Intel concedes server business to AMD ;-)

[Del Cecchi]

Rarely. I rarely get interesting files in emails (they're on web sites
also under PHB control). I was thinking more along the lines of strategy
and direction discussions, rather than the mundane bit-banging details.

Those are on notes databases. Detachable too.



del

 

[Robert Myers]

After a few billion$ were spent by others, why don't we flush a few more
of our stock holder's. Money's cheap.

I have no idea what Intel might do as a Plan B. I don't think there
is a Plan B. I think Itanium is it. They would have been better off
with Alpha? I think that subject's been beaten to death. They needed
something.

I guess Intel architects don't read CA? These issues were discussed there
whan Itanic was first announced.

I don't know what to say if you really think usenet posts to comp.arch
are the last word on a subject.

Itanic may have been on the launching pad before people discussed this,
but these people discussed these problems as soon as they saw the monster.
Apparently you don't think Intel's architects are as sharp as those who
publicly post to CA.

I don't know what you mean. NetBurst was the same flavor of wishful
thinking as Itanium: we'll figure out a way to get the compiler to
schedule it, even if the processor has the agility of an oil tanker.

Why does "made by IBM" make it bad? A proprietary architecture is a
proprietary architecture (Itanic much more so than PowerPC, in fact).
Why would one take applications from a more or less open architecture
(x86) to a closed one? Why would you trust your enterprise to Intel as
opposed to IBM? ...particularly when one has a tad more experience in the
market.

*Somebody* is going to be using *some* architecture not proprietary to
IBM to make high-end boxes. It doesn't matter whether I think it's
good or bad. It's going to happen. If Itanium dies, it will be
something else.

That was certainly Intel's plan. Too bad it was fatally flawed by a
turkey of an architecture. Intel needed a bust-out architecture to pull
anything like the off, but chose one that needed more than a little
invention than they could handle. Meanwhile everyone else scaled up their
performance. ...sorta what Intel (with x86) did to RISC.

Intel was not alone. Elbrus was going to conquer the world with its
VLIW architecture. I think Intel bought whatever was left of Elbrus.

Meanwhile, AMD swiped Intel's lunch money. Intel doesn't show pretty
poor signs of catching up.

For all that, Intel's feathers don't seem to be too ruffled. If it's
all a bluff, they bluff very effectively.

RM

 

[keith]

I have no idea what Intel might do as a Plan B. I don't think there
is a Plan B.

I agree (amazing, eh? . AMD's caught them flat-footed with no Plan-B.
Dumb! Arrogance!

I think Itanium is it. They would have been better off
with Alpha?

I certainly think so. ...but they didn't own Alpha, so that's a moot point.

I think that subject's been beaten to death. They needed something.

And needlessly pissed away their leadership on a dog, with *no* Plan-B.
Incredibly stupid. This is exactly what I've been saying for several
years. It's good to have you aboard! ;-)

I don't know what to say if you really think usenet posts to comp.arch
are the last word on a subject.

If you think the people who poat there are stupid, ignore them at your own
perill. When there is a rotten fish around, chances are someone
who's fished before will smell it.

I don't know what you mean. NetBurst was the same flavor of wishful
thinking as Itanium: we'll figure out a way to get the compiler to
schedule it, even if the processor has the agility of an oil tanker.

Perhaps you do know what I mean, after all. ;-)

*Somebody* is going to be using *some* architecture not proprietary to
IBM to make high-end boxes. It doesn't matter whether I think it's good
or bad. It's going to happen. If Itanium dies, it will be something
else.

Well, HPaq killing Alpha, and Intel screwing the pooch on Itanic... ;-)

I certainly *hope* there will be something else. Mono-cultures are boring.

Intel was not alone. Elbrus was going to conquer the world with its
VLIW architecture. I think Intel bought whatever was left of Elbrus.

Whoopie! ...and it got them???

For all that, Intel's feathers don't seem to be too ruffled. If it's
all a bluff, they bluff very effectively.

Intel's feathers are certainly ruffled. The biz is in the dumps, so how
this plays is anyone's guess, but Intel has *nothing* to answer AMD with.
Indeed I'm shocked they didn't figure it out *long* before now. Instead
of trying to come up with a proprietary memory system (or whatever), I
would have thought a few bux spent in extending the architecture they'd
all but locked up would have been prudent. I guess the eight-figure club
doesn't think the same way simple engineers do.

 

[Yousuf Khan]

For all that, Intel's feathers don't seem to be too ruffled. If it's
all a bluff, they bluff very effectively.

Mostly their feathers don't look ruffled because they've been able to
maintain their illegal incentives, which has kept manufacturers from
even trying out the competition.

Yousuf Khan

 

[rbmyersusa]

I agree (amazing, eh? . AMD's caught them flat-footed with no Plan-B.
Dumb! Arrogance!


I certainly think so. ...but they didn't own Alpha, so that's a moot point.

I guess I don't understand what might have been possible between Intel
and HP, but it's all moot because Intel wasn't going to go there, no
matter who owned Alpha.

something.

And needlessly pissed away their leadership on a dog, with *no* Plan-B.
Incredibly stupid. This is exactly what I've been saying for several
years. It's good to have you aboard! ;-)

This post in groups.google brings up a google ad for Itanium servers.
In practice, Keith, I've suckered you into helping keep the itanium
name in play. ;-)

If you think the people who poat there are stupid, ignore them at your own
perill. When there is a rotten fish around, chances are someone
who's fished before will smell it.

Any appearance to the contrary, I'm slow to think anyone stupid. I'm
often skeptical of the conventional wisdom, and c.a is heavy on
conventional wisdom. The conventional wisdom was that Intel was
betting long odds. That's not news to me, and, yes, it's easy to find
people to say I told you so.

tanker.

Perhaps you do know what I mean, after all. ;-)

For all that, NetBurst (and Itanium) can pull good numbers with a
properly-tweaked Intel compiler.

Well, HPaq killing Alpha, and Intel screwing the pooch on Itanic... ;-)

I certainly *hope* there will be something else. Mono-cultures are boring.

Ermph. Nobody is too impressed with my posts about google. The
mono-culture could be even more single-channel than anybody has dreamed
about since the 1956 anti-trust consent decree, only it won't be IBM,
Intel, or Microsoft that owns everything.

Even if you don't believe my predictions about everything becoming
centralized with thin clients, and even if you don't think it will be
google, the issue that is google's issue (power consumption) is going
to be everybody's issue, but I think you already know that.

Elbrus.

Whoopie! ...and it got them???

Since they made that acquisition late in the game, it does show that
they are determined.

Intel's feathers are certainly ruffled. The biz is in the dumps, so how
this plays is anyone's guess, but Intel has *nothing* to answer AMD with.
Indeed I'm shocked they didn't figure it out *long* before now. Instead
of trying to come up with a proprietary memory system (or whatever), I
would have thought a few bux spent in extending the architecture they'd
all but locked up would have been prudent. I guess the eight-figure club
doesn't think the same way simple engineers do.

By proprietary memory system, you mean not going with hypertransport,
which as we *all* know, isn't a memory interconnect. That was a safe
bet. I wonder who has the title of Czar of Anticompetitive Strategies
at Intel?

RM

 

[Basil Evrenidis]

So is this the

"No one ever got fired for going IBM line of thought"?

or the

"No one ever got fired for going Microsoft line of thought"?

ooops, they're the same thing. The type of response you'd expect from a
manager who has switched their brain to off.

 

[rbmyersusa]

No, it's the AMD can't possibly dominate the industry until it has more
capacity line of thought.

Those who bought IBM products can still get support from IBM, no matter
how old the machine. Got some dusty old deck that needs some ancient
compiler that ran on System 360? They still got it, 'cause they're
still here to have it. Who knows what the future may hold for AMD, but
Intel isn't going away.

I wouldn't count on Microsoft for yesterday's email, and I don't.

RM

 

[keith]

No, it's the AMD can't possibly dominate the industry until it has more
capacity line of thought.

Those who bought IBM products can still get support from IBM, no matter
how old the machine. Got some dusty old deck that needs some ancient
compiler that ran on System 360? They still got it,

Do you *really* believe that? Products are sunset all the time, with no
service.

'cause they're still here to have it. Who knows what the future may hold for AMD, but
Intel isn't going away.

If the hardware is compatable, does it matter?

I wouldn't count on Microsoft for yesterday's email, and I don't.

I wouldn't count on M$ with today's, but that's how they designed the
Win-World. ...somehow a lot of people do just that though.

 

[Robert Myers]

Do you *really* believe that? Products are sunset all the time, with no
service.

Oh, I worded my post to make it sound like hardware was supported
forever. What I meant to say was that software doesn't get stranded.
I've heard of cases where software has been stranded, but I haven't
heard of many.

If the hardware is compatable, does it matter?

What does "compatible" mean, and how do you test for it?

RM

 

[Del Cecchi]

Oh, I worded my post to make it sound like hardware was supported
forever. What I meant to say was that software doesn't get stranded.
I've heard of cases where software has been stranded, but I haven't
heard of many.

What does "compatible" mean, and how do you test for it?

RM

That brings up an interesting question. Say I had a copy of PCdos or MSdos,
and a copy of some software, say PCwrite or ezriter or something of vintage
1982 along with some files created with said software could I boot that OS
and run that software and edit those files (pretending I have somehow
attached a 5.25 floppy instead of the 3.5 thus having no media problem) on a
modern PC, circa the XP era?

Just curious.

del

 

[YKhan]

That brings up an interesting question. Say I had a copy of PCdos or MSdos,
and a copy of some software, say PCwrite or ezriter or something of vintage
1982 along with some files created with said software could I boot that OS
and run that software and edit those files (pretending I have somehow
attached a 5.25 floppy instead of the 3.5 thus having no media problem) on a
modern PC, circa the XP era?

Just curious.

I certainly wouldn't want to test that out. Even in the days when that
stuff was current, it was hit and miss about compatibility.

Even though all of the same instructions exist in the instruction set,
lord knows if for some reason there might be a timing loop in there
dependent on an instruction that now goes several orders of magnitude
faster.

Backward compatibility probably should be attempted beyond a few years
old.

Yousuf Khan

 

[Robert Myers]

That brings up an interesting question. Say I had a copy of PCdos or MSdos,
and a copy of some software, say PCwrite or ezriter or something of vintage
1982 along with some files created with said software could I boot that OS
and run that software and edit those files (pretending I have somehow
attached a 5.25 floppy instead of the 3.5 thus having no media problem) on a
modern PC, circa the XP era?

Just curious.

Well, you would ask.

My ancient DOS word processor will *not* run under the command prompt
under Windows XP because it claims there are insufficient FILES in the
config.sys file.

Now, Windows XP doesn't use config.sys, and that information should be
in \windows\system32\config.nt, but increasing the FILES there to a
level that should be much higher than necessary doesn't fix it.

Who knows. Maybe somebody here does. Fortunately, I still have an
XT-compatible that runs actual DOS.

RM

 

[keith]

Oh, I worded my post to make it sound like hardware was supported
forever. What I meant to say was that software doesn't get stranded.
I've heard of cases where software has been stranded, but I haven't
heard of many.

It happens all the time. It costs money to support products, even if
the bits are free.

What does "compatible" mean, and how do you test for it?

Are you saying that an AMD processor is somehow incompatible with an
Intel? ...that switching from one to another is a huge deal? COme on,
you can do better.

 

[Robert Myers]

It happens all the time. It costs money to support products, even if
the bits are free.

If you got the money, Keith, IBM got the time (and, in most cases, the
archival material).

Are you saying that an AMD processor is somehow incompatible with an
Intel? ...that switching from one to another is a huge deal? COme on,
you can do better.

Because you are so obsessed with the performance of hardware, you
assume that everyone else is, too. They're not. The software is more
important than the hardware, the people who run the software are more
important than the software, and the business processes they support
are more important than they are. An AMD processor is not an Intel
processor. Period. "Compatible" is a sucker word.

RM

 

[Rob Stow]

If you got the money, Keith, IBM got the time (and, in most cases, the
archival material).




Because you are so obsessed with the performance of hardware, you
assume that everyone else is, too. They're not. The software is more
important than the hardware, the people who run the software are more
important than the software, and the business processes they support
are more important than they are. An AMD processor is not an Intel
processor. Period. "Compatible" is a sucker word.

All that and you still didn't answer his simple question:
"Are you saying that an AMD processor is somehow incompatible
with an Intel?"

 

[Robert Myers]

All that and you still didn't answer his simple question:
"Are you saying that an AMD processor is somehow incompatible
with an Intel?"

An AMD processor is incompatible with an Intel processor because it
doesn't say Intel on the package and Intel will tell you to pound sand
if you have questions about it.

Does that work for you? It works for enough people with money to
spend and other things to think about.

RM

 

[Rob Stow]

An AMD processor is incompatible with an Intel processor because it
doesn't say Intel on the package and Intel will tell you to pound sand
if you have questions about it.

Does that work for you?

All it does is show that you know you don't have a leg to stand
on w.r.t. this compatibility issue and you are dodging the
question instead of giving a straight answer.


It works for enough people with money to

 

[rbmyersusa]

I use icc, and I use Intel chips. Life is too short to be fussing with
other chips without good reason. Compatible? Like I said, how do you
test for it?

RM

 

[George Macdonald]

I use icc, and I use Intel chips. Life is too short to be fussing with
other chips without good reason. Compatible? Like I said, how do you
test for it?

That old bait is getting really stale and smelly Robert. I was sure you
could do better.

 

[Del Cecchi]

I use icc, and I use Intel chips. Life is too short to be fussing with
other chips without good reason. Compatible? Like I said, how do you
test for it?

RM

Same way you test a microsoft service pack before rolling it out company
wide?

or for a serious answer you set up some test hardware and beat on it with
your business critical applications to see if it works or not.

del