Might be a book that even R. Myers can love :-)

[Yousuf Khan]

Jim Carlson and Jerry Huck's "Itanium Rising" book as described in the
following article:

http://www.shannonknowshpc.com/stories.php?story=04/06/14/7665801

Yousuf Khan

 

[Robert Myers]

Jim Carlson and Jerry Huck's "Itanium Rising" book as described in the
following article:

http://www.shannonknowshpc.com/stories.php?story=04/06/14/7665801

I made a post to comp.arch about this book with the subject line
"Stupefying hubris from Intel/HP about Itanium" to comp.arch on March 1,
2003. Del Cecchi had previously pointed out the existence of the book
on October 2, 2002, but it's not clear that anyone but me read (well at
least skimmed) it. In any case, I was the only one to say anything
subtantive about the book's contents. If I weren't so interested in
self-description (what people and companies say about themselves and
why), I would have been tremendously annoyed at the book.

Even leaving aside some reasonably subtle technological questions (some
of which have been discussed on comp.arch), _Itanium_Rising_ can't tell
the really interesting story because its authors would probably get
fired for even trying to tell it. Aside from building a processor with
an ISA that wouldn't be subject to any of Intel's cross-licensing
agreements, what did the principals in this drama really think they were
buying into? Even the history of the internal presentations that were
made would be fascinating. What did they think they knew, and when did
they think they knew it? ;-)

RM

 

[Yousuf Khan]

I made a post to comp.arch about this book with the subject line
"Stupefying hubris from Intel/HP about Itanium" to comp.arch on March
1, 2003. Del Cecchi had previously pointed out the existence of the
book on October 2, 2002, but it's not clear that anyone but me read
(well at least skimmed) it. In any case, I was the only one to say
anything subtantive about the book's contents. If I weren't so
interested in self-description (what people and companies say about
themselves and why), I would have been tremendously annoyed at the
book.

Okay, then maybe R. Myers might not like this book.

Yousuf Khan

 

[Robert Myers]

Okay, then maybe R. Myers might not like this book.

But I am intensely interested in self-description, I'm interested in
what makes grand technical initiatives succeed or fail, and I think it's
more interesting to say what you think is going on before all the horses
have crossed the finish line than it is to be a smug historian. .

Like the book or not like is more or less beside the point. I was
stunned that Intel/HP let such a daringly unrepentant bit of
self-promotion see daylight in view of the disappointing performance of
Itanium, but even that (fairly reasonable, I think) reaction is a
distraction. The fact is that the book _was_ published, and without
corporate gloss or apologia, as far as I know: just, here it is, the
most wonderful processor ever, just as we said it would be.

Leave out the technical issues. If Intel/HP have to climb down from the
fortress they've built around Itanium, how will they ever pull it off?
It would be like IBM admitting that maybe System 360 wasn't such a great
idea, after all (which, who knows, maybe it wasn't).

RM

 

[George Macdonald]

Leave out the technical issues. If Intel/HP have to climb down from the
fortress they've built around Itanium, how will they ever pull it off?
It would be like IBM admitting that maybe System 360 wasn't such a great
idea, after all (which, who knows, maybe it wasn't).

Wasn't it "Stretch" which wasn't a good idea... preceding S/360. I guess
there was a lot of folklore back then too.

Rgds, George Macdonald

"Just because they're paranoid doesn't mean you're not psychotic" - Who, me??

 

[Robert Redelmeier]

Leave out the technical issues. If Intel/HP have to climb down from the
fortress they've built around Itanium, how will they ever pull it off?
It would be like IBM admitting that maybe System 360 wasn't such a great
idea, after all (which, who knows, maybe it wasn't).

Whatever one thinks about the technical merits of S/360,
the commercial success was undeniable. The same could be
said of x86.

I very much doubt that IA64 (Itanium) will ever enjoy such
commercial success. It's more likely to go down that path
heavily travelled by Intel after 432 and i860.

-- Robert

 

[Robert Myers]

Wasn't it "Stretch" which wasn't a good idea... preceding S/360. I guess
there was a lot of folklore back then too.

Stretch just precedes my actually becoming conscious of computers in any
but the most theoretical of ways, so I know only the written record--no
folklore. Branch prediction, speculation, out of order execution,
hardware prefetch, and fused FMAC: less than two hundred thousand
transistors. What's not to like? .

RM

 

[Robert Myers]

Whatever one thinks about the technical merits of S/360,
the commercial success was undeniable. The same could be
said of x86.

I very much doubt that IA64 (Itanium) will ever enjoy such
commercial success. It's more likely to go down that path
heavily travelled by Intel after 432 and i860.

You may be right. I'm just having a really hard time imagining how this
is going to go down. Intel's goal is to 360-ize as much enterprise code
as it can (only for IA64, obviously). How well they are doing that, how
well they can do that, is something that I just cannot judge, although
it is apparent things are not going according to plan at the moment.

George Macdonald said something about his own experience with Alpha (and
by extension, with Itanium) that sounded absolutely pivotal: software
developers don't want to develop for a platform that isn't going to make
them money.

That's easy, you're thinking: pass on Itanium. Not so fast, buckaroo.
One future: x86-64=Open source, low rent, lots of volume, no margin.
Itanium=Proprietary, expensive, low volume, high margin. If you look at
the Intel branding ads aimed at corporate decision-makers, Xeon is
"productivity" and Itanium is "enterprise." Who wants to be
"productivity" when one could be "enterprise?" Especially if the "who"
is a manager, to whom "productivity" is something that is delivered by
nameless underlings. .

They'll never get away with it, you're sputtering, and you may be right,
but Intel shows no signs of abandoning its strategy.

RM

 

[Robert Redelmeier]

One future: x86-64=Open source, low rent, lots of volume, no margin.
Itanium=Proprietary, expensive, low volume, high margin.

Both entirely true.

If you look at the Intel branding ads aimed at corporate
decision-makers, Xeon is "productivity" and Itanium is "enterprise."
Who wants to be "productivity" when one could be "enterprise?"

I doubt even the dinosaur brains will swallow that swill.

Two words: "second source". No-one wants to be dependant on a
single supplier. PC vs Mac. S/360 succeeded mostly by offering
a uniform platform with a promise of continuation (backward
compatibility) that attracted development.

x86/WinNT (maybe Linux) holds that position now. What compelling
argument is available for IA64? What performance? 64 bits is
available ~painlessly with x86-64.

They'll never get away with it, you're sputtering, and you may
be right, but Intel shows no signs of abandoning its strategy.

I don't sputter. I know the market will decide. Mistakes get
punished, and stubborn fools commensurately.

-- Robert

 

[daytripper]

Whatever one thinks about the technical merits of S/360,
the commercial success was undeniable.

I think the technical merits were right up there as well.
What other system had a control store that required an air-pump to operate?;-)

/daytripper

 

[George Macdonald]

Stretch just precedes my actually becoming conscious of computers in any
but the most theoretical of ways, so I know only the written record--no
folklore. Branch prediction, speculation, out of order execution,
hardware prefetch, and fused FMAC: less than two hundred thousand
transistors. What's not to like? .

There was a famous encounter at an IBM conference where a question was
asked about Stretch and Watson Jr(?) himself was reputed to have denied its
existence: no such thing ever existed.

There's another story, from about that time, that IBM was so shocked to
learn that the CDC 6600 -- a perceived competitor at the time -- had been
conceived, designed, engineered and built in the space of 18months that
they shut their best designers/engineers up in a hotel for several weeks
and told them not to come out until they had a paper design... which became
the S/360.

Rgds, George Macdonald

"Just because they're paranoid doesn't mean you're not psychotic" - Who, me??

 

[Robert Myers]

x86/WinNT (maybe Linux) holds that position now. What compelling
argument is available for IA64? What performance? 64 bits is
available ~painlessly with x86-64.

The technical merits may not be all that important one way or the other.
If enough software developers feel they can forego Itanium, Itanium
will die. If a software developer decides to forgo Itanium and loses
enough important clients in the enterprise space because of it, the
software developer may be shut out of the most lucrative part of the market.

I don't sputter.

No offense intended, I assure you.

RM

 

[Felger Carbon]

IIRC the Cyber6600 came significantly after April '64 too.

According to the first sentence of Chapt 43 of Sieworek, Bell, and
Newell's "Computer Structures: Principles and Examples", the first
6600 was delivered in Oct 64. The 6600 project was begun in the
summer of 1960.

 

[K Williams]

There was a famous encounter at an IBM conference where a question
was asked about Stretch and Watson Jr(?) himself was reputed to
have denied its existence: no such thing ever existed.

That I can buy. There are many things done that one cannot talk
about. ...even a CEO.

There's another story, from about that time, that IBM was so
shocked to learn that the CDC 6600 -- a perceived competitor at
the time -- had been conceived, designed, engineered and built in
the space of 18months that they shut their best
designers/engineers up in a hotel for several weeks and told them
not to come out until they had a paper design... which became the
S/360.

I don't buy that one. ...sounds like cyber-legend to me. CDC and
IBM (and their customer set) had little to do with each other.
IIRC the Cyber6600 came significantly after April '64 too. It
*was* used for Plato in the '70s though.

 

[K Williams]

Robert Restampedesswrote:

Whatever one thinks about the technical merits of S/360,
the commercial success was undeniable. The same could be
said of x86.

Undeniable, indeed. That's *MY* basis for supporting K8.

IBM tried to kill S/360 with 'FS', just as Intel has tried to kill
x86 with Itanic. Both companies did it for internal reasons,
totally disregarding customer's wishes. "Amazingly", neither went
over well with the customer set. Both 3x0 and x86 are still with
us (after forty and twenty years, respectively). ...evolution over
revolution.

I very much doubt that IA64 (Itanium) will ever enjoy such
commercial success. It's more likely to go down that path
heavily travelled by Intel after 432 and i860.

....as AMD64 stampedes the mess Intel's tried to create. History
does repeat. Those who try to harness history will make money.
Those who try to harness their customers deserve an ugly death.

 

[George Macdonald]

I don't buy that one. ...sounds like cyber-legend to me. CDC and
IBM (and their customer set) had little to do with each other.
IIRC the Cyber6600 came significantly after April '64 too. It
*was* used for Plato in the '70s though.

I guess it's likely folklore but I know that when the 7074 was to be
replaced in a certain office of a multinational corp in 1967, the S/360 was
the obvious and natural replacement for the DP side of things; OTOH there
was serious consideration given to Univac 1108 or CDC 6600 for technical &
scientific work, which had often been done on a 7094... and often at
extortionate time-lease terms. IOW it wasn't clear that the S/360 could
hack it for the latter - turned out that it was dreadfully slow but near
tolerable... if people worked late:-( and got much better later. Certainly
the performance of S/360 fell way short of expected performance as "sold" -
I can bore you with the details if you wish.

The CDC 6000 Series didn't become Cyber Series till ~1972[hazy]; before
that there was 6200, 6400, 6500 and 6600... and there was the notorious
7600 in between. Dates of working hardware are difficult to pin down -
supposedly secret confidential data often went astray and announced
availability and deliverable were umm, fungible. The story is probably a
bit folklorish but no doubt that IBM was seriously threatened by Univac and
CDC in the technical computing arena.

Rgds, George Macdonald

"Just because they're paranoid doesn't mean you're not psychotic" - Who, me??

 

[Robert Myers]

I guess it's likely folklore but I know that when the 7074 was to be
replaced in a certain office of a multinational corp in 1967, the S/360 was
the obvious and natural replacement for the DP side of things; OTOH there
was serious consideration given to Univac 1108 or CDC 6600 for technical &
scientific work, which had often been done on a 7094... and often at
extortionate time-lease terms. IOW it wasn't clear that the S/360 could
hack it for the latter - turned out that it was dreadfully slow but near
tolerable... if people worked late:-( and got much better later. Certainly
the performance of S/360 fell way short of expected performance as "sold" -
I can bore you with the details if you wish.

The CDC 6000 Series didn't become Cyber Series till ~1972[hazy]; before
that there was 6200, 6400, 6500 and 6600... and there was the notorious
7600 in between. Dates of working hardware are difficult to pin down -
supposedly secret confidential data often went astray and announced
availability and deliverable were umm, fungible. The story is probably a
bit folklorish but no doubt that IBM was seriously threatened by Univac and
CDC in the technical computing arena.

Threatened? . The outlines of the folklore you report is the
folklore I started my career with: CDC (later Cray) for hydro codes, IBM
for W-2 forms. Lynn Wheeler's posts to comp.arch have helped me to
understand how it was that IBM sold machines at all, because, as far as
I could tell, they were expensive and slow, JCL was descended from some
language used in Mordor, and the batch system was designed for people
who knew ahead of time what resources a job would need (that is to say,
it was designed for people counting W-2 forms and not for people doing
research). My impression of IBM sofware was fixed by my early
experience with the Scientific Subroutine Package, and even the
compilers were buggy for the kinds of things I wanted to use--no problem
for financial applications, where there was (as yet) no requirement for
double precision complex arithmetic.

One is tempted to summarize the Stretch/360 experiences as: "How IBM
learned to love banks and to hate the bomb." In retrospect, IBM's
misadventure with Stretch might be regarded as a stroke of luck. An
analyst too close to the action might have regarded IBM's being pushed
out of technical computing in the days of the Space Race as a distaster,
but the heady days of "If it's technical, it must be worth doing" were
actually over, and IBM was in the more lucrative line of business, anyway.

RM

 

[George Macdonald]

I guess it's likely folklore but I know that when the 7074 was to be
replaced in a certain office of a multinational corp in 1967, the S/360 was
the obvious and natural replacement for the DP side of things; OTOH there
was serious consideration given to Univac 1108 or CDC 6600 for technical &
scientific work, which had often been done on a 7094... and often at
extortionate time-lease terms. IOW it wasn't clear that the S/360 could
hack it for the latter - turned out that it was dreadfully slow but near
tolerable... if people worked late:-( and got much better later. Certainly
the performance of S/360 fell way short of expected performance as "sold" -
I can bore you with the details if you wish.

The CDC 6000 Series didn't become Cyber Series till ~1972[hazy]; before
that there was 6200, 6400, 6500 and 6600... and there was the notorious
7600 in between. Dates of working hardware are difficult to pin down -
supposedly secret confidential data often went astray and announced
availability and deliverable were umm, fungible. The story is probably a
bit folklorish but no doubt that IBM was seriously threatened by Univac and
CDC in the technical computing arena.

Threatened? . The outlines of the folklore you report is the
folklore I started my career with: CDC (later Cray) for hydro codes, IBM
for W-2 forms. Lynn Wheeler's posts to comp.arch have helped me to
understand how it was that IBM sold machines at all, because, as far as
I could tell, they were expensive and slow, JCL was descended from some
language used in Mordor, and the batch system was designed for people
who knew ahead of time what resources a job would need (that is to say,
it was designed for people counting W-2 forms and not for people doing
research). My impression of IBM sofware was fixed by my early
experience with the Scientific Subroutine Package, and even the
compilers were buggy for the kinds of things I wanted to use--no problem
for financial applications, where there was (as yet) no requirement for
double precision complex arithmetic.

I remember, coming from working with S/360, getting my eyes opened when I
first saw a 6600 "installation" - terminals<gawp> (actually they called
them VDUs or something like that), in client cubicles, connected by wires
to the computer on a different floor... where you could actually page
through files. Clients got charged a bundle to use them mind you. I
recall saying to my colleagues at the time: "hey maybe one of those days
we'll all have one of those err, VDU thingys on every desk and we'll
program straight into the file on the computer and look at the results
there too - no more coding forms, punch cards or listings etc." They all
laughed like hell.

As for JCL, I once had a JCL evangelist explain to me how he could use JCL
in ways which wasn't possible on systems with simpler control statements -
conditional job steps, subsitution of actual file names for dummy
parameters etc... "catalogued procedures"?[hazy again] The guy was stuck
in his niche of "job steps" where data used to be massaged from one set of
tapes to another and then on in another step to be remassaged into some
other record format for storing on another set of tape... all those steps
being necessary, essentially because of the sequential tape storage. We'd
had disks for a while but all they did was emulate what they used to do
with tapes - he just didn't get it.

One is tempted to summarize the Stretch/360 experiences as: "How IBM
learned to love banks and to hate the bomb." In retrospect, IBM's
misadventure with Stretch might be regarded as a stroke of luck. An
analyst too close to the action might have regarded IBM's being pushed
out of technical computing in the days of the Space Race as a distaster,
but the heady days of "If it's technical, it must be worth doing" were
actually over, and IBM was in the more lucrative line of business, anyway.

So much for analysts - plus ça change....

Rgds, George Macdonald

"Just because they're paranoid doesn't mean you're not psychotic" - Who, me??

 

[Felger Carbon]

Ok, answer this question: Where is the money?

...even John Dillinger knew the answer! ;-)

Willie Sutton. Not John Dillinger. Gotta get yer Ne'er-do-well's
right. ;-)

 

[Robert Myers]

Robert Myers wrote:




Ok, answer this question: Where is the money?

...even John Dillinger knew the answer! ;-)

It is the style of business and not the plentiful supply of money that
makes banks and insurance companies attractive as clients for IBM.
Under the right circumstance, money can pour from the heavens for
national security applications, and it will pour from the heavens for
biotechnology and entertainment. Whatever you may think of that kind of
business, IBM wants a piece of it.

From a technical standpoint, there is no company I know of better
positioned than IBM to dominate high performance computing, the future
of which is not x86 (and not Itanium, either). Will IBM do it? If the
past is any guide, IBM will be skunked again, but there is always a
first time.

RM