You may know Steve Perlman as the man behind OnLive, but it's looking increasingly foolish to think the California entrepreneur is all about games these days. WebTV may be a thing of the past, but Perlman's company Mova provides advanced facial capture software to big names in Hollywood, and on June 4th, he told students at Columbia University that his incubator Rearden had developed a groundbreaking new approach to wireless technology. That last one's quite the claim, and details were scarce in June, but today the company's ready to explain how it will "completely transform the world of communications" with a little something called DIDO.
DIDO stands for Distributed Input Distributed Output, and there's a reason that word "distributed" is repeated twice -- basically, what Perlman is imagining is peer-to-peer, ubiquitous Wi-Fi. But where Wi-Fi becomes less effective as more access points compete to use the same narrow channels, DIDO promises to increase stability and maintain bandwidth the more points there are, because Rearden claims it has hit upon a way to give each user their own wireless frequency by tweaking the waveform. Remember those monster cloud servers powering OnLive? Here, they're calculating what Perlman calls "unbelievable amounts of mathematics" to deliver customized waveforms to each individual using the DIDO network at a given time, in effect allowing each to have their own private channel to the access point. Since the signals don't interfere with one another, additional DIDO access points simply add diversity to the network by allowing a user's device to connect to more than one access point at a time -- among other important boons we'll discuss after the break.
Perlman claims that while traditional wireless technologies have all been nerfed to reduce congestion, the use of carefully calculated waveforms per user means that there's nothing holding back DIDO in the same way; Rearden can boost the power, range and speed of the wireless connection while using simpler, more power-friendly radios for transmitters and receivers. "The complexity of a DIDO radio is closer to that of a walkie-talkie, rather than a cell phone," Perlman told us this afternoon. The entrepreneur claims they've tested prototype DIDO radios at speeds of up to 100MBps already, and (perhaps separately) with radios over 30 miles apart, and says a rural version of the technology is planned to bounce radio waves off the atmosphere for a 250 mile coverage radius. He also says the tech can achieve sub-millisecond latency at a distance of a few miles -- no mean feat, when compared to the dozens or hundreds of milliseconds it can take existing cellular technology to respond. That would make DIDO particularly useful for applications where latency is crucial -- like OnLive itself.
Now, Perlman readily admits that the tech takes some serious computation on the server end -- and the fact that a centralized server is required, period, threw us for a loop -- but he argues that the model has a number of benefits. Since the system depends on the ubiquity of public DIDO access points, he's toying with a business model that would have each user chip in a few bucks for access to the general network and the ability to roam anywhere they'd like, and he'd use that central server to mandate that everyone pays their dues. Without the need for internal computation, each dumb access point would theoretically be cheap to deploy, too. Finally, should there be a need to redeploy DIDO's (flexible) spectrum for other uses, the company can ensure compliance with the flip of a switch -- turning off every one of the system's software-controlled radios.
It's taken Rearden ten years to build ten DIDO prototypes and a massive vision surrounding their use, and as with any technology this early on, we're a bit skeptical that it will take off. "In time, this will displace cellular, and it will displace Wi-Fi, and here's the shocking thing -- it'll displace wires for connectivity in your home," Perlman told us, and like any ambitious dreamer, we're afraid he's getting a bit ahead of himself. Apparently, the powers that be in Washington didn't think much of the idea, and Rearden's now talking to foreign providers about commercial deployment of the technology in lieu of domestic ones. Still, there were plenty who said OnLive was impossible, and Steve Perlman proved them wrong. We definitely want what he's selling, if he can deliver the goods.
Sources: Steve Perlman, DIDO White Paper