HDR Discussion Thread

[Kris Deering]

Originally Posted by Geoff D

I definitely think that SDR viewing settings & conditions compound the problem, I've mentioned it plenty of times before. But when people say that x HDR movie is so dark that it's unwatchable (which still happens even now, check the OUATIH thread) then I also think that the tone mapping crunching down the entire image - and thus making content already at or on SDR-style levels of average brightness look incredidark - might still be in play.

It's reasonable enough to assume that any decent tone map should do what we think it should do and just follow the PQ curve until it goes above what the TV can do, and thereafter it clips or maps accordingly. But, if I may dare say so to the great (and I mean that) Kris Deering, I think it's a bit naive to assume that this is a constant across every model and every brand of TV, given the user experiences that I come across on these boards. And I spend a lot of time on here, lemme tell ya!

As for calibration, are you kidding? Not that it mightn't solve a few underlying issues like this that people have, but 99.9% of the people who contribute to these boards have never had a TV pro-calibrated in their life and theyre not going to start now.

There is always a chance that the tone map is being too aggressive, but as a calibrator I have measured A LOT of flat panels for HDR calibration and it is extremely rare to see one that has any type of roll off coming anywhere near the SDR range of the signal. What you typically see down there is poor ST2084 tracking that is too dark. Then if you take a movie that isn't bright to begin with (like OUATIH), it is exacerbated. Then any other display mechanisms in place may add to that as well (dynamic modes for contrast). I obviously can't say this is the case for everything out there, but it does seem to be prevalent in a lot of sets I've measured. Then you couple the fact that the SDR range is still 2-4x's lower in output then their SDR setup, and you have everything working against you.

I'm actually surprised that HDR display manufacturers don't try to error more on the high side of 2084 as this would probably lead to a lot more happy customers. But too often I see it on the low side unfortunately.

[Geoff D]

I'm not separating tone mapping from PQ tracking when I think of this problem, if they're totally separate then of course I'll defer to your experience but even so, that the lower end can be rectified via calibration is not something you're going to find a lot of love for amongst the average joes/janes. They think calibration means copying the settings from rtings. So poor tracking, poor mapping, poor whatever, it's still having an impact on this content separate from the perceptual effect of them having SDR on torch mode.

It's ironical that ultra-bright ultra-colourful HDR has proved to be the least of tone mapping/tracking/whatever's problems though. Not that stupidly bright gear like The Meg and Pacific Rim haven't proved troublesome, but - as I think you're referencing - it's like the mapping/tracking/whatever is so geared up for expecting such extreme content that it shits itself when given something akin to SDR in an HDR container. From Goodfeathers and Arrival to Solo and TLJ to OUATIH and Knives Out, this issue has never gone away.

[Lee A Stewart]

Samsung Display to terminate all LCD production by end of 2020

Samsung Display will end all production of LCD panels in South Korea and China by the end of 2020, the company said in statements to Korean media and Reuters.

https://www.flatpanelshd.com/news.ph...&id=1585650177

[PaulGo]

Before: What's a CRT?
Soon: What's a LCD?

[ray0414]

Originally Posted by Lee A Stewart

Samsung Display to terminate all LCD production by end of 2020



https://www.flatpanelshd.com/news.ph...&id=1585650177

Best part of the article is the 1st known pictures of Samsungs oled prototype from last week.

[LordoftheRings]

Originally Posted by PaulGo

Before: What's a CRT?
Soon: What's a LCD?

Cathode Ray Tube
Luminous Crystal Display

* https://www.photoworkout.com/best-hd...ware-business/

[mrtickleuk]

Originally Posted by PaulGo

Before: What's a CRT?
Soon: What's a LCD?

My concern would be that they've spent many years, selling LCDs with a backlight that uses QLED technology, and calling these LCD-with-QLED TVs "QLED TVs". We all know they are not true QLEDs in the sense that the Quantum Dots do not themselves emit the light.

Are they actually saying they will no longer even make the LCD-with-QLED TVs, not even for the bottom of the market? edit: Ah. I should have read it properly
"Samsung Electronics, the consumer electronics division, is not expected to stop sales of LCD TVs in the near term. The company is already sourcing LCD TV panels from other panel makers, including AUO and CSoT (TCL)."

[Geoff D]

Originally Posted by Geoff D

I'm not separating tone mapping from PQ tracking when I think of this problem, if they're totally separate then of course I'll defer to your experience but even so, that the lower end can be rectified via calibration is not something you're going to find a lot of love for amongst the average joes/janes. They think calibration means copying the settings from rtings. So poor tracking, poor mapping, poor whatever, it's still having an impact on this content separate from the perceptual effect of them having SDR on torch mode.

It's ironical that ultra-bright ultra-colourful HDR has proved to be the least of tone mapping/tracking/whatever's problems though. Not that stupidly bright gear like The Meg and Pacific Rim haven't proved troublesome, but - as I think you're referencing - it's like the mapping/tracking/whatever is so geared up for expecting such extreme content that it shits itself when given something akin to SDR in an HDR container. From Goodfeathers and Arrival to Solo and TLJ to OUATIH and Knives Out, this issue has never gone away.

And, just to reiterate that it's not only ultra hyped up SDR that's skewing people's reactions, there are several examples from just the last couple of days in the threads about the new Star Wars UHDs wondering why the Dolby Vision streams look so much brighter than the HDR10 discs do on their TVs. As these transfers are virtually SDR inside an HDR container then they're vulnerable to this sort of problem.

So even when comparing HDR to HDR the dynamically mapped DV version is kicking the ass of the in-house tone mapping/tracking/whatever that's pulling down the HDR10. Not that the DV is making it much brighter than it should be, it's just making it look like it should and I really do think that'd be enough for many to be happy with them w/ref to the brightness.

Are some of these folks using older TVs with less than stellar mapping/tracking/whatever? Yes, but as the extant UHD user base is predominantly made up of people with older and/or less than stunningly spec'd TVs then that's why these complaints keep popping up with regularity.

[edit]

And one of the people is reporting that even when using forced DV to send the HDR10 discs thru the DV processing chain they also look much brighter, so even though this isn't actually applying any dynamic mapping at all the Dobly EOTF tracking is still markedly superior to the TV's own internal processing.

[Agent Kay]

Originally Posted by ray0414

Best part of the article is the 1st known pictures of Samsungs oled prototype from last week.

They have shipped more Oled displays than anyone worldwide

[Kris Deering]

Originally Posted by Geoff D

And, just to reiterate that it's not only ultra hyped up SDR that's skewing people's reactions, there are several examples from just the last couple of days in the threads about the new Star Wars UHDs wondering why the Dolby Vision streams look so much brighter than the HDR10 discs do on their TVs. As these transfers are virtually SDR inside an HDR container then they're vulnerable to this sort of problem.

So even when comparing HDR to HDR the dynamically mapped DV version is kicking the ass of the in-house tone mapping/tracking/whatever that's pulling down the HDR10. Not that the DV is making it much brighter than it should be, it's just making it look like it should and I really do think that'd be enough for many to be happy with them w/ref to the brightness.

Are some of these folks using older TVs with less than stellar mapping/tracking/whatever? Yes, but as the extant UHD user base is predominantly made up of people with older and/or less than stunningly spec'd TVs then that's why these complaints keep popping up with regularity.

[edit]

And one of the people is reporting that even when using forced DV to send the HDR10 discs thru the DV processing chain they also look much brighter, so even though this isn't actually applying any dynamic mapping at all the Dobly EOTF tracking is still markedly superior to the TV's own internal processing.

I believe it. The factory calibrations for Dolby Vision modes are typically very good out of the box. But that is because Dolby requires a measurement that really only takes about a minute to get a fantastic result. HDR10 is typically all over the place. DV modes also rarely have any adjustments, they are nearly the equal to a "director's intent" mode for HDR. Given that the low end of flat panels for brightness is typically around 700 nits (OLED), when properly setup/calibrated, there should be little to NO difference between DV and HDR10 in the reference range of the set (first 100 nits). But clearly that is rarely the case.

[Geoff D]

That's exactly it dude. In terms of pure TV spec then there shouldn't be any problem for even the most basic HDR TV to be able to handle these kinds of restrained 'SDR inside HDR' style presentations that barely get above that first 100 nits, or content that has bright speculars but average brightness set to SDR style levels, but as their respective approaches to mapping/tracking/whatever are so squinky then it can end up grossly distorting this kind of imagery.

[Penton-Man]

Originally Posted by Geoff D

master compressionist

Master engineer, A Passing -
Jim Houston was a compassionate man in his personal life as well as a class act in his professional life.

Online (in a non-public group) he was an influential, prolific contributor - paragraphs worth of detailed, evidence-based posts regarding UHDTV color, HDR and other related topics in this hyperlinked community professional group (DTID) - https://forum.blu-ray.com/showthread...s#post13948582
where he, among others, like senior video engineers from consumer television manufacturers the likes of Philips, Sony, etc. brainstormed aspects of HDR, WCG before it came to fruition at the consumer level. At the time, his comments carried over to discussions on this forum - https://forum.blu-ray.com/showthread...yz#post8249641

Plus -

“He wrote influential papers on topics ranging from the color fidelity of High Dynamic Range images….”, see the article in Deadline Hollywood - https://deadline.com/2020/03/jim-hou...ry-1202894852/

P.S.
Not mentioned in the above ^ Deadline Hollywood article and unbeknownst to the public-at-large is that after Technicolor departed from the Stage 6 D.I. facility on the Sony Pictures lot, he took on the singular lead responsibility of rebuilding the facility to upgrade the systems for data, networking and video to next generation all-4K capabilities and to provide a digital backbone for the entire studio for 4K DI and mastering which was quite pioneering back then.

That evolution/accomplishment led to the foundation for many of the contributions by folks (https://forum.blu-ray.com/misc.php?d...osted&t=189830 ) about titles and such in this other popular, multi-million (11 mill to date) viewed thread on Blu-ray.com - https://forum.blu-ray.com/showthread.php?t=189830

[Penton-Man]

Originally Posted by PaulGo

The look of 5247 film stock, including TECHNICOLOR processing had specific characteristics which can be summarized as: Vivid, strong saturated colors. Strong contrast. Typical dynamic range: 8-9 stops.

Yes, vintage film stock dynamic range ^.
As a comparison to modern day acquisition (which included Vision 3), I posted this chart somewhere on this thread back in spring or summer of 2018 -



Anyway, for continuity, I shared the discussion with the guys over at thee other HDR site here – https://www.avsforum.com/forum/465-h...l#post59411088

[Geoff D]

So you resorted to posting on there during your enforced 'social distancing' from this place? Heh.

[Waboman]

Was the PM put in quarantine again?

[Geoff D]

As to Mr P's question - that he posed in a much more, ah, direct direct way on AVS instead of the usual drip-feed blu-ray.com style - aboot 5247 and dynamic range affecting how the UHD of Top Gun might look: we already have some movies shot on that stock and/or the 5294 which TG used for interiors which have appeared on UHD. And yeah, it's fair to say that they might not all have a gigantic increase in dynamic range over the existing SDR versions but they still offer enough to make the HDR rendition more worthwhile than not e.g. CE3K, Ghostbusters, Blade Runner, Alien, Apocalypse Now etc.

[DisplayCalNoob]

Originally Posted by Geoff D

That's exactly it dude. In terms of pure TV spec then there shouldn't be any problem for even the most basic HDR TV to be able to handle these kinds of restrained 'SDR inside HDR' style presentations that barely get above that first 100 nits, or content that has bright speculars but average brightness set to SDR style levels, but as their respective approaches to mapping/tracking/whatever are so squinky then it can end up grossly distorting this kind of imagery.

Are you guys suggesting, that DV dynamic metadata and or dynamic tone mapping has no advantages over static, low nit or high?

[Geoff D]

My point there is that most display tech *itself* - in terms of sheer nits, contrast etc - can easily handle these kinds of low APL presentations on paper, even the crappy supermarket special TVs, but it's the processing/mapping/tracking/whatever that's letting the display tech down, and sometimes drastically affecting this kind of low APL HDR content by making it look much too dim and dark. In these cases then dynamic tone mapping is a godsend, but if the in-house mapping/tracking/whatever worked as to how many professionals and enthusiasts wanted or assumed it should work (by tracking the first 50% of the PQ EOTF e.g. 100 nits accurately and then mapping or rolling off the rest according to what peaks the TV can do) we'd have less problems with "too dark" HDR IMO.

[mrtickleuk]

Originally Posted by Agent Kay

They have shipped more Oled displays than anyone worldwide

Yes, when you add in mobile telephones and other very small displays. That's not a fair comparison though.

[mrtickleuk]

Originally Posted by Geoff D

My point there is that most display tech *itself* - in terms of sheer nits, contrast etc - can easily handle these kinds of low APL presentations on paper, even the crappy supermarket special TVs, but it's the processing/mapping/tracking/whatever that's letting the display tech down, and sometimes drastically affecting this kind of low APL HDR content by making it look much too dim and dark

There is another very very important facet to the "SDR within HDR" problem. All LCDs have to max out the backlight in HDR mode. Blacks are no longer black, you get milky grey, halo-ing around subtitles etc. It looks horrible, and takes a lot more power to display the same thing, but worse (so is very bad for the environment). This is why if it's going to be SDR, it's better to just let it be SDR.
You could even say "if it's not very HDR, it's better to just let it be SDR" and then we have a debate about what "very HDR" should mean. But that is a hornet's nest