Slashdot Mirror
Intel and HP Commit $10 billion to Boost Itanium
YesSir writes "Support for the high-end processor that has had difficulties catching on is coming in from its co-developers Intel and HP. 'The 10 billion investment is a statement that we want to accelerate as a unified body' said Tom Kilroy, general manager of Intel’s digital enterprise group."
So as I'm reading this there's a big plug for AMD Opteron just below the article. This would appear to me to be the threat to the Itanium, the same which effectively has killed big iron -- inexpensive commodity hardware. Sink a few thousand into Opteron systems and run what you already have, or sink far larger amounts into some gobble-de-gook system which won't run, except under software emulation, what your multiprocessor system does. Sorry HP/Intel and everyone else dumping money down this rabbit hole, I think you've lost the plot. Today's super computers are parallel computing down with 64bit Gen x86 processors, like the AMD Opteron. The glue is in the software, not in big fat chunks of expensive silicon.
if still not convinced, i might have a few meg of core to sell you
A feeling of having made the same mistake before: Deja Foobar
Too bad HP won't spend $$$ to bring back the Alpha.
I miss architecture diversity....
seems just a bit too late. they should donate to help feed some starving children not starving platforms.
Am I the only one who doesn't see the point in this?
Disclaimer: I'm not hyping Northern Colorado as being "the next Silicon Valley". Intel is taking over the old Celestica plant next to the HP campus in Ft. Collins, Colorado, and AMD is looking to open up about 200 jobs in the same area (being Ft. Collins). Interesting move... http://circuitsassembly.com/cms/content/view/2709/ 94/
Itanium has been taking share from both IBM power and Sun Sparc.
True but can they compete with the UltraSparc T1 (which has 32 threads compared to Intel Itanium's 2 threads)?
No Sigs!
I saw a documentary on TV a while ago about the impact that Moore's Law has had on the world economy.
Essentially it stated that:
. The big economic powerhouses of the world are based largely on IT
. A key driver for the developemnt of IT is the continual improvement in computer power (and corresponding drop in price)
. When Moore's Law hits the basic physics of silicon, and they can't make any more faster chips, then this economic driver stops
If Intel and so on don't keep on pushing Itamiums then unless AMD can keep ahead we are all in trouble
The chip was made to compete with "Big Iron" servers - the only problem is that it is marketed to the windows consumer market, and that is who looks at it when making purchasing decisions. AMD has really started to eat up this space, and if Intel does not start to turn this boat around fast they could really get hurt when 64bit CPUs are commonplace.
>"Itanium has been taking share from both IBM power and Sun Sparc."
Uhh, it could hardly lose share could it? If it lost any share the product wouldn't exist. What, did they double their share from 1 to 2 users?
Ten billion is an awful lot to throw away on this loser chip.
I mean, few people actually WANT to run a different chip (and thus a different OS and versions of apps) in their data centre, compared to their desktops. They used to do it, because it was necessary. Now it isn't necessary, so people don't want to do it. Intel's only hope is to try and get people to use it EVERYWHERE, on their desktops too. But there aint no hope of that either.
What's the point of running "Big Iron" and/or Itanium if we have to deal with hacks/patches and headaches to run real world production applications like SharePoint, SQL and other Office collaboration suites?
The similarity of the two marketplace failures is amazing.
Are there any advantages to using the Itanium over an Opteron or Athlon 64?
"It ain't a war against drugs.it's a war against personal freedom" --Bill Hicks
This appears to be an utter waste of capital (to the tune of $10 Billion) in a feeble attempt to rescue something from nothing. Yes, the Itanium costs are sunk costs, but there's absolutely no way that any of the companies involved will see any resonable ROI on this route.
Itanium missed its window of opportunity--it's time to move on.
Anyone want to tie this into their $10 billion push?
[Fuck Beta]
o0t!
This is more along the lines of post-mortem muscle contractions.
I'm sure that SOMEONE out there is willing to pour money down the toilet for this platform. And they'll make HP/Intel very very happy.
Then again, there's people who're into snorting drain cleaner too...
Chas - The one, the only.
THANK GOD!!!
Competition In Action!
"This is a $140 billion opportunity on hardware. It's dwarfed by the opportunity in software and services on top of that," Kilroy said. "There's a reason there's $10 billion of investment in play."
And 1400% profit, too! Nice.
Large Man with Dead Body: Here's one.
The Dead Collector: That'll be ninepence.
The Dead Body That Claims It Isn't: I'm not dead.
The Dead Collector: What?
Large Man with Dead Body: Nothing. There's your ninepence.
The Dead Body That Claims It Isn't: I'm not dead.
The Dead Collector: 'Ere, he says he's not dead.
Large Man with Dead Body: Yes he is.
The Dead Body That Claims It Isn't: I'm not.
The Dead Collector: He isn't.
Large Man with Dead Body: Well, he will be soon, he's very ill.
The Dead Body That Claims It Isn't: I'm getting better.
Large Man with Dead Body: No you're not, you'll be stone dead in a moment.
The Dead Collector: Well, I can't take him like that. It's against regulations.
The Dead Body That Claims It Isn't: I don't want to go on the cart.
Large Man with Dead Body: Oh, don't be such a baby.
The Dead Collector: I can't take him.
The Dead Body That Claims It Isn't: I feel fine.
Large Man with Dead Body: Oh, do me a favor.
The Dead Collector: I can't.
Large Man with Dead Body: Well, can you hang around for a couple of minutes? He won't be long.
The Dead Collector: I promised I'd be at the Robinsons'. They've lost nine today.
Large Man with Dead Body: Well, when's your next round?
The Dead Collector: Thursday.
The Dead Body That Claims It Isn't: I think I'll go for a walk.
Large Man with Dead Body: You're not fooling anyone, you know. Isn't there anything you could do?
The Dead Body That Claims It Isn't: I feel happy. I feel happy.
[the Dead Collector glances up and down the street furtively, then silences the Body with his a whack of his club]
Large Man with Dead Body: Ah, thank you very much.
The Dead Collector: Not at all. See you on Thursday.
Large Man with Dead Body: Right.
"To those who are overly cautious, everything is impossible. "
Intel and HP spend untold sums of cash developing and rolling out a chip that comparatively few use. Thus, the market has effectively told them that there is not a large need for this behemoth. So how do they respond? A pledge to spend $10 billion more? How does this make sense again?
Itanic is dead. RIP. Game over. Hasta La Vista (no pun intended).
Game, set, match. Etc., etc., etc...
Seems to me that HP are better off keeping their piece of the $10 billion. You gotta spend money to make money but I fear this isnt the best way to improve their bottom line in the short term. Which is exactly what needs to be done since Carly got the shaft. I think they're taking their eye of the prize.
serenity now!
The Itanium admittedly has great FP performance _per clock cycle_, but that's about the only nice thing anyone can say. $10bn?!? Talk about throwing good money after bad!
"The history of science is cluttered with the relics of conceptual schemes that were once fervently believed and that have since been replaced by incompatible theories." -Thomas S. Kuhn
Tell you what.
You give me a billion dollars and I'll kick each of you as hard as I can in the balls.
Ten times.
I'm pretty sure that in 5 years each and every one of you will look back and wish you had taken my option.
I'm trying not to be a jerk about this, but the person who posted the "bring out your dead" monty python skit hit the nail right on the head.
1q2w3e4r5t6y7u8i9o0pqawsedrftgthyjukilo;p'azsxdcf
so what's the point? Lack of 32-bit support nearly killed it out of the gate. Then they added software 32-bit emulation that sucked, and no one wanted it. Then they added 32-bit support in the hardware. Still nobody wanted it. Now they're going back to software 32-bit emulation. Sooo...how many enterprise servers really want to be running a WinCPU?
I'm just a dumb IT droid, but this makes no sense. Unless the $10 billion is going for bribes.
we will end no whine before its time
. ..
You also have to spend money to lose money. The trick is getting something to come back. HP are already doing things with AMD processors so you gotta figure there's some real head-scratching going on among the workforce at HP.
A feeling of having made the same mistake before: Deja Foobar
"IA-32 EL is OS-based and is only available after an OS has booted,"
http://www.digit-life.com/archive.shtml?2006/0125
Betcha money it's not any form of Unix.
we will end no whine before its time
How do you know they aren't planning this as some method of helping bring an end to wars? If they get the pentagon buying Itanium equipped missiles, just think what they could do!
A feeling of having made the same mistake before: Deja Foobar
Sure, it is a huge sum of cash and perhaps the 'shareholders' might get more short term benefit out of investing the same sum of money into commodity microprocessor R&D but the itanium could eventually pay off in a big kind of way. It seems that most people posting here are just as impatient as shareholders when it comes to results, they want them NOW! Good things can't always manifest themselves in a short period of time and I think it is impressive that Intel & HP continue to invest money into something that has yet to produce any tangible benefits over existing architecture. I'm willing to bet that x86 isn't the omega to processor design ideology, and itanium may not be either, but Intel & HP seem to believe it is a step in the right direction. Very few people that post here have the knowledge necessary to even begin assessing whether such a design may ever pan out and it appears the jury is still out among those who have the capacity to decide. Meanwhile Apple continues to recieve gratuitous praise for releasing shiny white computers with chamfered corners. Maybe if Intel & HP invested 10 Bn into cosmetic processor design they would be recieved more favorably with the press.
In spite of all the negative publicity, Itanium is quite far from dead. The recent corrections in path make a lot of sense. What really put Itanium out of orbit was Intel's decision to use Itanium in even the small and medium systems. This meant lost marketing focus, and some lame architectural decisions for x86 compatibility. Itanium has nothing in common with x86 except its made by Intel.
.Net Framework 64-bit and Sql Server 2005. (Not in Exchange Server, Biztalk Server etc. Earlier we even had Windows XP running on Itanium. Sigh!).
It seems the finally found the market:
Last week Intel went back on x86 compatibility, only software emulation. Makes sense, the market for Itanium is big iron. It is way to expensive for anything less. And the users better run 64-bit Itanium optimized code to get their money's worth.
Microsoft trashed all Itanium plans for the small and mid segment. They will support Itanium only where it makes sense in their product line, just Windows Server,
Intel's Motherboards supporting both Xeon and Itanium have now been postponed to 2009. This makes sense too, Itanium customers won't be interested in saving a few thousand bucks on commodity motherboards.
And finally 10 billion $ pumped in; good news. I'd think Itanium will be back, by 2008. Architecturally, it is nothing to laugh at atleast. It is just that it lacked everything else, platform-compiler-apps support.
Life is just a conviction.
Truth be told, IA64 is a fantastically better architecture than IA32 or x86-64. Some of it's current caveats, for example, suboptimal software support and high costs, are not due to it's technical qualifications or drawbacks. Once the architecture reaches a critical mass and reasonable market acceptance, these issues should disappear. (more chips -> more people will target software for it, more chips produced in volume -> less cost per chip, etc.)
It's other caveats, for example, poor compiler support, are issues that need to be considered carefully. I'd like to specifically address the poor compiler support. I am not concerned about this issue for the following reasons:
1. Compilers can improve easily, with a recompile. If the architecture achieves a critical mass, then more people and organizations will justify the time and effort to improve compilers on the architecture. Not only can they improve, but taking advantage of such improvements would not require replacing hardware, which makes it an issue of time.
2. The architecture is much more realistic about the guarantees that it's willing to make as a processor. One of the early complaints, was that initial generation of compilers for IA64 would generate, on average, 40% NOPs. It's important to consider a few details when regarding that statement.
A. First, each clock cycle could allow the execution of up to 3 concurrent operations.
B. Second, the architecture is not inserting extra NOPs transparently into the pipeline, as almost all modern processors do in the event of a pipeline data hazard. This fact can be viewed different ways.
i. Most modern processors have to evaluate wether to insert a pipeline stall every single time that an instruction is executed. This is, essentially, wasted work because such a computation could be done by the assembler, however, it does spare the processor the burden of loading useless NOPs into the pipeline and the cache. On the other hand, minimizing the logic that a processor has to complete per cycle generally decreases the minimum amount of time necessary per clock (meaning that it could scale to higher clock speeds.)
ii. The immediate question is, does reading all these NOPs out of memory cause a bigger hit to performance, than making the processor calculate the data hazards? Personally, I don't know. But, let's consider the idea for a moment. On both processors, let's assume that the instruction cache is fast enough to deliver data without wait states, assuming the cache has the data. When your processor is prefetching well, then the NOP issue shouldn't be a big issue. (Except for the fact that the NOPs will now be in the binary, making the binaries larger. I consider this a moot point given the inexpense of modern storage.) When your prefetcher can't anticipate correctly, though, I think the IA64 loses. Both IA64 and other modern architectures have branch predictors, so I suspect unanticipated branches which cause a pipeline flush (unavoidable) and unanticipated cache fills (unavoidable) will be mitigated roughly equally, But because the IA64 has longer instructions that aren't quite as dense, the IA64 will stall longer. Btw, I'm ignoring data stalls, to simplify my argument and because I don't think the architectural differences in the IA64 will significantly impact it. I'd enjoy being corrected on this point.
The IA64 includes a predicate register, which stores the results of comparison instructions. Instructions in an IA64 'bundle' can be qualified to be executed conditionally, based on the condition of a certain bit in the predicate register. This allows the IA64 to avoid some branches. The compiler/assembler can pack a bundle which includes the appropriate two instructions, each qualified to execute for different states of the predicate register. Essentially, the processor is simultaneously issued the commands for both p
fnord.
Something smells fishy to me. $10 billion is alot of money for a marketing campaign.
Assuming that each Itanium chip retails for roughly $1,000, Intel/HP could simply give away 10,000,000 chips for the investment they're making. Do they really think that there will be enough demand for these chips between now and 2010 to make up for that kind of marketing expense?
I have a hard time believing they will actually spend anything near this amount on marketing, even if the campaign is successful.
The most rabid believers in American Exceptionalism are the exact same people whose policies are destroying it.
Man, that was a sweet processor. I recall comparing my spanking new DEC Alphastation to the Cray down at San Diego Supercomputing Center in 1995, and there was just about no difference. That machine flew.
Funny thing how Digital's hardware dominance seemed to just dry up and blow away, tho'. I seem to recall in the 80s and 90s it was the place to be if you were a hot and ambitious hardware hacker. Wonder what happened?
Here's a quick bit of math for thought:
Let's say that Intel contributes half, so USD 5 000 000 000.
Let's say that Intel nets USD 5 000 per chip (probably WAY overestimating sales price and underestimating costs)
Intel would need to sell 1 000 000 chips to make this additional investment break even.
This excludes opportunity cost, cannibalism of existing Xeon sales (though, it's probably the other way around), and probably a host of other things.
It looks like sketchy math to me. To me, it seems obvious that the Itanium will be increasingly pushed into niche processing markets -- and even there, the few benefits that the Itanium presents will be continually reduced as x86 moves up market. Faster FP? Better RAS features? Better scaling? Those can and will be bolted onto Opterons (and probably even Xeons) over time.
In other news .. flogging a dead horse, to cost 10 billion dollars.
It's fairly obvious that the Itanium has been a failure. But then why so much interest from so many companies? From TFA, it's not just HP and Intel. I've heard that the architecture is good in theory, but bad in practice - and my own experience supports the latter. Maybe this is a desperate push to finally turn the theory into practice. Perhaps they should invest that 10 billion in a compiler that can actually support the Itanium's architecture.
In any case, it's an uphill battle now - the Itanium is not looked upon favourably by most people I know. Right now, AMD has the most rational offerings for general-purpose computing and I wonder if IBM will market the cell (or a variant thereof) to the HPC market. Interestingly, both those designs are not dependent on the high-end market to survive. Anything recent that was dependent on that market seems to have failed - even the beloved Alphas. In that context, this investment does seem dubious.
For some reason I'm thinking that $10 billion is probably more than they've ever made on the Itanic.
No, no, no. It is, of course, retirement money for the original designers and even bigger money, or golden rain as I prefer to see it, for the big-honchos who financed it. They must be acknowledged for the untrodden highway they made.
When the golden rain stops pouring over the decision makers, they might use the remainder to hire new designers.
This all according to my own belief and superstition.
Jokes aside, of course there will be overseas prgrammers! Even the CPU flagship Alpha was to a large extent made overseas; in Barcelona for that matters. No, don't be fooled by Fawlty Towers. Que?
Even Rolls Royce are made overseas nowadays! Goddammit.
> I work a lot of overtime in a high-stress, tight deadline job. Once you get into that kind of downward spiral, how do you find another job?
That's an obvious one: you quit this job before looking for a new one.
> I'd quit if I had a choice, but I really need the money
I wonder what you do with all the overtime pay? Sometimes a good career has to be organised, and this starts with having some money in the bank for situations like this.
Proper planing can also reduce the level of stress you are experiencing...
Itanium2 systems are among the top in transaction processing. asp?resulttype=all
http://www.tpc.org/tpcc/results/tpcc_perf_results
and THE top one for clusters.
It makes sense for such an inventmen to go to
a) improving the fabrication facilities - achieving lower defect rates
and reducing price;
b) improving the fabrication process - aiming at higher clock rates
Remember also the recent announcement that an Itanuim CPU will no longer contain essentially a whole IA-32 CPU.
~velco
10 Billion? That means it is just as important to humanity as nuclear fusion? WTF?
Apple will buy Intel and make it's own CPU's
This just makes me insane, I know it was already mentioned several times that people wish HP would put this kind of effort into reviving the Alpha. But to read about them putting this much money into a piece crap like that Itanium after the way they chucked out the Alpha, is expecially galling when you consider that in HP's own internal testing, Alpha EV8s and 9s consistently wipe the floor with even the latest Itaniums.
my old sig is obsolete, and I haven't come up with a stupid enough new one yet
I'd have thought that it was possible to rearrange the deck chairs for a lot less than that. Maybe it includes the musicians' salaries as well, though.
What really surprise me it that the Slashdot crowd give it such a negative attitude. Even though I have no direct interests in Itanium, I really enjoy continued investment in the IA64 architecture for many reasons. It is much more fun with more living architectures. I thought the Slashdot crowd was interested in technology, but sadly that seems not to be the case.
AMD is starting to kick Intel's pants in the most lucrative arena, small- and medium-sized servers. Instead of trying to compete technologically in that area (as opposed to just marketing), they're throwing good money after bad into a failing/failed architecture which only makes sense for a few highly-specialized applications. If it weren't for the fact that most holders of Intel stock know next to nothing about the industry, I would expect a cry for a change of leadership.
Sure, there are a few supercomputing-type applications where the Itanium really, really shines - but they're sufficienty specialized that Intel just doesn't move a very large number of CPUs.
Like I've said before, Intel is in a bind because of its own laziness and arrogance. Look at one of the primary advantages of the A64/Opteron architecture - the on-die memory controller. More memory bandwidth, lower latencies, and a memory subsystem that scales with the number of CPUs. Big-iron vendors proved that technology long before AMD decided to use it. Yet Intel has always enjoyed the superior manufacturing side of the business - if *anyone* could afford to have put those extra transistors on the die, it was Intel. Since they're almost always a step ahead of AMD in making smaller transistors, they had the *ability* to do something along those lines long before AMD did - but relied on the old tradition of more megahertz and lots of marketing. I don't think that this move is much different, they're putting their efforts in the wrong direction.
steve
Oh, you're not stuck, you're just unable to let go of the onion rings.
Everything you've said about processors? SGI used to be the KING of 3d graphics processing. What happened? Cheapass PC hardware caught up, broke even, and eventually lapped SGI's technology.
The Alpha still rocks. It just happened to take the rest of the industry better than a decade to catch up.
Unfortunately, there's no modern Alpha to flog the x86 with.... and all SGI has to wag over the competition is gigs of texture memory (for the price of a good sized whitebox render farm).
Recently an article was published on anandtech that puts the itanium in a new light: it's actually very efficient in terms of die area utilization. Combine this with Intel's recent announcement that they were scrapping the hardware x86 compatibility on the itanium, which takes up a fair bit of die space, and you have a very small core of the sort that's absolutely perfect for multi-core applications.
Itanium needs a lot of cache to function well, for reasons that the aforementioned article describes, but it's not unreasonable to assume that intel's shared cache technology from Yonah will make its way into Itanium.
This thing might be trying to compete with chips like the Ultrasparc T1.
Look, the only reason why Intel and HP keep sinking money into this dog is because they've already spent incredible amounts on it and have been telling everyone, for about a decade now, that it's the Future of Enterprise Computing.
They seem unable or unwilling to face up to the fact that the Itanium is a complete albatross, and there's really nothing admirable about that.
there's a typo: Intel and HP commit 10 billion to booze and women, that's the title, I have no idea what this "Itanium" thing is and where it came from.
Disclaimer: I'm not hyping Northern Colorado as being "the next Silicon Valley". Intel is taking over the old Celestica plant next to the HP campus in Ft. Collins, Colorado, and AMD is looking to open up about 200 jobs in the same area (being Ft. Collins). Interesting move... http://circuitsassembly.com/cms/content/view/2709/ 94/
Ft. Collins has certainly seen a lot of high-tech in the last few years. Itanium was largely developed at HP Fort Collins, and now most of those engineers are Intel engineers. There's LSI Logic, Agilent/Agavo, HP, Intel, and of course all of the research that comes from a major university (CSU).
Colorado could very well become the tech center of the mountain region - in the Boulder area alone, there's IBM (Niwot), StorageTek (Broomfield), Xilinx (Longmont), Sun (Broomfield), Level3 (Broomfield), Ball Aerospace (Superior/Louisville), NIST, NOAA, and quite a bit more. There's also Qwest, Lockheed Martin, Raytheon, Qualcomm, Kyocera, and a number of other companies with significant operations in the area.
Interestingly, one of my friends worked at the former Celestica facility in Fort Collins; another works at HP in the Linux division.
Major University + Low Taxes / Land = High Tech. That's why AMD is in Dresden.
The chips are way over priced and too under performance for people to spend the money. No marketing campaign can fix that.
Granted, the Itanium was launched before the market/fabrication processes were really ready for it.
I do not believe that is has deserves the flack it is catching in here recently. It's a killer FP unit, and in fact one of the fastest general purpose CPU's out there, it's current price/performance ratio is where the real problem lies. However I predict this will change in the near future, here is why:
First off, the itanium is heavily dependent on huge and fast cache memory, there is no out-of-order execution, so cache misses are a bitch. This is much more pronounced in the Itanium because IA64 instructions tend to take up more than "just" twice the space (compilers for itanium tend to insert a lot of NOP's). Intel has during the last two years invested healivy in cache structure research, this has paid off, Intel has become extremely adept at producing cache memory. Further more if properly fed with data, the Itanium can processes 6 instructions per cycle against the opterons 4 AND due to its conditional oprations it can get the job done with fewer branches.
Secondly, the itanium core is SMALL, with its L1 cache its roughly 21m transistors, which is about half the size of the Opteron or the Xeon, and with the decsion to drop x86 hardware support from the core, we might even see the Itanium drop below 20. Multi-core processors will be where the Itanium will shine, and multi-core processors is the way the market is going.
IMO Itanium has time on it's side.
I agree with your post. Plainly off-topic, but I could not resist your .sig!
Of couse He could. Being both fully man and fully God he knew both sides of perfection and imperfection. Now we all know He was without sin; so if you think overheating a burrito is a sin, the answer is no; but then there's no accounting for taste. :)
The proper formulation of the question is "Could God ... ?" exercise such power if He so desired. This is the power of the Trinity. It separates the spiritual world from the material and addresses the question of the spiritual in the material.
This is all really well described in the first chapter of the Gospel of John.
Oh, ya, I forgot. This is Slashdot. Moderate to -5: proseltyzing
I changed the title to better reflect the truth.
Big iron is just fine. The deal is that Intel had delusions as to what it meant and tried to apply PC standards to a world which lives at a much higher level.
PowerPC archietecture is alive and well in true big iron. It will even find its way into IBM mainframe technology. Server farms are not big iron, they are glorified PCs with a little bit more reliability yet suffer from all the issues PCs normally do.
The big iron at work here, Tandem, IBM Mainframe, and iSeries have no appreciable downtime and the majority if not all recent hardware failures were transparent to all users. Hell some of the hardware failures are transparent to support staff as IBM usually calls first.
A processor does not make you big iron any more than disk drives make a server. Its a mindset that isn't happening at Intel and one HP lost a long time ago.
* Winners compare their achievements to their goals, losers compare theirs to that of others.
What they are getting into Aerospace propulsion now?
He who knows best knows how little he knows. - Thomas Jefferson
People like to talk about how Itanic, as it were, is a flop. It is, but not because it's not a good processor. Itanium is a very cool architecture with features long-time in coming. For instance, used properly, branch predication can be a HUGE boost to performance, and it's proven itself to be so when used properly on the Itanium.
The first problem is one of marketing. HP/Compaq is a screwed-up company, the merger of two wholy incompatible companies that could never work together properly. Put this together with the fact that they canceled Alpha, another great processor, and you can see that selling Itanium is more about politics than engineering. The next problem is pricing. For a single-chip solution, Itanium is awesome, if you don't count the fact that you could buy multiple Opterons for that price and achieve more performance with properly threaded code.
There are, of course, technical problems. Itanium is a heat monster. They didn't design it with power consumption and heat dissipation in mind. Did you know that the Itanium's top speed isn't limited by wire delays like it is in most other chips? No. It could actually run a lot faster, were it not for the fact that they can't get the heat off the chip fast enough. Another problem is the compilers. Static scheduling has its limitations, but the real limitation is that Itanium compilers can't manage to do even decent scheduling. It's too complicated. Much of Itanium's performance is theoretical. Given a small piece of C code, you can recode it in assembly and get it to run 10 times faster. If only the compilers were as smart as the assembly coder.
Itanium was a great idea. It's just being executed poorly, and the R&D is being put into the wrong place. The architecture is there. It's great. Now, get the price down, design it for lower heat dissipation, and get some people working on that damn compiler!
Damnit. I had better replace those pads soon...
Sticking feathers up your butt does not make you a chicken - Tyler Durden
Consider: One of the compiler issues has been the ability to schedule all four pipelines with instructions that are useful, instead of no-ops. Now, consider using a method like the T1 does, where you have four sets of VLIW threads, each with on average 3 instructions. You could get away with executing the four threads with 12 pipelines on average. In effect, you can take the no-ops from one set and fill them with instructions from another thread, and keep the pipelines chugging. If tied together properly, it would have binary compatibility with current Itanic code, make use of today's ineffecient compiler generated code better, and make the arch work much more effeciently with OS threads ala the Sun T1. Given that the overall core (not including x86 and cache) for the Itanic is fairly small, something like this could probably be done very effectively and push the Itanic ahead.
If they're not, theres a chance your air's gonna get polluted :(
deliver in pallets of slightly used 20's please, to the lobby of my bank. details on request.
if this is supposed to be a new economy, how come they still want my old fashioned money?
This is an old ploy. I'll bet they scrap the chip within 24 months. With Alpha, Compaq did the same - went public with huge statements of commitment to test the waters. In less than 12 months, they announced the death of Alpha.
- The Kessel run is for nerf herders. I can circumnavigate the entire Central Finite Curve in a lot less than 12 parse
It's obvious that they know a lot we don't, but everyone here is too arrogant and pastywhite to admit that
maybe
just maybe there is a difference between what you read in a C|net article and what's available to the public, and what's going on in intel's labs, and within the company's boardrooms.
You are forever on the outside track.
Grandparent quoted (or invented) completely unsubstantiated claim about Itanium taking market share.
Who do you get to be an expert to tell you something's not obvious? The least insightful person you can find? -J Roberts
There doesn't seem to be anything very compelling about Itanium. It doesn't clock particularly fast compared with other competitive hardware, it only does 4 FLOPS per cycle like most other current processors, it isn't very compatible with other software, it's difficult to develop for, and it doesn't even have a nice looking case. WHAT is so special about the Itanium that keeps Intel pumping money into it?
Opteron = Redundant Array of Inexpensive Processors. With this kind of money being spent on a rat hole I won't be buying any more Intel or HP stock anytime soon.
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
That's kind of odd. Loveland was the first site for an HP shop outside of California back in '62, and after they moved here (to get lower-priced tech workers) they built a mini-tech economy that included LSI, Celestica, Flextronics, Advanced Energy, and scads of others in the area. But now, HP has shut the Greeley and for all intents and purposes the Loveland plants, is drastically cutting the Fort Collins plant, Celestica's been outsourced to Mexico, both Flextronics plants are long-since gone, AE is downsizing. I work right beside the big AMD design center in Longmont and we joke about how there are fewer cars there every day. I'm very surprised that they'd move into an area where high tech seems to be moving OUT as fast as it can, because they've created a localized expensive workforce exactly like they were trying to escape when they moved here forty years ago. At one of my previous jobs at a hardware place, I was working with five people I'd worked with before, every one of them at a different place, one at *two* different places, and as I was working there several other people moved through that I'd known from other jobs as all five of the people I mentioned moved on to other jobs. The nomadic tech community here is pretty cool, but feels incestuous sometimes.
Nostalgia's not what it used to be.
Rumour has it that intel is going to bring out new versions of the itanic with RISC features such as hardware out-of-order and speculative execution to try to get the performance up to a reasonable level on non-scientific (i.e. business) workloads.
itanic came from late 1970s supercomputer designs.
intel wanted to take over the (64-bit) world with itanic and they almost managed. They, together with HP PHBs, convinced the technically-illiterate PHBs who ran SGI to cease development of their own 64-bit RISC CPUs. itanic was promising jam tomorrow and no one was going to be able to compete. itanic has been pushed into smaller and smaller niches as time has gone on. Once it was the 64-bit CPU to rule them all. Now it's just for "supercomputers."
intel and SGI gave a 10240 itanic cluster to NASA. The competition was an Opteron cluster from Sun which was cheaper, faster, cooler, less power-hungry, so intel paid for it to save face.
The UK Atomic Weapons Establishment just ordered an Opteron-based Cray for nuclear weapons research.
If you want to see a VLIW design done right, look at the Transmeta Crusoe processors, or the Sun MAJC.
If someone is really dumb enough to throw money down that toilet I just hope they remember to flush often between handfuls of bill. Sorry to see IA64 go without even ever having had a chance at it, but even viable 64-bit veteran architectures like the Alpha have gone down that drain.
The Opteron is much cheaper than 75% of the Itanium.
;) ).
:).
So far I've seen a fair number of reports from people that the Opteron actually works much better than the Itanium for most real world tasks using real world system configurations. I haven't seen a single report saying otherwise (other than from PR announcements).
A dual core Opteron only requires 233 million transistors and a 200 square mm die.
Whereas a low-end 3MB cache McKinley Itanium 2 has 221 million transistors for _one_ core. The more recent 1.6GHz Itaniums use up to 410 million transistors taking 374 square mm (still one core)!
So I wouldn't say the Itanium 2 is a better architecture for floating point.
I bet that the Itanium 2 design is much better in parallelizable floating point tasks than non-parallelizable tasks. If I'm right, then that those very tasks will be fairly easy to run well across multiple cores anyway.
The IBM POWER5 has 276 million transistors and IMO it makes far better use of them than the Itanium (of course the IBM eServer p5's 36MB off-chip L3 cache might help a bit too
AMD Opteron (TM) 180 2 cores, 1 chip, 2 cores/chip
spec cfp2000 rate base=32.3
spec cint2000 rate base=35.0
transistors = 233M x 1
AMD Opteron (TM) 280 4 cores, 2 chips, 2 cores/chip
spec cfp2000 rate base=68.7
spec cint2000 rate base=71.8
transistors = 233M x 2
AMD Opteron (TM) 880 8 cores, 4 chips, 2 cores/chip
spec cfp2000 rate base=129
spec cint2000 rate base=134
transistors = 233M x 4
IBM eServer p5 570 (1900 MHz, 4 CPU) 4 cores, 2 chips, 2 cores/chip (SMT on)
spec cfp2000 rate base=125
spec cint2000 rate base=74.4
transistors = 276M x 2
HP Integrity rx4640-8 (1.6GHz/9MB Itanium 2) 4 cores, 4 chips, 1 core/chip
spec cfp2000 rate base=77.9
spec cint2000 rate base=72.5
transistors = 410M x 4
Given that, I'd say there is really very little reason to go Itanium (especially since HP's commitment to Tandem, OpenVMS seems rather questionable, a pity actually).
Either Intel is nerfing their Itaniums (by going single core with so many transistors for cache vs multicore with less cache) or their architecture just doesn't work well in practice. In any case, if you're going to pay a lot for non-x86, might as well go IBM POWER5.
Otherwise, the Opterons are pretty decent. I'll leave the Intel x86 figures for someone else to do
It's protected by extensive IP patents. That was its goal. Not to be innovative or great or whatever. Intel was afraid of the clones (AMD, Cyric, etc) so they had to build a new architecture that no one would be allowed to clone.
HMS Itanic.
"Speaking the Truth in times of universal deceit is a revolutionary act." -- George Orwell
I know I surely would if I were a big investor.
How could they be so stupid?
Even Intel mentioned they were expected by shareholders to make $26 billion last year with Itanium sales and only made sales targets in the mere millions??
Not to sound flamebaitish but how many billions upon billions have Intel/HP invested in the Itanic errr ITanium? To me I see it as a way to say "Hey! We just blew billions into this project and were are going to get a return whether you like it or not!" and being totally ignorant about sunken costs or what the market prefers.
For $1 billion they could resurrect the alpha and take over the whole market. IT sounds silly to say this but the alpha is the only chip that is fast enough to get corporate clients to switch and can run winx86 software nominally fast. If not then bring back PA-Risc for some of their high end systems.
THe market has responded many times over and over again that they dont want the Itanium. Yet, stupidly and arrogance designed to protect a few executives self image (perhaps their job) means 10 billion more lost.
I think Sun and IBM are getting quite a kick out of it.
If a mere mortal known as a middle class person such as myself or 99% of those reading this made such a dumb decision about investments at work we would be canned in a second.
Intel purchased a small Israeli chip firm to make the new Pentium Dou core that is used in Apple's powerbooks...err macbook pros. Very efficient chip and low power usage produced with a cap only in the mere millions. That just goes to show how poor of an investment it was.
This makes no sense. Just like in the game of poker you need to leave the money on the table and walk. Its gone and your not going to get it back.
http://saveie6.com/
These CPU and systems are not your run of the mill white box generic PC server etc.
These CPU's run and maintain alot of business infrastructure.
If you've run a computer in a SAN enviroment, this CPU makes sense.
The data throughput capability alone should convince most people what this CPU is intended for.
If you want sustained DATA i/o that will saturate a SAN director switch, then this type of CPU will clobber your XEON/OPTERON cpu without the need to have multiple boxes to acheive something close.
No Xeon/Opteron that I know have multi-rope I/O and when then do, they might be able to match this CPU
PowerPC archietecture is alive and well in true big iron. It will even find its way into IBM mainframe technology.
I think you mean the POWER architecture. 'Power Pee Cee' is Apple marketing jargon.
Brilliant. Carly's been gone for a while, but it seems the shit heads she brought on board are still making (stupid) decisions there. This Itanium/Itanic thing is just the latest round in the long slow HP death spiral.
RS
Shoes for Industry. Shoes for the Dead.