Retina Displays and 4K TVs May Not Be Worth the Trouble
Scientific American article -
Why Retina Displays and 4K TVs May Not Be Worth the Trouble: Scientific American
Your digital screen has more pixels than ever, but all that visual detail comes at a cost
Comments
I received an email from a friend who said she couldn't access the article from Scientific American. Possibly other people can't either, so I'm posting the article.
Why Retina Displays and 4K TVs May Not Be Worth the Trouble
Your digital screen has more pixels than ever, but all that visual detail comes at a cost
When Apple unveiled its Retina screen on the iPhone 4, the world gasped. “There has never been a more detailed, clear, or viewable screen,” read a review on the tech Web site Engadget. “Staring at that screen is addictive,” said Wired magazine.
What they were reacting to was the superhigh resolution. The iPhone 4 packed in 326 pixels per inch (ppi)—pixels so tiny that you can't discern them at standard viewing distance. Apple went on to incorporate Retina displays into the iPad (264 ppi) and MacBook Pro laptops (227 ppi).
So began the Resolution Wars. Recent phones from Samsung, Nokia and HTC pack in 316, 332 and 440 ppi, respectively. Google's Nexus 10 tablet leapfrogs the iPad with 300 ppi.
And now the television industry has joined in. It is pushing 4K sets—that's four times the resolution of high-definition TV. Four times.
Now, up to a point, higher resolution really does look better. Yet there are some footnotes.
Low-resolution graphics look no better on a high-resolution screen. If you've programmed an iPhone app, you know that it doesn't look any sharper until you reprogram it for the sharper screen. Until then, the phone just applies pixel doubling (substitutes four pixels for every one on the lower-res screen), which doesn't improve sharpness.
In fact, they look worse. You may remember that when HDTV came out, standard-definition broadcasts actually looked worse than they did on standard TVs. (They still do.) Well, guess what? Same thing happens on other screens.
In theory, standard-res graphics on a high-res screen look exactly as sharp as they always did, thanks to pixel doubling. Yet as many MacBook purchasers discover with dismay, pre-Retina graphics look worse on Retina screens. This might be because a standard screen smooths out gaps between pixels, but on a Retina screen the gaps are so tiny, the subtle smoothing goes away.
In any case, the problem is especially severe on the biggest app of all: the Web. Few Web sites have been rewritten to accommodate Retina-type screens, so their graphics usually look awful.
Bigger = slower. Even if Web designers do get around to designing high-res versions of their graphics, those files will be bigger and therefore slower to load. On cellular phones and tablets that dole out Internet service by the megabyte, they are also more expensive. Do we want to wait longer and pay more to have those sharper Web sites? Shouldn't we be able to choose?
Already our Internet providers impose monthly data limits. Do we really want each Web site to eat up, say, four times our monthly data?
Sharp text should be automatic but isn't always. The previous points do not apply to text. Text is not graphics. Whenever a program or Web page displays text, Apple's Retina software automatically delivers extremely sharp characters to your screen.
Unfortunately, that's true only if the software companies use Apple's prescribed text-handling routines, and not all of them do. For example, documents in Adobe's InDesign layout program look horrible.
4K TV broadcasts? Forget it. The Retina-zation of television is particularly absurd. No cable or satellite company will send out 4K broadcasts because, in the bandwidth space of one 4K channel, providers could send out four HDTV channels. (Companies already send out low-res versions of HDTV channels to conserve bandwidth.)
The data required for a 4K video is also too great for DVDs, Blu-ray discs or Internet streaming. So what, exactly, will you watch on a 4K set?
If you buy a Sony 84-inch 4K set ($25,000), the company will loan you a hard drive containing 10 Sony movies in 4K.
That's it? We're going to ship hard drives?
The hardware we need for our superhigh-resolution future is here. Now we need to figure out—on our phones, Web sites and TVs—how we're going to squeeze in all that high-res content.
SCIENTIFIC AMERICAN ONLINE
Test superhigh-res marketing claims with math:ScientificAmerican.com/apr2013/pogue
Great rebuttal, Tallest Skil~!
Is this sarcasm, or are you agreeing that the article's fallacies speak for themselves?
"I want to go to the Moon."
"No, that's too hard; let's stay in LEO."
I mean…
I was just passing on the Scientific American article for what it's worth. My response was to the rude 'Wahh Wahh' that you directed at me.
You? Oh, no; apologies if you perceived it that way. It was a 'summary' of the content of the article.
Sony is planning 300GB Blu-Ray:
http://www.bbc.co.uk/news/technology-23492609
For film, 4K is pretty close to the master resolution so they can sell collectible movies as being the original master quality. For the most part, consumers won't tell the difference but with bigger discs, entire TV series can fit on one disc along with extras. Blu-Ray at that size would work for inexpensive archival purposes too.
I don't think that optical media will work for consumer use at that size though. It would take hours to burn a 300GB disc and it just means more chances for disc errors during burning. For commercial distribution, there's the hurdle of getting people to adopt all new Blu-Ray players, which was bad enough the first time.
I think retailers will push 4K displays because they need to drive sales. This also helps drive sales of things to support 4K like new HDMI cables, media boxes etc. I don't think it's needed and I don't think people will see the benefit for the most part but the alternative would be sticking with what we have and retail tends not to stand still here when there's something new they can push. This happens sometimes like with 8-bit color. That's actually something that should have changed by now because 8-bits per channel only gives you 256 variations for each color and this is now being spread over displays that will have 3840 pixels in one dimension. 12-bit would give 4,096 variations, although 10-bit would suffice as it's not likely to have just one color spread over a large area. It's a harder sell though.
What I think will happen is we'll see the 4K displays before we see the content and it will be sold alongside the visual upgrades and new form factors that comes from the new display tech. I think 4K content distribution will remain a very small market for a long time and mostly be limited to special broadcasts. Some live sporting events get broadcast in bars on large TVs so for the odd occasion, it'll be a benefit. Even then, I suspect most will broadcast in 1080p and it'll be upscaled.
Internet speeds will improve eventually. 4K needs around 20Mbps+ for H.265 broadcast so it's doable. For the bulk of disposable TV, I don't think there's a benefit for them to move to it. The best focus for TV content is just to move away from 24/7 broadcast to on-demand. On-demand is actually easier to fill a 24/7 broadcast because people can watch whatever they want in a library whenever they want. Even news channels don't have to air 24/7, just when there's news but there's no wasting a dedicated channel.
Display manufacturers - move to 4K, OLED/IGZO, no harm in having nicer displays
TV broadcasters - stick with 720p/1080p, move more to on-demand
movie distributors - either stick with 1080p on Blu-Ray or move to 4K masters on new Blu-Ray. Already on-demand online really but it would be nice to see some reform on the licensing to avoid things like this:
http://www.theverge.com/2013/4/30/4287902/netflix-losing-almost-1800-titles-from-its-streaming-library-starting-tomorrow
They need to find a way to pass the costs direct to consumers without removing the barrier to entry (the pay-to-watch barrier) kind of like free-to-play games. If you like it, you make an in-movie purchase or something and it can be automatic e.g after 20% watched, it starts the transaction but allows you to back out before it's done.