Australian news website ITwire, has an article up about the MPEG’s announcement of the draft standard of their next generation of video codec, due to replace H.264 over time. Hopefully now that we’ve mostly settled on H.264 as the “one codec to rule them all” it will be a comfortable transition to the next generation.
The main benefit? The same as all codec evolution has provided: we trade off more power in the encoding and decoding end, to improve quality at ever lowering bandwidths. H.264 is two-to-four times more efficient than MPEG-2; and the proposed new codec is expected to be twice as efficient as H.264:
“There’s a lot of industry interest in this because it means you can halve the bit rate and still achieve the same visual quality, or double the number of television channels with the same bandwidth,” said  the chair of the Swedish delegation and organiser of the meeting, Per Fröjdh, manager for visual technology at Ericsson Research”
Of course, nothing is too simple. In the same way that H.264 is also known as MPEG-4 Part 10, and MPEG’s Advanced Video Codec (AVC) – H.265 will also be known by then names of “High Efficiency Video Coding” (HEVC) and MPEG-H Part 2. In fact the official MPEG name is High Efficiency Video Coding. The H.2xx series is a European standard for the same thing.
How important is this? In the next year or two it will have zero effect on our working lives. Over time, as the standard moves from draft, through the patent search/pooling phase and finally through to licensing, devices will start to adopt it; brewers will gain the ability to play it; silicon will be struck for hardware decoding in future generations of mobile graphics chips (and not-so-mobile as well); and then probably cameras based on the new codec.
The standard is expected to be finalized in 2014, and that will signal the start of adoption.
Longer term, doubling the efficiency of the codec means we can half the bandwidth for the same image quality we haven now, which is what I bet most people will do, or improve the quality at the same bandwidth. It will make HD streams more viable, and an “HEVCCAM” camera (not announced by anyone ever) would be attractive as the future 24Mbit record rate would equal a 50 Mbit record rate now. Twice the image quality for half the same bandwidth.
So, be aware of it, keep tracking it through the process, and determine when it needs to have an impact on your work life. At a minimum 2-3 years. There’s always the chance the MPEG-LA will make the license terms unattractive, but to date they haven’t gone down that path.
9 replies on “New MPEG Standard – H.265 – What does it mean?”
Mobile video is a much bigger concern to all parties involved now than it was 5 years ago. Back then services like Netflix, Amazon and iTunes weren’t doing on-demand video in any serious capacity, and delivery devices like the iPhone were JUST being released. So I guess what I’m saying is that there’s a lot more incentive now to push this standard into use than there was during the migration from MPEG-2 to MPEG-4. For Apple especially this could be a very big deal in delivering content on a lot of fronts, so I expect them to get in on it as early as humanly possible.
Besides iTunes content delivery, isn’t AirPlay a streaming h.264 implementation? As the Retina implementation spreads across the rest of the Mac lineup, the ability to push more pixels will become more important, though ironically people may still be waiting for for their Televisions to break past the 1920×1080 resolution barrier…
As you say, there are two ways to look at this for content distribution. Either existing quality at lower bit-rates, or higher quality at comparable bit-rates. With most ISPs being owned by Cable and Satellite providers these days, there’s not question that those providers aren’t invested in providing cheap bandwidth for services like iTunes or Netflix, which could become a compelling alternative to traditional TV for more and more people in the years ahead. Honestly, now that it’s becoming more and more clear that the internet will ultimately become the future of media distribution, Telcos should be forced to sell off those assets for the same reason that the Hollywood studios of the 50s were forced to get out of the theatre business. Innovation could be happening in media distribution on the internet that’s being blocked by the very companies that stand to loose money.
As it relates to iTunes and AppleTV, I’ll be interested to see what Apple does with the “extra” headroom. They eliminated SD material from the 2nd generation model, and in preferences you can choose to download either 720 or 1080 when renting or buying movies and tv shows. By the time h.265 arrives (potentially only a year and a half away) I could see them dropping 720 support entirely, and perhaps just offer two options for viewing, LT and HQ: both 1080, but one at a halved bit-rate, and the other at the current bit-rate meaning much improved picture and sound. I think Apple is motivated to provide higher quality material, as it’s in their interest for there to ultimately be no reason for someone to buy physical disks.
I’ve done a couple of bind tests I most people already can’t tell if what they’re looking at is 1080p AppleTV or BluRay. Double the quality at that file size, and I think the niche of people for whom BluRays will remain a “value added” will drop to almost nil. That said, they still have to work on higher quality audio, subtitles, supplemental features, etc. if they’re ever going to completely eliminate the disk.
Yikes! That’s enough for now…
“though ironically people may still be waiting for for their Televisions to break past the 1920×1080 resolution barrier…”
If the over rumoured Apple iTV doesn’t do this from day one; I would be gobsmacked…
Beyond my general doubts about Apple producing a full-on TV…
I think the reality is that 4K (presuming that will be the next standard) sets will be the niche of professionals for the time being. Why?
1. Cost. The rumoured 4K sets being released later this year are coming at a $20,000 price tag. Though Apple’s supply chain a ability to bring new technologies to market is enviable, even they can’t make panels at the sizes people want for a cost anyone outside of content creators would be willing to pay. Just look at Apple’s gradual implementation of Retina screens across their Mac lineup. Some were expecting the whole lineup to be refreshed with Retinal displays, but the 15″MBPwR is a trial balloon. It’s clear it’s going to take a year or two for Apple to bring down the cost of 2x resolution screens enough that they can put them on MacbookAirs without raising the price AND refine the manufacturing process to be able to technically build “Retina” screens at 27″ sizes for iMacs and Cinema Displays.
2. From a content standpoint, the ability for traditional cable/satellite providers to push 4K material is dubious, considering most stations are still pushing either 720p or 1080i material (and heavily compressed at that!). And while Apple COULD push 4K download options, the bandwidth required even under h.265 would make it a bandwidth cap slaughter.
3. Perceived benefit to viewers. The truth is that most people aren’t either aren’t watching large enough TVs, and/or are watching them from too far away to even be able to see a increase in sharpness. See this chart –
http://hd.engadget.com/2006/12/09/1080p-charted-viewing-distance-to-screen-size/
As far as the screens go I think, and I could be wrong, that they are produced about the size of a garage door (or there a bouts). So if that is correct (?) it would be not much more that sticking 4 iPad Retina screens together (or cutting them to size). I don’t see how that should cost anywhere near $20,000. Ofcourse the greedy bastard ‘production quality’ monitor makers will spin the shit out of it, but, the real problem I see is how do you get a graphics card to run the bloody thing!
A 4K 27″ iMacPro – yum yum!
My understanding is that any sheet (and I think you’re correct about the size) comes with bad pixels. The smaller the screen you’re cutting, the more units you can get out of it regardless of bad pixels. At 27 inches, enough bad pixels ruins the whole sheet. Or so is my understanding…
The pixels would be so small would you see them? Bad pixels varies with each seller; Dell have a no bad pixel policy for their better monitors, whereas Apple don’t really care. I’de still buy one though… 🙂
If you look at HEVC site. It says they picked BBC/Samsung implementation to H265. So that means Samsung will be getting a lot patents.
The way MPEG licensing works is via a central body – MPEG-LA (Licensing Authority) with fair and equitable licensing of all patents in a pool.
Ah, just as computers are getting powerful enough to comfortably edit and manipulate H.264 natively here comes H.265 to bring our machines to a crawl again. The vicious circle of tech life continues.
Marcus R. Moore,
I don’t think what consumers can actually see is as important as what consumers are led to believe they can see. The perception that bigger is better, plus keeping up with the Jones’, plus TV set makers eventually ceasing to make HD sets (just like they killed SD sets) makes me think that 4k sets are inevitable. I think it will be close to 10 years before it becomes common but I think it’s inevitable than for no other reason than TV makes and video camera makers need to keeping coming out with newer and ‘better’ products to sell in order to say in business.