Spansion Inc.

views updated

Spansion Inc.

915 DeGuigne Drive
Sunnyvale, California 94088
Telephone: (408) 962-2500
Toll Free: (866) 772-6746
Web site:

Public Company
1993 as Fujitsu AMD Semiconductor Ltd.
Employees: 7,500
Sales: $2 billion (2005)
Stock Exchanges: NASDAQ
Ticker Symbol: SPSN
NAIC: 334413 Semiconductor and Related Device Manufacturing

Spansion Inc. is a Sunnyvale, California, manufacturer of Flash memory, which is used to store data in a number of electronic products, including wireless phones, personal digital assistants, automotive subsystems, and networking equipment. One of the fastest growing segments of the semiconductor industry, Flash memory is similar to DRAM found in personal computers. It is able to write and rewrite data, but Flash memory, which is more expensive than DRAM, has the added ability to store and retain information when the power supply for an electronic device is turned off. As the largest Flash memory company, Spansion concentrates on the integrated electronics market; its customers are the leading original manufacturers of automotive electronics, consumer electronics, and handsets. Spansion is a public company listed on the NASDAQ.


In 1992 "Negative Gate Erase" technology was introduced. This was the first way to maintain erase performance in an electronic device without the need for external power, thus setting the stage for Flash memory. A year later, two of the giants in the semiconductor industry, Advanced Micro Devices Inc. (better known as AMD) and Fujitsu Ltd., joined forces to create a joint manufacturing venture called Fujitsu AMD Seminconductor Limited (FASL), marrying AMD's Flash memory technology with Fujitsu's process technology and high-capacity manufacturing operations. FASL quickly became a leader in the field, responsible for a number of innovations. In 1996 the company introduced the industry's first 2.7volt flash device, offering the lowest voltage to that point in the Flash market. A year later FASL took the technology a step further, bringing out the industry's first 1.8volt Flash device. That same year, FASL was responsible for a breakthrough in the automotive electronics field when it unveiled the first Known Good Die (KGD) Flash memory. It was capable of operating under extreme conditions and temperatures as high as 145 degrees Celsius, an important factor when a device is incorporated into an engine or other automotive components where high temperatures are achieved. Another FASL technical achievement of 1997 was the introduction of the simultaneous read/write feature, which for the first time allowed for the reading of data while a Flash device was conducting program or erase operations. Simultaneous read/write became a standard for NOR Flash chips, earning EDN Magazine's "EDN Innovation of the Year." (NOR stood for the programming expression "not or." NOR chips could read data at high speed, yet were extremely accurate. They were ideal for running the software used in mobile phones and other devices. A second type of Flash memory, NANDstanding for "not and"was better suited for digital cameras, MP3 players, and other electronic devices. In simple terms, NOR and NAND were two different ways computers searched data and connected the Flash memory cells. NOR chips offered better quality and faster speeds but were more expensive than NAND chips, which weren't as reliable but could store a lot of information at less expense. In a basic cell phone, for example, NOR chips contained key programming code while NAND chips were used for mass storage of less critical information, such as a user's favorite ring tone.

FASL continued to break new ground in the Flash memory industry in the final years of the 1990s. In 1998 the company unveiled the first page-mode flash device, which permitted the fast read speeds required in high-performance applications. A year later FASL took the technology a step further by creating the first burst-mode Flash device, offering even faster read speeds.


FASL's breakthroughs continued with the start of the new century. In 2000 the company offered the first product to combine Flash and SRAM in a single chip. The Multi-Chip Product (MCP) was another key advance for the mobile handset market, allowing for smaller devices by delivering a complete memory subsystem in a single small chip. Then in 2001 FASL introduced MirrorBit technology, which led to the first fundamental change in non-volatile storage technology in the semiconductor industry in more than 20 years. It allowed for the storage of two bits per cell while maintaining the same level of performance. Another FASL breakthrough in 2001 was the development of the first Flash memory device with a 32-bit data bus, resulting in extremely quick system throughput, which was especially useful in automotive applications and printers. The year was also noteworthy because of sagging Flash memory prices, which forced FASL to cut back production at some of its facilities. In 2002 FASL brought out advanced sector protection, providing an even greater level of protection against loss of data.

AMD and Fujitsu deepened their relationship and their commitment to the Flash memory market in March 2003 by establishing FASL LCC, a company that brought together the Flash memory operations of both parent corporations. AMD held a 60 percent controlling interest in the joint venture. AMD also contributed a recent acquisition, Coatue Corp., a startup company that had developed a way to use polymer technology to build arrays of nonvolatile memory cells. As a result, the technology was able to produce a memory chip that combined Flash memory's non-volatility and low power consumption with the speed of DRAM. This new class of polymer-based technologies, Coatue claimed, would be able to create memory chips that handle applications conventional memory chips could not accommodate, including solid-state video storage, instant-boot computers, and single-chip personal computers.

The new FASL would also begin to market its products under a new brand name: "Spansion." according to in an article written at the time of the announcement, "By further separating itself from Fujitsu and AMD, the new company will be able to form industry alliances and take the necessary steps to compete in the cutthroat market for flash memory. At its inception, FASL LLC will represent the world's second-largest flash memory provider, behind Intel Corp. Although market researcher Semico Research expects flash dollar sales to overtake DRAM in 2004, driven by cell phones, PDAs, and the like, flash pricing has continued to decline, with a 30 percent drop in price per megabyte last year." Out of the gate, the new company, based in Sunnyvale, California (with another headquarters located in Tokyo), would be worth an estimated $2.5 billion. It enjoyed only limited independence, however. Although it maintained its own sales operation, FASL's parent companies would be its only distributors, with AMD given priority status in the Americas and Fujitsu in Europe. The parent companies split the Asian market.


Dedicated to enabling, storing and protecting digital content, Spansion is the largest company exclusively focused on Flash memory solutions and has one of the most diverse and comprehensive Flash product lines on the market.

FASL LLC was headed by Dr. Bertrand Cambou, who served as both president and chief executive officer. French-born, he held an engineering degree from Supelec, Paris, and a doctorate in electronics from Paris XI University. He was involved in the development of terminals for secure e-transactions while working at In-genico, and then went on to Gemplus International SA, a Smart Card maker, where he became chief operating officer and co-president. Before taking charge of FASL, he joined AMD in 2002 to become group vice president of the Memory Division. He also had 15 U.S. patents to his credit.

The new FASL did not let up on product development. In 2003 it used second generation Mirror-Bit flash technology to introduce the industry's first 512Megabit NOR Flash memory chip, the highest density NOR Flash device available in the market. A year later, in June 2004, FASL took the name Spansion LLC, and in the fall of that year announced a new kind of Flash memory chip that combined NOR and NAND architecture. The new ORNAND chips could serve both the NOR market for flash chips, lead by Intel and Spansion, and the fast-growing NAND market, dominated by Samsung and Toshiba. According to the San Jose Mercury News, "Spansion's new trick is to modify its NOR flash chips so that they can mimic the features and lower price of NAND flash. In doing so, the company's ORNAND flash chips could be used in either the NOR or the NAND markets." The hope was that Spansion chips would find their way into a greater number of cell phones, which as they grew more sophisticated, adding cameras and music capabilities, required greater data storage. Spansion offered developments on other fronts as well in 2004. It unveiled the 1.8-volt, 80 megahertz burst-mode Flash memory chip. Combining the lowest voltage of any 256-megabit Flash device on the market with two-bit-per-cell technology, it was ideally suited for wireless applications.

The final quarter of 2004 proved difficult for Spansion, which had to contend with price-cutting competition and as a result reported an operating loss of $39 million, much to the displeasure of AMD's CEO, Hector Ruiz. He told analysts the performance "makes me want to puke." He also made it clear that he was open to anything that would increase value to shareholders, a comment that most observers interpreted as a signal that Spansion would either be sold or spun off. If so, Spansion would no longer be a drag on AMD's earnings, which reported a net loss of $17.4 million in the fourth quarter, and AMD would be in a better position to challenge its bitter rival, Intel, in the computer processor field. Barron's explained why Spansion was becoming such a nettlesome problem for AMD: "Spansion eats up lots of capital, and its main rival is Intel, which puts AMD in a strategically vulnerable spot. Since flash contributes a modest 7% of Intel's revenues, compared with 40% for AMD, should the urge strike, Intel can brutally slash prices. In that event, Intel may bleed a little, but AMD will hemorrhage massively. And when the money-losing Spansion needs more dough, AMD, as one of the joint-venture owners, has to pony up." Barron's also contended that if AMD separated itself from Spansion, "Intel may not feel the urge to compete as aggressively. In short, Spansion may be a more valuable business if AMD isn't a majority owner."

2005 IPO

Spansion reported a weak first quarter in 2005, as the Flash memory market continued to experience an oversupply of product and eroding prices, and sales were down 29 percent over the same period the previous year. Essentially, the die was cast for Spansion's future. AMD announced in April 2005 that it would spin off the unit in a public offering of stock. All through the summer and autumn of 2005, Spansion and its lead underwriters, Citigroup Inc. and Credit Suisse First Boston, prepared for the stock offering. In November the company was reorganized as Spansion Inc., and in the meantime company representatives met with potential investors at road show presentations. It proved to be a difficult sale, however. The company was hoping to price the stock between $16 to $18 per share, but as Spansion continued to post quarterly losses, investors grew increasingly wary. The company lowered its asking price to the $13 to $14 range, but on the eve of the December offering, the price to institutional investors fell to $12. In all the sale raised $506.4 million, a far cry from the $750 million figure once bandied about, but still the largest technology stock offering of the year. Given Spansion's string of quarterly losses, however, merely completing the offer was considered a success.


Fujitsu AMD Semiconductor Ltd. is founded.
The company introduces a simultaneous read/write feature to Flash memory.
The Spansion brand is introduced.
The company becomes Spansion Inc. and is taken public.

As Spansion entered 2006 as an independent company, its prospects began to improve. Flash chips remained a fast-growing market, and Intel had to reallocate some factory space in order to produce more chipsets because of a shortage. As a result, it produced fewer flash chips, presenting an opportunity for Spansion. Being in control of its own fate was also a major plus. "We're a much better capitalized company than before," Spansion executive Tom Eby told Investor's Business Daily. "And we eat, sleep and breathe flash."


Intel Corporation; Samsung Electronics Co., Ltd.; Toshiba Corporation.


"AMD Distances Itself From Flash Market," , March 31, 2003.

Brammer, Rhonda, "On AMD's Menu: Eating Intel's Lunch," Barron's, September 26, 2005, p. 21.

Clark, Don, "Memory Race Turns in a Flash," Wall Street Journal, March 7, 2005, p. B3.

Detar, James, "AMD's Shareholders Chipped In to Buck Up Spinoff Spansion's IPO," Investor's Business Daily, January 9, 2006, p. A05

Kharif, Olga and Cliff Edwards, "For AMD, a Thorny Dilemma," Business Week Online, June 23, 2005.

Ladendorf, Kirk, "AMD Considers a Future Without Flash," Austin American-Statesman, February 13, 2005, p. J1.

Takahashi, Dean, "Sunnyvale, Calif., Firm's Chip Could Shake Up Flash-Memory Market," San Jose Mercury News, November 9, 2004.