Talk:Ultra-high-definition television

Ultra HD Premium logo missing

The article states "their Ultra HD Premium logo (seen to the right)." I have removed that part for now. Dawnbandit1 (talk) 02:05, 17 January 2017 (UTC)

T220, and also the term "2160p"

I have proposals for two changes to the article.

1. The section in history regarding the IBM T220 should be removed, as that monitor has nothing to do with the UHDTV standard other than a similar (but not the same) resolution. The T220 is a monitor from the 2001-2002 era (i.e. long before the UHDTV specification). This wikipedia article asserts (without reference) that this is "the first UHDTV production monitor" (not sure whether that means "the first monitor designed for production of UHDTV content" or "the first monitor with UHDTV-class resolution to enter mass production"). Even though it long precedes the publication of the specification, of course it's possible that it was designed in anticipation of UHDTV, or that is was designed for video production for cinema, which has been working with 4K resolutions since before UHDTV came along. However, reading IBM's own press release, this is their description of their own monitor: "ARMONK, N.Y. - 27 Jun 2001: IBM announces the T220, the world's highest-resolution flat panel monitor, enabling photograph-quality imaging for science, banking, engineering, publishing, medicine and business critical visualization tasks." http://www-03.ibm.com/press/us/en/pressrelease/1180.wss This monitor really has nothing to do with UHDTV or cinema production, and associating it after the fact does feel a bit like trying to rewrite history simply because some person started googling for old high resolution monitors to write a "history" section here.

2. There is no legitimate reference for the term "2160p", just various "4K explained"-type articles whose authors decided to use that term. It seems obvious they are just mimicking the naming convention of past video standards (720p, 1080p), and I don't consider this a legitimate source for "2160p" being an "official" shorthand for resolutions like 3840×2160 or "4320p" for 7680×4320. I have not seen any endorsement of these terms from the likes of ITU or other sources we would use for "official" information about UHDTV. As such I think these terms should be removed from this article. "Some tech columnist somewhere called it that" is not good enough for it to be endorsed here, in my opinion. GlenwingKyros (talk) 18:52, 13 October 2016 (UTC)

Where does the 144Gbps uncompressed video bit rate in the info panel come from?

Frames of 7680 pixels wide by 4320 pixels tall (33.18 megapixels) by 24 bits per pixel by 120 frames per second comes to 95.6Gbps. Even 8B10B/16B20B encoded, that's still only 119.4Gbps. Even with 32 bit pixels, that's 127.4Gbps straight and 160Gbps 8B10B/16B20B encoded.

Graham.Fountain | Talk 11:11, 3 September 2019 (UTC)

It's HDR-capable. 12 bits per color plane at 120Hz: (7680 * 4320 x 12 x 3 x 120) / (10^9) = 143.327232 -- The Anome (talk) 11:39, 3 September 2019 (UTC)

It comes from the cited sources, which is always the first place you should look if you are wondering where anything on wikipedia comes from :P http://www.nhk.or.jp/strl/open2013/tenji/tenji03/index_e.html That being said, it is wrong, as noted above the raw calculation is 143 Gbit/s rather than 144, but the math ought to include blanking intervals anyway, which aren't really a set standard but the most common would probably be 8800 × 4500 @ 120 Hz 36 bit/px = 171 Gbit/s, using CTA timing.[1] GlenwingKyros (talk) 16:02, 3 September 2019 (UTC)

Blanking intervals! That certainly brings back memories. There's certainly some justification for sampling a little bit outside the edges of the frame internally within the very start of the image capture pipeline, and a logical way to carry that content within a near-standard bitstream, but I don't think consumer content is ever likely to contain that information. -- The Anome (talk) 16:56, 3 September 2019 (UTC)

Blanking intervals aren't some obsolete concept unfortunately; while their original purpose is no longer valid since we don't use CRTs, the intervals themselves are still used today in all consumer devices including HDTVs and UHDTVs, for transmitting secondary data, such as the inline audio stream, and all video metadata (including the HDR infoframes). At least, that's how it is in interfaces like HDMI, DVI, and DisplayPort. For TV type content, CTA intervals are almost universal. But I guess if the table here is interface-agnostic we could make a case for not including blanking overhead.GlenwingKyros (talk) 17:20, 3 September 2019 (UTC)

I did look at the cited sources and didn't see anywhere that specified 36 bits per pixel. I did see references to 12 bits, and I now know what that meant, i.e. 12 bits per colour, 3 equal weight colour values per pixel, so 36 bits per pixel. But that didn't leap out and now I know what it meant there's no point in going back and seeing if it was explicit and I just missed it or if it was implicit and assumed knowledge I didn't have. So while they do support what's in the article, they didn't do the job of being pedagogic as such - especially not that written in Chinese. Also, if the article is going to be encyclopaedic, I suspect there needs to be, at least, enough to verify a derived value like that in the article itself. Maybe I'm lazy, and just want to be spoon fed; but, in my book, that's a lesser sin than being rude to the users.

As to blanking intervals, all that is needed is to identify line and frame ends/starts. So - unless the display head has no FIFO queue to adapt the data transfer rate to any display burst rate it, specifically, implements - I don't see how, on its own, that could possibly add 20% ((171/143 - 1)*100). There's certainly no need to add periods of quescence into a digital data stream to allow for that. 16B20B encoding does add 25%, but it's not used just so values outside the 16-bit set can be used as delimiters. I admit, however, that I may be pointing out the absurdity of adding blanking periods as a small act of petulance.

The carriage of additional data streams, e.g. for sound and subtitles, etc., is out of scope to me. But I can see how, in many cases, it's a valid issue that wants a note somewhere; and, were it included in the given value, it would certainbly need explanation.

Graham.Fountain | Talk 10:23, 4 September 2019 (UTC)

I wasn't trying to be rude, it's just that my first instinct was to open each source and ctrl-F for "144" (and Google translate operated automatically, so actually I didn't even realize one was in another language xD), and the origin of the figure was immediately found that way. Sorry if my tone came across the wrong way. Also I only mention blanking intervals for consideration, despite the fact that it's true they add a lot of overhead (depending on the standard) and aren't strictly needed for physical operation of a display these days, they are used commonly (i.e. search for 17.82Gbps, which is 4K 60 Hz with CTA timing, the same standard), which is the timing used by virtually every 4K TV in existence). You're right, it's probably more trouble than its worth to account for those (and more difficult to cite, since it's a calculation rather than a page that says the result directly). Up to you guys if we care enough to include them.GlenwingKyros (talk) 15:33, 4 September 2019 (UTC)

Comparison of 8K UHDTV, 4K UHDTV, HDTV and SDTV resolution

What does exactly that image represents? I only see bigger sizes but for which constant feature? --Backinstadiums (talk) 17:08, 13 March 2020 (UTC)

Constant pixel size. GlenwingKyros (talk) 19:06, 13 March 2020 (UTC)

@GlenwingKyros: Is aspect ratio a third variable (together with display size and viewing distance) into play for the optimal viewing distance? Ottherwise, how does it affect the display quality? --Backinstadiums (talk) 19:33, 13 March 2020 (UTC)

Aspect ratio is not related to display quality or optimal viewing distance. GlenwingKyros (talk) 20:07, 13 March 2020 (UTC)

Remove list of UHD channels

Ultra HD broadcasting is no longer a novelty and therefore I don't think it serves much purpose to have a list of such channels. It seems to be just a target for vandalism these days than anything else.GlenwingKyros (talk) 21:38, 16 December 2020 (UTC)

I'm inclined to disagree. UHD is still very much in its infancy. Were this about HDTV, listing HD channels would be pointless (and impractical) as HD is now very widely broadcast and received, and is arguably pretty much the norm now. However, UHD is still being established and still very much a new and minority format (and not 'succeeded' by a newer one, yet); it's not widely broadcast and the channels listed represent only something like 1% of the channels broadcast (a very rough estimate but I think the right ballpark?). I do agree that, sadly, it's a target for vandalism but that shouldn't be a reason for its removal Satbuff (talk) 08:33, 21 December 2020 (UTC)

Honestly I just have a hard time believing such a list is really helpful to anyone or serving any purpose. Whether it's a minority or not compared to other formats, or how many there are comparatively percentage-wise really isn't the question for me. It's a large enough list numerically that I don't really think it's worth maintaining, as it's not really notable enough to warrant listing anymore. But that's just my opinion. GlenwingKyros (talk) 18:26, 21 December 2020 (UTC)

(5 months later) Edits to this list are pretty much exclusively vandalism still. Again I really doubt this list is providing any useful purpose to anyone. GlenwingKyros (talk) 18:29, 14 May 2021 (UTC)

The number (and proportion) of UHD channels IS (at least part of) the question for me as this shows that broadcast UHD is still a novelty, in its infancy, still developing, and so these 'pioneer' channels are notable. Granted there hasn't been a whole lot of genuine changes to the list in the past few months but that's hardly surprising in the circumstances. Again, while removal would prevent vandalism, vandalism is not a reason for removal. Satbuff (talk) 08:17, 26 May 2021 (UTC)