Between the floods, tornadoes, earthquakes, and $4/gallon gas, there’s so much doom and gloom in the air that the Indian River Silicon announcement that it is now formally
open for business came as an especially welcome piece of good news last week. The newly-formed employee-owned wireless silicon design house, composed mostly of Conexant’s former RF team, was brought to life shortly after their parent company decided to get out of the wireless business late last year. In my November 2007 editorial,
Brain Drain Pan, I lamented Conexant’s decision to disband its talented RF semiconductor design team, and worried that the talent pool that had taken decades to assemble would be scattered to the four winds within a few weeks.
Happily, I called this one wrong.
A brief conversation with Al Petrick, Indian River’s Senior Vice-President of Marketing (and one of its founders), revealed that the company was formed less than a month after Conexant laid off the team; and received their first contract less than three months later, in January 2008. With several more substantial jobs in the works, it looks like this scrappy little brain trust of RF know-how has a shot at survival, and perhaps even a bright future.
There are several reasons why I have good expectations of Indian River’s prospects, beginning with the depth of knowledge they have about both high-frequency ICs and the systems they go into. Besides its deep knowledge of Wi-Fi, the assembled brain trust has worked with most other wireless technologies including FM, GPS, Bluetooth, WiMAX, and 60 GHz Wireless PANs (802.153c). They also have lots of experience in the beam steering technologies that are optional in 802.11n and will be used widely for 60 GHz systems. Petrick explained that, rather than just design chips, their intention was to be a full-service design RF house that provided production-ready reference designs and even protocol and driver software.
Less obvious, but equally important, is that the Indian River team has spent a lot of time learning the many secrets of supporting multiple standards on a single chip without interference. This eclectic skill set is becoming increasingly critical as mobile hand sets and handheld computers are being asked to support multiple wireless interfaces in increasingly-compact form factors.
Although it might not be obvious at first, the company’s Central Florida location is one of its other big advantages. While not as well-known for its RF prowess as Cambridge or San Diego, the Central Florida area has several high-powered defense contractors (including Intersil and Lockheed Martin, and a few sources that cannot be mentioned) that maintain a stable of wireless talent which Indian River says it can draw upon for everything from mixed-signal, algorithm development to advanced antenna design.
Indian River’s choice to go with TSMC and IBM for foundry services was a good move since both fabs offer good support for RF. Although they expect to do most of their design work in CMOS, they have also made arrangements with IBM and Jazz to handle any designs that require SiGe. One of my few concerns for company success is whether its relationships with its fab partners are pretty solid so that they can weather the availability glitches that are common at both the top and bottom of a market cycle.
Indian River management team is to be commended on the quick moves and nerves of steel (not to mention some deep pockets) it took to keep Conexant’s RF team from unraveling as I’ve seen happen to many other great talent pools over the years. Like the small airports I fly from, design teams are rare confluences of important resources and people which often produce tremendous benefits for both the parent company and the community they serve. And, like those grass strip airports, they are often fragile entities which are under-appreciated, easy to plow under for immediate gain, and almost impossible to replace.
Comments? Questions? Other tales of rescued (or dismantled) design teams? Write me at
lhg at en-genius dot net or post your comments on our blog.
