Qualcomm has announced the "availability" of silicon that conforms to Draft 2.0 of the 802.11n standard, from its prospective purchase Airgo.
This is bizarre. There is no silicon available. There can't be, because you can't "conform" to a draft of a standard. And earlier this year, Airgo pointed all this out, rather huffily, to the other silicon vendors that were then pushing Draft 1.0 silicon.
What do you mean Draft 2.0?
Although Airgo was first to market with MIMO chips, it got sidelined to some extent by developments in the IEEE 802.11n standards group, which decided, late in 2005, to adopt a joint proposal from the four leading non-MIMO players, Broadcom, Marvell, Atheros and Intel, who have all promised silicon to meet the Draft 1.0 specification which came out fairly swiftly.
Airgo, was left with a bigger job (and, it claimed, a backward step) to meet that draft. It has looked like missing out on the draft N bonanza that is predicted for the first half of 2007 when, we are told, laptops will be fitted with Draft N chips. It has looked ripe for purchase since then.
Under Qualcomm's direction, Airgo still faces an awkward transition - hence the announcement of a Draft 2.0 strategy, in which it will not deliver any Draft 1.0 products.
But the Draft 2.0 product is very much a pre-announcement. Glenn Fleishman of Wi-Fi Net News points out that the IEEE 802.11n group has a list of 370 issues still to be resolved from its November meeting. That's only 12 percent of the original list of comments to Draft 1.0, and the group expects to sort them out at its January meeting, says Fleishman, but a Draft 2.0 won't be voted on until at least March 2007.
With no possibility of a Draft 2.0 design until after then, Qualcomm vice president Enrico Salvatori admitted to us that the Draft 2.0 silicon was not actually "available" was planned for sample quantities in the second half of 2007.
What about standalone 802.11n?
Qualcomm is known for handsets, not the kind of access points and laptop cards where Wi-Fi tends to be at the moment. The main reason to buy Airgo is clearly to build Wi-Fi into its phone chipsets to deliver dual-mode devices: "we are talking about fixed-mobile convergence," said Salvatori. Qualcomm's previous partner in FMC, Atheros, presumably goes by the board. Qualcomm will integrate Draft 2.0 Wi-Fi, backward compatible with IEEE 802.11a, b and g, as well as Draft 1.0, into its Mobile Station Modem and Snapdragon chipsets.
The Draft 2.0 announcement shows Qualcomm isn't jetissoning standalone Wi-Fi. In fact, standalone Wi-Fi chips are part of its integration programme: "We understand that the current 802.11n standard is not finalised yet," said Salvatori. That is why the company isn't going all-out for integration yet. "We will go to full integration only when 802.11n is finalised," he said.
In fact, with Airgo's Wi-Fi chips, Qualcomm may be moving into consumer gear besides handsets, said Salvatori. That's not a change in direction though - merely an expression of the fact that devices for communications and other tasks are converging.
Bluetooth, likewise, will be integrated into Qualcomm's core silicon, and the company plans to move with the standard there - delivering the successful Bluetooth 2.0 technology. The company will track future Bluetooth versions based on ultrawideband, but there's no hurry: "Bluetooth 2.0 delivers most of the requirements from operators," said Salvatori. .
Prepare for patent wars?
Any controversy about the "Draft Draft N" announcement may well pale beside likely intellectual property issues. Qualcomm is also one of the more litigious (as well as acquisitive) industry players, and gains a fine stack of MIMO-related patents with Airgo.
Qualcomm wouldn't comment on how it would be using these patents, but Salvatori acknowledged their value and said: "We will continue to push our business model, based on enabling OEMs delivering products based on our essential patents and IPR."
More revealing, perhaps, was this comment last year, from Airgo's chief executive Greg Raleigh. "802.11n is Airgo's technology," he said. "Our intellectual property is wrapped up throughout. There appears to be a market north of 2.5 billion chipsets, and 80 percent of the Wi-Fi market will be 802.11n by the end of 2007."
Whatever Qualcomm does in FMC, we'd be very surprised not to see it following up the potential revenue stream from the patents.