informitv / news
Opinion
Telco triple play or media gateway?
IPTV is now a hot topic in the telecommunications industry and infrastructure companies are making the case for converged internet protocol networks providing video, voice, and data triple play services. Will the service providers have it all their own way? How will the humble set-top box evolve in this market?
Paul Walsh of set-top box design company Vidanti offers his personal perspective.
Triple play and beyond! Who wants just the ‘triple play’ of converged video, voice and data? Make it mobile. Make it guess what I want to watch. Do it, and do it now!
There’s nothing wrong with futurism, but maybe we should settle first for really converging the triple-play services and truly integrate what we already have before bundling in another service.
How many companies will ultimately get to play in the set-top box game? Is it already an over-crowded market? Will companies be forced to exit due to lack of market share or be acquired and merged into bigger entities?
Analyst reports are already proving empirically that, if they all stay in the market, each set -top box vendor can expect to enjoy a minuscule portion of the pie and to re-coup the research and development expenses for, well, at least one day last year. So, it’s inevitable that many will retire and go home. The lucky ones will be acquired and exit strategies will be delivered, to at least some level of revenue multiples. Right?
Yes and no. The next year will see the exit of companies that have relied on MPEG-2 technology. It will also see the entry of many new ones who have found mastery over MPEG-4 AVC / H.264 and other low bit rate codecs. By and large, the former will have earned revenue from the video market, perhaps in closed systems such as the hotel sector. The latter will have earned revenue from providing home networking and broadband products.
So, for a while, the market will get more crowded rather than less, but the players will be different.
Muddleware
What about all those middleware solutions, each one different from the rest? Can the IPTV industry make real progress without some unifying standards for the software that manages the content assets at one end and tells you what’s available to view at the other?
It seems pretty unlikely. It is more likely that there will have to be some standardisation that still allows for individual company creativity. Just like the internet, in fact. The only problem is that such efforts are late in starting, and the delay could hold back the deployment and take-up.
Is IPTV going to be a version of the cable company or the satellite TV company— one where you buy your service and the technician comes around to install the boxes, complete with a security card that stops you doing illegal things? Or is it going to be like free-to-air TV, where you buy your own Freeview box and self-install?
Hybrid services
It must now be the consensus at least for Europe and large parts of Asia that there is no point streaming free-to-air TV channels over precious broadband capacity when there is a digital terrestrial multiplex to do that for you. It is not so much a prediction as a statement of fact. Slightly more interesting, but equally as likely, will be the hybrids that combine IP with cable and satellite.
The cable and satellite TV industries are concerned about IPTV. Not much, because they know that the barrier to entry for them is not that different. Want to be like a telco? Buy an ISP.
They can sleep easy at night in the comfort of the notion that ‘Telco TV’ is going to consist of a bunch of standard definition television channels that would otherwise have been transmitted from terrestrial, combined with a video-on-demand library full of second-rate titles.
Meanwhile, they can offer high-definition television. “Cable will always be superior,” they’ll insinuate; “satellite TV is the real cutting edge”. Yet inside each receiver, decoder or set-top is a video processor chip that is identical at the MPEG-4 H.264 level, and an internet protocol based system is no different.
For the next two years, while the typical broadband service provider is building out the infrastructure and managing those content agreements to the letter of the contract, it will be just like the cable company. Then, almost overnight, business cases showing the increased return on investment to the service provider of the retail model with self-install will become common.
It won’t be just the cost of supporting customers that swings it. The ‘bypass’ of content from the internet portals will reach a point where it will be unmanageable for any service operator. So the message will change. Just buy a box and connect in. Pay by subscription or per-hour. If it doesn’t work and your video content won’t stream, there’s always the user forum. It will be just like the broadband industry.
Emphasis will move away from the dedicated set-top box installed by the service provider. Broadband is just a pipe. It delivers entertainment services and most importantly lets you choose what you want to see by giving you the extra choice to search out your own content.
Now comes the problem. How do you get that content from your broadband access point to your television? What about the second or third TV in the other rooms?
Will it be wireless? It’s not going to be 802.11a/b/g WiFi that solves this, and it does not seem likely that it will be 802.11n. Extensions to WiFi standards may work, but they have to be interoperable with the standards. Make it proprietary and it will not be adopted.
A safer bet in Europe and Asia is Powerline Ethernet, where high-speed HomePlug adapters carry data communications over the household power supply. It is already possible to distribute H.264 encoded HDTV content over HomePlug. It is here, now.
Security will be a factor in both technologies, but it ought not to be insurmountable.
Quality
Quality of service will become a dominant concern. It will be an issue in the set-top box, in the home network and in the wide area network. It will become evident that this was never addressed let alone solved in the MPEG-2 set-top boxes. The best solutions will come from the broadband routers, where data packets have long been conditioned not to get in the way of voice packets, but it will still be hard to scale up to the video packets.
Then it will become obvious that, whatever you can do at the set-top box level, it will be oblivious to what the DSLAM thinks, and the DSLAM hardly knows or cares about the video head-end server. Then we’ll think about standards.
Security
Up to now, security has been a non-issue because existing subscriber numbers are far too small to make it sufficiently challenging to the leisure hackers and not commercially important enough for the denial of service hackers.
IPTV brings part of the PC experience into the living room and to the TV set, literally, in the case of Media Center PCs. Imagine the controversy and plummet of consumer confidence when those IP packets carry spyware and viruses. The PC user community will seem comparatively sophisticated when it comes to protecting their assets from attack.
Embedded systems
When you want to deploy your solutions in the millions you need to go beyond that and make each unit as cheaply as you can. Usually, this calls for dedicated, embedded systems that carry enough computing power and functionality to solve the problem, and at just the cost point that this entails. If it is a real-time problem such as streaming video, it needs to be a real-time embedded system.
That’s why the IPTV world in the next year will be about system-on-chip solutions. The key race will be to populate those with all the systems software that is needed to bring them to a state of equivalence with the expensive PC solutions. The key message is that it does not need to wait. The technology is available now.
Paul Walsh is chief executive of Vidanti, a company based in Cambridge, England, spun out of Conexant Systems, offering ‘white label’ product designs for IP set-top box manufacturers.
www.vidanti.com
RSS