Slashdot Mirror
Intel Shifting 64-bit Plans
OS24Ever writes "News.com has an article stating that 'Intel plans to demonstrate a 64-bit revamp of its Xeon and Pentium processors in mid-February--an endorsement of a major rival's strategy and a troubling development for Intel's Itanium chip' Is this the end of Itanium?" Looks like the rumors were true.
Well we all saw this one coming with all the delays on the Itanium.
Intel has already publicly admitted to having X86 processors with 64 bit extension in development. Also, take a look at microsoft, who refer to X86-64 as "64 bit extended architecture."
Everybody and his brother figured out long ago that Itanium is not something that will penetrate effectively into the desktop market. It's hot, expensive, incompatible, etc. It requires a ton of work to get code running smoothly on Itanium. Th only amazing thing is how long it took intel to admit that it had egg on its face!
I'm afraid that we'll see Microsoft leave AMD waiting at the altar for the Win64 OS only for AMD to find out that MS has eloped with Intel for their 64 bit X86 variant.
I doubt that the US justice department and antitrust will have any bearing on such a move.
Anyone else?
While I would love to own one of these, unless Intel releases chips with CT (the 64-bit stuff) and not LT (LaGrande Technology - Palladium/TCPA "security" support), I will be sticking with what I have or buying a PowerPC (GNU/Linux either way). Hopefully LT will be optional like HyperThreading is now, but given that many of the "features" it enables require control of the majority of the market, I seriously doubt it.
So what did you do with all you had was a 286 @ 16 MHz to use?
They don't 'need' 64-bit machines, the machines were built before the software was written. It is the way things work...
If I point out that you are incorrect, making me a foe does not make you any more correct.
I agree. I think Itanium will catch on given two things happen: first, Intel needs to lower the price - I just bought a 1U server and would have loved to have gotten an Itanium - but it's prohibitively expensive compared to P4s or Xeons. Second, the x86 emulation is going to have to improve if Itanium is going to gain acceptance as a desktop cpu (granted if this happens, there will likely be an Itanium clone much line athlon-64 and the opteron). Otherwise I think Itanium is likely to go the way of alphas - a great design but overpriced and with poor application support.
I suspect your probably right. 64 bit x86 is what Intel is going to use to counter the appalling sales pitch that "We're selling you a 64 bit processor while that Intel based machine only has a 32 bit processor". That's going to be heard far and wide once they find themselves being compared to Apple G5's and Opteron based systems. Itanium wasn't going to be the chip that helped that situation. What little most consumers have heard about it wasn't good regardless of whether or not it was true or fair (and it might have been both, that's beside the point) so Intel's going to get themselves in that 64 bit crowd (regardless of whether or not it means anything to the consumer using the computer and again, that's beside the point).
Intel knows as well as anyone that consumers prefer the bigger number and like to see the word "compatible" on things they buy.
Appended to the end of comments you post. 120 chars.
So what did you do with all you had was a 286 @ 16 MHz to use?
---
When all they had was a 286 @ 16MHz, they didn't do large-scale simulations of molecules on the computer, or design airplanes mostly on the computer. 64-bit machines already exist, and the software to take advantage of them already exists --- people want to be able to do the things they do on current 64-bit machines on commodity hardware.
A deep unwavering belief is a sure sign you're missing something...
The real question is have they finally dumped the stupid x86 instruction set in favour of a space/energy/coding efficient RISC set?
Well, as others have pointed out, that's passe. But the Itanium is whole new fancy thing building on the lessons learned -- "EPIC" (explicitly parallel instruction computing, or some such).
Sure it would take time and money but in the end [snip snip snip...]
Yep, time and money but in the end. That's Intel's plan with Itanium, and they've still got some hope for it. In the meantime, AMD was going to eat their lunch (and breakfast and dinner) with x86-64, and it'd be silly to just sit there letting that happen. Sure, they *could* just wave their hands frantically and gush about how superior the new is to the old and hope the market listens -- but as real-world experience shows, refactoring usually wins over rewriting from scratch.
Maybe this will help get MS off their asses to put out a 64 Bit OS for Non-Intel (i.e. x86-64) I have wondered why MS has drug itself so slow when they had an Itanium version some time ago soething contractual with Intel ?
:)
32 Bit vs 64 Bit MS operating systems , TWICE as many chances for bugs
No, this does not signal that Itanium is doomed.
I have to agree. What really signals that Itanium is doomed is the fact that no one is buying it.
But you gotta dig the irony: Intel is making an AMD-compatible processor.
One seriously cannot underestimate the significance of binary compatibility. Nowadays The external ISA is a silly detail anyway. Any processor worth the silcon it was made on has a RISC microarchitecture.
Whatever, man. I have G5 and Itanium2 machines at my desk. The HP Itanium2 runs Linux and WinXP 64-bit edition (which came out last June). The Itanium2 (McKinley) is an old slow one that crushes the G5 easliy on everything (using Intel's compiler) by factors of 2-3x. The new Madison Itaniums are substantially faster (look at the SPEC CPU benchmarks). The Itanium is far superior to anything else out there, it just doesn't run x86 code all that fast, and the GNU compiler sucks on the Itanium because the optimzier cannot get the VLIW right. The Itanium is just ahead of its time. And most people are too stuck in the x86 mindset to even see it. CPU buyers lose as a result.
Forgive me if I get a little bored by this 'revelation', I wrote about it in September:
:) )
http://www.theinquirer.net/?article=11668
And I followed it up a week later with this:
http://www.theinquirer.net/?article=11781
Ok, people seem to not taken me seriously then, so I'll reiterate. Prescott has 64 bit extensions built in. They use the AMD64 instruction set. This is because MS twisted their arm into it.
The question of when they turn it on is more a political one than a technical one, and that I don't know the answer to right now, most likely because Intel does not know either. They are in one hell of a bind. If Prescott is 64 bit, why should I pay 5x as much for an Itanic again? Oh yeah, a marginal performance gain on FP code, but a loss on Int. Whoopty-#&%^#-ding-dong.
It will be announced at IDF, count on that. When you can buy it, good question. My guess is that it will be an inticement for the first Prescott/EE buyers.
-Charlie
(As a self-plug, if you read the Inq, you would know these things
Intel started working on this years ago, but it got hushed up due to political issues between the Itanic team and the Prescott team. It's gotta be Opteron compatible at this point. If Intel management hadn't been stupid, then they could have announced their own x86-64 and put the hurt on AMD. At this point, their only option is to "embrace and extend".
I just did some work porting CFD code from IRIX to an Opteron system running Linux. The processor had to be 64 bit because some of the runs we do now require 15 GB of RAM.
I only did the porting work - I only have a vague understanding of how CFD works. So I can't say what percent of the runs require more than 4 GB of RAM, but I've gotten the impression that most runs require over 2 GB of RAM, which is enough to complicate things with a 32 bit OS.
software we used at my company required 64 bit precision for accuracy. That would be painfully slow on a 32 bit machine.
Note, that all modern processors already have 64/128bit extensions, which most compilers will use. 64bit processors won't be any faster at double-precision FP operations.
I imagine this process is well underway, and nearing completions for the Solaris 10 release.
Meanwhile a port of HP-UX is imminent if the Itanium tanks. Take the x86 port effort + 64-bit clean IA64 version and mix together and you get the Opteron optimized version (well, it's a wee bit more complicated than that...)
So we'll have Darwin, *BSD, Linux, HP-UX (probably), Solaris, Windows NT 5.2, zow! All that's left is for Apple to port the GUI, and we'll have a cool platform for the future.
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
The reason IA-64 is destined for the scrap heap is the price. Given that I do some memory-intensive things I wanted a 64-bit machine for my new server. I bought an amd64 box simply for one reason: cost!
amd64 is a clever extension of x86 and probably the biggest win are the new registers - that results in quite a performance boost. But IA-64 was just pretty damn cool.
I wish M$ didn't strongarm Intel. If the current headaches of the myDoom worm have taught us anything: architectural diversity is a good thing. I would rather there be 3 64-bit architectures out there (IA-64, amd64, and PPC64); heck bring back Alpha! It means that its one more stumbling block to writing virii.
FWIW, my amd64 box is fast and handles huge jobs easily. With a couple of gigs of swap thrown on it even my Lisp code runs fast (amazing).
See, hearing about things like this pisses me off.
... it's kind of depressing.
... it hasn't taken off. The real pisser, though, is to think that the dominant 64-bit architecture of the future might essentially be i386 with more and bigger registers. Hopefully at least IBM will step in with its POWER-based solutions. Man, if I ever get drunk and start bitter rants, I swear it'll be about processor architectures ...
When I think of all the nice system lines that have died off because their parent companies decided "Well, we could just have Intel make our 64-bit chips, and then make money selling systems", and all the technically nice architectures that are basically dead now because of decisions like that (MIPS, Alpha, et. al)
I mean, I wouldn't mind if Itanium had been more successful. It was actually neat to think of Digital's EV8 team building SMT technology into Itanium. (Is this the work that's been manifested as HT on P4 on Xeon-class machines?) Especially since EPIC is supposed to make things so much different. But
Andy Glew (the designer of the Pentium Pro) on EPIC vs. normal architectures:
"Yes, but the IA-64 EPIC is not a modern architecture -
it is a design by committee, with microarchitects who believed
religious dogma instead of thinking.
At least some modern microarchitectures have made optimization
easier than in their predecessors. Apart from some egregious
glass jaws (mea culpa), P6 was often less sensitive to optimization
than the P5. The compiler folks complained that their unoptimized
code often ran as fast as their optimized code.
AMD's K7 and K8 continue in this vein.
This is one of the reasons I jumped from Intel to AMD:
the Intel P6 is philosophically a lot closer to the AMD K7 and K8
than it is to the Intel Pentium 4 (Willamette, Prescott), or Itanium.
Pentium 4 is fragile, just like Itanium."
Look at how many trouble Intel has trouble admitting they are wrong and following AMD on something. Intel is so used to AMD copycating them on everything, they don't seem to know how to deal with the shoe being on the other foot.
So far Intel has followed AMD onto DDR memory, after dragging their feet for a year. Now it's happening with 64 bits. Next expect to see it with integrated memory controller, desktop dynamic power management(like quick 'n cool) and hypertransport. I'm sure when they come around the technologies might be similar, but they'll have some other name for it. Hopefully, Intel doesn't try the old Microsoft embrace and extend.
Argh!
This is very bad news for a computing purist.
We get to live with the ugliest hack ever, for
yet another 20 years.
Please burn all CPU designs that contain the
A20 gate.
Bram Stolk http://stolk.org/tlctc/
Will Intel's extensions to the x86 architecture be compatible with AMD's. Or will fat binaries that can execute in x86, x86-64 Intel and x86-64 AMD be necessary?
cheap labor conservatives - they want to keep you hungry enough to be thankful for minimum wage.
OK, how about a modern load/store architecture that isn't starved for registers?