From Multimedia Futures, a sister publication In the last year, a great deal has been written about the future of interactive television, but what about the future of the television itself? An interactive television set-top box is outwardly indistinguishable from a cable television set-top decoder. On the television screen are several virtual buttons to help […]
From Multimedia Futures, a sister publication
In the last year, a great deal has been written about the future of interactive television, but what about the future of the television itself? An interactive television set-top box is outwardly indistinguishable from a cable television set-top decoder. On the television screen are several virtual buttons to help the user interact with the programs on offer. Inside, the set-top box contains anything up to 20 chips to handle functions such as managing the network and video decompression. There are some 15 or so chips to handle digitization, data, video compression, operating systems and interaction with the user. Real-time decompression of MPEG video usually requires one to two Giga-operations per second, which typically requires three chips. This is the standard model and for the most part, set-top box manufacturers and interactive television service developers are happy to allow it to rest. But Sultan Zia, vice-president of Digital Equipment Corp’s Video & Interactive Information Services unit, which is charged with building itself a business around the potential market for interactive television, says that although he is actively involved in developing the set-top market for his company, he sees only a brief future for the set-top as a separate box from the television.
Within a year, he says, the number of chips is going to be dramatically fewer. If the size of the component board that runs the digital interactive show is made smaller and considerably cheaper, he says, the possibility of bringing the whole thing into the television set is then a real one. It can be objected that, with a few exceptions, the video cassette recorder never made it into the television, so why should the set-top, but the marginal cost of the former is much higher. Zia thinks the demand for interactive television, spurred no doubt by the popularity of the multimedia personal computer and the Internet terminal, will push the television into a new personal computer-graphics dimension. There is no way this side of year 2000 for anything but a premium television set to include components to handle the sort of graphics found in today’s high end games. But that, says Zia, is no reason why low-end interactive-ready televisions should not be out within a couple of years – perhaps with add-on graphics boards to handle games available as separately-priced upgrades. Companies like General Instrument and Scientific Atlanta are fighting it, he says of the two largest set-top box manufacturers in the world. Their concern is understandable in that the emergence of an interactive-ready television would leave the likes of Sony Corp, Philips Electronics NV and Hitachi Ltd in the driving seat. Graphics extensions and decoder chips would remain a market for the box-makers and Scientific Atlanta Inc confirms that in the long term we’re aiming for high level graphics boxes. The company admits the number of chips required to run a digital set-top box is already dropping and expects it to be down to one or two chips within three to four years.
By Morgan Holt
For others, the change has already started: at the tail-end of last year, PowerTV Inc unwrapped Eagle, its first interactive set-top box on a single chip. Interfacing with a PowerPC-based processor, the Eagle chip can play audio and MPEG video, although real-time decompression still requires a separate chip set. Thanks to a dedicated few at Microware Inc, the David Digital Audio-Video Interactive Decoder operating system, based on the proven OS-9, already fits into only 128Kb of memory while still supporting MPEG2 on the PowerPC processor. But Scientific Atlanta is sticking to its guns. Despite all the talk of digital, analog will have a long life, says director Bill Brobst. An analog infrastructure such as conventional cable television is mature and cost-effective, he says. Some people held out a hope of deploying digital transmission cost-effectively, he says, alluding to the ‘real’ c
ost of installing a digital interactive television network. Although companies such as Silicon Graphics Inc have promised digital decoders at around $350, the hidden cost of installing digital trials is intimidating. In addition, digital standards are still up in the air. MPEG 4 is currently being debated and may even have been overtaken by the fractal compression techniques from Iterated Systems Inc. The Davic Digital Audio-Visual Council is only now settling on a voice, video and data standard, hoping that it will have set something in stone by June. And the convergence of industries has ensured that the International Telecommunications Union and International Standards Organization are already involved. In fact, the fast- changing nature of interactive transmission technology and its digital standards should guarantee a few good years of business for the set-top manufacturers. A television usually has a life of at least 10 years, meaning that customers are unlikely to commit themselves to an all-in-one set that may be out of date in 18 months. That said, one look at the frenetic makeovers of multimedia personal computers shows how far the novelty of technology can drive the consumer. Even so, market trials of interactive television throughout the US and Europe have shown little enthusiasm for what has been offered, leading advertisers away from the medium.
Not a happy one
Many are choosing to spend their new media budgets on the Internet instead. And, as predicted by Scientific Atlanta’s Brobst, interactive television is a lot more expensive than anyone expected. There is a mismatch between the technology and the readiness and capacity of the marketplace to create sufficient demand for the whole range of interactive services, says Luke Spikes, managing director at consultancy Spikes Cavell. A customer who has been pushed into a corner, he says, is not a happy one, an opinion echoed by the vice-chairman of the DAVIC management committee, Al Kovalick: Market success for interactive television in the US doesn’t have anything to do with what’s going on in DAVIC. What’s slowing the market is the Internet. Everybody’s looking at what’s going on there and saying, ‘Gee, let’s do this first and then worry about interactive television.’ Interactive-ready televisions may sound like a neat idea, but it could turn out to have been last month’s idea: anyone for the Internet-ready television?