SI
SI
discoversearch

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: L. Adam Latham who wrote (84983)7/6/1999 2:00:00 PM
From: Tony Viola  Respond to of 186894
 
Adam and thread, don't know if you could say there's anything new in this article, but some numbers cranking by the author on what AMD may be able to build, or not build, for CPU chips:

infohq.com

InfoHQTM - Computer Buying Advice

Can AMD Deliver the K-7
Athlon ?

An InfoHQ Editorial

On the eve of the launch of AMD's "revolutionary" K-7 , many
individuals and analysts wonder if AMD can deliver the new CPU in
quantity ( AMD announced the K-7 CPU (now named Athlon) was
shipping to vendors on 23 June and that systems would be available
in the 3rd quarter of 1999). While there has been much speculation,
we believe that a careful study of the K6-3's history and a review of
AMD's factory capacity may be the key to the answer.
One of our burning questions for the last 3 months has been, " Why
hasn't the K6-3 been produced in sufficient numbers to meet
demand?
Since the introduction of the K6-3 on February 22 1999, the CPU has
been in very short supply. We have seen very few new K6-3 systems
advertised in our local newspapers, and as a general rule, secondary
retailers have few, if any, K6-3 CPUs for sale. In the midst of this
shortage, AMD continued to increase the speed of the K6-2 to 450
MHz (February 27) and then to 475 MHz (April 5) and launched
several new laptop CPUs including: the K6-2P at 350, 366, and 380
MHz (March 9) and the K6-3P at 350, 366, and 380 MHz (May 24).
From the above we can conclude that it was not AMD's intention to
manufacture the K6-3 in large quantities or they were having such
significant quality control problems manufacturing the new CPU that
they chose instead to increase the speed of the K6-2 and to
manufacture laptop CPUs to cover this shortfall. Mr. W.J. Sanders III,
Chairman and Chief Executive Officer of AMD, stated in his Letter to
Shareholders in AMD's 1998 Annual Report dated 1 June 1999, "I
expect that an increasing proportion of production for the market
segments served by AMD will be devoted to the AMD K6-III
processor family throughout the remainder of 1999". Does this mean
that finally the K6-2 line will be phased out in favor of the K6-3 that
was announced in February?
Our second major question is, "Does AMD have the factory capacity
to manufacture the K-7 in sufficient numbers?". Right now, AMD
has only one factory producing all its desktop and laptop CPUs, Fab
25 in Austin, Texas. Per AMD, the factory is capable of producing
5,000 wafers per week -- 250,000 wafers per year and was operating
at 80% of its capacity in May (a wafer is roughly equivalent to 100
CPUs). AMD is also completing its new Fab 30 in Dresden, Germany
which is supposed to be up and running by the end of this year and
will also be capable of producing 250,000 wafers per year
(manufacturing capacity was stated in the above referenced Letter to
Shareholders).
So the burning question is, "Can Fab 25 produce enough Athlon
K-7s to meet demand through January of 2000?". We think not.
Assuming AMD can get the one factory up to a 95% production
capacity, that would mean they could produce 2,187,500 CPUs in the 7 months from July 99 -
January 2000 (25,000,000 annual CPU production x 15% additional factory capacity x 7/12
months of production). Assuming that any discontinued K6-2 capacity is used to produce K6-3s,
AMD would only produce 312,500 K-7 CPUs a month. Which means that the majority of the
new K-7 CPUs will be used in new systems from large OEM manufacturers like Compaq and
IBM. < Actually Mr Sanders has been quoted as saying that there would only be 1 million K-7s
shipped during the 1999 calendar year> .
We assume that the low K-7 production will result in a similar scenario as the release of the
K6-3. Even though the K6-3 was announced in February of 1999, AMD still had not reached
reasonable production levels by June of 1999. So even though AMD announced the release of
the K-7 on 23 June, we will not see reasonable production numbers of the CPU until the new
Fab 30 comes on-line in January 2000.
< InfoHQ Note: Obviously, we can not predict what AMD's management will do with any
certainty. However, if AMD can survive financially until its new Fab 30 begins production, they
will be able to double their CPU production, which should make them a much more competitive
company in the future.>



To: L. Adam Latham who wrote (84983)7/6/1999 4:28:00 PM
From: MONACO  Respond to of 186894
 
This guy Dan Niles must be quite the gambler (unless he has inside info), willing to lay his "whole credibility" on the line one week before Intel releases the numbers that can make or break him (IMO... I got to believe he missed something in his calculations)....M