Wireless USB edges towards market
A US start-up has developed a second-generation wireless USB (wUSB) chip that it hopes will help spark mass uptake of the short-distance wireless point-to-point connection.
The market for wUSB products has been stalled for two years by a combination of immature technology, evolving technical standards and a developing regulatory environment. In the meantime, both start-ups and established players have continued to absorb tens of millions of dollars in development costs to serve a market that may or may not take off.
“I think the current swathe of companies have probably had between $300m to $500m, if you take all the venture-capital investments in the start-ups, plus the spending of established companies such as CSR, Realtek and NXP,” said Marty Colombatto, chief executive officer of Staccato Communications.
Staccato’s strategy is to build a single-chip wUSB transceiver, including the radio-frequency circuitry, on a standard CMOS process, rather than going for a two-chip approach using a radio chip built on a silicon-germanium process. This should enable the company to take advantage of declining costs as the process matures, yields rise and volumes increase.
The company has received $77m of investment in three rounds, plus undisclosed strategic support from partners, and expects to seek a further round of funding to take it through to the end of next year. For that investment it has so far developed a number of test chips, including one to prove the company’s radio interface, plus a first-generation chip made in 110nm CMOS and its second-generation chip in 65nm CMOS.
But it is an increasingly expensive business to introduce chips for a new communications standard and see them through to mss market take-up.
“If you spoke to the general managers at CSR or Broadcom they would tell you that they would earmark between $75m and $100m to get all the way to a production-worthy state,” said Colombatto.
Jeff Chang, vice president of marketing at Staccato, added: “Since we decided we would take the hard road [of a single-chip design] we had a number of challenges to overcome, not least getting the radio to work in CMOS. We also had to deal with different power modes, and chasing evolving standards as the WiMedia requirements coalesced and the worldwide regulations evolved.
“So even though we were successful in doing a single chip it came a little late,” he added.
Colombatto argued that the reason that wUSB has not been seen in many products to date, despite its high profile at events such as the Consumer Electronics Show in January 2007, is that the chips have been too expensive, at prices of around $15 to $20 each in volume.
“At about $5 a piece you enable the chips to get introduced into products,” he said. “The Wi-Fi market took off at $5 and it was the same with Bluetooth. I was at Broadcom when Fast Ethernet gave way to Gigabit Ethernet, and that market began to work at $5.
“The real point is that RipCord 2 is the first device to hit the price, power and size requirements for a company such as Canon to build it into a still camera."
Colombatto denied that wUSB had missed its market window, despite the rise of alternative approaches to achieving the same ends, such as high-speed Bluetooth and cheaper Wi-Fi.
“It is tempting to say that the opportunity may have been missed,” Collombato added, “but I think the history lesson is that we were too optimistic and aggressive about when products would hit maturity. RipCord2 [Staccato’s second-generation chip] is proof positive that we have now hit that point. All the same potential exists – it has just been shifted by two years.
“2009 or 2010 will be when we hit the inflection point and go to volume.”