Ok, what did I miss about p4 2GHz in 01? Do they have "minor" issues
that should last for 1.5 years?
07:34am EDT 20-Sep-00 Credit Suisse First Boston
...............................................
Summary
With two weeks to go, it appears that Intel will report revenue at least in
the 8-10% sequential range, with double-digits still possible.
While the quarter started off very linearly, and was on track for 12-15%
growth, a tardy summer revival in Europe and not enough Celeron processors
have limited the potential upside we projected on 8/17/00.
However, demand has started to "lift off" with particular strength from
servers and laptops, and stronger European orders for 4Q. Supporting this
are memory orders and a renewed build-out of Micron's Lehi fab.
It appears that most of the pessimism about 3Q00 is already in the stock, so
based on the longer-term outlook (as corporate upgrades now begin), we
reiterate our Strong Buy and $100 price target.
........................................
Intel is the world's largest semiconductor company and the largest supplier
of microprocessors (CPUs) with an estimated 80% unit share of the PC
processor market. The company is also the largest supplier of Flash memory
and PC chipsets and a leading supplier of low-end networking solutions.
Intel in Final Stretch of a Wild 3Q-Intel Guidance Intact
If you didn't have your seatbelt on with Intel in 3Q, you probably feel dizzy
and confused. Guess what? We believe Intel had the same nausea. 3Q00
actually has been a more linear quarter than usual this year. Normally, 3Q
is very back-end weighted, with the month of September making up nearly 50%
of Intel shipments. This year started off looking like Intel only needed 35-
40% of revenues from the month of September to hit its guidance for 3Q00.
Given normal patterns, this meant that by early August it appeared Intel was
on track to hit 12-15% sequential revenue growth, with the possibility that
Intel would be accused of underestimating demand...again. Part of this was
due to the assumption that the new price cuts that were initiated in mid-
August by Intel right before its Developer Forum would coincide/result in a
pick up in shipments. However, no one seemed to have told Europe, which up
until ten days ago looked anemic this quarter. Yet, just when it looked like
the vultures on Intel might be right, corporate and European demand appear to
have been resuscitated. In Asia, it appears that original forecasts will be
met, although with not much upside-Gigabyte and Microstar look to be fine,
while Asustek may grow 50-60% on a month-to-month basis for September.
The bottom line is that Intel looks to be on track to report 8-10% sequential
revenue growth this quarter, with 10 days still left. This would fall out on
the high side of Intel's guidance, but short of our revised number for 12% (
modified on 8/17/00). ..................
Two Culprits: A Late "Lift-off" and Not Enough Celeron
We believe there are two main culprits for the fluctuating 3Q signals and the
likelihood Intel will beat consensus, but not blow out this quarter:
A later summer lull, with Europe waking up later than expected.
Not enough Celeron allocation for 3Q.
As mentioned, normal demand exiting the summer tends to pick up in late
August and build in September for the seasonally strong 4Q. Unfortunately,
there was no sub-$1000 phenomena of 1998 or the FreePC/e-Machines push of
1999 to fuel an abnormally robust back-to-school season. In addition, it
appears that Europe remained in siesta longer than expected-possibly due to
the oil shock. Whatever the reason, we believe the "lift-off" in processor
shipments (as Intel refers to it) didn't occur until the second week of
September........................
Four years ago, if Intel didn't have enough of a lower speed product, it
could always "downbin"-that is re-mark a higher speed product and sell it at
the lower speed. Unfortunately, with Intel's new segmentation strategy, it
is not so easy to take a Pentium-3 and resell it as a Celeron. ............
Competition Unable to Take Advantage; Intel ASP Looks to Be Up
Fortunately for Intel in 3Q00, we believe AMD also had a similar
misallocation between its slot and socket lines, thereby preventing AMD from
taking market share at Intel's expense. While in the high-end ($800-900
Athlons), AMD appears to be cutting prices very aggressively (slashing $125-
200 per chip) to motivate users, it appears that OEMs continue to prefer
AMD's Duron chips as a secondary source in the lower end products. Thus,
even with the aggressive AMD price cuts, it appears that Intel is not losing
market share in the high-end. Given a richer mix of laptops and servers,
along with the lower Celeron allocation, means that Intel's ASPs should be up
slightly.
Meanwhile, AMD appears to have its own issues (in terms of motherboard
infrastructure support and long-term manufacturing) and has been aggressively
cutting prices in order to make this quarter's unit forecast of 3.6 million
units. It appears that thanks to Compaq, AMD will hit its unit target...this
quarter.
Manufacturing Fine at Intel
Concerns regarding Intel's production also seem to be misplaced. It does
appear Intel's yields are not as high as it hoped for to date with its 1.13-
GHz (and to a lesser degree its 1-GHz) processors. However, this really has
no material impact to either 3Q or 4Q. The issue is mainly one of pride.
For the first time in a decade, AMD has closed the gap in terms of processor
speeds now that it is getting good yields on its own 1-Gig chip. However,
being successful in the microprocessor market is no longer just about having
the fastest chips (as it was in the 1980's and early 90's), it's also about
having the supporting chipsets, motherboards, and marketing clout. With
Intel and Via resolving their lawsuits (except for the lawsuit designed to
prevent Intel IP from getting to AMD), and Via working on its own strategy,
we believe AMD has lost a critical portion of its chipset support. However,
the current Intel yields do mean that in the near term Intel can not turn up
the heat on AMD by accelerating its processor introduction roadmap. We do
not believe that this will have an impact in 2001, as based on feedback form
equipment vendors, we believe Intel will be more than on track by the time it
releases the Pentium-4 in October. More importantly, there appears to be no
credibility to recent contentions that Intel's designs (mainly the Pentium-4),
cannot hit the 2-GHz level next year...................................... |
