Spears & Munsil 4K UHD HDR Benchmark (Test Discs) May 15, 2023

[Stacey Spears]

Originally Posted by Geoff D

I dare say you need to double check the Panny, and it's not just me who noticed it as Vincent Teoh picked up on it as well, it's clipping the upper end of highlight information in DV and it should be spectacularly obvious with all that heavy nittage going on in your montage, chapter 2 should show it in all its glory. When I first mentioned this DV clipping problem in the 820 thread [url]https://forum.blu-ray.com/showthread.php?

Just looked at the chapter 2 with the UB90000 and OPPO on the LGB7, no clipping at all. Maybe player vs. TV-led?

On the tone map ramp, again, they look the same. At 1000, both players don't show much above 1000. Then when I switch to 10,000, both go all the way up to 10,0000.

Maybe the 9000 is different than the 820? I will test both on the Z9D tomorrow. The clipping on chapter 2 is the same with the OPPO and Panasonic last time I checked. Its a Sony / Dolby issue. DV should never clip! I need to look at the EDID in the Sony to see what it is telling the DV BD players to do. Then I can compare with LG because the LG does not clip in player-LED either.

[Geoff D]

Thanks for checking. I'm pretty sure Vincent noticed it on the 9000 when connected to an LG OLED in his review, he also recently noted that the Pioneer LX500 seems to have the exact same problem, again not connected to a Sony display at all. But you're getting it clear all the way to the big ten thou so there's no arguing with that!

[Stacey Spears]

Originally Posted by Geoff D

Thanks for checking. I'm pretty sure Vincent noticed it on the 9000 when connected to an LG OLED in his review, he also recently noted that the Pioneer LX500 seems to have the exact same problem, again not connected to a Sony display at all. But you're getting it clear all the way to the big ten thou so there's no arguing with that!

One thing I can do is use the monotonicity pattern and actually measure the pixels coming out over HDMI. I just need to hookup the Accupel. Really need to finish the app that takes automated measurements with it. This way we remove the display from the equation.

[rickardl]

So... in the colorspace eval image, what should be blinking/be visible in this part of the image?

Note that I am special case here, I am using a panasonic 820 for downconverting to a 2013 Samsung plasma....

[Geoff D]

I had the same thought. In my case the Y High and Cr Low boxes are blinking, on both the Panny and OPPO. Wat dis meenz?

[Geoff D]

Originally Posted by Stacey Spears

Just looked at the chapter 2 with the UB90000 and OPPO on the LGB7, no clipping at all. Maybe player vs. TV-led?

On the tone map ramp, again, they look the same. At 1000, both players don't show much above 1000. Then when I switch to 10,000, both go all the way up to 10,0000.

Maybe the 9000 is different than the 820? I will test both on the Z9D tomorrow. The clipping on chapter 2 is the same with the OPPO and Panasonic last time I checked. Its a Sony / Dolby issue. DV should never clip! I need to look at the EDID in the Sony to see what it is telling the DV BD players to do. Then I can compare with LG because the LG does not clip in player-LED either.

Just to go back to this, I realised that the tone map ramp won't show any clipping at source *because* it's in HDR10, which is not the problem so the ramp is irrelevant: it's the specific DV output where the clipping is happening on the Panny decks. You said chapter 2 still looked fine and I can only take your (very respected) word for it, but even if it is the LL mode causing it then it's still a problem. The Panny doesn't have a manual setting for selecting player or TV led IIRC, are you able to force player-led on the Panny into the LG using some EDID trickery to see how it reacts?

[INdetectableMAN]

Originally Posted by Geoff D

What you're seeing is the Optimiser turning the PQ curve into its own thing

I have come to the conclusion that the video that this disc brings quickly reveals the problems of color volume when the OLED technology aims to increase the production of brightness beyond a point

The Panasonic OLED 2018 55FZ800 EU is a panel that produces around 750 nits, and the tone mapping that it does has the objective of maintaining the brightness instead of preserving all the information.

However, the Panasonic UB9000 player has a different approach, and when activated, the tone mapping does not result in such brilliant highlighhts, but in return, as you are seeing in the photographs, the volume of color increases in the bright areas.

Of course, this happens obviously because I imagine that the video is done conscientiously to put the limit of its possibilities to the screens, otherwise, with UHD movies I at least had not been aware of it.

So many thanks to the authors of the video, because it has served me in a few minutes to learn a little more, and understand that more important than the maximum brightness, is without a doubt the balance between brightness and volume of color, and OLED technology has a problem with this

[Auditor55]

Originally Posted by INdetectableMAN

I have come to the conclusion that the video that this disc brings quickly reveals the problems of color volume when the OLED technology aims to increase the production of brightness beyond a point

The Panasonic OLED 2018 55FZ800 EU is a panel that produces around 750 nits, and the tone mapping that it does has the objective of maintaining the brightness instead of preserving all the information.
However, the Panasonic UB9000 player has a different approach, and when activated, the tone mapping does not result in such brilliant highlighhts, but in return, as you are seeing in the photographs, the volume of color increases in the bright areas.

Of course, this happens obviously because I imagine that the video is done conscientiously to put the limit of its possibilities to the screens, otherwise, with UHD movies I at least had not been aware of it.

So many thanks to the authors of the video, because it has served me in a few minutes to learn a little more, and understand that more important than the maximum brightness, is without a doubt the balance between brightness and volume of color, and OLED technology has a problem with this

That's why I don't see OLED as the future of display technology.

[Geoff D]

Originally Posted by INdetectableMAN

I have come to the conclusion that the video that this disc brings quickly reveals the problems of color volume when the OLED technology aims to increase the production of brightness beyond a point

The Panasonic OLED 2018 55FZ800 EU is a panel that produces around 750 nits, and the tone mapping that it does has the objective of maintaining the brightness instead of preserving all the information.

However, the Panasonic UB9000 player has a different approach, and when activated, the tone mapping does not result in such brilliant highlighhts, but in return, as you are seeing in the photographs, the volume of color increases in the bright areas.

Of course, this happens obviously because I imagine that the video is done conscientiously to put the limit of its possibilities to the screens, otherwise, with UHD movies I at least had not been aware of it.

So many thanks to the authors of the video, because it has served me in a few minutes to learn a little more, and understand that more important than the maximum brightness, is without a doubt the balance between brightness and volume of color, and OLED technology has a problem with this

Well, yeah. The luminance is literally what contains the wider, taller colour volume and on some displays you'll get one or the other, but not both. Some people have tried to explain such things already: https://forum.blu-ray.com/showthread...4#post16416134 and with consumer OLED the WRGB panels are what hurts the colour volume in brighter areas the most, that by adding a white sub pixel along with the other three you get more brightness and less stress on the blue pixels' shorter lifespan but when it comes to brighter images then they can quickly lose colour volume. The Optimiser will bring back the volume by compressing the range but, as mentioned upthread, it comes at the cost of turning it into a more linear SDR-esque response curve that looks much less HDR-y on a light cannon of a TV.

It's not a coincidence that the legendary Sony BVM-X300 monitor has a 'true' RGB OLED panel so it maintains the colour in the highlights, but when being driven as hard as they are in the professional HDR monitoring environment then they're not going to last forever, hence the creation of the dual-LCD HX310 model to supplant the X300.

[Stacey Spears]

Originally Posted by Geoff D

Just to go back to this, I realised that the tone map ramp won't show any clipping at source *because* it's in HDR10, which is not the problem so the ramp is irrelevant: it's the specific DV output where the clipping is happening on the Panny decks. You said chapter 2 still looked fine and I can only take your (very respected) word for it, but even if it is the LL mode causing it then it's still a problem. The Panny doesn't have a manual setting for selecting player or TV led IIRC, are you able to force player-led on the Panny into the LG using some EDID trickery to see how it reacts?

Just to be clear, when I sent LL into the LG it does not clip. It is specifically a Sony issue. At least with OPPO where I can force LL vs. tunneled.

The tone mapping pattern is what I am planning to use for the DV encoding test I am working on. Hopefully I can share it with you in a couple of weeks to test. It will be small so you can probably burn to a DVD-R, certainly a BD-R to play on the Panasonic. The OPPO can play straight from the HDD.

Both the OPPO and Panny clip on chapter 2 on the Sony in DV, but the detail returns if you pull contrast down, which is the wrong thing to do. I plan to demo this to Sony/Dolby in a couple of weeks. Basically i don't think DV tuning is optimal on the Z9D.

[Stacey Spears]

If you use the quantization rotate pattern and switch from 600 to 10,000 nit version, you will see more and more white in the center of the Cb/Cr ramps as you go up on OLED. Not so much with LCD. That is the white sub pixel doing its thing.

[Stacey Spears]

Originally Posted by INdetectableMAN

I have come to the conclusion that the video that this disc brings quickly reveals the problems of color volume when the OLED technology aims to increase the production of brightness beyond a point

The Panasonic OLED 2018 55FZ800 EU is a panel that produces around 750 nits, and the tone mapping that it does has the objective of maintaining the brightness instead of preserving all the information.

However, the Panasonic UB9000 player has a different approach, and when activated, the tone mapping does not result in such brilliant highlighhts, but in return, as you are seeing in the photographs, the volume of color increases in the bright areas.

Of course, this happens obviously because I imagine that the video is done conscientiously to put the limit of its possibilities to the screens, otherwise, with UHD movies I at least had not been aware of it.

So many thanks to the authors of the video, because it has served me in a few minutes to learn a little more, and understand that more important than the maximum brightness, is without a doubt the balance between brightness and volume of color, and OLED technology has a problem with this

I recommend you compare the DV version to the 1000 or 600 versions on your display with and w/o HDR optimizer. The 600 should result in no tone mapping by the display since its below its capability. The 1000 will have minor tone mapping. Both of those still go beyond the gamut the display is capable of, though the Grand Prismatic should be within it. The peacock feather and quantum dot go beyond P3. The quantum dot fill up the 2020 CIE.

The goal of the montage was to cause trouble. Help them develop better algorithm for future content and displays.

[Geoff D]

Originally Posted by Stacey Spears

Just to be clear, when I sent LL into the LG it does not clip. It is specifically a Sony issue. At least with OPPO where I can force LL vs. tunneled.

The tone mapping pattern is what I am planning to use for the DV encoding test I am working on. Hopefully I can share it with you in a couple of weeks to test. It will be small so you can probably burn to a DVD-R, certainly a BD-R.

Both the OPPO and Panny clip on chapter 2 on the Sony in DV, but the detail returns if you pull contrast down, which is the wrong thing to do. I plan to demo this to Sony/Dolby in a couple of weeks. Basically i don't think tuning is optimal for that combo.

That's what I mean, the OPPO isn't clipping the DV image at source using LL so using that mode on the LG won't show this problem. What I'm saying is that Panasonic players specifically, using LL, are hard clipping DV to something like ~2500 nits and the info can't be brought back using the contrast control. That's why I aksed if you could somehow force LL on the Panny into the LG to see what effect that has, it seems to be a very specific problem to LL on the Pannys and perhaps the Pioneer LX500 as well, on what looks like a Sony OLED?:

Pioneer UDP-LX500 4K Blu-ray Player Review (vs Panasonic DP-UB9000) - YouTube

(go to about 8min 30secs).

I look forward to the patterns, I should be able to play them off of USB right? MEL versions anyway? (but those will be fine for what I want to do with them). I've already got some LG DV demo clips that play from USB just fine on the OPPO.

[Stacey Spears]

By not having a tone / gamut mapping standard, every display looks different. It is fundamental problem of HDR. DV tries to solve this, but still has a few bumps to smooth out.

[Stacey Spears]

Originally Posted by Geoff D

That's what I mean, the OPPO isn't clipping the DV image at source using LL so using that mode on the LG won't show this problem. What I'm saying is that Panasonic players specifically, using LL, are hard clipping DV to something like ~2500 nits and the info can't be brought back using the contrast control.

That's why I aksed if you could somehow force LL on the Panny into the LG to see what effect that has.

This should be possible with an HDFury. You can capture the EDID from the Sony and place it onto the HDFury and then this should force the Panasonic into LL mode on the LG.

I can't figure out what I did with my HDFury. I tried to find it the other day.

[Geoff D]

I'm not putting money forward to help you buy another one, the disc was expensive enough!

[p5browne]

A question here:

We all know USB settings are different to HDMI settings on the TV, but what actually happens to the BD Player HDMI Output, re BD Disc versus USB Input?
Are there any changes to the signal taken from the USB Input, versus what is taken from the BD Disc? The TV we can calibrate, but a BD player, not at all. Just maybe some electronic wizardry that affects the signal?

[Geoff D]

One would hope that anything played from USB through the player would go through the same decoding circuitry as the discs, though that's not a given: when Panasonic's UHD players first appeared they would not play anything encoded with HEVC over USB whatsoever, it literally took six months for it to be added via firmware for mp4 files and a full year for MKV/TS HEVC file support (even then they're not a great player of files, but Panasonic never intended them to be).

[Stacey Spears]

Originally Posted by rickardl

So... in the colorspace eval image, what should be blinking/be visible in this part of the image?

Note that I am special case here, I am using a panasonic 820 for downconverting to a 2013 Samsung plasma....

The PQ tracking section won't work in that case since you are converting to SDR. In that case, use the SDR 709 version of the pattern.

If the image is being scaled to 1080p, that section won't work either. Needs to be 1:1 for that to work.

The bottom section, again, use the SDR version. If color decoding is correct, the red and green boxes should be equal. This does not work on current HDR displays due to tone mapping. Need brighter displays for that to work.

[Stacey Spears]

Originally Posted by p5browne

A question here:

We all know USB settings are different to HDMI settings on the TV, but what actually happens to the BD Player HDMI Output, re BD Disc versus USB Input?
Are there any changes to the signal taken from the USB Input, versus what is taken from the BD Disc? The TV we can calibrate, but a BD player, not at all. Just maybe some electronic wizardry that affects the signal?

For Blu-ray, the HEVC decoding happens in the player. As does conversion from 4:2:0 to at least 4:2:2. For USB, those both happen in the display. And pretty much every display does the wrong upsample for 2020, they apply 709. (not color conversion, chroma upsample) This will result in a half pixel shift between luma and chroma. I believe LG fixed this in 2019, not sure if anyone else has.

For Dolby Vision, if you have Sony, the tone mapping is done in the display for Dolby Vision over USB, but done in the player because it is in Low Latency mode. For LG, both can be done in the display.

There may be other paths they take as well, difficult to know for sure.