Intel follows the margin

Hey,

Anybody sobered up yet from the over-the-weekend I-told-you-so parties
yet? ;-).

Before you decide to load up any more than you already have on AMD and
to short any more than you already have on Intel, consider this:

http://www.forbes.com/newswire/2004/...tr1366203.html

quote

Intel brings in a better profit margin with Centrino, a bundle of three
chips, than it does on other mobile chips.

"Centrino commands a brand premium," said JMP Securities analyst Krishna
Shankar.

snip

With Centrino, Intel is trying to convince home laptop users that
portability is as essential as raw computing speed.

/quote

So, here you are, you've got four lines: a "high-end" server line, a
real-world server line, a desktop line, and a portable line. Unlike
what some people think, your only goal in life is to make money, not to
dominate the world.

Your high end server line has had problems from every direction
practically from the beginning, and it becomes more clear all the time
that, if you ever do get there, the margins you expected probably won't
be. Your real-world server line and your desktop line are both under
price and performance pressure, probably both in niches with
monotonically-decaying margins. What do you do?

You convince the world that what they really need is what makes you the
most money, that's what you do. There are other reasons for Intel to
make some of the moves they've made, but they all just conveniently
point toward Intel selling more of what it makes the most money on.
It's probably not too late to get out of those positions you just took.
Most of the rest of the world still has a hangover. ;-).

RM

Robert Myers wrote:
Anybody sobered up yet from the over-the-weekend I-told-you-so parties
yet? ;-).

Before you decide to load up any more than you already have on AMD and
to short any more than you already have on Intel, consider this:

http://www.forbes.com/newswire/2004/...tr1366203.html


So, here you are, you've got four lines: a "high-end" server line, a
real-world server line, a desktop line, and a portable line. Unlike
what some people think, your only goal in life is to make money, not
to dominate the world.

Your high end server line has had problems from every direction
practically from the beginning, and it becomes more clear all the time
that, if you ever do get there, the margins you expected probably
won't be. Your real-world server line and your desktop line are both
under price and performance pressure, probably both in niches with
monotonically-decaying margins. What do you do?

You convince the world that what they really need is what makes you
the most money, that's what you do. There are other reasons for
Intel to make some of the moves they've made, but they all just
conveniently point toward Intel selling more of what it makes the
most money on. It's probably not too late to get out of those
positions you just took. Most of the rest of the world still has a
hangover. ;-).

I'm not sure what you're getting at? Are you saying that Intel makes more
money on Centrino than on any other line of processors? I don't think the
article says that, all it says is that it makes more money Centrino than on
any other line of *mobile* processors. I seriously doubt that Centrino is
carrying Intel's other lineups.

Yousuf Khan

Yousuf Khan wrote:

snip


I'm not sure what you're getting at? Are you saying that Intel makes more
money on Centrino than on any other line of processors? I don't think the
article says that, all it says is that it makes more money Centrino than on
any other line of *mobile* processors. I seriously doubt that Centrino is
carrying Intel's other lineups.


Grrr. Mumble. Grumble.

Which of Intel processors carries a "brand premium"? Itanium does, sort
of, but customers usually don't pay it with any sense that they're
getting what they want for their money. Xeon has a reasonably
profitable brand premium, for a while at least, but it's under serious
attack.

The only sense in which anybody who really matters at Intel cares about
competing with AMD would set much too low a standard: making money. The
competition with AMD for desktop/server business is important to the
present, but Intel would like it not to be for the future.

Intel has surveyed the prospects in the server and desktop processor
business and decided they aren't as good as they used to be, so they are
looking to make their future elsewhere to every extent possible. I can
find other articles, like the puff piece about Barrett in
_Business_Week_, that say the same thing in different ways: the future
is everything _but_ what desktop and server processors. Or, at least it
would be if Intel could make it so.

If anything, P4/Xeon is carrying Intel right now. That's not a good
spot for Intel to be in. Intel isn't going to defend P4 because it's a
losing battle. It has assiduously kept the Xeon brand independent of
architecture. Dump P4, attach the Xeon label to something else as
quickly as you can, and persuade everybody that what they really want is
a laptop.

It's natural for people in technical groups to think of technical issues
as driving business. If a particular technology takes a nosedive, so
will the previously wildly-successful careers of lots of
narrowly-focused techies. An individual technologist doesn't have to
think that way, and neither does a company. When confronted with an
unexpected challenge, you can bear down harder under the old rules, or
you can try to change the rules. Intel is trying to change the rules.

RM

Robert Myers wrote:
Which of Intel processors carries a "brand premium"? Itanium does,
sort of, but customers usually don't pay it with any sense that
they're getting what they want for their money. Xeon has a reasonably
profitable brand premium, for a while at least, but it's under serious
attack.

The only sense in which anybody who really matters at Intel cares
about competing with AMD would set much too low a standard: making
money. The competition with AMD for desktop/server business is
important to the present, but Intel would like it not to be for the
future.

Intel has surveyed the prospects in the server and desktop processor
business and decided they aren't as good as they used to be, so they
are looking to make their future elsewhere to every extent possible.
I can find other articles, like the puff piece about Barrett in
_Business_Week_, that say the same thing in different ways: the future
is everything _but_ what desktop and server processors. Or, at least
it would be if Intel could make it so.

Well yes, if Intel wants to avoid strong competition, then at the moment the
laptop market is where it is at right now. But that's just a temporary
thing. The other markets it's starting to look weak in nowadays were all at
one time some of its strongest cash cows: desktops and servers. How is Intel
going to avoid the same fate in the laptop market? The last thing Intel
wants to do is put all of its eggs into one basket (the laptop basket) and
then a few months down the road, price competition enters that market. The
laptop market right now is a relatively boutique business compared to the
desktop, so it's filled with rich people with more money than brains. But
once the mystique of laptops wears off (you still see people get all
googly-eyed when they see a laptop, like as if they've seen a Ferrari), the
only thing that will matter is price.

Right now people might be willing to pay the premium for a Centrino for its
cachet. But the Centrino is much more expensive than Celeron, and even
P4M's. Even if people don't migrate to Intel's competition, they just have
to look across the aisle at the other Intels, the P4 and Celeron laptops, to
see that laptops can be had for a lot less. When Pentium-M replaces the
Pentium-4 as the next mainstream processor, people will still see that price
is lower if you don't pay for the Centrino branding.

If anything, P4/Xeon is carrying Intel right now. That's not a good
spot for Intel to be in. Intel isn't going to defend P4 because it's
a losing battle. It has assiduously kept the Xeon brand independent
of architecture. Dump P4, attach the Xeon label to something else as
quickly as you can, and persuade everybody that what they really want
is a laptop.

Well the Pentium brandname itself is also quite independent of architecture.

It's natural for people in technical groups to think of technical
issues as driving business. If a particular technology takes a
nosedive, so will the previously wildly-successful careers of lots of
narrowly-focused techies. An individual technologist doesn't have to
think that way, and neither does a company. When confronted with an
unexpected challenge, you can bear down harder under the old rules, or
you can try to change the rules. Intel is trying to change the rules.

Technical people look at the same factors as the general public, they just
have a lot more knowledge about the details.

Yousuf Khan

Yousuf Khan wrote:

snip


Well yes, if Intel wants to avoid strong competition, then at the moment the
laptop market is where it is at right now. But that's just a temporary
thing. The other markets it's starting to look weak in nowadays were all at
one time some of its strongest cash cows: desktops and servers. How is Intel
going to avoid the same fate in the laptop market? The last thing Intel
wants to do is put all of its eggs into one basket (the laptop basket) and
then a few months down the road, price competition enters that market.

It's a safe bet that Intel is not going to put all its eggs into one
basket if they have a choice.

The laptop market right now is a relatively boutique business compared to the
desktop,

A state of affairs that Intel apparently wants to change.

so it's filled with rich people with more money than brains.

Just like gamers who will spend their last nickel to get that last bit
of performance, rich people with more money than brains are a market of
limited size. Intel probably understands that, too.

But once the mystique of laptops wears off (you still see people get all
googly-eyed when they see a laptop, like as if they've seen a Ferrari), the
only thing that will matter is price.


Back to cars again, eh? Marketer story from a Bob Colwell (former Intel
Chief Architect) presentation:

"Everybody here who owns a Lexus raise your hand."

Hands go up.

"Why did you pay so much for a Toyota?"

snip


Well the Pentium brandname itself is also quite independent of architecture.


Good thing, no? Celeron is architecture-independent, too. Not all
brands are positioned to command a price premium.

snip


Technical people look at the same factors as the general public, they just
have a lot more knowledge about the details.


If you find yourself away from an internet connection and the telephone
for a few minutes, maybe by a quiet lake on a sunny day in mid-summer,
you might want to squander a bit of such a precious moment on examining
that assumption as carefully as you can.

Don't squander a really good moment on it, but you might also want to
consider this proposition as if it were a subject for college debate:
RESOLVED: "A good brand name is more valuable than a good architecture."
Don't jump to conclusions. Try both sides of the argument.

RM

Robert Myers wrote:
The laptop market right now is a relatively boutique business
compared to the desktop,

A state of affairs that Intel apparently wants to change.

Which if they succeed, will end up opening them up to pricing competition.

But once the mystique of laptops wears off (you still see people get
all googly-eyed when they see a laptop, like as if they've seen a
Ferrari), the only thing that will matter is price.


Back to cars again, eh? Marketer story from a Bob Colwell (former
Intel Chief Architect) presentation:

"Everybody here who owns a Lexus raise your hand."

Hands go up.

"Why did you pay so much for a Toyota?"

Yes, that's a good quote. Here's another quote heard recently about Intel's
move to can Pentium 4 and replace it with Pentium M:

"All we hear are unsubstantiated promises about how intel is going to do so
well with its next series of blunders."


Technical people look at the same factors as the general public,
they just have a lot more knowledge about the details.


If you find yourself away from an internet connection and the
telephone for a few minutes, maybe by a quiet lake on a sunny day in
mid-summer, you might want to squander a bit of such a precious
moment on examining that assumption as carefully as you can.

Remember I just got back from Bangladesh, where I had to do everything with
dialup and Google Groups. I qualify for having survived "roughing it". :-)

Yousuf Khan

Yousuf Khan wrote:

snip


Remember I just got back from Bangladesh, where I had to do everything with
dialup and Google Groups. I qualify for having survived "roughing it". :-)


That was the hardest part of your journey? Glad you're back safely.

RM

"Yousuf Khan" wrote:

"Why did you pay so much for a Toyota?"

Yes, that's a good quote.

But inappropriate and unnecessarily insulting. In the US, several
cars made by Toyota are only available under the Lexus nameplate.

chrisv wrote:
"Yousuf Khan" wrote:


"Why did you pay so much for a Toyota?"

Yes, that's a good quote.


But inappropriate and unnecessarily insulting. In the US, several
cars made by Toyota are only available under the Lexus nameplate.


There is no need for anyone to feel insulted. The point is that people
will pay for a brand name and will pay a premium for the right premium
brand name, not that people who pay a premium for a premium brand name
are suckers.

RM

Robert Myers wrote:
Yousuf Khan wrote:
Remember I just got back from Bangladesh, where I had to do
everything with dialup and Google Groups. I qualify for having
survived "roughing it". :-)


That was the hardest part of your journey? Glad you're back safely.

What you expected me to be robbed at gunpoint or something along the way?
:-)

Yousuf Khan