Two days ago Hector Ruiz stepped down as CEO of AMD, while
Dirk Meyer—the existing president and COO—is moving up to take his place.
Will this change of leadership ease AMD’s woes? More specifically, can it turn the tide on
bringing products late to market, losing market share, and losing millions of
dollars.
Not likely, in my opinion.
And this is not a matter of waiting to see what the new regime will
do. Meyer is an engineer and was already
responsible for the day-to day operations of AMD. As such, he shares the blame for AMD’s
problems.
The real problem is Intel’s prowess in low-cost
semiconductor manufacturing and the Intel ability to outspend AMD by billions
of dollars. The other culprit is Jerry
Saunders, the feisty founder of AMD and the one who years ago decided to take Intel
head-on by copying the line of Intel processors that power the PC market. That stubborn decision led to the demise of Ruiz. AMD is too undercapitalized to best Intel in
a processor development race or win a fight for lasting manufacturing
supremacy.
AMD’s recent success was not about its own muscle; they
caught a sleeping giant off guard.
Intel failed to sense the change in the PC market’s changing
requirements away from constantly improving performance and the attendant
increased consumption of power. Data
centers were experiencing increased costs in electricity to power the
electronics as well as the cooling systems required to keep the electronics
from burning up. Cries to Intel to
concentrate on lower-power devices as well as increasing performance fell on
deaf ears. Hector Ruiz was listening, and AMD responded with a
line of processor chips that met performance requirements but consumed much
less power.
When Intel realized what was happening, their new CEO, Paul
Ottillini, turned on the afterburners to develop processors that did more than
equal the AMD product line. The Intel
processor development teams and their manufacturing arm introduced devices that
surpassed the AMD offering. No one in the semiconductor industry can
compete with Intel in the PC market.
The semiconductor units of both Motorola and IBM, working in
partnership, tried it in the late 1980s with the Power PC processor line. They also caught Intel in a deep sleep. The Power PC was a generation ahead in
manufacturing technology compared to what Intel was bringing to the PC
marketplace. Within 18 months, Intel had
introduced a line of processors that was a generation ahead of the two
intruders. Keep in mind that Motorola
and IBM were no slouches in the semiconductor world at that time and both had
deep pockets.
What is AMD’s solution?
The company is basically a one-trick pony in regard to the markets served. Years ago, AMD abandoned their embedded
systems business to concentrate on the PC market. Today, they need to reverse that move. They lack the capitalization to effectively
complete with Intel. AMD will only be
minimally profitable if it does nothing but pick up the market crumbs that
Intel leaves behind.
AMD must become a “fabless” company, meaning they outsource all manufacturing to another company, called a foundry, which focuses solely on semiconductor manufacturing. Two such companies exist in Taiwan, Taiwan Semiconductor Manufacturing Company (TSMC) and United Microelectronics Company (UMC). AMD would then be free to sell its existing manufacturing plants and stop building new ones.
Next, AMD would have to enter new markets such as the
embedded systems business they once pursued.
Marketing influence—something missing from AMD the past few years—would
need to take hold in order for new-market penetration to succeed. This would be a formidable task, since the AMD
culture is accustomed to an engineering and manufacturing orientation. And prospects for a dramatic culture change
are improbable given the company’s new management.
Lastly, AMD needs to recognize that the acquisition of ATI was
a huge mistake. The price was too high,
but it’s time to stop throwing good money after bad. To continue to compete in the PC market
against Intel, who also makes graphics processors, is as futile as competing
against Intel with the Pentium class processors. The day will come when Intel will offer a
central processor and the graphics processor all on one integrated chip that
will achieve a more cost-effective solution.
One of the first things Meyer needs to do is sell ATI. Yeah, a move that that would take two big
brass ones, but he can’t stop the hemorrhaging while AMD is still in the PC market.
Will all this happen?
Who knows? Only time will tell us
what recovery plan the new AMD management will devise.
AMD is not undervalued; rather it is fairly valued for now
but the upside reward is minimal while the downside risk is great. Given its weak competitive stance and lack of
capital, there’s even a chance the company could go under.
Smart investors will wait to see what happens.
Copyright by Bill
Durrenberger July 18, 2008 All rights reserved.
Disclaimer: The statements and information presented in this article reflect
the opinions of the author and do not constitute a recommendation to buy or
sell specific securities. Before
purchase of any investment, the buyer should consult a financial adviser.
