Original URL: http://www.theregister.co.uk/2013/01/11/h265_mpeg5/
Next-gen H.265 video baked into Broadcom's monster TV brain
March to MPEG-5 for mobes, 4K screens reaches silicon
Vid 720p video at 30 frames per second and less than half a megabit a second: that's the promise of the H.265 vid compression tech, which is up for ratification any day now and already has chips ready to decode it.
The codec scales from bandwidth-starved mobiles, allowing one to watch streamed TV in the park, to monster 4K Ultra HD tellies. Getting the tech into silicon could deal H.265's rival video compression systems a deadly blow.
Broadcom's new BCM7445 chip has a quad-core ARMv7-A processor to run apps, games and other stuff one expects from a modern TV or set-top box. But it can also chomp through video encoded with H.265, properly known as High Efficiency Video Coding until it is rubber-stamped as an official MPEG and ITU standard. The chip is aimed at Ultra HD "home gateways".
For those of us on smaller screens or simply not at home, the H.265 codec squeezes twice the video into the same channel, or the same video into half the bandwidth, as the ubiquitous H.264 (MPEG-4) coding which dominates today despite Google's ongoing attempts to get everyone using its VP9 standard.
VP9 is a successor to VP8, and like its predecessor it can be used without paying a royalty fee. It's already embedded into Google's Chrome browser to play back HTML5 video files in WebM containers, but despite VP8 being around for a couple of years and free it has not challenged the dominance of the non-free H.264. The promoters of H.265 including Cisco hope that momentum continues, as this underwhelming video tries to show:
Broadcom also has intellectual property in H.265, and at CES this week announced chips supporting the standard and pushing it to 4096 x 2160 (4K or Ultra HD) and 60fps - even The Hobbit was only shot at 48 frames per second and most films happy tick over at 24 - and it’s the chips which matter in this business.
Decoding video in software is possible, but it's hard work so most devices - phones, tablets, TVs and Blu-Ray players as well as personal computers - offload the mathematics to specialist chips that consume less power than a general-purpose processor running the numbers and generally do a better job*. But developing those chips is a significant investment, and companies won't spend the money unless they're very confident about the future of the codec - the very point which undermined VP8.
Cisco and Broadcom will back H.265 (aka MPEG-5, aka HEVC), and device makers will have to bet on the codec which they think is going to dominate for the next few years. This will almost certainly push VP9 into software decoding at best and make Google's technology the power-hungry option for video playback. ®
* One can raise this example against the long-term success of software-defined radio, but that's an argument best had over a pint or two.