Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: Tenchusatsu who wrote (79597)	4/21/1999 3:25:00 PM
From: Paul Engel	Respond to of 186894

Ten & Intel Investors - Here's a good article on Intel's Internet directions - past & present. With the Level One acquisition, Intel might "commoditize" the network industry. Imagine Intel applying its hardware design discipline to produce these cutting edge "routing engines". Applying its own version of Moore's Law to network components, Intel might produce a new generation of ASICs that is twice as fast in a short period of time. When the fastest ASIC on the market is an off-the-shelf component made by Intel, the "hot startups" that we mentioned earlier will not be so hot anymore. Anyone will be able to buy the components needed to build a high-speed router. Paul {===================================} ixix.com "Intel Inside" the Internet April 20, 1999 Tawei Liao I. Introduction For years microprocessor speed was the most important hardware feature of a computer. A PC had to be fast in order to run the latest spreadsheet and word processing programs. That is no longer true; the Internet has changed the computing world. Right now, the killer application is the Internet. And the most important aspect of this killer app is the speed of the network, not the microprocessor. The speed of microprocessors will remain crucial in servers that power web sites and in high-end scientific and engineering workstations. But for the average consumer that uses his computer to surf the web and check email, the speed of the modem is far more important than the speed of the microprocessor. Andrew Grove is not asleep at the wheel. The chairman of Intel wrote about the "threat or promise" of the Internet in 1996 in his book "Only the Paranoid Survive" [1]. He already knows that the Internet represents a "10X" force that Intel cannot ignore. It is a strategic inflection point that has forever altered the competitive landscape of the entire computer industry. The last time Intel faced a tidal wave this big was in mid-1980s when the Japanese took over the world's memory chip market. Grove made the decision to get Intel out of the memory chip business and focus on making microprocessors. Intel emerged from that crisis stronger than ever. Since then it has become the world's leading maker of microprocessors. Can Intel remain competitive in this era of the Internet? That is the question we shall attempt to answer. II. "Intel Inside" the Internet Intel has several answers to the network revolution: 1.Win the segment of the network market closest to the microprocessor. 2.Get into the network industry by making routers and switches. 3.Create more demand for microprocessors by offering "host-based" solutions. 4.Become the leading component supplier for network equipment manufacturers. At first these four approaches seem coherent and quite reasonable. But as we shall see later on when we examine them in more detail, they are full of contradictions. And only success in the fourth approach will have a lasting impact to the company. First, Intel entered the segment of the network industry closest to the microprocessor: the market of Network Interface Cards (NIC). A NIC is a hardware component on a computer that handles connections to a local area network. Intel has done well in this market. The other big player is 3Com. Both companies account for roughly 70% of the NIC market. Second, believe it or not, Intel actually tried to compete in the network market by selling its own line of routers and switches [2]. This effort is commendable, but totally out of touch with reality. Intel has never been successful at making or selling end-systems. Its core competency has always been semiconductor components. This strategy failed of course. Intel simply cannot compete with Cisco in making routers or switches. Third, Intel's marketing department has been working hard trying to offer the microprocessor as a network solution. It has been pushing for host-based ADSL and cable modems. For example, on November 12, 1998, Intel made an equity investment in Integrated Telecom Express, Inc (ITeX) [3]. ITeX is a maker of host-based ADSL (Asymmetric Digital Subscriber Line) modems. Sometimes called a "software" modem, its product uses "the power of the main processor rather than the separate microprocessor and memory normally required by a standalone DSL modem device." [4]. This software solution lowers the cost of an ADSL modem. Intel is very interested in this technology because this software only runs on Pentium processors. It takes advantage of the new instructions in the Pentium III designed to speed up multimedia and communication functions. Intel is trying to create demand for its Pentium III processors by helping ITeX succeed. A similar announcement for host-based cable modems was made on February 16, 1999 at a cable convention [5]. Host-based solutions using the microprocessor addresses the consumer Internet access market. It is a small step forward, but it is not enough; Intel needs another strategy that aims at the other 90% of the market. Capitulation to the Internet revolution came on March 4, 1999 when Intel acquired Level One Communications [6]. This acquisition is aimed at "bringing customers advanced networking capabilities by offering increased bandwidth and functionality through silicon integration." It will make Intel into a leading component supplier for Internet equipment manufacturers. This strategy elegantly extends Intel's core competency in semiconductor design to the network market. Imagine Intel applying Moore's Law to produce ever-faster network components. This is the only strategy that will help Intel succeed in the Internet revolution. III. A peek into the future The 800-pound gorilla of the network industry today is Cisco Systems. There is a slew of startups that aim to compete with Cisco. Companies such as Juniper, Nexabit, NetCore, Pluris, Avici, etc all claim to have terabit switch routers. There is one thing that all these startups have in common: they all design their own ASICs (Application Specific Integrated Circuits). These ASICs are the "routing engines" that enable a router to process millions of packets per seconds. Higher routing speed can only be achieved through advances in hardware. Juniper, for example, claims that its M40 router is able to handle 40 million packets per second. This is due entirely to their superior ASIC design. With the Level One acquisition, Intel might "commoditize" the network industry. Imagine Intel applying its hardware design discipline to produce these cutting edge "routing engines". Applying its own version of Moore's Law to network components, Intel might produce a new generation of ASICs that is twice as fast in a short period of time. When the fastest ASIC on the market is an off-the-shelf component made by Intel, the "hot startups" that we mentioned earlier will not be so hot anymore. Anyone will be able to buy the components needed to build a high-speed router. What we just described should sound very familiar to everyone. This is what happened to the PC industry. The open architecture of the IBM PC with its off-the-shelf components made by Intel and an operating system from Microsoft "commoditized" the industry. PC clones flooded the market and drove IBM out of the PC business. Is this the future of the network industry? Is Cisco Systems destined to become the next IBM? Will Dell be making high-speed routers using Intel components? This is an exciting and unpredictable future. The one thing we are certain of is that Intel's entry into this market will forever change the nature of competition [7]. IV. References [1] A. Grove, "Only the Paranoid Survive," Bantam Books, 1999 [2] Intel Product Description [3] ITeX Press Release. [4] News.com Article. [5] News.com Article. [6] Intel Press Release. [7] T. Liao, "Cisco's Strategic Inflection Point." Internet ConneXion, Inc