Intel found to be abusing market power in Japan

On Mon, 14 Mar 2005 21:25:57 -0500, Yousuf Khan wrote:

George Macdonald wrote:
As for Intel/AMD, you have a good point about the symbiotic relationship:
Intel is now in the strange, never before seen, situation that they
actually, crucially need the cross-license agreement to survive in the new
x86 world - absolutely no question of living without it. AMD has, of
course, just as much need and I wonder if they would even think about
taking civil legal action against Intel for their marketing sins.

Why does Intel absolutely need the cross-licensing agreement? The only
thing I can think of is that they'll need it for the access to the
x86-64, but what else?

Isn't that enough? I didn't think you'd be the one to need convincing
about x86-64 as a necessary component of future PCs. Beyond that I'm not
sure but I'm pretty sure that AMD has some other patents which might be of
interest, e.g. large L1 cache efficiency.

--
Rgds, George Macdonald

On Mon, 14 Mar 2005 17:02:19 -0500, George Macdonald
wrote:

absolutely convinced that Intel's enthusiasm for "platformization" based on
the "Centrino success" is a total misread. People don't buy notebooks
because of Centrino.

Do you mean people buying notebooks in general or people specifically
buying Centrino when they buy a notebook?

--
L.Angel: I'm looking for web design work.
If you need basic to med complexity webpages at affordable rates, email me
Standard HTML, SHTML, MySQL + PHP or ASP, Javascript.
If you really want, FrontPage & DreamWeaver too.
But keep in mind you pay extra bandwidth for their bloated code

On Wed, 16 Mar 2005 03:57:18 GMT,
(The little lost angel) wrote:

On Mon, 14 Mar 2005 17:02:19 -0500, George Macdonald
wrote:

absolutely convinced that Intel's enthusiasm for "platformization" based on
the "Centrino success" is a total misread. People don't buy notebooks
because of Centrino.

Do you mean people buying notebooks in general or people specifically
buying Centrino when they buy a notebook?

I mean the supposed "success" of Centrino as a marketing exercise. IMO
people, and I'm including myself here of course, don't go out to buy and
say: "I gotta have one of them Centrino jobs"... unless maybe they're too
stupid to own a computer in the first place. Features are what sell a
system; in the case of Centrino, one of the most important features, given
that its performance is satisfactory, is the power management & battery
life which is mainly down to the Pentium M and its chipset; the NIC is a
throwaway which doesn't matter. In fact given the choice I'd avoid an
Intel-based NIC for just about any alternative.

--
Rgds, George Macdonald

Robert Myers wrote:
There are no spoils of war to divide up. Opteron didn't open up
any
new vistas. The business is in a period of decline and
consolidation.

Huh? Since when? This is the biggest and most profitable end of the
semiconductor industry, of course there's spoils of war to divide up

here -- lots of it. Of course, it's all AMD's to gain and all
Intel's to
lose.


Tell that to Wall Street. Don't waste your time with me.

Don't worry, Wall Street knows it too. You're the only one that isn't
aware of it.

The high tech sector may take decades to get back to where it was
pre-bust. These are _not_ good times.

Even in a downturn, we're talking about tens to hundreds of billions of
dollars in sales here. Most of it currently going to Intel, obviously.
Is it a wonder why so many startup companies always try to get into the
x86 market, despite having to compete with Intel? Some firms (eg. VIA)
make steady income in this market with less than a 1% marketshare in
it. It's related markets, like chipsets and peripherals are also huge
money makers for companies.

There's this thing called efficient market theory. You think you got
the landscape scoped better than the market, go make yourself rich.

Please do explain this "efficient market theory" of which you speak. I
guess one of its theorems is "The x86 market is efficient only if Intel
makes the majority of x86 processors"?

Were I AMD, I would use this situation to reposition myself
vis-a-vis
Intel by way of some agreements cut on the sly in some
out-of-the-way
place, not by going to court. That's what I would do. What AMD
does,
of course, is completely independet of my guess as to what they
should
do.

That's also illegal. It also falls into the same anti-trust laws,
it's
collusion. It's no different than if AMD and Intel decided to one
day
start pricing their processors exactly the same as each other at
exactly
the same time -- this would be price-fixing. The other thing would
be
marketshare-fixing.


That's how your mind works, apparently. AMD can sue Intel (not
smart,
in my opinion), or it can come to some kind of agreement about the
rules of engagement. Such a thing could be a consent decree
supervised by a judge. It doesn't have to be public and it doesn't
have to be illegal. AMD and Intel would be colluding to keep, say,
Via out? I'd hope they wouldn't be so stupid.

There is something seriously wrong with your grasp on reality -- that's
how my mind works? Seriously, Robert that's how everybody's mind works.
There is not a single person who wouldn't describe what you were
suggesting as anything other than collusion. There is no such thing as
coming to an agreement about rules of engagement between two companies
-- those rules already exist, they are called the competition laws.
You're supposed to compete with your competitors, freely, fairly, and
to your fullest extent possible.

Yousuf Khan

George Macdonald wrote:
I thought it was the U.S. military that insisted on the 2nd source
for
x86s?

No, I've always heard it was IBM that insisted on it. When IBM was
first making the PC, there really wasn't much of a market for PC's in
the military. The military would've been on mostly mainframes and
terminals at the time. Though there were other personal computers at
the time, they were mostly hobbyist or gaming machines (e.g. Commodore
64).

Intel has always been the principal suppiler of chipsets into the
notebook
market almost without exception. How many non-Intel chipsets are in
notebooks with an Intel CPU?

Well, that's mainly because Intel has been driving most chipset
manufacturers out of its market in general, not just in notebooks.
There used to be a time prior to this when you'd see just as many
variations in notebook chipsets as you see in the desktop market, even
on Intel processors.

Everything I read indicates that Intel's marketroids see it as a
grand
marketing coup of brand/name recognition... nothing of the sort IMO.

Yeah, there was a study done by AMD to find out if it should also adopt
platformization too. They found that most people really couldn't care
less if the machine was Centrino or not, all they cared about was the
laptop's own brandname. The only subsets that did care about Centrino
were techies, and secondly airport travellers who are bombarded with
Centrino ads in airports.

Speaking of airports, I took my laptop to the airport once, and I
couldn't find any access points most of the time.

If I lived in Europe I'd have taken a serious look at the
Fujitsu-Siemens
AMD systems but they are not an option in the U.S.... possibly
because of
the Intel rules/incentives (choose one)?? Things *are* changing
though:
until recently, it was difficult to get a WinXP Pro system with an
AMD
notebook.

Actually, as the Japanese authorities said, one Japanese company was
forced to limit marketshare of non-Intel processors to these levels:
90% Intel in Japan, 70% Intel in Europe, and 80% in the rest of the
world. So assuming similar agreements with other Japanese makers and
given that Europe had the lowest marketshare requirements, it must have
been much easier to find AMD notebooks over there.

Yousuf Khan

YKhan wrote:

Robert Myers wrote:
There are no spoils of war to divide up. Opteron didn't open up
any
new vistas. The business is in a period of decline and
consolidation.

Huh? Since when? This is the biggest and most profitable end of
the
semiconductor industry, of course there's spoils of war to divide
up

here -- lots of it. Of course, it's all AMD's to gain and all
Intel's to
lose.


Tell that to Wall Street. Don't waste your time with me.

Don't worry, Wall Street knows it too. You're the only one that isn't
aware of it.

For the past year, INTC has tracked the sector, down about 10% over the
past year. AMD has outperformed the sector, up about 10% over the past
year. If there is reaction to recent news from Japan, I don't see it
in the chart. AMD had quite a nice gain from September to December of
last year, but most of that gain is gone. AMD _is_ selling at about
twice the P/E of INTC, which means there's considerable optimism built
into the current price. Whatever explains that price, I doubt it is
the prospect of AMD suing Intel.

I'm not much of a chart reader, but if it's obvious to Wall Street, it
should be obvious in the chart. Build your own at www.bloomberg.com.

The high tech sector may take decades to get back to where it was
pre-bust. These are _not_ good times.

Even in a downturn, we're talking about tens to hundreds of billions
of
dollars in sales here. Most of it currently going to Intel,
obviously.
Is it a wonder why so many startup companies always try to get into
the
x86 market, despite having to compete with Intel? Some firms (eg.
VIA)
make steady income in this market with less than a 1% marketshare in
it. It's related markets, like chipsets and peripherals are also huge
money makers for companies.


google "etrade AND broker" and go crazy.

There's this thing called efficient market theory. You think you
got
the landscape scoped better than the market, go make yourself rich.

Please do explain this "efficient market theory" of which you speak.
I
guess one of its theorems is "The x86 market is efficient only if
Intel
makes the majority of x86 processors"?


google "efficient market theory" (exact phrase). The first hit gives a
concise and accurate explanation.

Were I AMD, I would use this situation to reposition myself
vis-a-vis
Intel by way of some agreements cut on the sly in some
out-of-the-way
place, not by going to court. That's what I would do. What AMD
does,
of course, is completely independet of my guess as to what they
should
do.

That's also illegal. It also falls into the same anti-trust laws,
it's
collusion. It's no different than if AMD and Intel decided to one
day
start pricing their processors exactly the same as each other at
exactly
the same time -- this would be price-fixing. The other thing would
be
marketshare-fixing.


That's how your mind works, apparently. AMD can sue Intel (not
smart,
in my opinion), or it can come to some kind of agreement about the
rules of engagement. Such a thing could be a consent decree
supervised by a judge. It doesn't have to be public and it doesn't
have to be illegal. AMD and Intel would be colluding to keep, say,
Via out? I'd hope they wouldn't be so stupid.

There is something seriously wrong with your grasp on reality --
that's
how my mind works? Seriously, Robert that's how everybody's mind
works.
There is not a single person who wouldn't describe what you were
suggesting as anything other than collusion. There is no such thing
as
coming to an agreement about rules of engagement between two
companies
-- those rules already exist, they are called the competition laws.
You're supposed to compete with your competitors, freely, fairly, and
to your fullest extent possible.


I have been very slow to learn some things about life. One thing I
have learned is not to challenge how others perceive reality. Another
thing I have learned is that there is no such thing as how "everybody"
thinks.

When you say "there is not a single person..." you are either unaware
of how unlikely it is that such a statement would ever be true or you
are unaware of how unpersuasive bluster is in argumentation.

The circumstances are this: Intel believes its business practices are
legal. AMD believes otherwise. Both are presumably reading the same
law. That means there is a difference in interpretation of the law.
The two companies could agree as to specific interpretation of the law
with respect to specific business practices. _Then_ if Intel continued
with practices it has agreed are illegal, a lawsuit for AMD would be a
cake walk.

Why not just sue Intel now? No guarantee that AMD will prevail, even
with a judgment from Japan. Relief, if any, would be far into the
future. A carefully-drafted agreement, which need not be illegal,
could give AMD the level playing it wants sooner rather than later.

RM

George Macdonald wrote:
On Tue, 15 Mar 2005 05:17:25 -0500, Robert Myers

wrote:

On Mon, 14 Mar 2005 17:02:19 -0500, George Macdonald

snip

I'm sure VIA has found a niche in the developing economies but as
for
China, I'd think their processors would lack the oomph required to
do
Chinese caharacter sets. I've seen this in action and even a
"text"
document drags the CPU down horribly... not sure how it all works
out.

Taiwan and Japan seem to cope somehow. Back to the cost of capital
issue, building a microprocessor industry doesn't seem like a wise
investment for China, except to satisfy their miliary ambitions,
which
they do have.

I was talking about the relative power of current VIA CPUs and their
ability to handle the job... which drags an Athlon XP 2500+ down
pretty
badly. IOW I don't see how VIA satisfies the reqts for even the
basics of
word processing, browsing etc. in Chinese chgaracters.


You know how I hate to be tedious, but the going word is that (say) an
800Mhz P3 is all anybody would ever need. I suspect that VIA will come
up to that standard soon. Are you saying China never will (without
seizing Taiwan, of course) or that an 800Mhz P3 is inadequate to do
Chinese text?

If the latter...I have a hard time imagining how that could be so. I
suspect poor programming, but I wouldn't mind being educated. I mean
*are* there tasks (other than computer games and bad programming) that
an 800 MHz PIII can't currently handle, and is this a real example?

One possible answer is that even Via wouldn't get there without IBM's
help, but I'm not sure I believe that.

RM

On 16 Mar 2005 10:50:05 -0800, "YKhan" wrote:

George Macdonald wrote:
I thought it was the U.S. military that insisted on the 2nd source
for
x86s?

No, I've always heard it was IBM that insisted on it. When IBM was
first making the PC, there really wasn't much of a market for PC's in
the military. The military would've been on mostly mainframes and
terminals at the time. Though there were other personal computers at
the time, they were mostly hobbyist or gaming machines (e.g. Commodore
64).

It's my understanding that the military was *very* interested in using 808x
CPUs, to the extent that some of the licensees, possibly Harris IIRC, also
specialized in making "hardened" versions.

Intel has always been the principal suppiler of chipsets into the
notebook
market almost without exception. How many non-Intel chipsets are in
notebooks with an Intel CPU?

Well, that's mainly because Intel has been driving most chipset
manufacturers out of its market in general, not just in notebooks.
There used to be a time prior to this when you'd see just as many
variations in notebook chipsets as you see in the desktop market, even
on Intel processors.

You'd have to go a long ways back to find any substantial volumes of
notebooks with non-Intel chipsets though... my point being that right now,
Centrino is not making sales of chipsets that would not otherwise be made.

Everything I read indicates that Intel's marketroids see it as a
grand
marketing coup of brand/name recognition... nothing of the sort IMO.

Yeah, there was a study done by AMD to find out if it should also adopt
platformization too. They found that most people really couldn't care
less if the machine was Centrino or not, all they cared about was the
laptop's own brandname. The only subsets that did care about Centrino
were techies, and secondly airport travellers who are bombarded with
Centrino ads in airports.

AMD's "results" don't make sense to me: it's the techies who know that
Centrino doesn't mean anything... maybe self-proclaimed wannabe techies
don't know?:-)

Speaking of airports, I took my laptop to the airport once, and I
couldn't find any access points most of the time.

I haven't tried it so can't say what the situation is... though I hear that
many Euro airports introduced Wi-Fi as a "freeby"... until it caught on.
There are also places where corrective action is/was needed to eliminate
previously non-intrusive, and therefore undetected, abuse of the ISM band.
Apparently if you live near a hospital, fahgeddabatit for Wi-Fi - they fill
up the entire band for a distance up to about 2 city blocks... which
violates the rules but their needs are err, "important".

If I lived in Europe I'd have taken a serious look at the
Fujitsu-Siemens
AMD systems but they are not an option in the U.S.... possibly
because of
the Intel rules/incentives (choose one)?? Things *are* changing
though:
until recently, it was difficult to get a WinXP Pro system with an
AMD
notebook.

Actually, as the Japanese authorities said, one Japanese company was
forced to limit marketshare of non-Intel processors to these levels:
90% Intel in Japan, 70% Intel in Europe, and 80% in the rest of the
world. So assuming similar agreements with other Japanese makers and
given that Europe had the lowest marketshare requirements, it must have
been much easier to find AMD notebooks over there.

I wouldn't call it *much* easier in Europe - one extra supplier in
Fujitsu-Siemens and even there they were playing the system down as a
"home" job.

--
Rgds, George Macdonald

Robert Myers wrote:
Tell that to Wall Street. Don't waste your time with me.

Don't worry, Wall Street knows it too. You're the only one that
isn't
aware of it.

For the past year, INTC has tracked the sector, down about 10% over
the
past year. AMD has outperformed the sector, up about 10% over the
past
year. If there is reaction to recent news from Japan, I don't see it
in the chart. AMD had quite a nice gain from September to December
of
last year, but most of that gain is gone. AMD _is_ selling at about
twice the P/E of INTC, which means there's considerable optimism
built
into the current price. Whatever explains that price, I doubt it is
the prospect of AMD suing Intel.

Well, a large portion of AMD's stock market run up was probably due to
Intel's poor execution all of last year. And then there was a rumour
(started by Dell's management) that Dell might start using AMD. And
later there was the announcement (also by Dell's management) that Dell
wasn't going to use AMD afterall. But really, the stock market has
little to do with the business of any company: the stock market
operates more on human psychology than economics.

I'm not much of a chart reader, but if it's obvious to Wall Street,
it
should be obvious in the chart. Build your own at www.bloomberg.com.

snip

google "etrade AND broker" and go crazy.

snip

google "efficient market theory" (exact phrase). The first hit gives
a
concise and accurate explanation.

I'm starting to see a distinct stock market-centric view in your posts
now. While I'm talking about the actual business of these companies.
Probably explains our inability to figure each other out. As I said as
far as I'm concerned, the stock market isn't any indicator of economics
or business.

BTW, that definition of Efficient Market Theory is down below. It
sounds like a theory in the truest sense of the word, an academic
construct. It's an idealization that stock markets exactly reflect
economics, and economics are reflected in stock markets. Nothing can be
further from the truth here. Stock markets are subject to manipulation
by very rich players who are not interested in transmitting information
efficiently.

Definition: The Efficient Market Theory regards markets as
perfect. Efficient market theory implies that markets are efficient
transmitters of information that affects price. Efficient market theory
would indicate that the stock price reflects all available knowledge.
If the efficient market theory is true, then all kinds of stock
research is futile as the stock price fully "discounts" any information
or conclusion that can be drawn. The efficient market theory could be
extended to decipher the extent of "efficiency" of a market.

The circumstances are this: Intel believes its business practices are
legal. AMD believes otherwise. Both are presumably reading the same
law. That means there is a difference in interpretation of the law.
The two companies could agree as to specific interpretation of the
law
with respect to specific business practices. _Then_ if Intel
continued
with practices it has agreed are illegal, a lawsuit for AMD would be
a
cake walk.

Well of course there's a difference in interpretation of the law. How
often does an accused criminal ever admit that they are guilty? There's
never any end of excuses.

And equally, the accuser strongly believes that they are in the right.
That's why the legal system has judges to sort these interpretations
out.

Why not just sue Intel now? No guarantee that AMD will prevail, even
with a judgment from Japan. Relief, if any, would be far into the
future. A carefully-drafted agreement, which need not be illegal,
could give AMD the level playing it wants sooner rather than later.

Why is there no guarantee that AMD will prevail after a judgement from
Japan? A guilty finding in a criminal case can be used in a civil case
as irrefutable evidence.

In the meantime, Intel cannot make any further contracts within that
jurisdiction that restrict AMD from doing business.

Yousuf Khan

Robert Myers wrote:


You know how I hate to be tedious, but the going word is that (say) an
800Mhz P3 is all anybody would ever need. I suspect that VIA will come
up to that standard soon. Are you saying China never will (without
seizing Taiwan, of course) or that an 800Mhz P3 is inadequate to do
Chinese text?

I know a family of Hong Kong escapees that last year came to
Saskatchewan from Vancouver. They have 6 computers: a
workstation for the father to work on, a good gaming rig for the
kids, and four cheap "homework" boxes for the kids that the
father built himself. Those four have a 512 MB PC133 DIMM and an
866 MHz C3 ( fanless !) plugged into used Socket 370 motherboards
that I gave him.

They find those C3 machines to be quite adequate with Linux for
tasks like word processing, e-mailing, and web browsing - all in
Chinese. For school work the kids also boot into W2K (English).