Slashdot Mirror
Intel 64-bit Announcements at IDF
cribb writes "Some fascinating stuff is going on over at the IDF. Ever since the first sneak previews of the Opteron, there has been lots of uncertainty around its future, and that of AMD. AMD have bet everything on the success of their new 64-bit CPU, and with Microsoft severely delaying the release of a 64-bit Windows, and Intel complaining that 64-bit processing has no place in the desktop market, things were starting to look dim for AMD. However, after rumours around the 64-bit extensions of the Pentium 4 EE, it became clear that Intel is not willing to lag behind AMD in the 'innovation' department. Now comes the shocker: Intel boss Craig Barrett today anounced that Xeon-class 64-bit server CPUs codenamed Nocona will be coming out the second half of 2004. It isn't clear whether they will support AMD's Opteron AMD64 extensions. Barrett is quoted saying, 'There will be one operating system that will support all (64-bit) extended systems.' Maybe 64-bit computing is right around the corner after all, and we may even see compatible instruction sets from Intel and AMD! And does this mean that Intel will be dumping Itanium, which never caught on as expected in the server market, and forget the billions spent on developing it?" See some other articles at EE Times, and EWeek.
...and does this mean that Intel will be dumping Itanium, which never caught on as expected in the server market[?]...
I'm sure it was an interesting intellectual exercise, and that they learnt a lot.
--
FreeNET user? Comfortable with the adverse selection?
There will be one operating system that will support all (64-bit) extended systems.
He's right. It's called Linux.
Intel may add an instruction or two but one version of windows, or linux will run on both chips.
Intel wants to save some pride and distance themselves from "AMD64" as much as possible.
This is a good thing, whenever someone plays catch up, they alwasy seem to develop a better product than if they were at the top. Take for example how IE6 has slowed improvements while other browsers continue to create. A little competition is a good thing.
Because it will sink the Itanic
Nyuk Nyuk Nyuk
-Master Switch, one more element in the machine
"Intel complaining that 64-bit processing has no place in the desktop market," Can't 64-bit processors use more memory at a time than 32-bit processors? I don't understand...
As for one operating system, who? They in cahoots with Microsoft, after Microsoft dragged it's feet on AMD? Sounds like collusion, anti-competitiveness, and all that.
A feeling of having made the same mistake before: Deja Foobar
Intel had to play 'catch-up" by incorporating MMX into the Pentium when NexGen was plotting on incorporating their own SIMD system (which became 3DNow!) but this time, they really got screwed over. They had planned on Itanium taking the 64-bit market over, and did not figure on AMD's x86-64 at all. What really did Intel in this time around was that AMD was doing what Intel had traditionally done, continue the backwards compatibility long past any logical point and not only making it work, but making it attractive. This is the mis-step that brought Motorola down from it's "king of the desktop CPU" position, when they released the 88k as the "next-generation" CPU rather than focus on delivering better 68k's. The division of resources back then is a step Motorola never really recovered from. I wonder how Intel will do on it.
Karma Whoring for Fun and Profit.
Oh yeah, it's on my list of Top 10 I.T. stories along with the new version of $MS_PRODUCT: is it worth the upgrade? and Linux: ready for the desktop? At least I was finally able to take off Latest Netscape skull raping by AOL off of it.
Mix the failings of Usenet with the shortcomings of the World Wide Web and the result is slashdot.
For all the power of new processors, you'll need a new version of Windows that will take more resources to do the same things.
Compare Win95 vs XP on a P4 and you'd see. Then again, we all use OSS right?
I think it is a pity, that the alpha processor (that was once the best processor) had to die, just because HP and Intel wanted to succeed with their Itanium processor (and are now failing).
Felix
/dev/earth not found. Reboot?
It's nice to see that Intel's not just sitting back on past glories and thinking that'll solve everything for them. With AMD and Intel getting so competitive, and comparable products from both coming out so close together, it can only benefit the consumers.
"Although this means that Intel could bring a 32/64-bit chip to PCs soon, Barrett said the company has no plans do so in the near future."
Right, so introducing a 32bit/64bit "server chip" is absolutely NOTHING like introducing a "desktop chip". They still clearly are pretending that they are not competing with AMD's strategy. Who are they kidding?
Linux? FreeBSD?
I'm still far more interested in the Sandal Platform(1/2 way down page) and what convergence can do for me. I still dont see 64 bit applications (real applications, not just games) catching on for a while...
The Doormat
If you're not outraged, then you're not paying attention.
So after the Apple 'first 64 bit desktop' campaign we get to see an AMD 'first 64 bit desktop' _and_ an Intel 'first 64 bit desktop' campaign?
In the mean time my 1998 vintage Mesh/Alpha desktop system (no, it's not a server, it was sold via consumer magazines in the UK) is still running happily with 64 bit Linux... and that was hardly the first either, an honour that probably belongs to someone like Sun.
Beep beep.
I had a similar issue when they were just delving into RISC and producing good ole x86 chips at the same time. They decided to scale back the RISC and dive into x86 and it worked out for them, they recognize the need to research both and look forward and move, although some money is lost, lessons from the Itaniums will go on even if they do die, which I doubt. Intel will do what it needs to survive and most likely stay king of the desktop market.
Cheers for AMD and their success wit x86-64.
Completion is best for everyone in this game.
While it sounds like something out of the Lord of the Rings trilogy, one must wonder exactly what he meant by that comment. MS just recently released the beta of XP for the 64-bit Athlon. If there's going to be '... one operating system that will support all (64-bit) extended systems' are they referring Linux which to my knowledge is the *only* OS/kernel that supports almost every danged chipset/platform out there? To date, no one has seen an MS offering for 64-bit x86 other than the beta Athlon released already mentioned.
It'll be one hell of a backtrack if they do drop Itanium. Yet it will be hard to keep Itanium viable with another 64bit chip that is (presumably) much better at handling x86 code.
What this really signals is that Opteron, and AMD64 are really quite impressive indeed. It's billions that Intel will be dropping so they can compete with it, and you don't make that sort of move unless you're really very very worried.
As to whether they will be compatible with AMDs extensions: I suspect Intel won't be ale to bring themselves to that. The "One operating system will support all 64bit extensions" sounds more like a deal has been cut with Microsoft to make the 64bit version of windows work with Intel's 64bit extensions as well of those AMD. In practice I suspect that means Intel will be very close to AMDs extensions, with a few quirks, and the intention of trying to grab the market and drag things away with their own extra extensions with newer chips.
Could this be behind the slowness of 64bit windows for Opterons?
Jedidiah.
Craft Beer Programming T-shirts
It's actually a 32-bit chip with some horrendous rounding errors.
EE Times is also reporting that Intel may be pushing a new kind of RAM interface to compete with existing DDR and RDRAM. At 2 Gbit/sec per wire, this is about twice the speed of current RDRAM and four times the speed of DDR SDRAM. But, more interestingly, this is a point-to-point architecture - unlike the traditional bus architecture, when you add more memory modules you can get more bandwidth. Also notable is that simultaneous bi-directional communications happens over a single wire. Infineon and Samsung have made test chips, and results are to be released at the International Solid State Circuits Conference today.
I wonder how this figures into their processor/chipset roadmap...
HIV Crosses Species Barrier... into Muppets
The world has changed, there is no more goatse.cx and this will soon be followed into oblivion by the Itanium
........
sux to be Intel
* Carthago Delenda Est *
If you have one of AMD's 64 bit processors you can get a prerelease version of the operating system to try out. Info & a signup link are available here.
Actually, it was made clear during the Q&A at IDF that the instruction set would be compatible with the AMD64 instruction set that AMD pioneered and which Microsoft has already built a 64-bit version of Windows around. Intel will undoubtedly have some 'additional' instructions included, making theirs a superset of AMD64 but the main point is that you will be able to buy one version of 64-bit Windows and install it on either an AMD or Intel-based machine. Now its' just a matter of timing. I would not expect MS to do the full release of their 64-bit Windows until Intel has the matching hardware in the pipeline, curtailing AMD's current lead in that market segment.
======= ~\_/~\_O Burmese
Sorry, but mainstream 64-bit was around last year's corner (AMD64/G5).
Posting anon due to affiliations.
Anyone that didn't think Intel had 64-bit Workstation and Desktop chips "in the pipeline", as it were, must be sitting in a cave humming with their fingers in their ears.
The production pipeline on these sorts of products. take years, so this was not a knee jerk reaction. If you look very carefully at what Intel has actually officially said the whole time, you'll see that they simply said they would provide a solution when the appropriate OS support and perceived need becomes available, and that is EXACTLY what has happened here. What do you know, Steve Balmer announces Windows XP 64 now has support for these "Xeon" extensions. These things don't happen over night.
It is still a fact that most people DO NOT need 64-bit computing in any way shape or form, but one mistake that Intel did make is the fickleness of the vocal minority and AMD fanbois.
Also, if you think that the existing Prescotts don't already have these extensions (just disabled at the moment), you are also kidding yourself.
Itanium -- incompatible with existing software, expensive, great speed, not catching on in market
Opteron -- compatible, inexpensive (relative to Itanium), also great speed, initial sales are good
Result: Intel releases server-class 64-bit x86 CPU.
Bottom line: Itanium is dead.
Rule #1 -- Politics always trumps technology.
It's asking a question about a statement in the story blurb.
First, with the release of AMD's 64bit processors, Intel dismisses 64bit desktop computing. Now after AMD's tremendous success, they've had a change of heart. How convenient.
Do Not Eat iPod Shuffle
The The Inquirer has some pretty decent (if biased) coverage of this.
Essentially there will be a single OS for the two (Intel and AMD). Unspoken is that Intel's implementation is AMD64 ISA, but a different technical architecture. If it's compatible, who cares. Secondary confirmation via Ars Technica
AMD is the only X86 chipset manufacturer offering 64bit notebook chips. They're clearly seeing the light and hitting a market that Intel's been struggling in for close to 4 years. Intel's claims of no need for 64-bit personal computing is just a smokescreen for their 64-bit failures. As technology advances we will have 64-bit personal computing... and a few years or decades later we'll 128-bit personal computing. Intel just doesn't want to lose face to AMD since AMD is first to market and posting profits.
News.com article
Intel's 64 bit extensions are compatible with AMD's. You will be able to run the same 64 bit OSes on them. Intel's 64-bit capable Xeons are Noconas, which are Prescotts in a Xeon package.
I work for Intel, but I do not speak for Intel. My opinions are not necessarily the opinions of Intel Corporation.
Jonathan Pearce jonathan@pearce.name
3EAAFB2A http://www.jonathan.pearce.name/
Try again, 64 bit is very useful for lots of things. Keep in mind that when you 'offload to a 3D card' as you so easily put it, you're using a largely specific-purpose processor. This means that you've got to be in the canon of algorithms that the hardware-maker thought you'd use. A general purpose 64bit is very useful.
In other news, Intel has bought all rights to the Nintendo 64 ;)
got sig?
Well, many calculations in academia are done on "desktop" computers. Some of the calculations done in the lab I work in can easily gobble up more 4GB of RAM. A couple of weeks ago we were looking into our options to address this problem at a reasonable price. Speed would also be nice when you have to cruch that many numbers, but if you don't have the RAM you can't even wait longer to get the results.
Not counting legacy PC architecture goofiness, 32 bits currently provide a 4G addressable space. So, apart for power-users, servers, hardcore gamers and trendy techno-posers, what's the advantage of running 64-bit systems? Sure you can make biggest calculations in one instruction, but overall you have to move twice as much data around to achieve the same thing if you have less than 4G or RAM.
Yes I know 4G of RAM is getting increasingly common, but is it really needed? just because Windows is as thick as a whale omelette doesn't mean you need that much to achieve the same result.
Honestly, I could understand the need to have more than 8 and 16 bit processors, to make multiprecision calculations less necessary for common things and to avoid segmentation kludges, but for the majority of people (i.e. people running Word and Excel, and playing Minesweeper a little), I don't see the interest at all. Better have good fast cheap 32-bit systems than expensive, underused 64-bit ones. Unless of course future versions of the Windows require that much power, which doesn't even seem likely for the short term.
"A door is what a dog is perpetually on the wrong side of" - Ogden Nash
There was a time when people got by with 8-bit and 16-bit processors. True, 32-to-64 is a much bigger leap than those jumps were, but technology keeps advancing.
My old modem had two processors and a slew of analog filters - now modems are just codec that allow the main cpu to do all the processing. Same is true with many sound cards (they used to have DSP coprocessors or specialized hardware). My old laptop had a MPEG decoder card, but now that's done by hardware. (get the trend I'm trying to show yet?). Pretty soon a video cards that is just a D/A will satisfy most users, like the equivalent WinModems.
I know 64 bit CPU's can access lots more memory.
How else does 64 bits make things better/faster?
(simple language please, i'm a techno idiot)
Its nice to see Intel eating a healthy portion of crow. After all, just a few months ago Intel was telling the press the they saw no need for a 64-bit desktop processor until 2010 or so! And even now, their justification for being so late is that they were waiting for the tools and infrastructure for 64-bit to become available. I guess that explains why they didn't release the 80386 until Windows 95 was released. ;^)
CHeers!
"The only good windmill is a tilted windmill."
You do realize that failed business ventures can be written off on taxes, right?
Just because Intel dumps Itanium doesn't mean they lost billions. They just chose the 'financially better' path.
Intel won't say if it has licensed AMD's x86-64 extensions. But Barrett seemed to hint that Intel's technology will be somewhat less than completely compatible with AMD's instruction set.
"For the most part, (software) will run on both systems," he said. "Intel has some (things) unique to Intel, which we will make sure people write, port and tune to."
--
Sigs are for geeks
i've been happily running linux on my alpha since 1997. ;)
Heck, who needs 32-bit computing? Why don't we just ditch our Pentiums and Athlons and go back to 286s? After all, 16-bit computing is enough for everyone, right?
Wake up bro. Smell the coffee. 64-bit processors are the future. And, one day, they'll be standard on new PCs and 128-bit processors will be the future. It's called progress.
Perhaps you preferred it when computing was all punch cards but some of us quite like these new-fangled PCs that are all-singing and all-dancing and that have visual displays. I know, I know, it's so "modern" of us, but we're living in the 21st century and proud of it.
"Accept that some days you are the pigeon, and some days you are the statue." - David Brent, Wernham Hogg
If IA32e runs the OS and software that is already out there, it has to be compatible with AMD64. Intel being as huge as they are can completely skirt the fact that they have been *forced* to adopt AMD's strategy for moving x86 forward. Not that it really matters to their bottom line.(Dictating the instruction set has suprisingly little economic value.)
The Slashdot crowd should however be clear on what is in fact happening at the instruction set level.
It also appears Intel is going to more or less get a pass on how badly they miscalculated with IPF (Itanium). Maybe they'll succeed in turning it into a few billion a year of a business and leave it at that. But clearly after all the investment the industry has poured into it, Itanium is amounting to very little. (E.g. one has to peg SGI as a clear loser here.)
The issue of supporting 64-bit address space on x86 was almost certainly hotly debated within Intel. There is much commentary on the company to be had here. That they spent so long pushing the Itanium boulder up the hill is a great fault. That they are adopting AMD64, the instruction set the software vendors and customers actually want, is a testament to their flexability and attention to the market.
Intel's shortcommings in the IA32 to IA64 switch were the following:
In the meantime, AMD took the evolutionary path and provided the 64-bit capability from desktops, to middle tier servers and higher end machines. They implemented an architecture that directly executes the IA32 but that was extended to the much needed now 64-bits. The performance / price ratio are much better than that of Itanium's and compilers were much easier to come about since the x86 ISA is a well known one.
There is no surprise that AMD made the right strategic move to provide the needed missing link in the evolution of the popular (but crappy) x86 ISA to the 64-bit arena. There is no surprise either that heavy weights such as IBM, Dell, SUN and even HP -- who pretty much designed Itanium -- put some of their eggs in their AMD busket.
And there is no surprise that Intel realized after the fact that it should had provided the missing step and it is now playing catch up.
Isn't unbridled competion good? The pervasiveness of Intel forced the AMD and the RISC designers to do their best to improve their own designs which now in turn are forcing Intel to improve its own?
The same story with UNIX/Linux and MS windows.
People need decent alternatives to chose from. Forced monolithic single-vendor solutions are bad for everyone.
Itanium here Billions wasted on effort Cash flushed down the drain
There's a growing sense that even if The Future comes,
most of us won't be able to afford it.
-- Lemmy
someone kill him.
The other big news today was AMD's announcement of the HE and EE (wtf they mean is anyone's guess) of low-power Opterons. With these lines you get a full-scale Opteron that only puts out 35 or 50 watts! True they're expensive as heck, but they seem perfect for blades and other large-scale installations where power and AC requirements cost more than the CPUs themselves.
More information: AMD, Intel at xbit
Discussion: AMD, Intel at Ace's
I've always thought it unlikely that Intel would be caught off guard by AMD's Opteron. I think Intel could have announced this earlier, but wanted AMD to become overconfident with its Opteron and spend oodles of cash etc. on developing public awareness of 64-bit computing, explaining what it is, convincing people that it's worth the upgrade, etc. Then, after AMD (who is already cash-strapped) puts all its eggs into the 64-bit basket, Intel finally comes out and says "Thank you for raising public awareness about 64-bit computing for the desktop for the past year, AMD. Now that you have no more money, we will now announce our 64-bit chip and compete with yours." Here's a list-form of Intel's strategy:
1. AMD comes out with Opteron.
2. Intel waits.
3. AMD spends all its money and resources on promoting 64-bit computing, thinking this will make Intel look obsolete and make themselves the chip-maker of the future.
4. Intel waits.
5. Intel releases own 64-bit computing and takes over the market that AMD spent all its money developing.
6. (AMD pulls out empty pockets and holds them like wings and wonders what happened:) ?????
7. Profit for Intel!
8. I cry.
The Register compares Itanic to the i432: "Bob Colwell, chief architecture honcho for the chip that saved Intel in the mid-1990s, the P6 (Pentium Pro), described the i432 as 'a wonderful research project masquerading as a bad product'."
Since Sun said they would support Solaris on the AMD 64-bit platform, if Intel's instruction set is the same or a super set, does that mean we'll get 64-bit Solaris on Intel?
Still, Sun and Apple should merge, port Solaris to the PowerPC, layer Darwin on top of Solaris and start shipping low TCO multi-user systems (e.g. E5k) that use the Sun-Ray stations to deliver the Mac OS Interface & apps to hundreds of users, finally breaking the MS monopoly on the desktop!
Wrong. 64-bit computing is ten years old with the Alpha, including PCs running GNU/Linux. Not to mention the later UltraSPARC, PA-RISC 2 and MIPS workstations.
And today we already have the PowerPC G5.
This all proves Wintel is the biggest drag in Informatics.
Leandro Guimarães Faria Corcete DUTRA
DA, DBA, SysAdmin, Data Modeller
GNU Project, Debian GNU/Lin
Sure, 32 bit is fine NOW but in the future there will be more intensive software that will require more RAM.
This is just a guess but I wouldn't be suprised in "Longhorn" will be native 64 bit. Maybe Microsoft is waiting for the hardware to catchup so their inefficient code can take advantage of more memory. (I know, it's a cheap shot at Microsoft)
I'm doing many simulations, and so are my fellow students. Modern CAD packages for doing MEMS, nano-tech work with high resolution scream for more RAM. 2GB is barely sufficient, and anything I can feed it is a worthwhile sacrifice. None of our labs can afford Itanics. But we sure can and do need more than 4GB (3GB if windows). I've been advising people to get Opterons whenever they are about to upgrade their systems in order to have an upgrade path in mind.
All bow to his Noodliness!! His Noodle Appendage has touched me!
For whatever reason, the world still tends to ignore anything but Intel and friends' processors. Intel and AMD working on negotiating a 64-bit desktop standard does not represent the beginning of 64-bit on the desktop. IBM and Apple make a pretty nice 64-bit desktop machine thats been out for months.
Like them or not, this is a really strong strategic move for Intel. AMD played its hand in the open the entire time, while Intel waited to time its announcement until just after the Opteron launch.
Now corporate shops will hesitate to commit to AMDs 64-bit chip because of speculation about whether its instruction set will be compatible with Intel's.
Pair that with Microsoft's delayed 64-bit Windows launch, and you've got another well played round from Intel.
It's likely that this was only a contingency plan, though. I'm sure Intel would have liked the world to move to Itanium, but this is their second choice of outcomes. AMD gains some market share for a brief moment, but then potentially has to backpeddle to design Intel compatibility into their already completed 64-bit offering.
This is a manual virus. Copy it to your sig and help me spread!
apparently they are compatable: http://arstechnica.com/news/posts/1077051210.html
Everyone needs to keep in mind that Intel are positioning this as a Xeon able to use a 64bit address space. They have made it clear that the Itanium actually does 64 bit processing, and have implied that this does not.
:-?), but you are still dealing with 32bit data, with 32bit instructions, 32bit registers and 32bit cache.
In short, you can access big memory (although the limit is 1TB; I thought the 64bit address space went higher than that
I'll admit to being an AMD otaku, but this still doesn't do all the things the Opteron and Athlon64 do. No 64bit processing, no higher bandwidth, no more registers, no high-precision floating point work, no on-die memory controller, no high-speed interconnect like Hyper Transport.
There seems to be some question as to compatibility with x86-64 as well, but that needs to be cleared up. Put simply, Microsoft aren't going to support another 64bit instruction set, so it's x86-64 or IA64, nothing new. If it is just x86-32 with larger addressing, then it's not even a true instruction set anyway and so MS don't need to make big new changes.
Finally, I'd like to point out that it can't be another unique 64bit instruction set, since the Linux kernel currently has no preparation for any new instruction sets from Intel. Even if Intel have 10,000 patches to get a new instruction set working and submit them to Linus, he won't be happy, and it'll take them years to get them all in. In short, if they want to see Linux running on their 64bit Xeons, they need to be x86-64, IA64 or x86-32 with addressing fixes. They'd be stupid to make a 64bit server chip and ignore Linux support these days. Gun->Foot.
AMD screwed up and set the default integer size on their X86-64 to 32 bits. This causes all sorts of problems
with pointer math since most software assumes that an integer is large enough to do pointer math.
Their are always good sounding arguments in favor of every bad decision; it there weren't, sane but ignorant people wouldn't make bad decisions.
If Intel sets the default integer size on their X86-64 to 64 bits they stand a chance of winning the marketing fight with AMD, even though they are late to the dance.
okay for everyone who does not know and is posting about "I dunno why u need 64bits yuk yuk". there are more general purpose REGISTERS available to the system. In compilers and OS's you keep having to do this silly trick of "load/store" data from/to the very very few registers that are availble to you. So this means your OS/application/net/system EVERYTHING stops so that you can go all the way out to memory to grab something, load in into a register to finally get back to work. God forbid it is not in the L1 cache, else you will spend hundreds of clock cycles sitting there doing nothing. some of the RISC chips like MIPS had like 20 general purpose registers they kicked ass because the OS used a few, the apps used a few and when you had to switch between apps/kernel space you did not spend tons of time load/stores..... ssssooooo amd64 have a huge bonus in speed because of all the extra registers available to it in 64bit mode. This will become faster and faster as the gcc folks optimize the compiler to start using more of them. Sorry I just am tired of everyone posting that it is a waste because they dont use 16Gb of main memory. ps: posted from a dual operteron 242. it kicks
--jboss
'Maybe 64bit computing is just around the corner.'
Yup, can't wait for a good SG...oh...
well...maybe soon Apple will...oh...right...
well SURELY AMD will come out with...oh...you don't say?
Why is it that just because 'Intel' hasn't come out with a widely accepted product, many people act as if it doesn't exist???
Don't park drunk, accidents cause people.
The post says:
Barrett is quoted saying, 'There will be one operating system that will support all (64-bit) extended systems.'
WRONG. Barrett did not say this, a Microsoftie did. (Makes sense since Intel doesn't make OSes)
The article says:
Intel's approach is compatible with AMD's, the Microsoft representative said. "There will be one operating system that will support all (64-bit) extended systems," the representative said.
Note that this refers to "(64-bit) extended systems" -the pure, native (non-extended) Itanium 64-bit version of Windows Server 2003 which is already shipping is, a will always be different from the x86-extended code base since it takes advantage of the EPIC instruction set.
to another - nocona - www.nocona.com - they make boots... boots for people with BIG feet...
really big feet.
Oh, that's fine...
Maybe one day we get 128-bit desktop computing, but now 64-bit will have to do, I suppose...
I guess we cannot always have what we want, but we may get what we need and so on.
This is the same thing that we hear from some moron (yes you) whenever a product such as a 64 bit processor comes out. The answer is, because developers will create programs that use the resources of a 64 bit processor, and make it necessary for so called "desktop users." That's the thing about software developers, they're always saying "if only I could add more (blank)" Well now they can.
I had a 80386/16 box to use in 1992-1993 (senior year of HS) it ran Win 3.1 with the W32 extensions, so it was out there well before win95 was.
Please send all UCE to scally@devolution.com so I can f
Dell, apparently. Since Dell has continued to be exclusive Intel, in the face of the onslaught of AMD64 PCs, you can pretty much imagine a call from Dell to Intel going something like this:
Dell: "Those 64 bit processors are very interesting, we get calls asking abou them."
Intel: "The Itaniums? Well, yes, we've put many years and millions into them, they should stir some interest."
Dell: "No, I'm refering to AMD."
Intel: "But you don't make systems with AMD processors."
Dell: "We haven't, yet."
Intel: "Oh, uh, we'll have something ASAP and I swear it's not going to be exactly like AMD but almost as good, uh, yeah, that's it! Real soon now!"
Dell: "Good to hear it."
A feeling of having made the same mistake before: Deja Foobar
When I saw 'IDF' I immediately thought 'Israel Defence Forces'? What do they have to do with Intel?
"100% AMD Compatible"
64 Bit Extensions
From the Intel FAQ Site:
Q9: Is it possible to write software that will run on Intel's
processors with 64-bit extension technology, and AMD's 64-bit capable
processors?
A9: With both companies designing entirely different architectures, the
question is whether the operating system and software ported to each
processor will run on the other processor, and the answer is yes in
most cases. However, Intel processors support additional features, like
the SSE3 instructions and Hyper-Threading Technology, which are not
supported on non-Intel platforms. As such, we believe developers will
achieve maximum performance and stability by designing specifically for
Intel architectures and by taking advantage of Intel's breadth of
software tools and enabling services.
Quidquid latine dictum sit altum viditur
So, with this, more or less, Itanium may or may not be out of the big picture. From what we're seeing now, it appears as though Itanium will remain a high-end enterprise class chip.
But you have to wonder... what on earth was Intel thinking? Ever since its announcement, intel has hinted that Itanium would eventually migrate down to the low-end desktop market.
But, it wasn't x86 compatible by a longshot, and had no intentions of ever being hardware-compatible with plain old x86 CPUs. Without backward compatiblity, there was close to zero chance of intel ever capturing the desktop market with it (it was a completely new architecture. there had been no software written for its new instruction set to date).
But then you realize that intel broke their most sacred tradition by breaking backwards compability. Suddenly, "intel-compatible" wasn't "intel-compatible" anymore. Moving from x86 to Itanium would be like moving from x86 to SPARC/Alpha/PowerPC.
And SPARC, Alpha, and PowerPC are all more powerful than the current Opteron chips, and cheaper by several orders of magnitude (specifically PowerPC).
In other words, it would be more likely for everybody to migrate over to Apple that it would be to move to Itanium. And it would be cheaper too.
With these latest announcements, I'm hoping that intel has finally adopted the x86-64 bandwagon and cooperated with Microsoft and AMD. (Imagine if WinXP-64 worked on two architectures.... and the compatibility nightmares it would cause...)
Either way, the scores are as follows:
Sun/DEC - 6/10 Have been using 64-bit for years. Yet, nobody seems to want it.
IBM/Apple: 8/10 - Successfully brought 64-bit to market, but launched without a supported full-fledged 64bit OS
AMD - 10/10 - Openly allowed developers to develop with the x86-64 sim years ago. Launched x86-64 before anyone else backed with full Linux support, and windows support in open beta. Successfully penetrated consumer, mobile, and enterprise markets simultaneously.
Intel - -5/10 (yes, negative) - Created an expensive proprietary system with no backward compatiblity, and is cumbersome to work with. It flops. They still don't have a 64-bit desktop processor. Their only successes are made by copying AMD.
-- If you try to fail and succeed, which have you done? - Uli's moose
I own 2 opteron servers and am looking to buy a 3rd. Low cost and great speed.
I'm pretty sure HP and Intel signed an agreement to have the Itanium as their only 64bit processor for quite some time. This will violate this agreement, unless Intel payed HP the penalties then Intel is going to be in some deep legal shit.
Moo!
Barrett is quoted saying, 'There will be one operating system that will support all (64-bit) extended systems.' Maybe 64-bit computing is right around the corner after all, and we may even see compatible instruction sets from Intel and AMD!
;-)
My 100 shares of AMD say "they damn well better!"
I'd rather be a conservative nutjob than a liberal with no nuts and no job.
history repeats itself.
Whether Apple innovates in the hardware department is debatable. But they are pretty good fortune tellers. Let me count the tools they brought first to the home PC user.
.
.
.
1. 64 bit computing
2. Bluetooth
3. Firewire
4. 802.11b/g
5. USB
6. DVD/CD Writeable [got tired of linking]
100,000,000. SCSI
cogito ergo oro
Read the article more closely:
"the Microsoft representative said. "There will be one operating system that will support all (64 bit) extended systems," the representative said."
They are talking about their own operating systems, and the statement applies to "64-bit extended systems"
To translate: MS will ship one OS for both the AMD and Intel versions of 64-bit X86 extensions. Since MS has already committed to supporting AMD's 64-bit extensions, this pretty much forces Intel to use the same extensions.
MS will also probably continue to have a version of Windows supporting Intel's Itanium-based 64-bit platform, which is not based on X86 extensions, but has a whole new instruction set architecture (IA64).
Net-Net: Microsoft is not going to have three 64-bit versions of Windows, they will most likely have two - one for the AMD (and now Intel) 64-bit extensions to X86, and one for IA64.
This has nothing to do with Linux.
I still don't understand what makes AMD64 better than Itanium, Spac64, alpha, and ppc64.
...any more than IBM would ditch Power4/5 architecture, just because they have a commodity market x86 chip with 64-bit address extensions (Opteron).
In the 'big iron' enterprise market against RISC where Itanium is beating everything handily (check out the latest TPC-C list Top 10 where Itanium holds spots 1,3,4,7,10 (5 out of the Top 10 are Itanium systems running a mix of Linux, HP-UX and Windows on HP and NEC systems), Itanium is gradually out-selling all of the big RISC opponents like Power4. Note that IBM is certainly not spending the money to put up an Opteron cluster for the TPC-C test(no 32-way or 64-way scaled solutions for it on the horizon) even if they got good enough results (which they wouldn't) if they can't beat Itanium 2 right now with the high-margin Power 4. No doubt they'll have a run at Itanium again this year with Power 5.
But there's no way that Opteron OR a 64-bit Xeon plays in the big high thoughput space, so people that assume Intel would get rid of Itanium simply don't know what they're talking about.
As for Itanium not selling, That's funny. Itanium sold over 100,000 cpus last year which is a big number for the enterprise server market (That's more than some other major RISC processors sold in 2003 (like Power 4)). If you don't believe me Google "Itanium" "100,000" and "Otellini" and you'll see lots of links to Intel pres Paul Otellini's announcement back in Nov that Intel would ship over 100,000 Itanium processors in 2003.
It's refreshing to read a post from someone who has a clue about the differences between Itanium and Opteron.....
It's hard to picture why there will ever be a need for 128-bit computing.
2^64 is 18446744073709551616. This is BIG. 17179869184 gigabytes. 16777216 terabytes of addressable memory. 16384 petabytes. This is basically the maximum amount of physical memory and the maximum size of one individual process's virtual memory mapping on a 64 bit architecture (yes, I know many current 64 bit implementations, including AMD64 are limited to 2^48 in practice; but the architectures can fundamentally handle both 2^64 physical and virtual addressing).
This is enough addressing that you can have 2.5GB of memory in a process for each man, woman, and child on the face of the planet.
And as to doing integer math larger than 2^64-- why? 2^32 is already overkill for most things.
Nope, I don't see "128-bit computing" becoming mainstream anytime soon. And it's far from clear 64 bit on the desktop is all that close, given the fact that A) the added code size contributes cache misses and saps performance, and B) there is not much done on the desktop now that requires more than 2^32 bytes of memory in a process, and C) not much stuff does math on quantities greater than 2^32 (4294967296). Keep in mind bank switching allows you to have more RAM than 4GB on all recent ia32 processors (2^36/2^40).
If we change architectures, it will be less about addressing limitations and more about the piss-poor quantity of registers available on ia32. More registers means more obtainable instruction-level parallelism.. this equals more work done on modern architectures.
that there are no 64bit x86 games (or big apps like Oracle) on the market for Windows, because *suprise* XP/server 2003 are still in beta. That won't prevent customers from stuffing more than 4gb of ram into their 64bit x86 systems, though. Oracle for for OS X could be finalized by the time Microsoft gets around to releasing 64bit XP. Or the next Halo patch, for that matter.
Yeah. I guess you missed the news that AMD is profitable again.
Linus isn't part of, and Intel didn't help start, OSDN, that's the parent of Slashdot. I had my OSD's confused.
Intel helped start (and Linus is an employee of,) OSDL. That one letter makes a heck of a difference.
Another non-functioning site was "uncertainty.microsoft.com."
The purpose of that site was not known.
Ten years ago, Win311 ran pretty well on a machine with 4MB. Now 128MB is recommended for XP. A linear extrapolation suggests that we'll need 4GB to run pretty well in 2014. I'm glad they're starting now so that the 64-bit CPUs will be really cheap when we truly need them.
The x86 architecture was not designed for 4+ CPUs on board. Itanium may not work well in a small-mid range environment but compare a 16-way Itanium to a 16-way Xeon and see what you get.
AMD believes that hyper-transport will solve this problem. We shall see. Where are my 8-way Opterons? Wern't they supposed to come out at the end of last year? I'm still bitter about the Athlon MP - which was supposed to allow for 4-way Athlons.
Do they get to vote each other off ?
AMD's first profitable quarter in 2 years means that the 64-bit computing market is now ripe, which is why Intel now enters the 64-bit market. Quite the timing, no?
An example from Reuters:
"Analysts have said they expect a major announcement from Intel this week on 64-bit computing, a technology that lets computers churn through doubly large chunks of data than the current 32-bit computers."
AMD turning the heat up
64-bit does not double the memory bandwidth, it simple means a bigger address space + 64bit integers, (although the Opteron does have high bandwidth this is nothing to do with 64-bitness)
I don't see "128-bit computing" becoming mainstream anytime soon.
Well, the designers of IBM's venerable AS/400 might disagree with you. Its architecture has been enabled for 128-bit computing since the early 1970's.
I think everyone seems too be taking this too far. I mean, this is a milestone, but then, look at the 32-bit chip, it didn't become main stream for like 10 years after it came out. As time goes on, people will need the extra performance 64-bit will give them, more and more people are starting to get into video processing for their own home movies, which, when you work with larges amounts of raw data, can take up quite a bit of room. This is not the ONLY reason, but it is one. I have a 64bit chip, but thats becuase I like to be on the cutting edge, its not like everyone is going to go to 64bit overnight. MS's first 32-bit OS wasn't until windows 95 came out, and 386's had been out LONG before that, give them time, the chips haven't even been out a year yet.
There's something about his name that goatse.cx could appreciate
I've heard all those arguments before.. sure, the numbers were different (8 bit is enough, 16 bit is enough, bankswitching solves the memory problem, codesize will destroy performance)
Fact is, once a larger wordsize and address space becomes available, peopel will find uses for it that you didn't imagine beforehand.
I agree with you on the memory addressing limitations, but you can't make the integer argument.
Floating point math is slow, really slow. Moreover, it's not accurate. You can't accurately express 2.5, for instance, in binary machine. With integers, you get back results as accurate as the data you provide.
I think that the combination of higher speed and accuracy of integers is a compelling argument for expanding usable range thereof.
Intel convinced HP to discard HP's PA-RISC and go with Itanic, what will HP do now?
Everybody's a libertarian 'till their neighbour's becomes a crack house.
This is just a preview of the upcoming 128-bit processor war!
340282366920938463463374607431768211456 bytes of addressable memory...powerpoint is really going to kick butt!
I wonder if he still feels that was a good strategic move? Likely so. I guess MS could have screwed them over even worse.
Clearly the market is most poised for 64 bit when
the RAM cost is low enough to justify a larger address space. A few months ago retailers don't want to say your computer costs $5000+ because it has 16 Gb. And what good is all that RAM if it is slow RAM?
In fact I'm watching a process running through slow RAM on a 2-year old machine right now. It's not swapping even though I have 256 Mb, a princely amount 3+ years ago. I like the power of not swapping but big RAM brings even bigger ambitions so big RAM has to be fast RAM.
Know your pads. One time pad: good for cryptography. Two timing pad: where to take your mistress.
While AMD have alway talking about new developpment and invits others to share, Intel keep all secret and try to act like no others exists (including there customers sometimes).
l ogy/64bitextensi ons/30083401.pdfn ology/64bitextensi ons/30083501.pdf
o ntent_type/white _papers_and_tech_docs/24592.pdfo m/us-en/assets/content_type/white _papers_and_tech_docs/24593.pdfo m/us-en/assets/content_type/white _papers_and_tech_docs/24594.pdfo m/us-en/assets/content_type/white _papers_and_tech_docs/26568.pdfo m/us-en/assets/content_type/white _papers_and_tech_docs/26569.pdf
Sorry Intel. There is no AMD words in your doc, but now all the worlds known that your IA32-e is no more than the AMD X86-64. For me you just act like a child!
Intel IA32-e documentation:
http://developer.intel.com/techno
http://developer.intel.com/tech
AMD x86-64 documentation:
http://www.amd.com/us-en/assets/c
http://www.amd.c
http://www.amd.c
http://www.amd.c
http://www.amd.c
How long Intel while wait before it make the same kind "new extention" compatible with HyperTransport ?
It seems to me like Microsoft possibly used its monopoly in operating systems to delay the adoption of Athlon 64 processors. Whether it is collusion or not is unproven but it smells like it.
On the downside, I kind of liked the IA64 architecture, I'm not a hater or a lover the x86 but I would have loved for IA64 to start taking off, it just looks like a clean architecture. We're talking about a 10year+ effort by one of the largest and most powerful companies in the world that is sputtering out. That seems so epic (no pun) to me in the post dot.com world where everything is about speed to market; how many 10year efforts are out there any more? I can only assume that this is kind of a death knell for IA64, especially if MS appears to be on the x86-64 bandwagon. Looks to me like HP and Intel are IA64's only supporters and they don't write that much software between the two of them. Can anyone offer some insight in to how the company that has beat much of the competition in to submission with price/performance has floundered IA64 so badly? Is there a fear of reducing their future margins by making IA64 affordable, like they won't be able to charge $3000 a chip? It's just such a talented group of people that make up Intel (I literally mean that and have tons of respect for them, they *know* how to build chips) I just don't understand it, I expected them to start turning the crank and be selling IA64s for the same cost a Xeons and at that cost we'd all be buying IA64s for our servers. This has to be more than politics. They simply bite off too much?
It was supposed to replace X86. Itanic will go same route. Repositioned and slowly fade into the sunset.
Help fight continental drift.
Sun will only support Solaris x86 in 32bit mode on the x86-64 systems. They'd be retarded to undercut the flagship UltraSPARC line of servers...
Wait 48 years and memory demands will be 2^32 times bigger than they are today.
No, I cannnot fathom what that memory would be used for either, probably a personal assistant AI, but time has proven the double-every-eighteen-months rule, so 48 years it is.
Granted, I don't mind using 64 bit processors in the meantime.
I remember reading about project "Yamhill" since early 2002. Intel has been developing their x86-64 chip since then. Obviously Intel has been waiting to see how OS vendors and the market would accept an x86-64 chip.
They also didn't want to prematurely flush Itanic.
When the shit hits the Itanic fan, Intel execs shouldn't be the only ones getting reamed for this huge blunder....HP execs should also share the blame.
-ted
Discuss.
"A great democracy must be progressive or it will soon cease to be a great democracy." --Theodore Roosevelt
But will Intels chip have an IMC????
I happened to see this story which quotes Intel CEO Scott Barrett as saying "Intel's 64-bit extension technology will be software-compatible with AMD's 64-bit extension technology." It also quotes one analyst saying "Intel will be a uniter rather than a divider, and that's very positive news."
Intel has enough money and market clout to wait and survive for a year without announcing a 64-bit chip (it's been about a year since Opteron, right? and Intel is still the major CPU brand). By waiting until AMD has spent all its money touting 64-bit computing to release its own 64-bit chip, Intel can deal a very devastating blow to AMD. If Intel steals AMD's 64-bit market with its humongous marketing account, what then for AMD?
A bruise for big ol' Intel who has lots of extra meat, yes, but Intel can easily survive, but if worse comes to worst this could be a dagger in the poor skinny ribs of AMD.
Guess we'll have to see whether I should take off my tinfoil hat, or buy an upgraded one...
It's hard to picture why there will ever be a need for 128-bit computing.
I'm sure plenty of people said the same thing about 32-bit (and even 8 and 16 bit) at some point in the past.
Aside from just the addressable memory is the ability to do larger math calculations in the larger registers. I've done some side by side comparisons of 32 and 64 bit compiled openssl on opterons and the 64bit version has a huge speed increase, very likely due to the additional size of the registers, and the additional registers that were added for 64bit mode.
Besides.. you can't say that the addressable memory that 64bit gives is more then enough... sure it's huge now, but in another 20 years or so it might not be. I remember getting my first PC 17 years ago... an AT&T PC6300, 8088 CPU (12MHz I think) with 640K RAM and 2x360K floppies. When I finally got a 20Meg (yes, meg) harddrive a year or so later I thought I'd never be able to use all that space. Now anything less then 250 or so gigs is a waste of time. RAM usage grows slower then drive, but it still grows. I don't buy a system with less than 2Gig of RAM anymore, and that's something like 3000x the amount of ram I had in that first pc. Keep extrapolating that out for another 20, 40, or 60 years.
- My favorite error message: xscreensaver, running on an old Sparc 5 w/ 8bit color: bsod: Couldn't allocate color Blue
If Intel releases a AMD-x86-64 compatible processor (i'm quite sure it is), say a permanent bye-bye to the possibility of mainstream CPU diversity we could have, had the x86 died as it should.
Binary compatibility... It sounds like a voice from the 70s.
Such thing is only convenient for Microsoft and its Windows.
Well, locked back to never-gonna-die-x86 (oh, please save me from those super-RISC core arguments and blahblahs alike)... Thank you AMD.
Intel complaining that 64-bit processing has no place in the desktop market,
then
... Craig Barrett today anounced that Xeon-class 64-bit server CPUs codenamed Nocona will be coming out the second half of 2004...
Maybe 64-bit computing is right around the corner after all, and we may even see compatible instruction sets from Intel and AMD...
whats your point.. desktops or servers..
Listening to people, you would think that Itanium is "just 64-bit" and that there won't be any difference between Itanium and Xeon-64. The truth is that the Itanium architecture has many advantages for high-end servers in addition to 64-bit. Xeon-64 will not replace Itaniums.
And 64-bit on the desktop? If you have 4 gig of
memory installed on your desktop now and that is insufficient, please raise your hand.
Wrong list.
432 and Itannic had full management support. The 432 failed becuase the market rejected it. Itannic appears destined for the same fate. The 960 wasn't rejected by the market, it was rejected by Intel management.
Ummm... Isn't (in binary) 10.1 = decimal 2.5?
.5 is 1/2. Of course you can't express many decimal numbers in binary. In decimal you can express 3/10 , but neither binary nor decimal can express 1/3. If you CARE about the base 10 decimal places (e.g. a financial application) then floating point is inappropriate.
In binary I can express anything that can reduce to something over a power of 2, and
But you can't say floating point is inaccurate. It is highly accurate in binary. If your application is doing pure math and is using the result of a floating point calculation to some purpose other than human eyes it's quite applicable.
And for many math functions in many languages (e.g. root exponentiation) decimal is simply not available. The Java math class is the only function in the Java libs that can extract the 12th root of an interest yield to determine the nominal interest rate. It accepts a float or a double, but not a BigDecimal. But so what. If you use floating point you'll still get a number out to many more decimal places than is needed for accurate rounding of the interest. I suppose if you're applying that interest to something like the national debt you might need to use double precision floating point...
P.S. if anybody has an open source bean that can do root exponentiation on the BigDecimal class please let me know
And does this mean that Intel will be dumping Itanium, which never caught on as expected in the server market, and forget the billions spent on developing it?
They can replace those losses when they sell a couple of those P4ees.
Learn something new.
Also keep in mind that even though a 64-bit processor with a 64-bit memory manager in an OS will allow for increased memory addressing, the chipset must also support the addressable memory. 8 GB is the limit in most boxes. I think what might be interesting in the future is the synergy between MRAM and full 64-bit processing . . . with MRAM cheap enough and abundant enough, virtual memory will be eliminated with applicatioins getting free reign across the entire addressable space that doubles as the entire storage space.
...will say "I knew it!" when Intel turns out to have not had anything ready and nothing ships in 2004.
However you'll point proudly to your chest and exclaim "I was that AC who knew it all along!"
But you have to wonder... what on earth was Intel thinking?
It wasn't x86 compatible by a longshot
Intel was thinking that Itanium would be enough faster than a native x86 to emulate x86 competively. The architecture was designed to make software emulation of x86 relatively efficient.
The trouble is, the required performance never arrived. Clock rates greatly lagged x86. Compilers have not been able to use the resources provided by IA64 effectively. Itanium hasn't been able to keep up with x86 when running native, much less in emulation.
Floating point math is slow, really slow.
Maybe you should check out some benchmarks on something newer than a 486, because most current CPUs can actually do floating point calculations faster than integer ones.
Some people are missing the point of x86-64, it isn't just being able to access > 4 gigs or perform 64-bit math in one instruction. The x86-64 architecture has some improvements over x86, like more general purpose registers, so in general, x86-64 compiled apps will be faster than the same x86 compiled app. So the next question is why do we need faster performance? You may not need it today, but generally software comes after hardware. So as more people have these processors, you can expect more applications that will take advantage of these apps. Even today, I sometimes find myself waiting on the computer, this should never happen (although I admit most of the time it's network limited and not cpu limited)
...it doesn't make it go any faster, but it sure looks like it could.
What most people are forgetting is that no matter what extensions Intel adds to the chip, they still have an enormous bottleneck called a NorthBridge chip. Last I checked, the AMD64/Opteron procs eliminated the need by moving the memory controller onto the chip and by utilizing HyperTransport.
With Xeon-64, in effect, Intel is now trying to shove twice as much traffic through that same bottleneck. Rrrrright. May the best chip win.
I have my money on AMD.
Texas Instruments is demoing the industrys first PCI-Express 1.0a compliant PHY running on a bridge. Buh Bye PCI (we wont miss you).
l
http://biz.yahoo.com/prnews/040216/dam006_1.htm
You're neglecting the difference between address space and physical memory. It may be a year or two before ordinary people buy more than 4GiB of RAM, but everyone benefits from having a larger virtual memory.
DVD burners are one of the few growth areas for PC peripherals. A DVD can hold more than 4GiB. A DV camcorder tape holds about 30GiB.
Pretend that you are writing a video editor. Would you rather mmap those huge files and let the OS VM handle paging chunks in and out, or would you rather write you own buffered I/O routines? Which will be faster, and which will have more bugs?
There are already applications used by ordinary folks which can benefit from 64 bit address space, even if the physical memory would work with 32 bit addresses. Large virtual memory spaces are a clear win.
"In the whole world, the market for 64 bit desktops is like 5-6 computers at most." - Some guy at IBM forget which, but i'm sure someone will append it here.
Seriously, anyone who says 64 bit computers aren't necessary are the same ones that tell me my 40GB HDD is huge. No, it's not. 40 MB once upon a time was huge, but now it's laughable. Haven't we learned that short sightedness brings problems? Like IDE adressing problems? How many times did we have to hack around THAT problem?
Trolls dont like to be Flamebait, because they burn so well. Protect our Troll heritage!
I just feel sorry for SGI, who are now facing trouble due to other companies decisions.
First they were skrewed arround when MIPS failed to break the GHz mark, and now by Intel for signing the death of the Itanium.
VENI, VIDI, VICI, DIXI
Is that a wise marketing move?
It's pretty clear that MS has been dragging it's feet to accoodate Intel. I believe this to be illeal. Would AMD sue MS over this?
Sigged!
The link in the message above was to 64-bit Windows Server 2003 but
Microsoft is also offering a download for a free 360-day trial of
64-bit Windows XP for desktop systems which can be downloaded here.
We still haven't seen where Itanium does not scale, whereas we know where the x86 has problems (too few registers,
Don't overlook that x86-64 adds 8 new general purpose registers. These are called R8-R15.
Remember "Bring 'em on"? *sigh
It is directly compatible with AMD's 64-bit implementation. Intel and AMD have a sharing arrangement with x86 dating back to when AMD first licensed x86 from Intel, which basically allows Intel to use whatever AMD adds to the instruction set. And vice versa of course. Generally whoever implements it first gets in a generation ahead (Intel with the SSEs, AMD with x86-64).
On another note, these new Xeons are based on the Prescott core, so it is now extremely likely that the existing Prescott cores all have the capability, just not turned on, like what Intel did with hyperthreading on the Northwoods. It's been clear from the start that Prescott is hiding some functions up its sleeve, as there are at least 10 million transistors that can't be accounted for with the increased cache and other added functionality, even when being very generous with the estimations.
Actually I never quite understood how some of the 32 bit motherboards are advertised as being able to use 8GB. Even if the memory controller could do that how would the CPU address above 4GB?
Sun will only support Solaris x86 in 32bit mode on the x86-64 systems.
Not true. I'm told we can expect Solaris 9 x86 64 bit support in the second half of 2004, in one of the quarterly Solaris update releases.
They'd be retarded to undercut the flagship UltraSPARC line of servers...
The SunFire systems have a different target market with different scalability, availability and managability requirements.
# init 5
Connection closed.
Oh...
Oops, too late ... already forgotten.
Beta doesn't count. If it aint in the box, it aint in the running.
AMD has done the heavy lifting designed an IA at the front end, and put the arch in the marketplace for people to start targeting code at it.
A year from now, people will be buying 64-bit Xeons running at 4GHz which will fly on the x86_64 optimized code... they were probably buying AMD before because Intel didn't have a similar offering, but now Intel can get their Gold-plated name back into the newly created market.
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
even SLES 8.x and RHES 3.0 are not rock solid on AMD64 servers. Sure, single CPU, fine. But you try to use 4-way on the AMD8XXX and you will have an interesting time.
I think Microsoft was having similar trouble trying to adapt Windows to run properly and without issue on this brand new hardware. Maybe AMD was dragging their feet addressing errata exposed by the effort?
But even then I feel that's a bit of a stretch. It's been in beta state for a long time now. At least they're offering security updates for it... that's a sign of commitment.
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
They would be expending far less effort than you would believe. The PPro and netburst architectures are pretty flexible. You update the chip to for larger width of registers in the register file, change the width of some parts of address decoding stages, add more address pins, and update the Microcode to understand the new mode/instructions.
I'm sure centrino (Pentium 4-M) was a more expensive endeavor then this Plan B.
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
Instant resume, configuration stored in the app itself, etc. etc.
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
Yeah, integrating a dual channel memory controller was probably a good idea.
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
I'm sure plenty of people said the same thing about 32-bit (and even 8 and 16 bit) at some point in the past.
The thing is, every time you double the width of the address bus, you square the amount of addressible memory.
Keep extrapolating that out for another 20, 40, or 60 years.
Ok.
Let's say that home computers circa 1984 needed 64kB of RAM, circa 1994 needed about 4MB, and a typical home user today needs 256MB. This is a factor of 64 increase every 10 years (and matches my experience/memory pretty well). If this pattern follows, home users will need more than 32 physical address bits in about 2027. 64 bits would be enough until about 2080. Of course you'd need a larger address space than that for swapping...
But that assumes an insatiable need, that memory requirements will increase exponentially forever. As with many things, I think a logistic model would fit better, that there is some unknown upper limit to the amount of memory we'll need (and there better be, since physics will eventually enforce it). I'd give estimates for that, but I'd have to pull a fourth data point out of my ass.
It's called SSE.
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
For whatever reason, Apple fans still tend to ignore anything but Jobs and Co.'s Reality Distortion Field. IBM and Apple releasing a 64-desktop chip does not represent the beginning of 64-bit on the desktop. BOXX and AMD made a pretty nice 64-bit desktop machine that was out for months before the first G5s shipped. Not to mention Digital's Alpha, which shipped years before that.
People do not need 64-bit computing for standard desktop computing applications - of course not.
That may be true if you're talking running mostly business applications and surfing the Internet. However, with the increasing use of home computers to edit digital still photos and digital movies downloaded from MiniDV/MicroDV digital camcorders, the demands on sheer computing power has climbed dramatically. Take for example editing digital still photos; with today's digital cameras capable to taking pictures with resolutions at four megapixels and larger, the result is picture files that runs into several megabytes in size, and editing them to change brightness/contrast, correct color problems, and so on takes a LOT of sheer computing power in both CPU speed and system RAM size.
The new 64-bit CPU's not only can do this type of heavy-duty processing in many cases better than 32-bit CPU's, but the ability of 64-bit CPU's to access very large amounts of system RAM means we can work with larger and larger multimedia files.
If your definition of 64-bit is a 32-bit operating system around a 64-bit chip, then the G5 is a 64-bit platform. Mac OS X 10.2.7 (and the upcoming 10.3) is not a 64-bit operating system. This is particularly frustrating because Apple's marketing machine has very carefully crafted their message to make a reasonable person believe the operating system is 64-bit, especially if you download and read Power Mac G5 Tech Overview (PDF). Apple says about the G5 version of Mac OS X that it runs all of your software -- and runs it faster -- with a version of Mac OS X Jaguar specially tuned for the PowerPC G5 processor, providing a seamless transition to 64-bit power. That's only the beginning of the smoke and mirrors. The 64-bit power only gives users two things: the operating system can address up to 8GB of RAM, though user programs are still limited to 4GB, and some of the G5 numerical hardware is available with a special version of GCC (3.3).
So you want to move hundreds of millions of customers to a new CPU design that's elegeant and does away with all the x86 junk.
How do you get around the chicken and the egg problem where there's no software developing as there's no customers, and no customers interested because there's no software?
Here's how: You start in a market that can afford to spend lots of money on developing their own software. The only way to enter in this market is to make your CPU run these customers' programs the fastest. Eventually you'll get enough software support to move this CPU down and sell to the medium and eventually the low end servers, but it'll take a few years. In this range, OS's will be building the capability to emulate x86 with little or no performance hit, and programs will be migrating to run natively in the new format. A few more years and many native apps later, you'll get high-end desktops/workstations migrating. Finally, the flood of average consumers will start rolling in.
How long will this process take? Let's look at the closest example of switching consumers to something that wasn't backwards compatible: DOS/Win3.1 to Windows NT. WinXP was the first "NT" version marketed to, and accepted by the average consumer. NT 3.1 was first released to the public in October '93. We have a full ten years to make the migration.
So you really shouldn't be comparing these processors just because their bits are the same. They're all 64 because very soon now 4 gigs of memory will no longer be enough for the desktop. The real comparison is x86 vs IA64 and the big question is whether IA64 really has what it takes to beat out x86 a few generations down the road.
The Itanium is still a kid, and it has a lot of battles ahead of itself before it's a consumer CPU, but IMO it has passed its first tests. I hated NT once also, but I've learned my lesson now.
Sure, sure, exponential growth and all. But you're not talking about 3000x the amount of RAM that's in a typical system today-- you're talking about 256 BILLION TIMES THE RAM that's in a typical system today.. that is available in a 64 bit system vs. a 32 bit system.
If one bit of memory fits on one silicon atom, it would take 127038750317144 kg of silicon (that's a lot, by the way!) to build a computer with 2^128 bytes of memory. So no, the exponential growth is not going to continue forever, dream on.
Of course you'd need a larger address space than that for swapping...
No you don't. You can have considerably more than 4GB of total virtual memory space on IA32 and other 32 bit architectures. You are just limited to (assuming no bank switching)
2^32 bytes of physical memory
2^32 bytes of (physical + virtual) memory mapped into any one vm context-- e.g. process.
Virtual worlds.
If you have every played a MMORPG, then imagine no zones. The complete world in memory.
If you think virtual worlds ends with MMORPG, you are mistaken.
Imagine storing a full movie, several clips all the editing tools you could need in memory.
The Kruger Dunning explains most post on
I think we'll see it quite soon --- imagine one database daemon using NUMA to map in a database image spread across many "simple" 64 bit servers.
Databases grow by leaps and bounds as we see fit to forfit more of our privacy to them, by dumping in more and more trivial amounts of data on a daily basis. Even if at some point we can address every single electron in orbit around every single atom that composes this planet, we'll outgrow it. Data storage will simply move off planet --- imagine data centers in orbit with databases that can detail entire star systems down to the last atom, with free sunlight to power them, and a dark backside to dissipate the heat.
Think Different. Think Big.
TurboD
the 64bit intel is a dog (performance wise) compared to the opteron at 64bit tasks.
Because their hacked 36-bit, giving them up to 64 gigs of RAM in some ugly paged mode. It works, but its slower then a true 64-bit solution.
You don't churn throught twice as much a *useful* information as 99.9% of all integers only need 32 bits (or less), so really the higher order bits are being discarded and 32-bit processors already have 64-bit floats. The quote is misleading and suggests that the 64-bit processor is going twice as fast.
Either you made a mistake in typing or you made a mistake period. 24000 is smaller than 100000. Thats one digit you missed there
The dangers of excessive individualism are nothing compared to the oppressiveness of excessive collectivism
If we change architectures, it will be less about addressing limitations and more about the piss-poor quantity of registers available on ia32. More registers means more obtainable instruction-level parallelism.. this equals more work done on modern architectures.
Modern architectures don't need more architectural registers. Register renaming takes care of the false dependencies and the spill/fill loads and stores can be eliminated from the critical paths of programs due to fast store forwading buffers and even more advanced techniques such as memory renaming.
The fastest integer machines on the planet are x86 based. 8 registers hasn't been a problem for quite some time now.
if intel engineers an x86-64 instruction set that is compable with AMD's, AMD still has one big advantage, they DESIGNED the instruction set, they have a more intimate knowlege of what they have built than intel, and hopefully know how to scale it better, since its their baby, after all.
-and occasionaly a giant moose.
Not really, most processors include more FP units than Integer units. That is why some machines may get better FP numbers, never mind that the spec FP and spec Int are 2 different magnitudes that are not 1 to 1 relative.
Most Intel machines have much better integer performing units, and are relatively weak with respect to other RISC competitors.
One word: memory-mapped files. In a 32-bit architecture you can only memory-map small files (there is a limit of 2GB per process), so any application looking to use memory mapping to make file access faster and/or simpler (and MM is a big win on both counts) can't do it. Instead we have to write a raft of code to duplicate what the OS does better (buffering, preload, write-behind...) just in case the user wants to open a couple of medium-sized files simultaneously.
With 64-bit addressing we can memory-map *all* files, regardless. It's a big win even if you only have 512MB of physical RAM.
Sadly the future is not as exponential as going from 64 bit to 128 bit, most kids do not seem to get what those magnitudes are.
:).
Besides there is really not much of a benefit for a 128 bit long datapath, at some point you can do much faster processors that are 32 bit for example than 64 bit, to the point that you can have technology very pipelined technology that can run at more than double the equivalent 64 bit datapath, therefore performing a 64 bit operation using 2x cycles of the faster 32 bit datapath is actually done faster than the 1x cycle of the 64bit datapath. However they are not as glamurous as the 64 bit native datapaths
Also people need to understand that complex structures like multipliers grow polynomially not linearly, this is a 128 bit multiplier is not twice as big as a 64 bit multiplier but much much larger (simpler terms would be squared increase).
Where did your 100000 come from? The article shows that with UIIIi alone, Sun outshipped Itanium by 2.5x.
Uhm... x86-64 isn't some modern marvel. It uses a prefix byte to indicate which opcodes are 32-bit and which are 64-bit. Hardly the most original way to tackle the problem. If nothing else, AMD actually ripped off Intel here by using all the original opcodes just with a new REX prefix.
All I know about Bush is I had a good job when Clinton was president.
First, your statements above are contradictory, in 1975 there was no Ada programming language, only a spec (steelman ??) that described what the language should contain (and not contain).
Also, it is not clear whether you meant that the 432 or Ada was a marketing failure (or both). Certainly the 432 was. OTOH, from its first release in 1980 or so, the Ada language has been far from a "market failure", despite there being no low-cost compilers for it and despite the limitations required by the SteelMan spec. Virtually all aeronautics, astronautics or critical communications software (Military or civilian) and weapons control software for the last 20 years was written in Ada (and not just in the US).
In addition, several commercial SW firms also found, even w/ Ada-83, that it allowed them to ship w/ far fewer bugs left for customers to find that code written in (Ugh!) C, as well as allowing bug-fixes using less than 50% of the developer resources than to fix bugs in (Ugh!) C.
As of 1995 the Ada language is much more oriented towards general programming, as well as being much cheaper to use than it had been. There has been a FREE (GPL) Ada compiler available since 1995 or so, and it is now (since version 3.2) integrated into GCC.
For more info on how Ada is being used and why it should be used for all new projects, see My small Ada site or David Botton's Ada Power site.Unlike what others have said, 64-bit computers are very important for games, especially 3d interactive ones. One of the most usual complaints about games is that they are not as interactive as they should. For example, lots of objects in 3d games can't be destroyed because they are simple decorations and not real game objects. 64-bit memory will open the path for huge game data present in memory at the same time, thus opening the path for really interactive games (especially RPGs).
The online games will also benefit greatly, because 64-bit systems will be able to host thousands of players in much less cost than todays 32-bit clusters.
Still won't fly!
Actually, I've always preferred them to be elephant ears. My g/f likes it...
Yes REX prefix byte is a small change in the instruction set, but a very nice one: it allow to extends the architecture to 64 bits (more longer registers, more addresses, etc...) while permit to execute IA32 userland code without problems.
The most nice part it that the change do more cleaning that hacking: Intel have a long history of adding new opcodes to IA32, but almost no one use it because binary programs have to run on all IA32 chips including the older one. So now the difference between older and new IA32 chips is so big that for time critical functions, programmers have to test the capability of the chip and to dynamicaly branche to a dedicated functions optimized for that chip. x86-64 make a hug reset to this entropy. This don't stop the process as new opcodes will exists in the futur, but at least it voids the last 10 years or so ugly hacks to the IA32.
In a wonderfull world, where all uses Debian source package and recompile it for each of his machines, the binary compatibility problem will be less important. But this even don't solve the problem, as the detection and specialization to a dedicated chip will be done while compiling.
The lesson is that it is not supportable to indefinitely add hack to an architecture. Sometimes it better to make a new backward compatible architecture. I doubt that 128bits computing will be the next thing that will trigg an architecture change. But without doubt architectur change will happens again in the futur to clean the older one and enable new capability.
And at this time the market will certainely follow the proposition that will have the bigger performance and smallest change ratio. This is exactly why IA64 is dead and x86-64 shine, forcing Intel to copy it and rename it IA32-e...
Actually IA-32e is something different from what AMD has done (IIRC). IA-32e is a 64-bit mode that defaults to 32-bit operands but still uses 64-bit addressing. This allows you all the benefits that a 64-bit processor would offer, without the unnecessary code bloat from having operands default to 64-bits.
As to your other comment, there are already differences between Intel's and AMD's implementation-- Intel has a CMPXCHG16B instruction while AMD does not. And I'm sure in the days ahead other differences will be revealed. It's unfortunate AMD didn't include a CMPXCHG16B instruction though.
All I know about Bush is I had a good job when Clinton was president.
I own an HP Athlon 2600 computer...so they must be interested in Athlon technology.
No! You are wrong, x86-64 and IA32-e both uses 32-bits operands. Other way there will not be compatible!
e _papers_and_tech_docs/24592.pdf
i ons/30083401.pdf
For AMD see the table 1.1 "Operating modes" page 43 of http://www.amd.com/us-en/assets/content_type/whit
For Intel see the table 1.1 "IA32-e modes" page 18 of http://developer.intel.com/technology/64bitextens
(I dont know why slashdot add sometimes a space in the URL, it's not in the original)
The CMPXCHG16B example just show that Intel continue there nasty game of adding opcode in a way that nobody can use it because thre need to run on AND chip too.
the G5 is not the power4, maybe but they do share almost the same architecture minus some cache and another processor on the die. The exacution units in the 970 are FROM the Power4, it is not really a different beast and it is a better processor than itanium because it can scale to different markets.
No contradictions if you recall the iAPX 432 timeline. The project started in '75, when Intel's only available processors where 8-bit. However, the 432 wasn't available until 1981, six years later.
Certainly in '75, Intel wasn't thinking to design a CPU for a language who's specifications were just beginning to be written. However, the 432 was designed to be highly secure and provide multiprocessor support, both things that appeared to be a good fit with Ada. The marketing types at Intel were probably jumping on the Ada bandwagon when they finally had silicon, but if you read Intel's own docs, the relationship (or hype) is there.
And yes, Ada has been a market failure. It is used in military applications because the government has dictated its use. In the vast majority of commercial markets, where there is a choice of development languages, it is hardly ever used. That isn't to say, it doesn't have some clever and/or valuable features, it's just failed to gain commercial acceptance.
more registers sure as hell makes things easier, and it will speed up execution even with all those "advanced optimizations" you mentioned. Even with register renaming you still have a load/store every 5th instruction or so, and even if those load/stores execute instantly you have a bunch of useless instructions clogging your datapath. With more program-visible registers fewer spills occur, leaving more room for real instructions in the piepline.
every _exit() is the same, but every clone() is different.
Ahh, I see, IA-32e isn't what I thought it was. :P
As far as CMPXCHG16B is concerned, it's not a nasty game-- Intel sets the standards for x86 instructions, and this just continues that. Was adding CMOVcc to the Pentium Pro a "nasty game"? I suppose CPUID was also "nasty". Why AMD didn't think to include CMPXCHG16B is beyond me.
All I know about Bush is I had a good job when Clinton was president.
You can quote me now:
16,384pb should be enough for anybody.
I know 16 bits per sample is CD quality, but audio people are kinda dissatisfied with that these days. 32 bit floats per sample is much nicer, and the high-end audio software can do that. You can do math on 32-bit numbers much faster with a 64-bit machine....one instruction instead of six, iirc. No worry about overflow if you multiply two 32-bits together.
You could argue that you only need this for "workstations," but lots of people are doing studio work on laptops these days, and lots of consumers like playing with the fancy tools. Besides, when you can get a low-priced computer with a gigabyte of RAM, it's not going to be much longer before the 4-gig limit starts to be a problem. Intel has been saying "you only need this for servers and workstations" ever since the 286, and they've been wrong every time.
Finally we'll have a low cost (free?) optimizing
C++/FORTRAN compiler suite for x86_64 other than
PGI 5.x.
Hands up those of you who went used Dell's site search for the following keywords: Opteron, AMD, athlon64 ;).
I doubt Dell has that much power over Intel. It's more like the other way round - Dell = "Intel: Server Division". But yeah I'm sure there was some pressure from Dell regarding the "problem".
It's even funnier when HP (The other Parent of the Itanic) started making noises about selling Opteron servers.
Even without these players as long as AMD didn't screw up (go bust, dud chip), the outcome was fairly predictable.
To a company running lots of x86 apps on x86 servers, switching to the Itanium is about the same pain and probably worse than switching to IBM POWER4. At least with IBM POWER4 you have a proven architecture - hardware, software all been there for years. With the Itanium, nope.
So when AMD comes along with the Opteron and 32 bit performance is good and SMP performance is outstanding, Intel has to come up with a suitable answer, and Itanium would not be acceptable anymore than Sun/Fujitsu SPARC or IBM POWER4.
Intel was probably just hoping that AMD would have problems coming out with enough chips, and indeed AMD did have problems but then IBM gave a helping hand with the SOI process... You see all those relatively low wattage PowerPCs? Well the Opterons are pretty low wattage too, the 80-90W specs seem more to ensure people don't cut corners (the different speed CPUs can't all be using the same watts).
I agree about CMPXCHG16B. AMD should have including it as it seem to help creating fast, explicit lock free algo. But to be atomic the CMPXCHG16B opcodes have to do bus locking. Because of that this don't make a hug difference at the end: The AMD implementation will have more opcodes but in a path where the bus locking is the slowest thing anyway...
"They'd be retarded to undercut the flagship UltraSPARC line of servers"
So what? TI has done a fair bit of UltraSPARC retardation for Sun in the past few years.
At the rate they were[n't] going, somebody should have nicknamed TI Eclipse or something.
Look for a big number writeoff and the phrase "sunk cost" on Intels 10Q filing. For an economist, sunk cost means never having to say you're sorry...:-)
The point is that Intel didn't come anywhere near their Itanium sales projections for 2003; they didn't make one tenth of that 100,000. Comparing actual sales against outrageously optimistic, one-year-old sales *projections* is either dumb or intentionally misleading, particularly when those sales projections were proved to be drastically wrong.
<Insert BSD-is-dead parody here>
As a mainstream processor, Itanium is dead. It has ceased to be. It may still have some use as a supercomputer or mainframe component, but even that's in doubt since its power consumption and heat output is significantly higher than Opteron.
Don't forget that these will be procs that can address 64bit (ie. lots of memory). Doesn't mean that they are actually processing data 64 bits at a time. The ALU is still only 32bit. This means that there is very little hit to compilers and OS's.
I won't debate that point. It may very well be so and is not worth researching just now.
Hmm ... I suppose one could say things like Lear Jets, Mercedes Benz autos and fancy yachts are market failures in the same sense. They do things out of the ordinary and require a deep pocket. (Of Ada, this is true of pre-1995 compilers and still true if one wants support, special features or just a validated (certified) compiler.
Not since 1995 or 1996. The US DoD did not extend the Ada mandate and in fact closed the AJPO in 1996. To repeat part of my original post, Ada has been in extensive use worldwide for aeronautics and astronautics (in fact my current knowledge is that every current airframe (civilian or military) runs on code written in Ada). The Paris Metro and part of the NYC Subway system runs on Ada code.
This may still be true, but I expect it to change. Ada-95 is eminently suited for general commercial development as well as systems programs and since GNU Ada has been available there are less stats available on market share than there are on Linux deployment.
To the best of my knowledge, Ada is being taught as a first programming language in some 200+ universities around the world, including in the US Military Academies.
It's only the most readable modern language and the one that is most likely to catch programmer errors at compile-time. These two factors alone make it the most cost-effective language to use for any project of substantial size, and for any type of project from writing commercial off-the-shelf apps (short time-to-market) to software that will live for decades (high maintainability), including system apps (like an operating system).
Speaking of which, I am in the first stages of designing a new OS (*nix-like) in Ada targeted to Intel/AMD architecture processors. I expect to make an announcement here when there is more in writing, but anyone that wants to discuss it or work on it from the first may reach me at toolmakr at buzco dot nyct.net.
Well gee, I guess the DoD should rethink the quadrillions of apps it has in ADA then. It was never really in the "market" unless you were programming for the military. It had a really nice 20 year run in that capacity too.
To the best of my knowledge, Ada is being taught as a first programming language in some 200+ universities around the world, including in the US Military Academies.
I know that at least a few Australian universities teach Ada first.
Are you trying to say that the respective CEOs of Intel, AMD, and Microsoft are NOT reading some unknown's post on slashdot RIGHT NOW!?! Are you implying that these same CEOs are not SHOCKED and AWED by the allegations contained in that same post?
I insist you retract your libelous words this instant, sir. The world will be brought to its knees by that post. Things will change. You wait and see.
http://www.tadpolecomputer.com/html/ they have been selling sparc laptops for a while.