Sign in with
Sign up | Sign in
Your question

anyone remember Itanium?

Tags:
Last response: in CPUs
Share
January 20, 2006 6:08:28 AM

Feel free to tell me to stop posting links, but i figure it will give people something to do, since the general consensus is that THG articles aren't doing it for us anymore.
Anyways, Cnet has a story about Itanium. Here’s a chart with a timelapse of projected revenue… I got a good laugh out of it.

More about : remember itanium

January 20, 2006 7:19:49 AM

if it had not been for AMD pushing Intel with performance all the way over the last 5 years we woul all be sitting here OC'ing our Itanium CPUs... Intel never wanted x86's to get as fast as they are now, and without AMD would have been happy to let them linger at much slower speeds to show off the performance advantage that there new baby Itanium had. The final killer was x86 doing 64bit, basically AMD forced Intel to write off $billions :) 
January 20, 2006 7:43:49 AM

The Itanium isn't that bad, there are several advantages it has.

Sure currently I'm running Opteron 270's but in a few years who knows ?

For some applications the Sun UltraSPARC has been good (the Opteron 2/4-way design almost mirrors it, the way memory bandwidth is multiplied)... for other applications the HP-RISC architecure has been rather popular.... which leads into the HP / Intel IA-64 architecture anyway.

Itanium has poor IA-32 (x86) performance, but that is a minor concern to most (almost all) the people using it, Consider that only 1% or so of the volume of the Itanium CPU has gone into the IA-32 component.

Originally when I saw the SANDRA results for it (even in IA-64) it made me seriosuly wonder why anyone would use it, so instead of 'bashing' it in forums (not saying you are) I decided to read up on it a little.

Only Sun and IA-64 are offering 128 bit wide system buses. Large boards with support for large amounts of memory (Over 32 GB) require slower speed buses just to work. (eg: Signal integrity).

Check out:
http://www.intel.com/design/itanium2/documentation.htm

Thanks for starting the thread on TomsHardware, hopefully a few people will reconsider the future of the IA-64. I reckon by 2010 (give or take) it will mature quite rapidly in comparison to IA-32 (x86) or even x64 (EM64T / AMD64).

This new Z-RAM cache concept might give IA-64 some problems down the road though, as one of the goals is to cram as large a cache as possible onto the chip. If 4-16 MB caches on 'consumer' chips become common it might just be the nail in the coffin.

Personally I think IA-64 has a future, just not 'tomorrow' or even 'within 12 months'.....

Contact Info available at: http://users.on.net/~darkpeace
Related resources
January 20, 2006 8:05:28 AM

I am not educated enough to comment on the chip, merely the irony of the failure of intel's marketing machine to sell it.
January 20, 2006 8:42:26 AM

I don't see how "pure 64-bit" is going to change anything that much.
16 to 32 bit (aka win3.x or 95 to 98se) didn't really change things that much, not for the average consumer anyways. We have 64-bit now, and there isn't much that actually takes advantage of it. I'm sure it will be the standard in a few years, but what is on the horizon that is so great that necessitates 64-bit? Vista?
The bigger question is, when 64-bit does become the standard, why will there be any monetary drive to switch to the itanium architecture? Why would software developers decide to start pumping resources into an architecture without a large install base?
January 20, 2006 8:44:02 AM

Where I live (Australia) Intel are not spending much (time or money) marketing the Itanium.

HP & Intel dev'd it together and most likely HP are the biggest users of the chip (as far as I am aware anyway).

The only things worth "marketing" are things being sold to the mass public consumers. I suspect that this is where Intel fell down, normally marketing is their (and Coke Cola, etc) best friend... in this case it is doing them more damage than good IMHO.

From a technical perspective, the IA-64 architecture has advantages.... try spending as much time reading about it as you appear to be laughing at it (and encouraging others to 'dislike' it without questioning the benefits).

Sure Itanium 1 and 2 are not 'fast' but they are 'powerful' in some areas, and actually cost effective at various 'heavy' tasks. Intel can do much better without even trying, but likely the next series of improvements will never be sold and only exist in labs.... They may even change the name of the processor for the next real IA-64 CPU (unless they try for an Itanium 3 to cut their research costs).

It'll take them 10 years or so of R&D before they are happy with it, but it will eventually reach a point where x86 / x64 won't scale well compared to IA-64 (due to compiler advantages mostly, and the fact that OoO execution is taking more and more silicon each CPU generation to continue scaling, what happens when it requries more than 20% of the space ?. A) They'll stop scaling in performance). Then their R&D billions will be returned 10 or even 100 fold.

Companies don't die, they either close, go bankrupt, are purchased by another company, and are considered immortal entities (compared to humans anyway).

People (fan boys) used to laugh at AMD once, long ago, and now AMD have their fans..... in time IA-64 will have fans... perhaps even you will dis' everything else except IA-64, and wear IA-64 menchandice when the time comes. Or perhaps when AMD start licensing IA-64 from Intel within the next 20 years... what would it take to convince you ?

The real irony is, people almost hate IA-64 as instinct, but the same 'educated' people would prob like the Cell (or other similar non-Intel) processors.

Check out the rather 'smart' article at:
http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=259...

There are not many tech websites left with authors who can actually get the people (readers) to think for themselves... instead of letting marketing or other such dribble do the thinking for them.

eg:

What do you think of nVidia GeForce 6200 'TurboCache' or ATI Radeon X300 'HyperMemory' ?

What if the same basic concept was used, but the GPU included some very high performance video memory (integrated on GPU, much like an L2 cache is now on CPUs.) and used dedicated Video RAM instead of system memory ?

Suddenly a concept only used on their lowest end, cheapest parts, most laughable parts looks like it is ready to be used on the highest performing GPUs that are 2-3 generations away from being manufactured and sold. Likely under a different name though. Just look at the next gen game consoles...

================================================

Perhaps digging up old TomsHardware articles comparing say.... the Intel Celeron 466 (MMX) to the AMD K6-2 500 (Original 3DNow!, MMX) in for example a game (AMDs' strong area right ?) like Quake 2, might remind you how the times change. (For the lazy the Celeron 466 outperformed the K6-2 500 in Quake 2 and significantly in Half-Life [1], which was derived from the Quake 2 engine. It also had a higher price:p erformance ratio than the K6-2).

Some of the older articles here on TomsHardware are "ancient tombs of knowledge", it is such a pitty they are sorted by year, then month, then week, then article.... one can not find what they seek easily here anymore.
January 20, 2006 9:00:46 AM

All I'm saying is: as a product (i.e. investment vs return) Itanium has been/is a failure. If, in ten years, everyone is making consumer level chips based on the architecture that intel spent billions to develop, then post here and I'll give you at pat on the back and say, "You were right."
Did you even look at the chart?
I'll take your advice and read about it for the three seconds that i laughed at it. How exactly am i "encouraging others to dislike it"?
It doesn't take much to convince a fanboy like me:
Quote:
Sure Itanium 1 and 2 are not 'fast' but they are 'powerful' in some areas, and actually cost effective at various 'heavy' tasks. Intel can do much better without even trying

is plenty. LOL
January 20, 2006 9:37:40 AM

I looked at the graph, yes.

Consider the 5 page article on it at Anandtech: http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=259...

Imagine a system with 2 multicore Intel Xeon x64 processors (next gen Xeon, not the P4 based one) and also 6 multicore (but more cores per chip) Intel IA-64 (Itanium 3/4 ?) processors.

Surely you can think of uses for such a sytem.... Even if it is only the next quantum leap in gaming.

Perhaps this link: http://www.chipgeek.com/procspec/procspec.htm ; will encourage you to realise there are computers out there not using x86 / x64 processors...... and the ones that are owe the x86 (IA-32) architecture to who ? (A: Intel)... (Although AMD, and others did extend it, and sell / lic some of their extentions to back to Intel).... In some respects Intel and AMD are still the same company (who do you think started AMD other than an ex-Intel engineer ?) working on the same thing, each 'sharing' research but competing with each other at the same time.

So many processor companies rise and fall, but Intel has the brightest future of them all (Their market share is well over 50%), yet you claim they are 'burning money' on IA-64 ? - Stop and think about it for a second.... or maybe a month if you get my gist. [:P]

Maybe there is a shadow micro-processor company 'INTegrated ELectronics' (Intels original name) still running the show, and Intel vs AMD is just one giant trick using some of the best business / psuedo economic-military tactics.

If everything is going virtual, we'll skip physical robots completely én routé to techtopia, and just get virtual workers in a virtual universe to do all our work for us. (Real world inputs --> Virtual Processes --> Real world output / benefits).

Just read the Anandtech article, think about it, and see if it changes your perspective on 'all that money'.
January 20, 2006 2:29:44 PM

Who says i play games?

I read the anandtech article; it's not bad. Here's a quote:

Quote:
If we ignore Intel's poor execution during the past months and the economic realities, and focus on the architecture, it is clear, however, that the Itanium has time on its side and is most likely the architecture with the highest potential.


I don't see how the other link, presenting cache sizes and retail prices, adds anything.

If you think "all that money" doesn't effect business click here

You are a piece of work.

Ladies and gentlemen may i present: The Itanium Fanboy
January 20, 2006 2:38:14 PM

Wasnt that the failed result of Intel's attempt to create a new metal and recoup their losses? :lol:  (Mixture of Iron and Titanium)
January 20, 2006 2:52:42 PM

Quote:
I don't see how "pure 64-bit" is going to change anything that much.
16 to 32 bit (aka win3.x or 95 to 98se) didn't really change things that much, not for the average consumer anyways.

Take a true 16 bit system and try and run todays software on it. Good luck. When they switched from 16 to 32 bit it didnt take hold overnight. Just like 32 to 64 bit, its going to take awhile for everyone, software and hardware companies, to catch up to make 64 bit shine.
It wont be too long and well be hearing about 128 bit cpus. "Why do we need that" people will say. At the time we wont, but as technology improves and the natural drive to make things faster, the need will be there. Who knows maybe in 10 to 15 years we could be working on 256 bit systems. Time will tell
January 20, 2006 3:00:04 PM

Unfortunatly although the itanium had potential, the lack of adoption has sent it the way of the Alpha. Great processor but no industry adoption.
x86-64 is probably the biggest killer for the itanium.
It's biggest weakness though is its integer performance (it has great floating point performance) and it scalability in high CPU count systems (its shared bus really kills it). Once opteron boards are out there that support 8 or more sockets, coupled with 4 core Opterons any advantage the itanium may have had will be gone.
The majority of servers out there run business systems (Payroll, finance, file servers, etc, etc), these are not floating point intensive in any shape or form.
The only major itanium customer at the moment is HP. HP-UX market share is shrinking, with the greatest growth going to Windows server/linux and AIX.
Companies are starting to wake up to the fact that x86-64 is almost as fast as Itanium and a much reduced price. Microsoft have announced that they will be supporting more server scenios on x86-64 than itanium.
I predict that Sun will eventually forget Sparc and go Opteron (with niagra a niche product). HP have announced support for Solaris 10 on their x84-64 blades and I wouldn't be surprised if HP port HP-UX to x86-64.
I have even heard that IBM may drop AIX for Linux which will run on Power5 and x86-64.
The future is ain't too good for itanium, unless they can drag a massive performance improvement out of the hat (dual core itanium is not living up to its promises as they can't get the frequency up)

Enough of my ramblings
January 20, 2006 7:45:22 PM

My uncle works for Intel, and more specifically the Itanium project. Spoke with hom just last week about it and as he says, it is almost exclusively for server side applications, not yet mainstream. (and was also by the way the first 64bit cpu, not AMD.) I still am not supprised that the Itaniums are low performers in the areas that you are talking about, but that is mostly because it was not designed to do that.
January 21, 2006 12:16:25 AM

Hello, you people!

I have to say something about this [wonder] package: the EPIC/Itanium.
It is not just another ISA; it´s much more than that (taking marketing/financial/implementation issues apart); it IS one of the greatest technological achievements, on what concerns computing paradygms. The Itanium (or "Merced" as its first code name was), is not a CPU, in the usual conceptual sense; it's, rather, a platform in itself, trying to address computing from the inside out (well, sort of...). The Itanium platform relies upon a very strong bet in compiler mastering: the use of [good old] VLIW (Very Large Instruction Word), which Transmeta also used with success (again, marketing/... issues apart), in its Crusoe & Efficeon lines. EPIC (Explicitly Parallel Instruction Computing) or I[ntel]A[rchitecture]-64, as Intel/HP designates it, nowadays, tries to perform - through a very advanced compiler - tasks that other CPUs have to perform in hardware. It's taking the hardware burden into the hands of the compiler, the software...
I think it's worthwile (for those interested, of course!), to have a glimpse of the Itanium's Architecture, for this was the first serious attempt of having computing through software & providing the programmer's [almost] complete control of the 'executing' hardware.
At the end of the day, no matter what will happen to it, EPIC/Itanium will remain as one of the brightest technological paradygms of all times... this century included.
January 25, 2006 8:19:55 AM

Quote:
This new Z-RAM cache concept might give IA-64 some problems down the road though, as one of the goals is to cram as large a cache as possible onto the chip. If 4-16 MB caches on 'consumer' chips become common it might just be the nail in the coffin.

Contact Info available at: http://users.on.net/~darkpeace


Z-RAM may just save the Itanium, one of the issues with the Itanium is the requirement for huge cache amounts as EPIC code is much larger than x86. The x86 code harks back from a day where memory was very expensive so one of the major design goals was very tight code, EPIC on the other hand has huge amounts of code imbedded as the execution order is set at compile time. If x86 processors double / triple cache, you can also bet Itaniums will... 50MB cache anyone?
January 25, 2006 8:25:53 AM

Quote:


Only Sun and IA-64 are offering 128 bit wide system buses. Large boards with support for large amounts of memory (Over 32 GB) require slower speed buses just to work. (eg: Signal integrity).



And that's a problem for them when a 8 way opteron server has a 512 bit wide bus to it's memory....
January 29, 2006 6:51:14 AM

People thought I was crazy for supporting the AMD Opteron once aswell, now I've got one and everyone else wants one.

'The Itanium Fanboy' - hehehe, not quite fitting of me currently as this week I've got no Intel systems at all. Nor do I personally own an Itanium 2 system. However, I do support the platform, just like some cling to the Alpha (even today). Very funny none the less...

I should print it on a shirt 'The Itanium Fanboy' with the logo and everything as a joke. :)  , wear it to work during dress down day, or even a small LAN event to see if I get out alive.

If something goes the way of the Sun UltraSPARC that would imply it was a success, not a failure btw. :p , I am yet to see another single socket CPU which can execute 32 threads at once if you get my gist. ;) 

True what is said about Z-RAM (above) too, it is a double edged sword, it can work for Itanium, aswell as work against it. It has more potential to help Itanium than any other platform to be honest now I think about it some more. (Not that Opterons with 6 MB L2 cache per core wouldn't rock, but Itaniums with 24+ MB of shared L2 cache would rock just as hard. ;)  )

The Opteron only has a 128 bit wide (in the respect is treats Dual Channel DDR as 128 bits wide) bus to memory, using NUMA it can have 2 or 4 such memory controllers that aggregate performance, or the (nVidia Pro 2000 series) chipset can 'interleave' the NUMA nodes (yes, that is the label Tyan used in the BIOS for it on the K8WE S2895) and only present one 'node' to the Operating System. There are pros and cons to each method.

Sun UltraSPARCs, Intel Xeons Itaniums, and other platforms don't lack the potential to do this, or something very similar, either. Sun have been doing it for years.

eg: http://www.sun.com/servers/midrange/sunfire_e6900/speci...
(Just a random Sun Fire server, "67.2-GB/sec. aggregate bandwidth", that looks good to me when compared to the apx 19.2 - 25.6 of a 4 socket Opteron solution). Then bring into account that SPARCs also have short pipelines and don't need huge memory performance. (unlike the Pentium 4 PreScott, which has a long pipeline and needs high performance memory subsystem to keep it 'well fed').

8 way Opterons are just 4 ways using dual-cores. The 16 way ones are just 2 x 4 way boards in one system using dual-cores with an interconnect. Check out: http://www.tyan.com and SuperMicro, etc.

The NUMA system used on the AMD Opteron was heavily inspired by Sun Microsystems and Alpha designs, at the basic level it almost mirrors them.

Recently, IA-64 just had another $10 billion injected into it (after this thread was started, and after my comments above), it is highly likely if Alpha, Sun & AMD (and several others) could build platforms to aggregate memory performance, then Intel surely can aswell. They've got FB-DIMMs, and possibly Registered DDR3 to lean on... You'd think they where preparing for something. [scratches beard].

We'll eventually hit a point (around 2010) where x86_64 requires greater percentage of 'CPU real estate' to be dedicated to decoding instructions (and associated) logic. IA-64 doesn't have this problem (or others, as the compiler does all the work) so more 'CPU real estate' can be dedicated to cores, cores which are smaller than x86_64 cores, thus more of them will fit inside a given area, and they'll have more room left over for cache... with the news of Z-RAM potentially being used as cache this is 'A Good Thing (tm)'.

If the current IA-64 core size grew 50% or even doubled, they could raise performance in any weak areas btw.... I don't think you realise how small these IA-64 cores (excluding cache) really are. They may even run at lower clock speeds than Athlon64 CPUs while doing more work per clock cycle. (much like the Athlon64 does compared to the Pentium 4).

You'd be able to fit between 4 and 8 (say 6) IA-64 cores within the space of a typical 32nm CPU, with enough room for a large (say 64 MB) shared cache to spare, and whatever other features (eg: improved integer performance) they want to cram into it.

Remember that 20% of the code does 80% of the work, and if you can fit 64 MB of code / instructions on the chip performance is going to scale (at least in single socket systems, until they get a nicer platform going for multi-socket IA-64 systems).

To me, IA-64 is just 'a work in progress' that Intel have managed to produce and sell as both the Itanium 1 and 2 (each with various models).

I am sure everyone here is aware how the IT market can change in '5 short years', the IT 'market' started around 1985. Sure computers are older, but the mass market, including business adoption of IT, I feel, started around 1985 with the introduction of 'high performance GUIs' [cough]. So the 'large scale' industry as a whole is only 20-21 years old. In another 5 years would be 25% older.

PS: If it catches on, and there is space in my sig, I'll update it to say 'The Itanium (& Opteron) Fanboy' !. :p 

The main problem is software support, 'real' development studios for IA-64, and encouraging developers to port applications over to it then optimize them for the platform.

Poor example of the top of my head: There is .NET 2.0 for x64 platforms, but there is not .NET 2.0 for IA-64.... Windows XP 64-bit Edition for IA-64 (not x64 Edition) is has also faded away.

When workstations (and associated) software start disappearing for a particular platform one would wonder about its future yes.... and perhaps the return on investment (so far anyway) has not been that great.... but $10 billion is nothing to be scoffed at. Esp when most suspect that around the time 32nm - 22/23nm CPUs are released x86/x64 will start to lose it's edge, when it does AMD (for example) won't just 'invent' x86_128 to counter act it, it wouldn't work. x64 (EM64T/AMD64) will be ample memory address wise for a very long time, but its performance may stop scaling around 2010.

x86 is holding CPUs back, and making leaps in performance difficult even now. In the last 2-3 years x86/x64 hasn't shown any major improvements in performance. We need to encourage developers to make software with isolated threads for it for performance to scale. (Adding more cache to x86/x64 won't help it much, and having 4 cores per processor won't provide the 4x fold increase in performance because of the software).....

.... you may as well introduce a new architecture if software needs to be recreated, market it to the world, phase out all support for x86/x64 and leave it to 'security risks' and poor backwards compatibility, (x86/x64 isn't scaling, but for different reasons to IA-64 which can have the 'training wheels' taken off then start scaling very well), then brainwash the world into IA-64.

Everything can be brought.... for a price. Intel have a very strong influence as they have vast amounts of cash compared to their competition. They are also the only ones who can make dual IA-64 / IA-32 (x86_64 EM64T) systems to run both sets of software.... phase it out gradually pushing various 'consumer benefits' the whole time.

Most people don't know much about computers, so if Intel start a large scale campaign they'll just follow along like sheep. Just look at the old Pentium campaign, some people actually think that 'Pentium' = 'Computer' and don't ask about anything else except 'the latest Pentium'.

Intel are removing the Pentium brand and have changed their 'Intel Inside' logo, which means they want to prepare to start a campaign for a new CPU (Intel Core Duo, and Conroe, etc in Q3 2006), then 'something even better' when they feel the timing is right. When every other non-x86 based platform is dead... totally dead... Intel can leverage 'alternative Intel' platforms without worrying about 'problematic' competition or negative feedback doing the one thing they wouldn't attempt because everyone else was doing it. (ie: Their 'Not Invented Here Syndrome' won't be an issue).
January 29, 2006 9:34:56 AM

If bunny pwns me then Lum-chan can pwn the bunny, or wear a bunny suit which would be even cooler: ;) 

(Yes, I know it is off subject, but I think people reading IA-64 stuff need a llittle pick me up every so often).

(Failed to paste, just open link in text editor using a fixed width font)
http://users.on.net/~darkpeace/asciiart/Lum-chan%20-%20...
!