Technology Quarterly | Difference engine

After 3D, here comes 4K

Home entertainment: A new television standard called Ultra HD is four times sharper than today’s best HDTVs. But providing content in this new format poses daunting technical challenges. And does anyone really need it anyway?

HAVING seen interest in 3D television fizzle, electronics firms are eager to find another blockbuster product that will get consumers to splash out. The development most are hoping will do the trick is a display technology known as Ultra High-Definition, which offers four times the resolution of today’s most advanced “1080p” HDTV sets. No question, Ultra HD provides stunning images—at least when displaying content created in the new “4K” video format. Unfortunately, only a handful of feature films (including “The Amazing Spider-Man”, “Prometheus” and “The Hobbit”) have been shot with 4K-capable cameras.

Will 4K follow 3D’s fate? It is far too early to say. But 4K television—far more than 4K cinema—faces some formidable obstacles. As with 3D television, the new 4K television format takes its lead from Hollywood. The existing wide-screen digital format used in cinemas is 1,998 pixels wide and 1,080 pixels tall. The 4K film standard has twice the resolution vertically and twice horizontally—ie, 3,996 pixels across the frame and 2,160 down—making it four times sharper all round.

By contrast, the 4K format used in television is slightly narrower, having 3,840 pixels across and 2,160 pixels down. The screen width was reduced to maintain the 16:9 aspect ratio of HDTV, with its 1,920-by-1,080 pixel count. This makes it possible to show existing video content that has been “upscaled” for Ultra HD without the need for black “letterbox” bands above and below the picture.

But who actually needs a super-sharp 4K television? The resolution of even an HDTV set with 1,080 progressively scanned lines (ie, continuously from top to bottom) is wasted on the vast majority of viewers. Most people sit too far from the screen to be able to see the detail it offers. A study done some years ago found the median eye-to-screen distance in American homes to be nine feet (2.7 metres). But researchers reckon that, given the human eye’s limited acuity, people even with 20/20 vision should sit no farther than 1.8 times the width of the screen away from it, if they are to distinguish the detail displayed.

At a distance of nine feet, even an existing HDTV set would need to have a screen of around 70 inches across the diagonal for viewers to benefit from the resolution they have paid for. With anything smaller at that distance, details simply blur into one another. There is no question that, with twice the resolution horizontally and vertically, a 70-inch Ultra HD screen would be pretty impressive from nine feet away, and would still provide resolvable detail at up to twice that size.

So the best argument for moving to Ultra HD is the trend towards larger screen sizes generally. Today’s HDTV sets begin to look spotty when their meagre 2.1m pixels are spread over screens greater than around 80 inches. With 8.3m pixels to play with, Ultra HD screens can be made twice as large before the pixels become too glaringly obvious.

Assuming you have the space and the budget for an 80-inch-plus TV, the next hurdle will be getting native 4K content onto its screen. In raw form, a two-and-half-hour film shot in 4K at the usual 24 frames per second contains 216,000 frames. With each frame of the film containing 8.6m pixels, and each pixel having 24 bits of colour information, the resulting video file contains 5.6 terabytes of data. Even with compression, transmitting such gigantic files over the air or by cable would require more bandwidth, at far greater infrastructural cost, than broadcasters can afford. Streaming a feature-length 4K file over the internet would run into similar bandwidth constraints. The internet connection needed would have to transport data at speeds up to a gigabit a second. Few homes have broadband that fast.

An alternative would be to distribute 4K films as Blu-ray discs, just as conventional 2K films are sold—or were until people switched to streaming them instead from Netflix, YouTube, iTunes and other online services. A 2K film etched on a Blu-ray disc uses all 50 gigabytes of its two recording layers. A 4K film would require a third or fourth layer. Even then, a more efficient compression method than the current H.264 standard would be needed to shoehorn a 4K film onto a Blu-ray disc.

Get ready for a squeeze

So a new compression standard is evidently required. How “lossy” could a compression method afford to be before it compromised the quality of a 4K picture? The international bodies responsible for compression standards have been discussing a successor to H.264 for the best part of a decade. The latest draft, known as High Efficiency Video Coding, is said to double H.264’s compression ratio without loss of image quality. Even if that proves up to the job, it will take years for the new compression standard (H.265) to be adopted universally. Many in the industry feel that if 4K television is to succeed, an entirely new way of encoding and delivering its content is needed.

Sony’s answer is to bundle a media server with its first generation of Ultra HD television sets. The server will come with ten films preloaded on its array of hard drives, along with a selection of short videos. Sony says the films included are direct copies of pristine 4K masters. But no one has said how customers will add fresh titles to their servers. This is obviously a stopgap measure, designed to encourage wealthy early-adopters to splash out on the latest video fad. (Sony’s 84-inch Ultra HD set costs $25,000, and buyers are also given a media server on loan.) But if Ultra HD is to be HDTV’s successor, then sooner or later cable and satellite-TV providers and streaming-video services will have to find a way to deliver 4K content reliably and cheaply. No doubt, they eventually will. The question is when.

The best guide is the penetration of HDTV. America’s first nationwide broadcast in digital high-definition was John Glenn’s lift-off in the space shuttle Discovery in 1998. It took another dozen years for HDTV to go mainstream. By that reckoning, it is likely to be 2025 before Ultra HD is in half of all American homes.

This article appeared in the Technology Quarterly section of the print edition under the headline "After 3D, here comes 4K"

A little faster, George?

From the March 9th 2013 edition

Discover stories from this section and more in the list of contents

Explore the edition