HDR Discussion Thread

[Richard Paul]

A journalist has said on Twitter that HLG will be added to the 2016 Sony HDR TVs with a firmware upgrade. The president of the Ultra HD Forum responded by saying that other major TV companies are planning to do add HLG with a firmware upgrade.

[Penton-Man]

Originally Posted by Richard Paul

A journalist has said on Twitter that HLG will be added to the 2016 Sony HDR TVs with a firmware upgrade. The president of the Ultra HD Forum responded by saying that other major TV companies are planning to do add HLG with a firmware upgrade.

Good to see that after all this time from last year the journalists are finally discovering HLG.

And Thierry at least lately addressing to the twitter folks, HLG signaling over HDMI as was outlined in the 2nd illustration last April to fellow Insider Robert Zohn…. http://forum.bluray.com/showthread.p...n#post12123572 and the tie-in post below that.

Note: the CEA in CEA 861 stands for Consumer Electronics Association but is now known as the Consumer Technology Association

P.S.
On a more advanced readership level…. regarding the acquisition and production workflow for HDR for live production, instead of utilizing the *gold standard* of 16 bit RAW production workflow (https://forum.blu-ray.com/showpost.p...&postcount=127)
like for feature film work where you have a lot of time (and funding), expect a more practical workflow like S-Log3 for acquisition and XAVC for compression.

[Penton-Man]

Originally Posted by Penton-Man

...for live production....expect a more practical workflow like S-Log3 for acquisition and XAVC for compression.

And not only for live production, but also for episodic television programming....

The Impact of HDR in Story Telling with Timothy Vincent - YouTube

where TV shows just don’t have the same luxury of time as feature motion pictures, I would also expect a lot of shows offering an HDR iteration to viewers to employ 10 bit XAVC compression with S-Log3 rather than RAW.

Perhaps not as fine in quality and future proofing as 16bit RAW and Dolby Vision, but still very respectable imagery.

[Penton-Man]

Originally Posted by Penton-Man

live

albeit in SDR, but Coronado and others caring about those who sacrifice to keep us free…..
http://fox5sandiego.com/2016/05/13/f...al-procession/

[MiC. UHD]

http://www.flatpanelshd.com/news.php...&id=1463138030


Great article!! Man I'm holding out hope for that update to hit both 2015 and '16 Sammy sets.

[bruceames]

Really makes you wonder if open-source HDR10 dynamic or DV will prevail on disc.

[Geoff D]

UHD should never have launched without dynamic metadata in the first place, but then if they'd taken another year or two to get UHD Blu out there would it have survived?

The CRI system (which is Technicolour's, I think?) is part of the UHD spec but without the provision in HDMI to actually ferry the dynamic metadata from one device to the other (heck, even the static metadata update was only added fairly recently) it's as good as useless.

I always hoped that players would be able to use the data on an internal level to provide a properly-mapped SDR output at source, thus avoiding the need for an HDMI update in that respect, but it seems like every SDR remapping system in the current spec (CRI, Dolby, Philips) needs a licence one way or the other so it's no wonder Panasonic and Samsung have done their own thing re: SDR remapping.

Still, I'm very pleased with the Panny's SDR conversion, even if I am flying blind as to dat accuracy. Ignorance, bliss, etc.

[MiC. UHD]

You would think open source HDR would prevail due to the lack of royalties but at this point I don't even know. Plus that article mentions that the industry is working on a third HDR format? It'd be nice if they can sort this out sooner rather than later.

[Penton-Man]

Originally Posted by MiC. UHD

http://www.flatpanelshd.com/news.php...&id=1463138030

Great article!!

Checking the date, essentially a rehash (without credit) of what we’ve been discussing days earlier since page 3 - https://forum.blu-ray.com/showthread...s#post12192776

apparently tipped to the AV writer by fellow Blu-ray.com member kristoffer.

Originally Posted by MiC. UHD

Plus that article mentions that the industry is working on a third HDR format?

The 3rd HDR format which you’re inquiring about was mentioned here - https://forum.blu-ray.com/showthread...g#post12206957

and has been intermittently discussed in detail (with the latest developments) for about a year now in other threads on this forum.

[Penton-Man]

Originally Posted by bruceames.

Really makes you wonder if open-source HDR10 dynamic or DV will prevail on disc.

After work when I get home I prefer not to stress out over such things and just watch my favorite TV show(s) with included commercial…..

DEADPOOL Hilarious Blu ray DVD Release Commercial - YouTube

[jono3000]

I was a bit annoyed that I'd bought my 4K TV too early and only have SDR, but I'm happy with what I'm seeing and in a couple years when the various HDR formats have settled down I will upgrade. Dynamic metadata sounds like the way forward, would existing UHD discs be compatible with this or would it require a specially encoded disc?

Sent from my Nexus 6P using Tapatalk

[Geoff D]

Originally Posted by jono3000

I was a bit annoyed that I'd bought my 4K TV too early and only have SDR, but I'm happy with what I'm seeing and in a couple years when the various HDR formats have settled down I will upgrade. Dynamic metadata sounds like the way forward, would existing UHD discs be compatible with this or would it require a specially encoded disc?

Sent from my Nexus 6P using Tapatalk

Dude, I was annoyed for the longest time about the 'outdated SDR 4K TV' thing but as it is I seem to be getting the best of both worlds: much increased dynamic range without having to forcibly max out backlight/contrast on my TV (as per static HDR metadata) and the WCG is coming through unmolested. I couldn't have asked for a better result TBH, and seeing as a new TV isn't on my horizon for a while this unexpected level of UHD performance will tide me over and then some.

As for existing discs, if I put my logic hat on (lemme just dust it off, it don't get used all that often) then I'd say that current discs aren't being mastered with any kind of dynamic metadata, just static MaxCLL and MaxFALL.

[PeterTHX]

Originally Posted by bruceames.

Really makes you wonder if open-source HDR10 dynamic or DV will prevail on disc.

DTS-MA wasn't able to be decoded for a year and a half after the launch of BD, now it's the standard lossless audio codec.

[Geoff D]

Originally Posted by PeterTHX

DTS-MA wasn't able to be decoded for a year and a half after the launch of BD, now it's the standard lossless audio codec.

Aye. While it sux!1! that so many UHD features are going to be added or adopted or employed properly further down the line, that really is the whole 'early adopter' thing in a nutshell. (I know I moaned endlessly before about the injustice of it all, but it's easier to be sanguine about it when things work out well for a change! )

It'll be interesting nonetheless to see which dynamic system prevails though. I hope Dolby gets it because their 12-bit encoding should help to counteract banding once and for all, but them thar licence fees might put some hardware manufacturers off, not to mention the Japanese brands who historically have only been interested in one set of proprietary standards: their own.

[MiC. UHD]

Originally Posted by Penton-Man

Checking the date, essentially a rehash (without credit) of what we’ve been discussing days earlier since page 3 - https://forum.blu-ray.com/showthread...s#post12192776

apparently tipped to the AV writer by fellow Blu-ray.com member kristoffer.

The 3rd HDR format which you’re inquiring about was mentioned here - https://forum.blu-ray.com/showthread...g#post12206957

and has been intermittently discussed in detail (with the latest developments) for about a year now in other threads on this forum.

Wow ummm ok how can I respond appropriately? First off, me sending a link was simply my small contribution to sharing some information with fellow members on the forum, especially since my comment was specifically in regards to my 2015 Samsung receiving the update to make dynamic metadata possible on my set. Now I apologize if my contribution to the thread was an inconvenient rehash of information that you and others have apparently discussed for day and days, God forbid I attempt to do more than simply just read what everyone else writes.

Unfortunately I'm not a "retired insider" and as much as I try to keep up with ever changing technology, at times I'm not as informed as I'd like to be and sometimes I run into situations where I have to try to decipher what the hell "SMPTE" is and what ST2094 even means. Maybe to you it's as simple as 1+1 but I'm still trying to understand it all for myself. Sorry my level of understanding doesn't appease you and you felt the need to cite my shortcomings.

Maybe another year of intermittent discussions will help bring me up to speed.

[Penton-Man]

relax, didn’t mean to cite your “shortcoming”.

Not to worry about the technical aspects and docs, so don't feel badly as I’ve heard/read more than one 'pro' calibrator incorrectly refer to ST 2094 as ST 2096.

[PS3_Kiwi]

Just when I thought I was beginning to understand all the UHD jargon, and thinking that I was clear in my mind what I needed to buy to play Blu-ray 4K UHD HDR discs, it seems that there is going to be several different types of HDR. I was going to start my collection this coming week when The Revenant is released in New Zealand, but not sure what to do now. Should I go ahead and start my collection, or wait until this whole HDR fiasco is sorted out.

[Richard Paul]

Originally Posted by Penton-Man

And Thierry at least lately addressing to the twitter folks, HLG signaling over HDMI as was outlined in the 2nd illustration last April to fellow Insider Robert Zohn…. http://forum.bluray.com/showthread.p...n#post12123572 and the tie-in post below that.

It is good to hear that the next version of HDMI will be completed this summer.

Originally Posted by MiC. UHD

You would think open source HDR would prevail due to the lack of royalties but at this point I don't even know.

CE companies have to pass the cost directly to consumers so they tend to be concerned with royalty cost. Samsung was concerned enough about it to go out and make a dynamic metadata system that would be free. I think that the Samsung system has a reasonable chance but the studios will likely decide based on many factors including business connections.

[James Freeman]

When we move Luminance (compress or expand) we screw the Chrominance, they are interdependent to one another.
In other words, you can't lower luminance (brightness) without completely re-calculating the chrominance (color) with complex matrices for accurate results.

Nobody knows for sure what the Samsung or the Panasonic UHD Player does for HDR to SDR down conversion because there are no HDR P3/2020 test patterns yet.

Here are several ways converting HDR to SDR in the same color space of 709:
* Adding gamut down-conversion makes it even more complicated and un-accurate.

1. Original HDR image.


2. Clipped down to 100nit.
This way will preserve accurate luminance and colors below the clipping point.
But the highlight information is completely gone.


3. Compressed Highlights.
Preserves dynamics but de-saturates the compressed highlight color.
Colors and luminance are accurately preserved below compression point.
This is IMO a good solution.


4. Compressed highlights and added saturation.
Preserves dynamics and saturated back to look like the original.
The problem is unlike 3, it saturated below compression point also, so the accurate colors get saturated too.
We don't want to loose skin tone color accuracy just so the sky looks slightly more saturated.


5. Compressed whole range.
This might be best for color to appear the same but the luminance is also compressed at lower levels.
This gives an HDR look on an SDR TV, but it could be too dark for daytime watching.

[Geoff D]

Thankfully the Panasonic can output SDR 2020 and there are HDR-encoded P3/2020 test patterns, the ones on the Sony discs. They're 100-nit windows and not the full breadth of the saturation range but they can at least be used to check whether 2020 is being adversely affected by the SDR conversion.

I plotted the SDR-converted 2020 RGBCMY points in HCFR and the gamut coverage I got was very similar to that of a 2016 HDR Sony (as measured by AVF using their $2.5K HDR signal generator). I've said before that I don't know whether the lower saturation points (which are arguably more important than the 100% brightness ones seeing as they relate more to real-world content) are tracking to 2020 as they should, but in the absence of any further test patterns on disc and the Panasonic's inability to play HEVC USB files I'll just have to hope.

FWIW I also plotted those HDR 2020 patterns in SDR 709 and they tracked extremely well to the smaller gamut, the dEs were higher (as to be expected) but on the actual CIE chart it was bang on.