Fab 30 starts production in Q3: Read Jerry's speech at yesterday's meeting:
I requested a copy of Jerry's speech from Drew Prairie, an ideal PR guy we are lucky to have at AMD -- professional, quick and pleasant to deal with. Since the speech is already in public domain, so post it here shall cost Drew no trouble.
Highlights:
*******************************************************
The K7 will be introduced at frequencies up to 600 megahertz.
In the second half of this year, we will begin production in Fab 25 on 180-nanometer technology, also with aluminum.
We plan to commence the migration of AMD K-7 production to Fab 30, employing 180-nanometer technology with copper interconnects, next quarter.
******************************************
CHAIRMAN AND CHIEF EXECUTIVE OFFICER
AMD
1999 ANNUAL SHAREHOLDER MEETING
NEW YORK, NEW YORK
APRIL 29, 1999
Two days from now, AMD will observe the thirtieth anniversary of its founding.
[Slide: AMD 30th anniversary logo]
In 1969, when I founded AMD with seven colleagues from Fairchild Semiconductor - where, in the original incarnation of that company - the monolithic integrated circuit was born, we saw limitless opportunities in the semiconductor industry. Thirty years later, AMD is a major player in a very different industry. Today, while huge opportunities still remain, the competition is infinitely keener and the marketplace is infinitely less forgiving.
Decades ago, I first advanced the notion that semiconductor integrated circuits were the new crude oil of the information age. It was a particularly prescient statement, and today the global semiconductor industry is the most competitive industry in the world: giant American, European, Japanese, and Asian companies compete vigorously to secure a growing and profitable portion of this industry that generated revenues of $135 billion in 1998.
Today, information technology, the offspring of the union of semiconductor technology and empowering software, is driving the American economy, which in turn is pulling the world economy. Improvements in productivity from the application of information technology are creating more wealth than the world has ever known. The microprocessor is the single most important contributor to productivity improvement in the world.
AMD's long-stated purpose is to 'empower people everywhere to lead more productive lives.' Coupling our purpose with our mission since inception to provide 'building blocks of ever-increasing complexity to reduce the cost, improve the performance, and shorten the time to market for manufacturers of electronic equipment for communication and computation' has made competing in the market for microprocessors - the ultimate building blocks - our manifest destiny.
The microprocessor business is the largest segment of the semiconductor industry, and the segment is dominated by Microsoft Windows compatible microprocessors. The profit opportunity in this segment is enormous.
The dominant alpha male gorilla in this market segment is Intel.
[Slide: mean and ugly gorilla]
According to Dataquest, Intel's total 1998 semiconductor revenues at $22,675,000,000 exceeded the combined semiconductor revenues of the next five largest U.S. companies combined!
[Slide: pie chart showing Intel plus Motorola, Texas Instruments, IBM, Lucent, and AMD]
We are one of those five companies, and the only one to compete head-to-head with Intel in microprocessors.
Why?
Because that's where the money is - and we're uniquely qualified to go after it!
[Slide: gorilla sitting on a pile of moneybags]
From Intel's 10K report for 1998, we learned that on $21.5 billion of microprocessor sales, they earned operating profits of more than $9 billion - or a 42 percent operating profit margin, while on the other $4.7 billion of revenues, they generated an operating loss of hundreds of millions of dollars. Clearly, Intel will continue to aggressively defend the moneymaking machine that their monopoly enables.
In 1991, with our introduction of the Am386 processor, we challenged the monopoly that Intel had established through their earlier breach of our 1982 technology agreement. Our success with the Am386 family and the successor Am486 family generated the highest operating profits in our corporate history. Our success also triggered the wrath of our much larger competitor, a raft of lawsuits, and a clone label.
[slide: angry gorilla]
We embarked on a course to resolve all the legal issues and establish AMD as the nucleating point for an alternative solution to an Intel monopoly based on differentiated microprocessors within the Microsoft Windows standard. After the commencement of sampling of our first K-86 processor more than four years ago, we articulated our long-term goal of attaining 30 percent unit market share of the global market for microprocessors for Microsoft Windows computing in the PC market by the end of 2001.
What we are trying to accomplish is a very hard thing to do. We are passionately committed.
[slide: Kennedy paraphrase]
With acknowledgement to President John F. Kennedy, 'Let every observer know, whether he wishes us well or ill, that we shall pay any price, bear any burden, meet any hardship, support any friend, oppose any foe, in order to assure the survival and success of our microprocessor business.'
Why?
Because the market for microprocessors for Microsoft Windows computing continues to represent our greatest opportunity for profitable growth and increasing shareholder value. Accordingly, my formal remarks today will be primarily focused on the processor business.
In a year that saw the largest decline in worldwide semiconductor industry revenues since 1985, microprocessors for Microsoft Windows computing were one of the few bright spots. Today, even as the overall industry recovers, AMD's fortunes are largely determined by our performance in the microprocessor business. In the final quarter of 1998, AMD-K6 family processors accounted for two-thirds of our revenues, and all of our growth and profits.
In May of 1998 we introduced the AMD-K6-2 processor with 3DNow! technology. Our 3DNow! technology is simply a better idea that provides dramatic improvements in detailed 3-D imaging, smoother video playback, and near theatre-quality audio to enable the most powerful visual-computing platform.
[Slide: AMD-K6-2 processor]
For the first time in our history, we had a new, differentiated processor, fully compatible with the Microsoft Windows computing standard that offered a clear and compelling advantage for consumers in a contemporaneous time frame with our competitor. The response of the marketplace validated our belief that our customers' best interests are served by having two very good sources working very hard for their business and providing an enhanced range of solutions for consumers.
By the end of 1998, eight of the world's top ten manufacturers of personal computers offered systems based on AMD-K6 family processors.
[slide: top 10 PC OEMs]
Despite the well-publicized yield problems of the first quarter of 1999, our customer momentum continued, and today nine of the top ten manufacturers of personal computers are AMD customers.
[Slide: IBM, Compaq, Toshiba logos]
We are also now supplying products to the top three mobile PC manufacturers -Toshiba, Compaq, and IBM.
Both unit shipments and revenues from AMD-K6 family processors increased dramatically in 1998, as unit volume more than trebled, and AMD-K6 family processor revenues nearly trebled to $1.25 billion. We shipped more than 13 million AMD-K6 family processors, including more than 8.5 million AMD-K6-2 processors with our innovative 3DNow! technology.
We achieved dramatic gains in market share. In the final quarter of 1998, AMD captured a 16 percent share of the worldwide market for processors for Microsoft Windows computing - more than double our market share for the same period of 1997.
In the bellwether North American retail sector, our market share grew throughout the year.
[slide: U.S. Retail PC Market Share by Processor]
In January of 1999, for the first time ever, our market share in this sector exceeded Intel's, and in February surpassed 50 percent! In March, even in the face of the media blitz around the PIII and a shortage of high-speed processors from AMD, we achieved a record 39 percent share of the portable market, and had a 45 percent share of the desktop market.
[slide: U.S. Retail Portable Market Share by Processor]
Our share of the sub-$1,000 market - the fastest-growing sector - remained above 60 percent.
We are mounting a successful challenge to an enormously rich and influential company that we can neither out-invest nor out-produce. We are challenging with the most powerful competitive weapon of all: innovation. Or, to put it another way, with better ideas.
To provide competitive performance at a lower price, we created the 100-megahertz Super7 platform in support of the AMD-K6-2 processor with 3DNow! technology. The value proposition implicit in this idea was so compelling that we secured industry endorsement in support of a new infrastructure. A host of motherboard manufacturers, chip set companies, independent software vendors, and even Microsoft rallied in support of an AMD-evolved infrastructure. In so doing, we took our first steps toward becoming a virtual gorilla.
[slide: wire-frame gorilla]
Our successful experience in building independent infrastructure support for the AMD-K6 family was an essential first step as we move on to the forthcoming AMD-K7 processor which entails further differentiation from the Intel proprietary - i.e., monopoly - infrastructure. The determination to be the nucleating force for an alternative platform represents simultaneously one of our greatest opportunities and toughest challenges.
The success of the AMD-K6-2 processor with 3DNow! technology enabled AMD to return to profitability in the third quarter of 1998, and to report record revenues and increased profits in the fourth quarter.
Today we have the broadest customer base and the strongest product line-up we have had in our thirty-year history - and yet, we just reported our largest quarterly loss ever!
What's wrong with this picture?
In the first quarter, unit shipments of AMD-K6 family processors plummeted to 4.3 million units from 5.5 million units in the prior quarter. Fierce price competition compounded the problem. With no help from our non-microprocessor product lines, total first-quarter revenues declined to $632 million from $789 million in the fourth quarter and we incurred a horrific loss of $128 million.
What happened?
As the dominant processor supplier, Intel sets the pace of the processor roadmap. It controls the schedule of introduction of higher-clock-speed devices, and in response to competition from the AMD-K6 processor family, it has accelerated the pace of the clock speed race. From a system performance standpoint, the clock speed of the processor is not an all-important consideration any more than revolutions per minute - RPMs - is the definitive measure of performance in an internal combustion engine. The definitive measure of performance in an internal combustion engine is horsepower. Unfortunately, there is no directly analogous metric for "computational horsepower" in a PC. Numerous benchmarks exist for comparing PC processors, and they are subject to much marketing manipulation. Clock speed has thus become the principal battleground of the marketing war for consumer acceptance. AMD is committed to matching Intel clock speeds even as we provide more system performance at a given price point.
Throughout 1998, Intel aggressively accelerated the pace of new product introductions, including rapid introduction of ever-higher-clock speed devices. Late in the year, it became increasingly clear that we would have to produce AMD-K6-2 processors capable of operating at 400 megahertz in volume to support our top-tier customers in the all-important Christmas season.
To support our customers, we pushed both our process technology and our AMD-K6-2 processor design to achieve maximum clock speed. Processor clock speed is a function of both the chip architecture and the process technology employed to manufacture the device. Tuning the process and the design to achieve maximum clock speed while maintaining acceptable yields is an ongoing challenge that ideally involves a series of engineering iterations to achieve optimum results.
Pushed by Intel to match an ever-faster pace of new product introductions, and pulled by the demands of our largest customers for competitive higher-clock-speed devices, we encountered yield problems last December that carried over into the first eight weeks of 1999 with a devastating impact on our aggregate production in the first quarter.
By the end of the first quarter, changes implemented to tune the design and the process technology were paying off. In the five-weeks of fiscal March, we shipped as many AMD-K6 processors as we did in January and February combined. Shipments of devices at 400 megahertz or higher clock speeds nearly tripled in March over January and February combined.
Even as we struggled in the first quarter, we added Gateway and Toshiba to our roster of top-tier customers. The yield issues that limited our production in January and February are now behind us, and our customer relationships, new and old, remain intact.
That brings us to today, where we are in a stronger position in process technology, production capacity, and platform competitiveness than at any time in our corporate history.
[slide: AMD R&D spending]
The level of effort, the size of the investment, and the extent of the creative initiatives to arrive at this point are unprecedented for a company of our size. R&D spending is expected to reach $750 million this year.
[slide: Fab 25]
The total investment in our Texas megafab, Fab 25, is expected to reach $1.8 billion by year end.
[slide: Fab 30]
Our Dresden megafab, Fab 30, will represent a $1 billion investment in the same time frame. These on-going investments are the principal contributors to the continuing escalation of our breakeven revenue level, which now stands in excess of $800 million per quarter.
Several questions thus arise.
First, can AMD attain the revenue levels necessary to attain and consistently sustain profitability?
And second, can AMD achieve profit margins adequate to achieve positive EVA to sustain the on-going investment demanded in our capital-hungry industry?
The answer to both questions is absolutely yes - but only if we can maintain our phenomenal success in the consumer sector of the PC marketplace and increase our penetration of the commercial sector. Corporations buy a lot more computers than consumers do, and the current pricing structure in the commercial sector supports higher processor pricing. Today, our participation in this segment is insufficient to impact Intel's ability to extract monopoly prices there. This enables Intel to subsidize their ability to be extremely aggressive where we do compete.
[Slide: AMD-K6-III processor]
The recently introduced AMD-K6-III processor with 3DNow! Technology and an innovative tri-level cache capability is competitive or superior in performance to the Pentium II and to the Pentium III clock-for-clock on relevant benchmarks, and is priced lower than both. In comparable technology, the K6-III should have a manufacturing cost advantage over the aforementioned Intel offerings.
We are ramping production of the AMD-K6-III processor to meet strong demand. We are currently shipping 400- and 450-megahertz versions. Generating volume shipments of K6-III processors at clock speeds of 500 megahertz and higher represents a significant challenge, including perhaps a migration to the next generation of process technology with 180-nanometer feature sizes.
Today, all of our processor production is on leading-edge 0.25-micron (250- nanometer) technology. We have successfully developed 180-nanometer, six-layer, aluminum interconnect technology to remain at the leading edge. We plan production on 180-nanometer technology in Fab 25 next quarter. We plan to make further inroads into the commercial sector, particularly in the small and medium business segments with the AMD-K6-III processor.
We have considerably more ambitious plans in the commercial sector for our next-generation product. We want to bring competition across the range of computers for the enterprise sector. We want to change the landscape of the computer industry. The AMD-K7 processor will be the first seventh-generation processor on the market.
What constitutes a new generation?
A new generation should offer:
· a new micro-architecture;
· higher clock speed capability for a given process technology;
· a faster and more powerful bus;
· more powerful floating point performance;
· higher multimedia performance.
[slide: AMD-K7 logo]
The AMD-K7 processor delivers all this and more.
We are on track to introduce the K7 in June. The K7 will be introduced at frequencies up to 600 megahertz. Today we are demonstrating a system powered by an AMD-K7 processor running at 600 megahertz featuring an AMD-developed motherboard and chip set. We previously demonstrated this system at Ce-Bit in February.
Based on initial product sampling, we are encouraged by the extremely positive feedback from customers. Nonetheless, the challenges we face in assuring the success of the AMD K7 processor are substantial and unprecedented in our history:
· It is a completely new design incorporating a number of better ideas, including a 200-megahertz bus, industry-leading floating point performance, and other enhancements to be announced upon introduction.
· The device will employ a module requiring a new infrastructure, including motherboards, chipsets, and BIOS. Once again, we must nurture and build a supporting infrastructure capable of supporting customers around the world. [slide: wire frame gorilla adding flesh and fur]
· The AMD-K7 processor features six layers of metal interconnects.
[slide: AMD-K6-2, AMD-K6-III, and AMD-K7 on 0.25-micron technology]
We are currently producing the AMD-K7 processor in Fab 25 employing 0.25-micron technology with aluminum interconnects. In the second half of this year, we will begin production in Fab 25 on 180-nanometer technology, also with aluminum.
[Slide: AMD-K7 on 0.25-micron and 180-nanometer technology]
· Meanwhile we must bring a new megafab in Dresden - Fab 30 - on line to build the most sophisticated product in our history employing the most advanced technology we have ever developed. We plan to commence the migration of AMD K-7 production to Fab 30, employing 180-nanometer technology with copper interconnects, next quarter. We believe this technology will enable production versions of the AMD K-7 running at clock speeds of one gigahertz or more by the end of next year.
Today, for the first time, we are demonstrating an AMD K7 processor in the Kryotech Super G computer running at 1,000 megahertz! This AMD K-7 gigahertz demonstration system is based on a standard pre-production K7 card module with an AMD-K7 CPU. The module is configured with 512 kilobytes of Level 2 cache RAM. When cooled using the Kryotech thermal solution, the processor operates at 1,000 megahertz - that is, one gigahertz.
Our first three decades have been quite an adventure. The adventure continues. I am sure you share my fervent wish that the next decade will involve more thrills and fewer spills. I can't promise you that. We are pushing the envelope of technology. We are on the frontier of science. We will continue to push for ever-greater performance at ever-higher clock speeds. The acceleration of the rate of technology advance is truly incredible.
In 1994, in a speech at the PC Outlook Forum entitled, "Microchips beyond the Millennium," I predicted that AMD would be producing a processor manufactured in 180-nanometer technology with 20 million transistors and 600-megahertz clock speed in 2001! The AMD K7 processor achieves this and more today!
The advent of the Internet as an integral part of everyday life has profound implications for total semiconductor demand. E-commerce is expected to be the main driver for semiconductor demand well into the next millennium as networked PCs and PC-like appliances proliferate to create a new universe of interactive communications. AMD's vision is to provide the most powerful visual computing platforms at affordable prices based on differentiated solutions enabled by AMD processors. Our customers and our infrastructure partners want us to succeed. They are helping us to deliver on the promise of that vision.
[Slide: fully developed wire-frame gorilla]
In this, the age of the Internet, the AMD-K7 processor represents the greatest opportunity for AMD ever.
Carpe diem!
[Slide: green gorilla]
|
