Spansion changes include ex-KLA-Tencor exec as new CEO
As NAND oversupply significantly harms Spansion and the NOR market, forcing the supplier to explore its alternatives for a sale or merger, Spansion names John Kispert as CEO, replacing Bertrand Cambou one day after his resignation.
By Suzanne Deffree, Managing Editor, News -- EDN, February 4, 2009
Spansion Inc is betting on new management as it continues to restructure and explore alternatives that include a sale or merger.
The flash maker, born out of a joint venture between AMD and Fujitsu, Tuesday announced that Bertrand Cambou had resigned as president and CEO and from the Spansion board of directors.
Boaz Eitan -- an executive VP, member of the board, and the ex-president of Saifun, a non-volatile memory provider Spansion acquired in 2007 -- was named interim president at that time.
However, the company said it was actively looking to fill slot and that effort did not take long, as Spansion named John Kispert as CEO and member of the board this morning, just one day after Cambou's resignation.
Kispert is a former president and COO of KLA-Tencor and is credited by Spansion as having played an instrumental role in the operational and organizational restructuring following the 1997 merger between KLA Instruments and Tencor Instruments. Before joining KLA-Tencor in 1995, Kispert held several senior management positions with IBM.
"As the new CEO, in addition to the strategic and restructuring initiatives, I plan to create a winning strategy for the company that leverages Spansion's market leadership, rich intellectual property portfolio, and strong customer relationships," Kispert said in a statement. "My first priority is to ensure that Spansion capitalizes on the company's tremendous strengths to bring value to our stakeholders."
NAND oversupply significantly harms Spansion, NOR market
The management changes come after Sunnyvale, Calif-based Spansion on January 15 announced it was looking at its options, including, but not limited to, opportunities to merge with or sell to similar US or foreign businesses. The company also said at that time that it would delay making the interest payment due that day on its outstanding 11.25% senior notes due 2016. Under the indenture governing the 11.25% notes, a failure to make an interest payment is subject to a 30-day cure period. If Spansion does not make payments at the end of its 30-day period, it could be forced to file for bankruptcy.
"Should the company be forced into bankruptcy several alternatives are available, the best of which will be chosen by the bankruptcy arbitrator," Jim Handy, an analyst with Objective Analysis, said in a research note this morning. "One would simply be for the company's debt to be restructured and business would continue as before. The most dramatic opposite stance would be for the company to be liquidated -- all the employees would be furloughed and assets would be sold off. We see neither of these as likely.
This is good news for the company's customers, who, in a liquidation, would be forced to redesign several products to use competitors' parts."
Handy noted that while Spansion has world-class manufacturing capacity in Aizu-Wakamatsu, Japan, and in Austin, that is unlikely to appeal as intact fabs to anyone who is not involved in the NOR business.
"Although an argument can be made for the sale of Spansion's manufacturing capacity since it consists of hard assets, Objective Analysis believes that the asset that is most likely to be sold to the company's advantage would be their intellectual property," he said, pointing to key charge-trapping intellectual property that is likely to be needed by all flash makers, both NAND and NOR, at 25 nm and tighter processes. "It is quite possible that some licensing agreement will gather Spansion enough cash to see the company through its current difficulties, however, any purchaser would be a visionary with the patience to wait a year or two before seeing any revenue stream."
Handy reminded that under Cambou's guidance Spansion focused significant attention on the future, battling a stagnant memory market and attempting to "blaze new trails" for the supplier's NOR technology by developing ORNAND, a cross between NOR and NAND flash, and focusing on using MirrorBit flash as a power-saving replacement for DRAM in servers.
"It's too early to tell whether this last will be accepted -- that process is likely to take another two years or more since the approach is revolutionary and acceptance will be slow," he said. "ORNAND could have done very well in a typical NAND market, one where NAND makers dropped their lower-density product in pursuit of sales of more profitable high-density devices. The trouble is, there's such a huge NAND glut today that all NAND makers are taking orders for low density parts that they would have otherwise abandoned. It's better to sell an unprofitable part than to leave their fabs idle."
That glut has harmed Spansion in multiple ways. Handy reminded that major NOR makers profit by selling their highest density parts. These parts are under extreme price pressure because their key market is camera phones and these designs can be worked to either use a large NOR or the combination of a small NOR plus a large NAND, he said. "Some designs are even converting to NAND alone. Unless the NOR is sold at a very low price, today's cheap NAND is likely to capture the bulk of the design's flash revenues," Handy said.
"If Spansion's visions for the future fail to become significant market drivers, NOR could lose many of its high density markets to NAND/NOR combinations or NAND-alone designs," he continued. "NOR would then take on the role of the low density chip sold at a low price point. Once this happened, then density increases would slow and a growing number of devices would become pad limited when migrated to more advanced processes. This would drive the market toward serial, rather than parallel, NOR chips.
Anyone who has been in the memory business for more than 20 years will notice that this is exactly the same process EEPROM followed."
According to Handy, today's NAND oversupply has accelerated difficulties that were already in store for NOR. "This is a very difficult time for all NOR companies, and especially for Spansion," he said.
-
Spansion has been working on a true NAND product to replace ORNAND. It is called ORNAND2. It is supposed to ship this year. If it ships, that would be great for Spansion.
J. Boone - 2009-4-2 09:28:00 PST -
Spansion is trying to make a true NAND product to replace its ORNAND. They call it ORNAND2. It is supposed to ship this year.
J. Boone - 2009-4-2 09:13:00 PST
