Itanium finally passes Alpha at HP

[Yousuf Khan]

Itanium sales have finally surpassed Alpha sales at HP. Looks like it's
mostly in the OpenVMS market though. Most OpenVMS customers are entrenching
around Itanium now. The Alpha-Tru64 market still seems to be volatile.

http://www.techweb.com/wire/story/TWB20040820S0005

Yousuf Khan

 

[E.S.]

Itanium sales have finally surpassed Alpha sales at HP.

Where did you find that in the article you quoted ?

 

[Alan Walpool]

E> Where did you find that in the article you quoted ?

Looks like someone posted before reading the article. It appears the
only thing the article discusses is an HP claim the performance of
openvms on itanium has surpassed the alpha. HP would love to kill the
alpha line, and it probably would save them some money on alpha
development. The alpha was dead when compaq brought it. Nice systems
but bound to fail eventually for many different reasons.

I have an old alpha system. Very nice computer, but now it is a
collectors item. You can tell by the low prices used alpha systems
fetch on e-bay that the alpha is pretty much history. With 64 bit
coming online in the commodity market it is just a matter of time. Sun
is trying to make the switch before it is too late.

Whatever. 64 bit has finally arrived!!! It goes to show that being
first does not mean you have a winner. The marketing of computers is
not like the Olympics ;-)).

Later,

Alan

 

[Nick Maclaren]

Looks like someone posted before reading the article. It appears the
only thing the article discusses is an HP claim the performance of
openvms on itanium has surpassed the alpha. HP would love to kill the
alpha line, and it probably would save them some money on alpha
development. The alpha was dead when compaq brought it. Nice systems
but bound to fail eventually for many different reasons.

All things are transitory. But it is as false to say that the Alpha
was dead when Compaq bought it as it is to say that Compaq killed
the most successful RISC architecture. The situation was that it
would have needed a massive change in approach to stop it fading,
but it is quite possible that would have changed it from a small
player to an x86 replacement. And it is possible that it would
have sunk even faster if that were attempted. We shall now never
know.


Regards,
Nick Maclaren.

 

[Yousuf Khan]

Where did you find that in the article you quoted ?

Oops, you're right upon further rereading, I find they were talking about
surpassing the performance of, not the sales of. Oh well, my bad.

 

[John Savard]

You can tell by the low prices used alpha systems
fetch on e-bay that the alpha is pretty much history.

Pity I live in Canada and not the United States. It'll be awkward for me
to take advantage of one of those bargains - I'll have to wait till
someone in Canada wants to get rid of his Alpha.

John Savard
http://home.ecn.ab.ca/~jsavard/index.html

 

[John Savard]

Whatever. 64 bit has finally arrived!!! It goes to show that being
first does not mean you have a winner. The marketing of computers is
not like the Olympics ;-)).

Well, the Itanium is also a really big chip with low yields.

64-bit has finally arrived, since AMD *was* first with something - first
at making it affordable.

John Savard
http://home.ecn.ab.ca/~jsavard/index.html

 

[Rob Stow]

Pity I live in Canada and not the United States. It'll be awkward for me
to take advantage of one of those bargains - I'll have to wait till
someone in Canada wants to get rid of his Alpha.

John Savard
http://home.ecn.ab.ca/~jsavard/index.html

1.) The Alpha servers/workstations available on E-Bay are
seldom with processors faster than 233 or 266 MHz.
In other words, only the really ancient stuff is being
sold on E-Bay - so its no surprise that the prices are
low.

Components for upgrading more modern Alpha servers, such
as 1 and 2 GB Memory upgrades for Alpha servers, by contrast
are selling for big bucks. People are willing to pay big
premiums to keep there Alpha servers are alive and well -
hardly a sign that the Alpha is history.

2.) Check the "for sale" newsgroups for your province or city.
Even in Saskatchewan (sk.forsale) we occasionally get local
sales of Alpha systems comparable to what is available on
E-Bay. However, those too are 233 and 266 MHz systems
almost all the time.

3.) What's wrong with using E-Bay but just limiting your search
to Canadian sellers ?

 

[Bernd Paysan]

64-bit has finally arrived, since AMD *was* first with something -
first at making it affordable.

And I thought that was Nintendo (sorry, couldn't resist ;-).

 

[Yousuf Khan]

Pity I live in Canada and not the United States. It'll be awkward for
me to take advantage of one of those bargains - I'll have to wait till
someone in Canada wants to get rid of his Alpha.

One of the projects I was on last year was busy buying new Alphas too.

Yousuf Khan

 

[Alan Walpool]

John> On Sat, 21 Aug 2004 23:46:18 -0500, Alan Walpool

John> Well, the Itanium is also a really big chip with low yields.

John> 64-bit has finally arrived, since AMD *was* first with
John> something - first at making it affordable.

Good point AMD 64 bit was first when it comes to price, and is still
the low price leader at the current time. Looks like the intel 64 bit
x86 processor is not going to compete with the AMD low end 64 bit
processors when it comes to price.

Sorry but forgot to mention the 64 bit powerpc that has been around
awhile. Powerpc is probably cheap but Mac is not cheap. To be complete
Sun has 64 bit also but moving to AMD 64 bit. Itanium is in there
somewhere.

I still like alpha good to hear there is still a demand for this
processor.

Curious - are there any other 64 bit processors that are still on the
market and being actively used?

Later,

Alan

 

[John Mashey]

Curious - are there any other 64 bit processors that are still on the
market and being actively used?

1) In 1Q92, SGI shipped the Crimson product, which used a 64-bit MIPS
R4000, albeit with 32-bit software. DEC shipped Alphas in systems
later that year, and {HP, Sun, IBM, in some order or other} over the
next few years.

2) R4x00s were shipping in Nintendo N64s around 1996.

3) There are of course lots of this family still around, and still
shipping ... in particular, folks like PMC-Sierra and Broadcom &
others sell them ...
and they're been in most CISCO routers for a long time, as well as
lots of laser printers, set-top boxes, and other products. See
http://www.mips.com/content/Ecosystem/Licensees/ProductCatalog/licensees
for companies with current licenses (not all of which use 64-bit, but
many), but many of these licensees sell the chips to other companies.

4) With all due respect to general-purpose computing, I suspect that
most of the world's 64-bit micros are *not* in general-purpose
computers. It is somewhat ironic that 64-bit micros have been there
in HPC and *embedded* for over a decade, and have taken soooo long to
get into the desktop and mid-range [and of course, are being viewed by
the press as "something new".] AMD did a very nice job extending
32-bit X86 to 64-bit, almost exactly analogous to what {MIPS, HP, IBM,
Sun} did. Of course, 64-bit super-computers have been around ~30
years.

5) Toshiba's sampling a TX//99-core based one for $45 apiece in 100s,
and I think NEC VR4133s are <$30. There may be cheaper 64-bitters out
there [I just haven't looked lately.]

 

[Alan Walpool]

John> 4) With all due respect to general-purpose computing, I suspect
John> that most of the world's 64-bit micros are *not* in
John> general-purpose computers. It is somewhat ironic that 64-bit
John> micros have been there in HPC and *embedded* for over a decade,
John> and have taken soooo long to get into the desktop and mid-range
John> [and of course, are being viewed by the press as "something
John> new".] AMD did a very nice job extending 32-bit X86 to
John> 64-bit, almost exactly analogous to what {MIPS, HP, IBM, Sun}
John> did. Of course, 64-bit super-computers have been around ~30
John> years.

John that was interesting reading. I was not aware that 64 bit had
such a large foot hold in embedded systems. Amazing but my take on the
above notes is why did it take x86 so long to make it to 64 bit? Looks
like intel was never going to move from 64 bit unless forced to do so.
I guess this is one time competition forced the x86 desktop to 64 bit.
I wish I had purchased some AMD stock when it below 10 bucks a share
several years ago. ;-)). That is how it goes.

Thanks,

Alan

 

[David Kanter]

Well, the Itanium is also a really big chip with low yields.

What makes you think Itanium has low yields? IIRC >70% of the die
area is cache, which is very easy to repair.

64-bit has finally arrived, since AMD *was* first with something - first
at making it affordable.

That is true.

David

 

[Bernd Paysan]

I wish I had purchased some AMD stock when it below 10 bucks a share
several years ago. ;-)). That is how it goes.

Huh? Seveal years ago (2000-2001 frame), AMD was at one point up to ~50
bucks (actually, 100 bucks, that was before the split); last year, it was
down to 5 bucks or so, and now it's at 12 bucks, so not much away from your
"strong buy" price.

 

[Bill Todd]

Oops, you're right upon further rereading, I find they were talking about
surpassing the performance of, not the sales of. Oh well, my bad.

It's also wise to observe exactly who was doing the talking - Terry Shannon,
well-known HP shill.

The only figures I've seen from that presentation indicated that while the
top-of-the-line Itanic was able to beat lower-end Alphas in the traditional
'Vax Unit of Performance' (VUPS) metric, the 1.15 GHz Alpha was still
marginally faster, even in its previous-generation process and without using
the absolutely newest (1.3 GHz) models.

Of course, if Alpha hadn't been killed three years ago, the new Itanics
would have been competing (though the performance gap would have made that
word somewhat laughable) against EV8, with over twice the performance of EV7
plus 4-way SMT (offering another factor of close to 3 in commercial
workloads like TPC-C, according to the simulations performed prior to EV8's
demise - enough to leave POWER5 in the dust as well, though in EV8's absence
POWER5 seems a good bet to thorougly shame Itanic for at least the next 30
months in TPC-C).

- bill

 

[Nick Maclaren]

|>
|> It's also wise to observe exactly who was doing the talking - Terry Shannon,
|> well-known HP shill.

That is unfair. He is biassed, because he wouldn't get cooperation
if he wasn't, but he is not simply a shill.

One line I noticed from his presentation was "No denial of service
attacks... and more". Like the 13th stroke of a clock, that casts
doubt over the rest of HP's claims. I have seen that claim made
by many vendors over many years, and it is invariably a sign of a
presentation that is largely bullshit. The reason is that it is
provably equivalent to solving the halting problem.

No, I can't say WHICH of the other statements in HP's presentation
are bullshit, and which are largely true, but I am sure that quite
a lot will be the former - simply because of that inclusion. It
really IS that indicative - marketdroids please note!


Regards,
Nick Maclaren.

 

[Alex Johnson]

Of course, if Alpha hadn't been killed three years ago, the new Itanics
would have been competing (though the performance gap would have made that
word somewhat laughable) against EV8, with over twice the performance of EV7
plus 4-way SMT (offering another factor of close to 3 in commercial
workloads like TPC-C, according to the simulations performed prior to EV8's
demise - enough to leave POWER5 in the dust as well, though in EV8's absence
POWER5 seems a good bet to thorougly shame Itanic for at least the next 30
months in TPC-C).

This is untrue. The EV8 was reported as having vastly better
performance than Itanium or Pentium 4, and I was eagerly waiting for it
(almost drooling). But one thing you leave out is the Alpha team's
tendancy to keep projects going long after their planned release dates.
EV8 would not be availible realistically until 2006. By that time it
would be 4 threads for Alpha vs 2 cores*2 threads for Pentium 4,
Itanium, and Power. It would be at best comparable, if not behind the
times, since all three of those architectures would have their 2*2 by
2005. Alpha has a problem with "give me 2% more performance and you can
delay the project as long as it takes". The team recognizes this as
their downfall and points it out all the time.

Alex

 

[Peter Boyle]

2005. Alpha has a problem with "give me 2% more performance and you can
delay the project as long as it takes". The team recognizes this as
their downfall and points it out all the time.

sources for this?
Peter

Alex

Peter Boyle (e-mail address removed)

 

[Matt R.]

Itanium sales have finally surpassed Alpha sales at HP. Looks like it's
mostly in the OpenVMS market though. Most OpenVMS customers are entrenching
around Itanium now. The Alpha-Tru64 market still seems to be volatile.

http://www.techweb.com/wire/story/TWB20040820S0005

Yousuf Khan

QUOTE:
"The performance crossover point -- the point at which IPF will meet,
and begin to exceed by a widening margin, the performance of Alpha --
is expected to occur in the EV7z/Madison9M timeframe," Shannon said
referring to the final iteration of the Alpha family (EV7z) and a new
Itanium configuration (Madison9M.)"
(End of quote)

Well, this is interesting, isn't it? The Madison9M can finally beat
the lower clocked Alpha. That makes me wonder what the promise of
higher IPC rates for EPIC compared to other ISAs are really worth.
Especially when you take the fact into account that the Madison has
huge 9 MiB L3 Cache on die compared to tiny 1.75 MiB L2 of the EV7z.
On-die cache mostly helps a lot for performance.

That's what we have seen regarding single CPU performance as i.e. SPEC
INT/FP BASE 2000 results are higher for Itanium than for Alpha. So if
HP (ok not really HP but Terry Shannon) is still saying that with
Madison9M the performance level of Alpha is reached/overtaken they/he
must refer to SMP systems. Here Alpha has an advantage with its four
highspeed interconnects for direct CPU communication compared to the
shared bus of Itanium which seems to be a bottleneck.

So if performance for larger system is more dependent on interconnect
choice and its implementation I ask myself why a new ISA is neccessary
and why other ISAs had to die in favor of this new one. The point
seems to me not to choose an ISA with a theoretical advantage in ILP
rate but to implement good (low latency) CPU interconnects.


Call me an Alpha fanboy but I find it simply amusing that a 0.18 µm
Alpha EV7z finally gets overtaken in terms of performance from a not
yet released 0.09 µm Madison with 5 times the amount of on-die cache.
Well, I wonder when the point of "overtaken" would have been reached
if the original plans for an EV79 (0.13µm, 3 MB L2, 1.8 GHz+) had been
realized and not delayed and finally canceled by HP.

Sorry for OT, but when hearing those statements from the above quote I
really ask myself why Alpha is being replaced by Itanium.


Regards,

Matt