HDR Discussion Thread

[Penton-Man]

continuing on from - https://forum.blu-ray.com/showthread...e#post17362940 and expanding technically on the DVIQ process presented in laymen’s terms to the reporter here - https://www.youtube.com/watch?v=mxMh0CshOaY#t=5m16s

in greater detail: ambient light sensor -> produces ambient light signal -> processor produces metadata consisting of at least a minimum luminance value, a midpoint luminance value and a maximum luminance value of the input image to which thee ambient light intensity function is applied -> mapping input luminance values in a reference viewing environment to output luminance values in the target viewing environment

[Penton-Man]

Originally Posted by Waboman

Put a pole onstage.

pole or not next to the right front fender, regardless, it was a beautiful morning –
https://forum.blu-ray.com/showthread...g#post16218767

for filming –

[Geoff D]

Originally Posted by Penton-Man

continuing on from - https://forum.blu-ray.com/showthread...e#post17362940 and expanding technically on the DVIQ process presented in laymen’s terms to the reporter here - https://www.youtube.com/watch?v=mxMh0CshOaY#t=5m16s

in greater detail: ambient light sensor -> produces ambient light signal -> processor produces metadata consisting of at least a minimum luminance value, a midpoint luminance value and a maximum luminance value of the input image to which thee ambient light intensity function is applied -> mapping input luminance values in a reference viewing environment to output luminance values in the target viewing environment

TVs don't have an unlimited amount of headroom though, that's what perplexes me so about DVIQ. Yes, if it can ramp up the APL so that it now looks as it should vis-a-vis the reference level when being slammed by ambient light streaming into the room then that's a good thing, but if your APL has increased by, say, fivefold your maximum peak brightness sure as shit hasn't, thereby drastically constricting the whole "high dynamic range" thing, in terms of luminance anyways.

Not that people's experiences with HDR aren't already all over the place but this sort of tech is only going to add even more confusion about what something should actually look like, and as it comes with the Dobly seal of approval people will still think they're viewing some kind of optimal presentation w/ref to the original grade. In terms of the lighting conditions then it's optimal, but in terms of what's actually been graded? Nah.

I mean, I'm glad that people will be able to watch "HDR" whatever the viewing environment may be but that cinematographer you've referred to before, Geoff something or other, hit the nail on the head when he was aksed about HDR in one of his videos. He said that if you're watching in viewing conditions with anything other than the light from the display then it's not actually HDR as the ambient light will wreck it. He even said that the light from the TV reflecting off your face is enough to disqualify it, though that sounds a bit extreme unless you've lined your face with bacofoil or something.

[DisplayCalNoob]

Maybe the tone mapping that happens at the low end of the with metadata is what maintains that dynamic range that is essential to the HDR presentation.

[DanBa]

"In an ideal world, all Pay TV operators would transition from legacy broadcast or classic IPTV to all-ABR delivery so they can unify their delivery infrastructure. That is what Jacques Le Mancq, CEO of Broadpeak, suggested in April when his company announced the latest version of its nanoCDN multicast ABR solution, which enables live streaming channels to be multicast across a broadband network but then converted to unicast sessions in the customer premise so any multiscreen device can consume them without changes to its software. He believes there is a natural roadmap towards what he calls the mutualisation of operations around HTTP-ABR video, and he is not alone."
https://twitter.com/DanielBa78/statu...31131566206976
https://www.v-net.tv/2017/07/13/mult...-video-future/
https://dvb.org/wp-content/uploads/2...t_bluebook.pdf
https://specification-search.cablela...chnical-report
https://broadpeak.tv/solutions/multicast-abr/
https://broadpeak.tv/blog/broadpeak-au-ces-2020/
https://www.cisco.com/c/en/us/produc...78-740369.html







Multicast server = Transcaster (Unicast to Multicast)
Multicast gateway = Agent (Multicast to Unicast) = HTTP Gateway
M interface: IP Multicast-ABR Low Latency (ABR: Adaptive Bit Rate)
L interface: OTT HTTP ABR Unicast LL (DASH / HLS)
B interface: legacy OTT HTTP ABR Unicast





compressed video (linear TV, …) >> Internet (IP Multicast-ABR, …) >> next-generation Multicast-ABR STB: router + HTTP gateway >> LAN (Ethernet, Wi-Fi) >> TV with OTT HTTP streaming app: decoder > uncompressed video

compressed video (linear TV, …) >> Internet (MPEG2-TS, …) >> legacy IPTV STB: router + decoder > uncompressed video >> HDMI >> TV


As the HTTP Gateway (i.e. Multicast-ABR STB) deals only with HTTP streams, if a HTTP stream conveys HDR then the STB is HDR pass-through.
https://twitter.com/broadpeak/status...94452984303616




A TV station can broadcast in Multi-HDR (HDR10 or PQ10, Dolby Vision or ST2094-10, HDR10+ or ST2094-40, SL-HDR2, (HLG)):
https://dvb.org/webinar/hdr-dynamic-mapping/



Multi-HDR bitstream =
PQ10 bitstream
+ Dolby Vision metadata (optional)
+ HDR10+ metadata (optional)
+ SL-HDR2 metadata (optional)


As the HTTP Gateway (i.e. Multicast-ABR STB) deals only with HTTP streams, if a HTTP stream conveys Multi-HDR / HDR HFR then it is a Multi-HDR / HDR HFR pass-through STB.


A Multicast-ABR / DVB-I infrastructure is ready for now and future live TV:
. now: SDR
http://www.digitalbitrate.com/dtv.ph...ste=2&live=101
http://www.digitalbitrate.com/dtv.ph...&sec=0&lang=en
. future: HDR, Multi-HDR, and HDR HFR for sports
https://www.avsforum.com/forum/465-h...l#post49723889
https://www.avsforum.com/forum/465-h...l#post49984041
https://www.avsforum.com/forum/465-h...l#post49991817
https://www.avsforum.com/forum/465-h...l#post54845156
http://cognitus-h2020.eu/index.php/hfr/

[Penton-Man]

excusing the typo, for those who missed the respective booth at HPA’s innovation tech zone last week, there’s always this next Wednesday for local SoCalers in Burbank, scroll down to –
“DEMO by PIXELWORSK
Pixelworks is a leading provider of video and motion processing solutions that bridge the performance gap between content formats and device capabilities”-
https://www.eventbrite.com/e/coloris...dssbdestsearch

esp. good for those considering how to deal with judder in high contrast HDR scenes.

[Penton-Man]

Originally Posted by Geoff D

....In terms of the lighting conditions then it's optimal, but in terms of what's actually been graded? Nah.

Thee algos do the best they can which ultimately works better than watching original DV in a room full of ambient light which some videophile-unsavvy consumers unknowingly may (likely) have been doing.

One example of the adjustment of thee original PQ curve to account for an environment of more ambient light than in a grading suite (600 nits) –

Originally Posted by Geoff D

lined your face with bacofoil

Is that what we here call aluminum foil?

[Penton-Man]

Originally Posted by Waboman

fallen into your couch cushions.

no way, as it’s well secured to the wall behind the small end table - https://forum.blu-ray.com/showthread...n#post16243793

[Lee A Stewart]

https://www.es.com/DomeX/

[Geoff D]

Originally Posted by DisplayCalNoob

Maybe the tone mapping that happens at the low end of the with metadata is what maintains that dynamic range that is essential to the HDR presentation.

And you can see into the shadow areas when the room is brighter than the sun how, exactly?

Originally Posted by Penton-Man

Thee algos do the best they can which ultimately works better than watching original DV in a room full of ambient light which some videophile-unsavvy consumers unknowingly may (likely) have been doing.

One example of the adjustment of thee original PQ curve to account for an environment of more ambient light than in a grading suite (600 nits) –

Is that what we here call aluminum foil?

"Doing the best they can" just to accommodate joe/jane sixpack who don't want to close their curtains still means its truncating the HDR signal and thus negating the "HDR" part of it to whatever degree. It's fine if we have 100,000 nits (!) of luminance at our disposal as the graph indicates, but that pertains to EXACTLY what I said about TVs not having an unlimited amount of headroom. People are lucky to get 1000 nits out of OLEDs or the bargain basement LCDs that pass as "HDR", never mind 10k nits and certainly not 100k nits! So if the entire image is just 'shifted' upwards then most consumer TVs are going to run out of brightness real fast to convey the speculars. 'Daytime' HDR is basically SDR with knobs on.

[Kris Deering]

Originally Posted by Geoff D

TVs don't have an unlimited amount of headroom though, that's what perplexes me so about DVIQ. Yes, if it can ramp up the APL so that it now looks as it should vis-a-vis the reference level when being slammed by ambient light streaming into the room then that's a good thing, but if your APL has increased by, say, fivefold your maximum peak brightness sure as shit hasn't, thereby drastically constricting the whole "high dynamic range" thing, in terms of luminance anyways.

Not that people's experiences with HDR aren't already all over the place but this sort of tech is only going to add even more confusion about what something should actually look like, and as it comes with the Dobly seal of approval people will still think they're viewing some kind of optimal presentation w/ref to the original grade. In terms of the lighting conditions then it's optimal, but in terms of what's actually been graded? Nah.

I mean, I'm glad that people will be able to watch "HDR" whatever the viewing environment may be but that cinematographer you've referred to before, Geoff something or other, hit the nail on the head when he was aksed about HDR in one of his videos. He said that if you're watching in viewing conditions with anything other than the light from the display then it's not actually HDR as the ambient light will wreck it. He even said that the light from the TV reflecting off your face is enough to disqualify it, though that sounds a bit extreme unless you've lined your face with bacofoil or something.

This has always been a conundrum with displays and content though. It is why some calibrators use a 2.2 gamma instead of the standard 2.4. But it also applies to displays that don't actually have enough contrast to do a proper 2.4 or even 2.2 and use something like 1886 for perceptual gamma. It all boils down to the same thing really. But I agree that boosting the bottom end to compensate for viewing conditions is going to limit overall dynamic range, but as I've mentioned before this is inevitable. The video market is now going through the same issues the audio market had when higher resolution audio became available because of the EXACT same reasons. So eventually HDR will be crippled by the BRIGHTNESS wars instead of the loudness wars because the mass market doesn't watch TV in perfect environments. So dynamic range will continue to be crushed over time. Once again we are in a situation where we learned absolutely nothing from the past.

Originally Posted by DisplayCalNoob

Maybe the tone mapping that happens at the low end of the with metadata is what maintains that dynamic range that is essential to the HDR presentation.

The only way to make the dark part of the image usable is to brighten it up, which means that no matter what you are sacrificing overall dynamic range. You'd still get the highlight portion, but the differences between it and the normal range would be compressed.

[DisplayCalNoob]

Originally Posted by Kris Deering

This has always been a conundrum with displays and content though. It is why some calibrators use a 2.2 gamma instead of the standard 2.4. But it also applies to displays that don't actually have enough contrast to do a proper 2.4 or even 2.2 and use something like 1886 for perceptual gamma. It all boils down to the same thing really. But I agree that boosting the bottom end to compensate for viewing conditions is going to limit overall dynamic range, but as I've mentioned before this is inevitable. The video market is now going through the same issues the audio market had when higher resolution audio became available because of the EXACT same reasons. So eventually HDR will be crippled by the BRIGHTNESS wars instead of the loudness wars because the mass market doesn't watch TV in perfect environments. So dynamic range will continue to be crushed over time. Once again we are in a situation where we learned absolutely nothing from the past.



The only way to make the dark part of the image usable is to brighten it up, which means that no matter what you are sacrificing overall dynamic range. You'd still get the highlight portion, but the differences between it and the normal range would be compressed.

The Dolby rep, mentions metadata specifically for the ambient light sensor. Brightness increases in the midtones may do thr trick.

[Kris Deering]

Originally Posted by DisplayCalNoob

The Dolby rep, mentions metadata specifically for the ambient light sensor. Brightness increases in the midtones may do thr trick.

Again, compensating for room environment is not that difficult, even the Panasonic player has settings for this. But compression is now unavoidable in overall dynamic range. That is not necessarily a bad thing, because it was already compromised by the viewing environment as it was, I'm just stating it to be clear. Again, we've seen this all happen before with audio.

[Penton-Man]

Originally Posted by Geoff D

....'Daytime' HDR is basically SDR with knobs on.

Negative Waves in the Morning - YouTube

I mean if you don’t like using Dolby Vision IQ with your LG tv you can always disable the ambient light sensor manually and figure out better math for the Applied Vision Sciences Group for IQ’s use with consumers who refuse to watch in a light controlled room similar to a grading suite given their lifestyles. Next up for the group is developing a solution for –

Originally Posted by Penton-Man

how to deal with judder in high contrast HDR scenes.

meanwhile Geoffrey, positive waves - https://forum.blu-ray.com/showthread...e#post15629485
show your happy face once in a while like -

Originally Posted by Penton-Man

some members of the team on set –

[Penton-Man]

P.S.
Think of it this way, not everyone has the will or way to see an endocrinologist for their diabetes, they settle for a family practitioner.

[Penton-Man]

Originally Posted by Penton-Man

your LG tv

Webinar: February 2020 Update, HDR and LG C9 Calibration plus Ask Me Anything - YouTube

[Penton-Man]

Originally Posted by Penton-Man

positive waves -

Van Halen - Jump (Official Music Video) - YouTube

[Lee A Stewart]

NTHU Research Team Develops New Display Material

A research team led by Professor Hsueh-Shih (Sean) Chen of National Tsing Hua University has announced that they have developed a new quantum dot material which is more stable and provides more realistic color. The main limitations of the current quantum dot display materials is that they are unstable and easily damaged, and the team has rectified this by growing “shield” with approximately 1-nanometer at the crystal facets of each quantum dot.

The new material has a wide range of applications, including the screens used in televisions, computers, mobile phones, and cameras.

https://displaydaily.com/article/pre...splay-material

[Lee A Stewart]

Trouble in Supplying OLED Panels Feared to Block OLED TVs’ Way into Mainstream in TV Industry

As LG Display's mass production of organic light emitting diode (OLED) panel plant in Guangzhou, China was delayed for more than half a year, a red light was turned on for the supply of large OLED panels.

According to the related industry sources said on Feb. 18, the mass production time of the Guangzhou OLED factory, which LG Display announced would begin to mass-produce in August last year, has not set the date even in February. The plant has been running test operations for seven months since it stopped mass production with production yields lower than a desired level.

Meanwhile, LG Display's panel supply has been only half of a level it originally planned. Although late last year, 130,000 OLED glass plates should have been produced a month at the Paju plant in Korea (70,000 units per month) and the Guangzhou plant in China (60,000 units per month), actually, the actual production volume is standing at 70,000 units a month. Given that the Guangzhou plant can produce up to three 65-inch panels and two 55-inch panels per glass plate, 300,000 panels have not been supplied a month.

http://se-cu.com/ndsoft/error.html

[Waboman]

Originally Posted by Penton-Man

Van Halen - Jump (Official Music Video) - YouTube

I just saw David Lee Roth last night.