Point is that fancy "green" environmentally correct RFAB fab should be named "ROTTING FAB"
its just sitting there rotting away. Maybe they should turn it into a graveyard for the TI executives that thought it was a good idea. They are slowly losing their cell phone chip dominance as well. They are gonna do a lot more job cutting than this.
Get out the chainsaw Rich, but be careful dont cut off your thumb!
The semiconductor market is projected to shrink 11 to 22% in 2009, so TI shrinks the company by 12%, just as the street expects. Wall Street calls TI a well managed company. Since 2001, on the wireless front, TI says the future of wireless is "DRP", then in 2008 gives up on the modem chip that would require an integrated radio (which is the focus of DRP). Since 2000, TI invests heavily in broadband, and then sells its DSL business to Infineon in 2005/6. In 2007 TI decides that Medical Electronics will be the next big thing, then moves the VP of the medical group to head the DLP group in 2008. In 2008 TI decides to pursue analog and embedded DSP as it's core strategy, which is almost identical to the "analog attachment and DSP" strategy of 1997. (What is the difference, other than the integration level?) And unfortunately, rather than emphasize analog in 2002, when the margins were good, TI emphases it now, just as the margins are eroding. That is not TI's fault, but it is bad timing. Even Linear Technology has announced a layoff in the past month.
And of course, the state of the art submicron fab building. In 2004 TI announces it will build the fab in Richardson. In 2007, TI declares it will no longer pursue the development of its own digital CMOS process, and RFAB sits, with no equipment, empty on a street in Richardson, Texas. Far less than the 3 billion it would cost once equipped, but a massive waste nonetheless. The longer it sits there, the less useful life it has to pay for itself.
The question is not whether it makes sense for a company to shrink when the market shrinks. Any accountant would say that it should. The concern is whether the management of Texas Instruments can set a direction, plot a strategy, and manage the balance between the pressures of Wall Street and the investment needed to build a company that can survive beyond the next fiscal year. I contrast this behavior with the behavior of Intel in the 80s, when analysts thought they were crazy for building fabs in a down turn. Why did they build fabs? Because that was their business, and they believed they would be the number one semiconductor manufacturer when the downturn was over.
Of course, history says that Intel was right, the analysts wished they had bought Intel back then. Apparently, Intel management knew how to run their business better than Wall Street did.
Del Whitaker, Keh-Shew Lu, CS Lee, Ming Chiang, Dan Reynolds. All of the VPs who once successfully grew analog businesses inside TI from the ground up have retired. I am looking hopefully for signs that the new management is as capable. I am hoping they can run Texas Instruments better than Wall Street can.
Blog Doing Math in FPGAs Tom Burke 24 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...