[opendtv] Re: TV Technology: 4K for Broadcast: Is it Worth the Expense?

From: cooleman@xxxxxx
To: opendtv@xxxxxxxxxxxxx
Date: Sun, 02 Jul 2017 14:11:53 +0200

Manfredi, Albert E schreef op 02-07-2017 8:11:

Donald Cooleman wrote:

Dive into the archives TV Mark (Schubin) posted about this in the SD
vs. HD days.

Granted, but everything has to be put in its proper perspective. Large
screens of the future are likely to be just as significantly different
from those of today, as those of today are from 20 years ago. One
example would be floor to ceiling display, but another one is the
tabletop displays (the tabletop *is* the display) that we already see
in TV shows like NCIS. So it's a totally different ballgame. Visual
acuity has to be taken into account. Rule of thumb, ~1 arcmin of
angular separation starts to become noticeable, for humans with 20/20
vision (or less than great vision, but wearing appropriate
spectacles).

The point was that scaling engines are good enough to show little loss from the excercise of down- and upscaling. But one has to start with high quality source for that to be the case. So, 4K is only 'eliminatable' for distribution.

So, exactly like 20 years ago, we're talking about all types of
displays that have to render the image. Some do fine with SD, others
don't. In a broadcast medium, it's wasteful to simulcast for all types
of displays, while if over the Internet, you can optimize better, for
the transmission aspect anyway.

Yeah plenty of HD displays did not handle SD to well, one more reason we got the U60 close-out sale. The last series of Panasonic plasma, but the budget model, XGA only, both (panasonic) Plasma and the XGA's tended to do better on lower quality sources, and we have KPN TV AKA Digitenne, and that is when the definitively threw out the concept of Broadcast Quality. The original plan was for 40 channels, but actual testing revealed actual coverage area was only half the projected coverage, so higher FEC rate and half the available bandwidth. Now it has grown to 32 SD channels, so forget about passable HD if that is only for 9 channels;-). Of course it will be HEVC, so 265 not old hat 264 when KPN is forced to switch over to DVB-T2 in 2020.

Secondly Siebert is taking about all broadcasters, whereas Hayes seems
to speak for his constituency of US Big Stick owners.

I think the point is more like, for the next 20 years, what image
standards will make sense to have ready to go for consumer displays?
The answers to such questions always depend on one's assumptions.
Obviously, if we assume that screens will remain in the max ~60"
range, 1080p is fine.

If they offer actual 1080P, and not go the BBC route that cut the bitrate in half on satellite, to match the Terrestrial 'quality'. With the excuse that people to far away, so we don't have to actually offer HD. And that was still only 1080i not P. So actual HD not pre-filtered less than HD with a nominal HD raster. And if one starts with 4K capture. Though I prefer to see 8K capture and 4K delivery, or 8K delivery. As I am one of those pixelpeepers Mark Schubin keeps making fun of, lol. And that 13" 8K LCD at IBC two years ago looked mighty fine (no pun intended).

Somewhat related, remember when Craig thought that 4:3 displays would
always be popular, just because they were then? At the time I said no,
probably not, because LCDs do not have the restrictions of evacuated
glass bottles. There's no reason to assume squarish needs to survive.
Of course he argued and argued and argued, but the facts speak for
themselves. The same applies to size. Restrictions of today won't last
forever.

21:9 is now also gone, they are still there in small format, gaming monitors, and LCD makers are now able to cut panels to shape for retail applications, but in TVs they are gone. Perhaps,that's why Jeroen left TPV, LOL.

So, even with DVB-T2 one can't do much in 3 multiplexes (once there was
the promise of 6 following analog switch off).

That 6 was always dubious. In spite of all the hype, which was always
hard to believe, it makes sense that H.264 gains **most** over H.262
in restricted channels (as was always intended). If you want HD
quality, it's not likely that you get anything like a *persistent* 2x
improvement, H.264 over H.262. So, in an 8 MHz Euro TV channel, say
about 24 Mb/s, whether you use H.262 or H.264, roughly 8 MHz average
per channel, for 3 HD channels, sounds eminently sensible. But of
course, with H.264 you can also fit fast action in that 8 Mb/s
average, or 4 streams, reasonably well, especially because H.264 masks
blocking artifacts (H.262 does not).

Yeah but we did start with five.

Digitimes reported that LG was launching 82" OLEDs in Taiwan, but no
confirmation anywhere, no response from LGE. So for now LGD is sticking
to its 77" limit, though it can do larger, if there really is a client
(that is willing to pay for increased cost due to lower yield, I guess).

Yes, and in the late 1990s, Craig was arguing that HDTVs cost in the
10s of thousands of dollars, and were out of reach of the common
mortal. But even back then, the point was, spectrum-compatible HDTV
(i.e. digital HDTV) was being designed FOR the common mortal. Not as a
premium service. Now we have 4K sets already, available to the common
mortal, **even before** their size warrants the pixel count! (Another
point that took forever to get across.) The prices are totally
competitive already. Affordable, bigger sizes, are the next step.

NHK has shown 8K on 4 4K LG OLEDs don't remember if they were 55" or 65", but they were the thin ones that unfortunately were carried in a huge aluminium frame/bezel, so a very clear cross in the middle. Now if we can have little or no bezel, and the 'wall' form factor, 4x77" for example we are getting there;-). NDS/Cisco had a demo like that, but with more panels a number of years ago at IBC, I was not important enough for to be admitted. Their demo was focussed on the management/driving the content not so much the display side. LG is pushing this into museums and placed like that.

The regular 77" in the US just dropped 5K in MSRP from 2016 to 2017. As now there is the W(all) series model being this year's signature model at 20K. 2016 LG dropped its MSRP on the 77" from 30K €/$ to 20K just before the actual launch. LG OLED still a niche, but yield and production volume must be up there now.

Of course at the High-End vendors are finanlly beginning to care about brightness controll and master HDR and ow APL/near black performance, targeting the cinema market in case of Sony (Cledis microled) and Samsung 2.5mm pitch SMD LED's demoes at Cinemacon this year.

We have long been discussing indoor video use of direct-view LED displays, when they hit below 2mm pitch (the 4mm of a decade ago, was also impressive enough), then it was for HD now for 4K and 8K if one has the room or one goes well below the 1.2mm pitch demoed at ISE and CES.

Yes this is the 300-1000K end of the market, very, very high-end indeed. There was talk of Runco applying Leyard LEDs, but Runco got folded, when Leyard bought Planar.

Bert

----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:

- Using the UNSUBSCRIBE command in your user configuration settings at
FreeLists.org

- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word
unsubscribe in the subject line.

----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:

- Using the UNSUBSCRIBE command in your user configuration settings at FreeLists.org
- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word
unsubscribe in the subject line.

References:
- [opendtv] Re: TV Technology: 4K for Broadcast: Is it Worth the Expense?
  - From: Manfredi, Albert E

[opendtv] Re: TV Technology: 4K for Broadcast: Is it Worth the Expense?

Other related posts: