HDR Discussion Thread

[Agent Kay]

Originally Posted by Geoff D

I was gonna go with "Brother Numsie!" in reply to the gif but the gif is from Last Action Hero, isn't it?

Nope, it's the golden child!
And I read the age of ultron in the style of Wheel of fortune in my head also!

[Geoff D]

Right first time then

[LordoftheRings]

Originally Posted by lordoftherings

age?

Originally Posted by agent kay

of?

Originally Posted by geoff d

ultron?

[mrtickleuk]

Originally Posted by Geoff D

Why does + need ICtCp?

It doesn't. Its stated aim was to be a very simple add-on to HDR10 and be backwards compatible, and invisible to devices that only support HDR10. Adding ICtCp would completely break that.

If, on the other hand, you were to design a brand new HDR format and wanted, for whatever reason, to compete with dolby vision - something that HDR10+ is not going to attempt - then you would probably want to include support for ICtCp.

[DisplayCalNoob]

Originally Posted by Geoff D

Why does + need ICtCp?

The benefits in linearity and reducing crosstalk between colors with high color brightness, I believe will prove to be visually impactful. If consider the visual boost in image quality after a proper color calibration where saturation points in the CIE chart track accurately.

It means no bright reds drifting off to magenta, bright blues stay bright blue. A increased level of clarity in the image.

And the fact it was designed for HDR and is absolute like the EOTF used for DV/HDR10.

[DisplayCalNoob]

Originally Posted by mrtickleuk

It doesn't. Its stated aim was to be a very simple add-on to HDR10 and be backwards compatible, and invisible to devices that only support HDR10. Adding ICtCp would completely break that.

If, on the other hand, you were to design a brand new HDR format and wanted, for whatever reason, to compete with dolby vision - something that HDR10+ is not going to attempt - then you would probably want to include support for ICtCp.

ICtCp is currently being used by Netflix DV streams, Vudu recently joined. All current and upcoming encodes will use ICtCp.

It hasn't been adopted for disc, but my guess it is just a matter of time.

[DanBa]

Originally Posted by DisplayCalNoob

...
It hasn't been adopted for disc, but my guess it is just a matter of time.

Apparently, in this case, there is a conversion in the Dolby Vision Display Management module from Non-Constant Luminance YCbCr to Constant Luminance ICtCp, necessary for a higher quality color volume mapping to serve the diversity of HDR displays than the legacy YCbCr.
https://patents.google.com/patent/US20150248747

Originally Posted by WiFi-Spy

... ICtCp ...
It can also be used as color space for color volume remapping. ...

[Geoff D]

Originally Posted by DanBa

Apparently, in this case, there is a conversion in the Dolby Vision Display Management module from Non-Constant Luminance YCbCr to Constant Luminance ICtCp, necessary for a higher quality color volume mapping to serve the diversity of HDR displays than the legacy YCbCr.

Yes, I've suspected that having ICtCp be part of the Display Manager processing (if that is indeed the case) is partly what cleans up the dreadful chroma compression that I see on HDR10 layers of certain discs when later viewed in Dolby Vision, even on a disc that's got a metadata-only enhancement layer like Last Jedi. Just routing it through the Dolby processor alone is enough to subdue the chroma nastiness because of the improved colour handling in the HDR space.

Originally Posted by DisplayCalNoob

The benefits in linearity and reducing crosstalk between colors with high color brightness, I believe will prove to be visually impactful. If consider the visual boost in image quality after a proper color calibration where saturation points in the CIE chart track accurately.

It means no bright reds drifting off to magenta, bright blues stay bright blue. A increased level of clarity in the image.

And the fact it was designed for HDR and is absolute like the EOTF used for DV/HDR10.

I know what it is and what it does, I'm just wondering why this cheaper non-Dolby alternative to dynamic metadata would then turn around and pay Dolby to use their proprietary colour space. Not gonna happen. And adding it to the UHD Blu-ray spec in general is a long shot for much the same reasons. The BDA okaying a packet of additional metadata (which is all + is) is one thing, but okaying ICtCp is another matter entirely.

[DisplayCalNoob]

Originally Posted by Geoff D

Yes, I've suspected that having ICtCp be part of the Display Manager processing (if that is indeed the case) is partly what cleans up the dreadful chroma compression that I see on HDR10 layers of certain discs when later viewed in Dolby Vision, even on a disc that's got a metadata-only enhancement layer like Last Jedi. Just routing it through the Dolby processor alone is enough to subdue the chroma nastiness because of the improved colour handling in the HDR space.


I know what it is and what it does, I'm just wondering why this cheaper non-Dolby alternative to dynamic metadata would then turn around and pay Dolby to use their proprietary colour space. Not gonna happen. And adding it to the UHD Blu-ray spec in general is a long shot for much the same reasons. The BDA okaying a packet of additional metadata (which is all + is) is one thing, but okaying ICtCp is another matter entirely.

You know, at first I wasn't going to respond because I thought it was odd that you were asking this question. I was thinking to myself, he's pretty knowledgeable.

But, I figured ICtCp would be adopted by SMPTE. Oh well, that means a calibration can be done using ICtCp delta-e EOTF/CMS targets for disc playback.

Anyone know why contrast and it's affect on white and color tracking is never mentioned when discussing HDR. Wouldn't the possibility of white and color clipping exist, because there is information that exist within SDR range of brightness. If Rec 709 exist within the Rec 2020 container, shouldn't color clipping apply further up the gamut?

[Geoff D]

Whether I'm right or wrong re: ICtCp, being adopted by SMPTE doesn't suddenly mean a free for all on usage. Dolby's dynamic metadata was standardised as part of SMPTE 2094, but it's still a proprietary system....right?

[DisplayCalNoob]

Originally Posted by Geoff D

Whether I'm right or wrong re: ICtCp, being adopted by SMPTE doesn't suddenly mean a free for all on usage. Dolby's dynamic metadata was standardised as part of SMPTE 2094, but it's still a proprietary system....right?

Yeah, it is. I guess it all depends on Dolby willingness to share, and that could be one part they are not willing to share. I do know that Samsung was attempting to implement some parts that is in that topology Danba posted, the saturation boost.

[Geoff D]

Dolby have dripfed the basics to the wider industry, the rest seems to sit behind a paywall hence the creation of rival dynamic metadata systems in the first place (though certain functions and formulae are indeed similar). Will ICtCp be part of Dolby's largesse?

[Agent Kay]

First vesa display HDR set review I have seen

https://www.youtube.com/watch?v=Dmhr...ature=youtu.be

[DisplayCalNoob]

Originally Posted by Geoff D

Dolby have dripfed the basics to the wider industry, the rest seems to sit behind a paywall hence the creation of rival dynamic metadata systems in the first place (though certain functions and formulae are indeed similar). Will ICtCp be part of Dolby's largesse?

It's possible, I think their dynamic metadata and tone mapping is they see as there advantage, if ICtCp fits in that equation then maybe we'll never see it happen.

[Penton-Man]

Originally Posted by Penton-Man

One would assume the ‘Netflix calibrated mode’

Sony – BRAVIA – MASTER Series Netflix Calibrated Mode - YouTube

of the Sony Bravia Master Series Z9F and A9F presupposes that the consumer is viewing Netflix content in a Netflix recommended viewing environment (Surround light level: 5 nits) –
https://partnerhelp.netflixstudios.c...ng-Environment

no?

*b.t.w. for non-SMPTE membership who desire to avoid the fee to read the entire standard, to be nit-pickin precise, SMPTE ST 2080-3:2017 says “D65 ambient light level in the monitor environment surround shall have a luminance of 5.0 +/- 0.5 candellas per square meter” (nits)

So, what will the picture quality be of Netflix movies in the ‘Netflix calibrated mode’ when the users are not watching in a pre-determined ambient viewing environment (like 5 nits) because let’s face it, many consumers do not watch their movies in an industry standard mastering environment, however defined by the industry. Heck, even many videophiles aren’t compulsive enough to check surround room lighting with light meters and such and set their room lighting accordingly.

Conversely, what if the user is not always watching his/her Dolby Vision content in only one type ambient lighting environment? Well, at a recent event featuring prototype Sony Bravia Master Series TVs, it has been reported that there were several Dolby Vision modes with the Z9F. This would suggest that a PQ (perceptual quantization mapping function [adopted in SMPTE ST 2084]) compensation for changes in ambient light viewing conditions has been implemented in the Sony Z9F tv for Dolby Vision content, at least in these preproduction TVs. In case some readers didn’t understand what I alluded to the first time around in reply to LordoftheRings, this Z9F selectable viewing feature affords me the opportunity to further elaborate on a past heads-up post from last April…. https://forum.blu-ray.com/showthread...t#post14993439
by adding an addendum detailing method(s) which could be employed.

Addendum:
The intensity of the ambient light could be derived from measurements from one or more sensors attached to the device, user input, location data, default values, or other data. For example, even without a sensor, a user could select a viewing environment from a menu, such as “Dark” or “Bright” environment where each entry in the menu is associated with a predefined luminance value selected by the TV manufacturer. The goal of a Dolby Vision perceptual quantization mapping function adjustment is to map incoming intensity values to new intensity values by taking into consideration both the ambient light and the luminance characteristics of the target display.

Here is a graphic illustration of a method for PQ ambient light compensation resulting in an ambient light corrected curve (PQ’) as compared to the original PQ curve (in this case, worked out to < 1000 nits luminance, but could be extended with the formula) –



In a nutshell, what this all means is that thru menu selection in the demoed Z9Fs, Sony has offered a first step in automatically optimizing the picture quality of Dolby HDR for users when viewing in more than one type environment.

So, despite PQ not being scene referred like HLG, which doesn’t struggle nearly as much with different lighting environments, PQ (at least thru a Dolby Vision mode) is being personalized to different consumer viewing environments and I suspect that over time there will be further refinement in mapping function compensation formula and implementation by TV manufacturers.

[Penton-Man]

Originally Posted by LordoftheRings

https://www.youtube.com/watch?v=Gkyz...ature=youtu.be

Originally Posted by LordoftheRings

Penton is a man in constant exploration on everything development...all fronts, not just the western front.

As a glimpse/hint into the future of Top Guns (http://www.vfa101.navy.mil/) and to complement the F/A-18 Super Hornets in your above YouTube, how ‘bout also posting a YouTube of the F-35C –>

[alexanderg823]

Originally Posted by Penton-Man

So, what will the picture quality be of Netflix movies in the ‘Netflix calibrated mode’ when the users are not watching in a pre-determined ambient viewing environment (like 5 nits) because let’s face it, many consumers do not watch their movies in an industry standard mastering environment, however defined by the industry. Heck, even many videophiles aren’t compulsive enough to check surround room lighting with light meters and such and set their room lighting accordingly.

Conversely, what if the user is not always watching his/her Dolby Vision content in only one type ambient lighting environment? Well, at a recent event featuring prototype Sony Bravia Master Series TVs, it has been reported that there were several Dolby Vision modes with the Z9F. This would suggest that a PQ (perceptual quantization mapping function [adopted in SMPTE ST 2084]) compensation for changes in ambient light viewing conditions has been implemented in the Sony Z9F tv for Dolby Vision content, at least in these preproduction TVs. In case some readers didn’t understand what I alluded to the first time around in reply to LordoftheRings, this Z9F selectable viewing feature affords me the opportunity to further elaborate on a past heads-up post from last April…. https://forum.blu-ray.com/showthread...t#post14993439
by adding an addendum detailing method(s) which could be employed.

Addendum:
The intensity of the ambient light could be derived from measurements from one or more sensors attached to the device, user input, location data, default values, or other data. For example, even without a sensor, a user could select a viewing environment from a menu, such as “Dark” or “Bright” environment where each entry in the menu is associated with a predefined luminance value selected by the TV manufacturer. The goal of a Dolby Vision perceptual quantization mapping function adjustment is to map incoming intensity values to new intensity values by taking into consideration both the ambient light and the luminance characteristics of the target display.

Here is a graphic illustration of a method for PQ ambient light compensation resulting in an ambient light corrected curve (PQ’) as compared to the original PQ curve (in this case, worked out to < 1000 nits luminance, but could be extended with the formula) –



In a nutshell, what this all means is that thru menu selection in the demoed Z9Fs, Sony has offered a first step in automatically optimizing the picture quality of Dolby HDR for users when viewing in more than one type environment.

So, despite PQ not being scene referred like HLG, which doesn’t struggle nearly as much with different lighting environments, PQ (at least thru a Dolby Vision mode) is being personalized to different consumer viewing environments and I suspect that over time there will be further refinement in mapping function compensation formula and implementation by TV manufacturers.

Seems just like what the LG OLEDs have been doing for a few generations now with their DV Bright Room and DV Cinema modes.

[Penton-Man]

Originally Posted by alexanderg823

Seems just like what the LG OLEDs have been doing for a few generations now with their DV Bright Room and DV Cinema modes.

I don’t get out to the provincial forum threads nor the ‘professional reviewers’ websites all that much and other than LG OLED users reporting the feature, would be interested in knowing if anyone (who?, be it hobbyist or LG rep) before tried to explain the science behind a proposed Dolby Vision mapping function adjustment for different ambient light situations given the transfer function and process is display referred, which has always been considered an Achilles heel in comparison to HLG and which social media detractors to PQ have seized upon and emphasized every chance they get.

[Penton-Man]

Originally Posted by Penton-Man

Given the Fly Navy theme of thee above ^ pic, also on my mind today is the principled, courageous man who honorably served his country as a Naval Aviator and Senator, namely John McCain, who will R.I.P. overlooking the Severn River next to his former classmate and lifelong friend here.

[Penton-Man]

Seems the YouTube video next to that old hyperlinked -> Severn River post is no longer viable. Ergo, for those who missed the segment of that presentation from the Bits by the Bay conference first time around, try this – https://www.youtube.com/watch?v=nKf4208c_9E#t=26m43s