you were referencing apply only to the 2017 models???

LG 55C6P OLED Display
Oppo UDP-203 UHD Blu-ray Player (65-0131 firmware)
ELAC unifi B5 Speakers
NAD C316BEE Stereo Amplifier
Nakamichi Shockwafe 9.2 Sound Bar (3.2 firmware)/Xbox One X 4K Gaming Console/Apple TV 4K
RickD_99 is offline  
post #323 of 376 Old 08-14-2019, 09:55 AM
AVS Forum Special Member
 
sspears's Avatar
 
Join Date: Feb 1999
Location: Sammamish, WA, USA
Posts: 5,250
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 272 Post(s)
Liked: 460
Quote:
Originally Posted by Dominic Chan View Post
Thank you so much Stacey. It will take me some time to digest this, but if it’s not too much to ask, can you please also measure the following two separate sets:
- same settings as what was already measured, but with Display Type set to Basic Luminance (500 nits)
- as above, but with the Dynamic Range adjustment boosted to some suitable level.

I believe the three sets together will give people a very good idea of the behavior of the HDR Optimizer.

Thanks again.
I added Basic Luminance LCD to the xlsx. It is on the second tab.
Dominic Chan likes this.

Stacey Spears
Co-Creator, Spears & Munsil UHD HDR Benchmark
sspears is offline  
post #324 of 376 Old 08-14-2019, 09:56 AM
AVS Forum Special Member
 
sspears's Avatar
 
Join Date: Feb 1999
Location: Sammamish, WA, USA
Posts: 5,250
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 272 Post(s)
Liked: 460
Quote:
Originally Posted by RickD_99 View Post
Stacey I just performed the test you referenced above using your 2nd edition SMPTE bars and the green filter on my 2016 C6 OLED and I see only one big green bar...so maybe the errors you were referencing apply only to the 2017 models???
I know the B6 is wrong. C6 should be the same. What is your colour gamut setting set to? If it is correct though, that is a good thing.

Stacey Spears
Co-Creator, Spears & Munsil UHD HDR Benchmark
sspears is offline  
post #325 of 376 Old 08-14-2019, 10:12 AM
AVS Forum Special Member
 
RickD_99's Avatar
 
Join Date: Nov 1999
Location: Birmingham, AL USA
Posts: 1,540
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 292 Post(s)
Liked: 200
Quote:
Originally Posted by sspears View Post
I know the B6 is wrong. C6 should be the same. What is your colour gamut setting set to? If it is correct though, that is a good thing.
My color gamut is set to normal. The B6 has a different SOC than the C6/E6 models from 2016 so that probably accounts for the different results...

LG 55C6P OLED Display
Oppo UDP-203 UHD Blu-ray Player (65-0131 firmware)
ELAC unifi B5 Speakers
NAD C316BEE Stereo Amplifier
Nakamichi Shockwafe 9.2 Sound Bar (3.2 firmware)/Xbox One X 4K Gaming Console/Apple TV 4K
RickD_99 is offline  
post #326 of 376 Old 08-14-2019, 11:14 AM
AVS Forum Special Member
 
thirdkind's Avatar
 
Join Date: Mar 2002
Location: Philadelphia
Posts: 2,683
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 165 Post(s)
Liked: 140
I'm comparing the UB9000 and LX500 connected to my Sony A9F. On the Scaling HD pattern, the LX500 has weird chroma issues when set to 4K output (see attached images). I see this on both the HDR and SDR versions of the pattern. Looks fine when the LX500 is set to Auto output (1080p). Tried swapping cables and inputs, same result. Also looks fine on the UB9000 at 4K.

However, the same pattern on the S&M HD Benchmark 2nd Edition looks fine on the LX500 when outputting 4K.

Any thoughts? This is a replacement LX500; the previous one had sync issues. Trying to determine if it's another defect in the player.
Attached Thumbnails
Click image for larger version

Name:	IMG_1001.JPG
Views:	32
Size:	4.36 MB
ID:	2602886   Click image for larger version

Name:	IMG_0998.JPG
Views:	32
Size:	3.99 MB
ID:	2602888   Click image for larger version

Name:	IMG_0999.JPG
Views:	29
Size:	4.06 MB
ID:	2602890  
thirdkind is offline  
post #327 of 376 Old 08-14-2019, 12:58 PM
AVS Forum Special Member
 
sspears's Avatar
 
Join Date: Feb 1999
Location: Sammamish, WA, USA
Posts: 5,250
Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 272 Post(s)
Liked: 460
Quote:
Originally Posted by thirdkind View Post
I'm comparing the UB9000 and LX500 connected to my Sony A9F. On the Scaling HD pattern, the LX500 has weird chroma issues when set to 4K output (see attached images). I see this on both the HDR and SDR versions of the pattern. Looks fine when the LX500 is set to Auto output (1080p). Tried swapping cables and inputs, same result. Also looks fine on the UB9000 at 4K.

However, the same pattern on the S&M HD Benchmark 2nd Edition looks fine on the LX500 when outputting 4K.

Any thoughts? This is a replacement LX500; the previous one had sync issues. Trying to determine if it's another defect in the player.
There is a similar artifact with the OPPO into the LG. Any idea which decoder is in the LX500? Mediatek is in the OPPO. The artifact comes and goes as well, which suggests it is an un-initialized variable in the decoder chip FW. It does look a lot like the same bug. Maybe you can try switching between 444 and 422 as well along with native vs. scaled output. This would not be specific to a unit of an LX500, it would apply to all of them running the same FW. I am surprised your sync issue would be in one LX500 and not another.

The chroma diamonds don't exist on the scaling pattern on the 2nd edition. The 2nd edition is H.264 8-bit encoded while the new disc is H.265 10-bit encoded. Not sure if that will make a difference on the pattern.

Stacey Spears
Co-Creator, Spears & Munsil UHD HDR Benchmark

Last edited by sspears; 08-14-2019 at 01:03 PM.
sspears is offline  
post #328 of 376 Old 08-14-2019, 01:24 PM
AVS Forum Special Member
 
thirdkind's Avatar
 
Join Date: Mar 2002
Location: Philadelphia
Posts: 2,683
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 165 Post(s)
Liked: 140
Quote:
Originally Posted by sspears View Post
There is a similar artifact with the OPPO into the LG. Any idea which decoder is in the LX500? Mediatek is in the OPPO. The artifact comes and goes as well, which suggests it is an un-initialized variable in the decoder chip FW. It does look a lot like the same bug. Maybe you can try switching between 444 and 422 as well along with native vs. scaled output. This would not be specific to a unit of an LX500, it would apply to all of them running the same FW. I am surprised your sync issue would be in one LX500 and not another.

The chroma diamonds don't exist on the scaling pattern on the 2nd edition. The 2nd edition is H.264 8-bit encoded while the new disc is H.265 10-bit encoded. Not sure if that will make a difference on the pattern.
I might be misremembering, but I could've sworn in Vincent's review of the LX500 he said it had the same decoder as the OPPO. And it struck me as odd that last night was the first time I saw this issue because I know I've looked at this pattern previously, so thank you for confirming that it comes and goes on the OPPO. I was flipping back and forth between different colorspace settings in the LX500 to test chroma upscaling and that's when this first appeared; perhaps going back and forth between the setup menu and the disc triggered a bug in the firmware. I'll try flipping back and forth again and see if this issue resolves itself.

The sync issue I experienced with my previous LX500 was confirmed as a defect by JVB (it's a modified region free player). They tested my returned LX500 and had the same sync issues I did. The replacement had a single weird sync issue and that's when I first noticed this problem, and given my experience with my first unit, I thought it might be another hardware defect. Looks like that's unlikely though.
thirdkind is offline  
post #329 of 376 Old 08-14-2019, 04:22 PM
AVS Forum Special Member
 
thirdkind's Avatar
 
Join Date: Mar 2002
Location: Philadelphia
Posts: 2,683
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 165 Post(s)
Liked: 140
Quote:
Originally Posted by sspears View Post
The chroma diamonds don't exist on the scaling pattern on the 2nd edition. The 2nd edition is H.264 8-bit encoded while the new disc is H.265 10-bit encoded. Not sure if that will make a difference on the pattern.
Wanted to address this specifically. The disc info on the LX500 reports the SDR Scaling HD pattern on the UHD edition is 8-bit. Is that wrong?

While the 2nd edition doesn't have the diamonds, I'm not seeing the garbage pixels around the wheels that I'm seeing on the UHD edition SDR pattern at 4K.

I tried switching colorspace settings on the LX500, no change. Also tried a factory reset. Even left it unplugged overnight. Can't seem to unstick it.
thirdkind is offline  
post #330 of 376 Old 08-14-2019, 09:59 PM
AVS Forum Special Member
 
KC-Technerd's Avatar
 
Join Date: Jan 2004
Location: Liberty, MO
Posts: 1,709
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 212 Post(s)
Liked: 214
Quote:
Originally Posted by thirdkind View Post
I'm comparing the UB9000 and LX500 connected to my Sony A9F. On the Scaling HD pattern, the LX500 has weird chroma issues when set to 4K output (see attached images). I see this on both the HDR and SDR versions of the pattern. Looks fine when the LX500 is set to Auto output (1080p). Tried swapping cables and inputs, same result. Also looks fine on the UB9000 at 4K.

However, the same pattern on the S&M HD Benchmark 2nd Edition looks fine on the LX500 when outputting 4K.

Any thoughts? This is a replacement LX500; the previous one had sync issues. Trying to determine if it's another defect in the player.
I got a very similar result to your first image when I first tried the new disc on my Oppo UDP-203 with my new LG C9. I didn't take time to diagnose it, figuring I'd go back to it later. I did a firmware update on the TV since then (about a week ago). When I went back to the patterns again, I couldn't reproduce it. I was hoping the firmware update on the TV had fixed it. I'll probably do some testing on it again over the weekend or next week.
KC-Technerd is online now  
Sponsored Links
Advertisement
 
Reply Display Calibration

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


Forum Jump: 

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

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