The Silicon War Trilogy
Theft of Trade Secrets
When the Russians attempted to copy American semiconductor chips, it was considered a threat to national security. All intelligence assets on all continents were mobilized to stop it. CIA, DIA, NSA and FBI are alerted to the threat.
Theft Of Trade Secrets
The Silicon War Triology
Chapter 9 - National
Five hundred miles to the north of LA, the new tech sector called Silicon Valley was booming. With the "Space Race" in full swing and electronic systems being built by all DoD vendors, major civilian companies were formed to supply the needed electronic components. One of the largest was National Silicon Corporation (NSC) in Santa Clara, CA.
Leading this company was fifty year old, Charles S. Borke. He was appointed in the late 60's as President and CEO of National Silicon Corporation (NSC) and promptly moved their headquarters from New York, New York, to Santa Clara, CA, the heart of Silicon Valley.
To sweeten the deal for Borke joining for half his former salary, he was allotted substantial shares of National's stock. In addition, Borke brought his four existing management personnel from Fairchild as well as three others, each from TI, Perkin-Elmer, and Hewlett Packard to form the new eight-man management team at NSC, each with similar salary packages. This also started the "executive stock options" trend.
In 1970, National introduced 7400 Series of TTL Logic chips used heavily in printed circuit boards that were direct plug in replacements for the Texas Instrument TTL Series of products. How did they bring it out so fast. Simple, National just copied TI's.
In 1973, they did the same with the Motorola 4000 CMOS series product line. Again, National just copied Motorola's which was an exact copy of RCA Corporation's. However, no legal action resulted because this was how companies met the DOD's "we must have three sources" requirement.
Under Borke's expertise in operations and manufacturing, National introduced the industry's first standard high reliability line for military/aerospace. This enabled them to win many DOD contracts containing very large quantities of chips and as a result, NSC became the world's fourth largest semiconductor supplier.
"The building process at National and our plan to compete was straight forward. We knew exactly what we wanted to do, and who can do it"
- Charles S. Borke
As with all success, arrogance was justified as "confidence" and that affected many in Silicon Valley, especially at NSC. In one of the most infamous advertisement of this time was by NSC.
"Our message to the competition is simple and straightforward.
We've had it with namby-pamby, blue-sky advertising. From now on, National doesn't pussyfoot. We're going to take on the rest of the semiconductor industry and let the chips fall where they may.
We're the second largest manufacturer in just about every product category and we're going to let everyone know it.
We're also going to introduce some new products that will knock the competition right off their profit margins.
There are also a few things we're not going to do.
We're not going to make a lot of products nobody needs. That's Signetics' job.
We're not going to introduce a new, hotshot device that isn't even off the drawing board yet. Fairchild is much better at it anyway.
We're not going to promise a shipment for September that we couldn't possibly deliver before Christmas. That's TI's game.
And, we're not going to sit around on our ingots waiting for the second source business; Motorola's cornered the market on that one.
In short, we're going to be damned hard to compete with.
You know where nice guys finish!
National Silicon Corporation..."
Under Borke, NSC's manufacturing improvements was not based on design innovation but on the concept of "copy & improving" and standardizing manufacturing processes that were already established by other companies like Fairchild, Motorola, and Texas Instruments. And the quickest way to do this was by frequent raiding Fairchild's pool of talented employees.
In early, 1972, National added a Consumer Products Division. In the spring of 1974, along with Texas Instruments, National introduced the first all-IC hand held calculator. This consumer product exploded the mass manufacturing of semiconductor chips. And again the NSC design team was formed from TI, Rockwell, and Mostek personnel.
In summer, 1974, National added a Memory Products Division to handle the IBM mainframe memory market, and again it was staffed by raiding Intel, Motorola, and Fairchild. By the fall of 1974, National added a Microelectronic Products Division to handle microcomputers, single chip computers, and desktop computing markets.
This new division was headed by J. Daniel Asumi, Vice President Microelectronic Products.
Asumi was born in 1938 in Los Angeles, CA to Japanese Americans, and spent time in the Japanese concentration camps in California where his two sisters were born. Graduating 7th in his class with a BA from University of California at Los Angeles (UCLA) in three years, he interviewed and was hired on the spot by Fairchild Marketing in 1969. Three years later, he became their Director of Microelectronic and Consumer Products.
Fluent in both Americanized English and Hyōjungo form of Japanese, he quickly became valuable to Fairchild in negotiations, document translations, and plain old "entertaining clients."
His success in closing "pacific rim" contracts brought him to the attention of Borke as the lead candidate for the new VP of Microelectronic Products.
It was also widely know that Borke disliked Japanese, referred to them as "Japs" as all his age did as a result of their involvement in the Pacific during WWII. At 5 foot 7 inches, 210 pounds, 52-year-old Borke was not intimidated by anyone or anything until now. After a 2-hour interview, Asumi had passed with flying colors. At 6 foot 1, 185 pounds, Asumi was considered a "ladies man" and the war stories from Fairchild were something else. What amazed Borke the most about Asumi; he was trained in "Marketing" but had very solid "Operations" way of thinking. His instincts were almost the same as Borke's.
Lunch was light, golf was quit competitive, and the "accidental" meeting of Borke wife afterwards for drinks was really to get her opinion. It was a solid thumbs up. Asumi would be perfect if he just wasn't a "Jap."
The offer was substantial; negotiations were fun and lasted for three weeks. Borke felt Asumi was very good, well researched, prepared, and got to 90% of what Borke was going to offer. Borke considered this a good victory for him and more important, Asumi turned out to be an excellent negotiator as well.
Asumi won as well because his base salary was twice what he was currently making at Fairchild, he would also receive 35,400 shares of NSC stock and 125,000 stock options. NSC stock closed at 27 1/2, and Asumi now could be defined as a "millionaire." As Asumi signed the offer letter, he picked up his glass of wine and said "Dad, thank you for kicking me in the ass." It had been 4 years now since his dad had passed away and the loss was just now starting to set in.
Although Fairchild countered the offer, Asumi was not interested. At NSC, he would have free reigns to do whatever he wanted without the Fairchild political obstacles. Fairchild management decided that it would be in everyone's best interest if he would leave ASAP. And that's what happened. His office of 5 years was packed for him and he was escorted to the door.
NSC's Microelectronic Group quickly became the focus of all NSC major growth and strategic planning. Asumi was first assigned the calculator line and then the digital watch line. Asumi spun these off and formed the "Consumer Product Line."
Asumi was then assigned to lead a team to interface, setup, and implement existing designs from other companies and take over the chip manufacturing. By doing this, NSC not only obtained the chip designs but the manufacturing processes as well. Then the manufacturing processes could be used on additional chips designs.
One of these was the potential lucrative second sourcing of the Bowmar Calculator integrated circuits.
CHANDLER, AZ USA
FEBRUARY, 17TH 1975
To many, the "Bowmar Brain" defined the early pocket electronic calculator. Originally, an LED (light emitting diode) display manufacturer, Bowmar/Ali, Inc. (USA), found they could not sell their displays to the Japanese electronic calculator makers. Reviewing their business plan in 1970-71 and examining the new trends in the marketplace, Bowmar decided to design and manufacture portable calculators that would be sold by other marketing companies.
So in late 1971, they began shipping their first model, a hand-held unit, under their own nameplate. This model (the 901B) was also produced for other companies including Craig (model 4501) and Commodore (C110). Bowmar soon became one of the world's largest producers of electronic calculators, mainly pocket models, for sale by themselves and by other companies such as Sears, Montgomery Ward's, Radio Shack, etc. (under their own nameplates).
In the mid-1970s, as the calculator "boom" was in full swing. Bowmar could not get enough integrated circuit chips from their suppliers and that resulted in not keeping pace with the marketplace in low cost and new features.
Envious of the handsome profits being made by the assembly houses who put calculators together from other suppliers' components and then selling them at substantial profits, the giant US semiconductor manufacturers started to integrate their own companies vertically and moved heavily into the calculator assembly business. Texas Instruments was followed by Rockwell International and, then National Silicon. They brought with them a price advantage on the vital integrated circuits, plus greater technological manufacturing ability and, above all, the control and tactics over the supply of microcircuits to their competitive manufacturers.
Texas Instruments quickly moved into the number one position in the United States, which was by far the largest calculator market in the world, accounting for 40 per cent of all machines sold. It became clear that the only survivors in the big league would be those that were vertically integrated.
Therefore, Bowmar as an assembly house took the necessary plunge. They invested $7 million in a large semiconductor plant in Chandler, Arizona. However, by the time the facility came online, many manufacturers had developed overcapacity. Even the retail stores and warehouses were now stocked to overflowing with obsolete product.
Although Christmas, 1974 was expected to be a booming season, it wasn't going to materialize. Domestically, President Gerald Ford was presiding over "the worst economy since the Great Depression" as a result of the Yom Kippur war. Now everybody was strapped for cash due to the Yom Kippur triggered oil embargo, double-digit inflation and the deep recession.
Although Bowmar's problems were typical of the times, Peter K. Gopal, a long time Bowmar consultant approached NSC to second source the Rockwell calculator chip for Bowmar that they were currently using. It seemed that an outside team had copied the design and Gopal needed it to be manufactured in high volume quickly. The requirement of integrating an outside design with the experimental Rockwell Process was what brought Asumi to the meeting. The Rockwell Process had been tested before by NSC and the prototype chip was working, but the project had failed due the client's funding issues.
At lunch, Asumi and Gopal discussed the reverse engineering process that Gopal's outside team had used. Asumi expressed concern at to the viability of creating a working design this way. Gopal assured him that he had done this before and there would be no issues.
With this assurance, Asumi gave his blessing for the project start, created the paperwork for the NSC budget and funding requirements. Three weeks later, Gopal called him to say that the design was finished. How did they want to proceed?
As with all joint development projects, until all technical team members are in one room, the real facts aren't discovered. This was no different. Although Gopal had the design completed, it was drawn manually as a composite drawing showing all seven processing layers or masks, but it wasn't digitized or converted to computer data yet. Gopal stated that he had outside agencies that could do it, but the delay would be a month or two.
NSC's design team responded and said that they could get right on it and have it completed within the next two weeks. Once approved, they would also create the tooling or photo masks needed in the chip manufacturing process.
Therefore, with a quick estimate, Gopal's approval, and the implementation of a Project Change Order, NSC took over the process. Within one week, NSC created a series of 200 times size (200x) computer drawings, which were submitted to Gopal's design team for approval.
Within 4 days, the corrections and changes were marked up and submitted back to NSC for implementation. Finally, the "Go to Manufacturing" was authorized by Gopal and NSC shifted into overdrive.
By running a computer program on the Calma GDS 1 Computer Graphic computer, NSC created the tooling data that created the Pattern Generation Magnetic Tape. By filling out a production work order, NSC Design Team submitted the 12-inch Pattern Generation tape to the NSC Photo Mask department who would create the tooling for manufacturing.
Seven weeks later, Asumi contacted Gopal to inform him that the first batch of wafers would be ready for testing. Could he or a team member come over to assist?
Gopal and Gene Cavendish, a "moonlighting" consultant, who worked at Fairchild full time, met with the postproduction engineering team of NSC. Testing equipment was set up and a software program was loaded. As the microscopic needlelike probes were lowered to make contact with the test chip, everyone held their breath. Would it work? As it had been run many times before at Rockwell and at Bowmar's Shipping & Receiving, Cavendish executed the computer test program. The program ran thru a series of inputs to the calculator chip and then read the output results.
Within 3 minutes, the Functionality Test was completed. Cavendish simply said "Damn, not bad guys, it works." Although other tests would be run, the design, digitizing and tooling was correct and approved. High volume production was authorized.
On that early summer night in 1974, Asumi and Gopal had their first of many quiet dinners to celebrate the successful launch of a product. Gopal had bypassed the extortion of Rockwell's marketing and Bowmar would now have the volume of chips needed for Christmas, 1974.
In early July, Gopal traveled home to Singapore on vacation and there, he could see how the US economy was affecting everyone else in the world. It shouldn't have been a surprise to anyone that Gopal had second sourced the Rockwell chips. Rockwell had gone directly to Sears with a proposal to supply calculators directly, bypassing Bowmar altogether. Rockwell had now become a competitor of Bowmar's not just a vendor.
However, this vacation was also needed. One can only work so many 80-hour weeks nonstop. How many had there been? It's been two year now? Or was it longer? So it wasn't a surprise when his father said he had someone he would like to meet. Something about taking the Bowmar product line into Hong Kong and points west.
With calculators at the $14.95 retail level in the US, many of the Pacific Rim countries including India, Pakistan, and China could now afford the calculators. The need for math accuracy was an international need, and with a calculator on every desktop in the US, it had become a major competitive disadvantage for any company, let alone country to limit their math calculations to 10-4 or 0.0001.
With scores of companies rushing into the pocket calculator business as demand for this amazing new product soars. Prices began to fall as competition grew. The average price for a basic four-function model was now down to about $25 US. A few models, including those by Casio, Rapid Data, and Digitrex would sell for below $15. Many said when TI filed for a patent application for the hand-held calculator, the end of the US calculator market was near. It also should be understood, that no other country in the world would honor that patent outside the US.
Then came the techno-bomb shell, HP introduces their first pocket calculator, the HP-35, the world's first pocket calculator with scientific (transcendental) functions. Within 3 months, all manual slide rule sales plummeted worldwide, never to return.
No longer, would the worlds space agencies and vendors need to rely on the mainframe computer for their calculations.
This damn pocket calculator had just changed the world forever. Within the next 6 months, both Rockwell and TI came out with their own reversed engineered version.
It was early in August; when Asumi had lunch with Gopal to inform him that NSC had decided to enter the manufacturing of the calculators for their own needs and NSC could no longer supply Bowmar with the chips. Asumi recommended to Gopal that he needed to talk with a close friend who worked for Western Digital in Newport Beach, CA. Asumi would set up the meeting for Gopal.
Within hours of the initial meeting, Bowmar and Western Digital inked a contract to source Bowmar's complete product line. Overnight, Western Digital became the largest independent calculator chipmaker in the world and their stock tripled with the formal announcement.
Due to poor holiday sales and excess inventory, in early January 1975, Bowmar management calculated its after-tax loss had soared to $20 million on a year's sales of just $80 million. The company was left with no other course but filing for protection under the federal bankruptcy law. Many disagreed with the decision. Others, looked at the recession and said "It's time to move on."
However, it was in the men's bathroom at Bowmar that Gopal overheard two members of the board discussing the up and coming bankruptcy. This meant that Peter K. Gopal, President of Silicon System International was about to lost his major customer, Bowmar.
The next day, Gopal contacted Asumi with the news of Bowmar's decision and the need to halt everything. It wasn't "Oh, help! I need a job." It was simply "The Bowmar stock is going to fall like a rock and it's going to take down all the vendors as well, including WD."
"Shit. Look Peter, if you are free tonight for dinner, I would like to run this past my wife." As Dan handed the instructions on how to get to his house, he said "Get there early, say... 6:30. We can have dinner, and still give her time to think about what she would recommend."
It was over dinner that Asumi's new wife Niki was explaining the best way to take advantage of this information. "After all when the Bowmar bankruptcy is announced, the Western Digital stock will fall from 32 to 3 within an hour."
So we should just sell it outright." Niki continued "But to take advantage of this, you should buy as many WD Index Option Puts as possible."
As one of the new investment broker for one of the Brokerage houses in Palo Alto, Niki could easily set up an account for him and execute the transaction for him in the morning.
She added "We could also look into a short sell position as we get the market's opening prices for both companies." In the morning this was done.
Seven days later came the Bowmar Board's decision:
"Even those closest to the pocket calculator business have been overwhelmed by the blistering pace of developments and the free fall in prices. Indeed, so bloody has been the price war at the cheap end in particular that many a proud name has fallen by the wayside. Last week the second largest calculator manufacturer in the United States, Bowmar Instrument Corp, became the latest casualty and has filed for bankruptcy"...
In Feb. 1975 the journal "New Scientist"
Over the next 15 days, the three netted $386,484 using the stock market and their insider information.
In addition, Asumi agreed to manufacture a new "private labeled" version of the four function calculator chip for Gopal.
Gopal could now sell it to a new client in Singapore; his Uncle.