Itanium Will Only Be Partly Supported by Longhorn 234
ver.sicher.ungsvergleich writes "Although stopping short of pulling the plug entirely on Itanium, MS has said that Longhorn will only be able to work for a limited number of higher-end jobs. On the positive side, Microsoft does see a future for the chip, but that 'big iron' slot is not exactly what Chipzilla envisioned as Itanium's future."
Role for emulation? (Score:4, Interesting)
Will virtual X86 servers running on Itanium be an available option to supply services not supported by native Itanium code?
Re:Role for emulation? (Score:2, Informative)
Microsoft recently bought Connectix, makers of VirtualPC, ostensibly to use their system virtualisation technology in new Microsoft products.
You call February 19, 2003 recent? C'mon man, at least do some research. That's over two human years and 14 dog years. That's 23.46 technology years! This [google.com] Google search turns up links regarding Microsoft's purchase. This [microsoft.com] is the second link in the search!
Jesus christ, I knew slashdot was behind by a couple weeks when they reported things but you could usually rely
I think it's time to pull the plug (Score:5, Funny)
Where the hell is Netcraft when you need it?
Re:I think it's time to pull the plug (Score:2, Funny)
Re:I think it's time to pull the plug (Score:4, Funny)
Yay! (Score:3, Funny)
Intel is losing it's edge (Score:3, Insightful)
From talking to Intel folks quite a bit, it seems like there is a lot of blind pride on Intel's part in their product line and vision, and they dismiss most anything that I raise as an issue with their performance vs. AMD, and that's not a good sign to me.
Intel is not dying that's for sure, but they're going to have to make a course correction and not make another decade long mistake like itanium.
Jerry
http://www.cyvin.org/ [cyvin.org]
Re:Intel is losing it's edge (Score:3, Funny)
Brother, the Athlon 64 was the file that flattened that edge down to nice rounded stump.
Re:Intel is losing it's edge (Score:3, Insightful)
Re:Intel is losing it's edge (Score:2)
This comes hand-in-hand with multiple-cores per chip; server applications tend to have a lot of concurrency, so trading off some raw performance per core for the ability to fit 2-4 on a single chip with a sensible power envelope is a great win. So what if a single core has 30% less throughput if you have twice as many
What do you think their emphasis on power is? (Score:3, Interesting)
You may dismiss power consumption improvements, but if you think about it carefully, improving power consumption IS improving performance.
If you can halve the power consumption of a chip, it means you have the energy budget to now 'double' the power consumption of a chip, and possible double the performance.
Their netburst architecture hit a power wall; its pretty difficult to operate 120W CPUs. If they can get the same performance at 12W, and then incr
Re:Intel is losing it's edge (Score:2)
I hate 3 day weekends. (Score:3, Interesting)
Intel is in transition as far as processor direction, so there's no suprise here. Itanium has been dead for a while. The Microsoft "support" is there only because it's already been written and there probably is some support agreements already in place.
The real news would be what the sucessor to x86 will be.
Is this really a big deal? (Score:5, Interesting)
Re:Is this really a big deal? (Score:4, Interesting)
Should we be pleased that Itanium failed?
I mean, on one hand
Are there reasons other than poor support from Micorsoft for Itanics massive failure? Is it a poor arcitecture?
Like I said, I genuinely don't know.
Re:Is this really a big deal? (Score:5, Interesting)
In the world of good compilers originally envisioned by HP/Intel: No.
In practice: Yes.
Why? Because compilers aren't nearly as good as HP/Intel hoped but state of the art Out-of-Order processors are great. There is only so much theorethically possible ILP to extract in regular code, and good OoO chips extract most of it in an automatic fashion from existing code. So the hardware guys did a better job here than the software guys, and the Itanium bet on software.
To clarify on OoO processors doing most of the possible work in extracting ILP: Even if the instruction window was increased to infinity (that is, all ILP is always found) it would still not yield dramatically much better performance (I have seen estimates of about 25% best-case). So even with a perfect compiler there is just not much to gain, and we do not have perfect compilers. This very high level of performance in extracting ILP is what is forcing the new shift to TLP with architectures like the Sun Niagara.
I don't think we should be pleased that the Itanium failed. As I have often discussed in the past I think Intel really deserves a lot of credit, they are the undisputed top dog in the market, and despite that they are also one of the companies that consistently attempt new different approaches in high-profile products. Neither the Itanium nor the Netburst (which really is interesting and innovative technology) worked out well, but it is trying things that makes technology move forward.
That's not to say that AMD is a bad company, they managed to make the best x86 implementation yet, which is great (though I still consider the K7 to have been the golden age since their pricing structure truly was incredible then).
Re:Is this really a big deal? (Score:2)
Re:Is this really a big deal? (Score:2)
Yeah, this is to some extent true. OoO chips are a nice deal though since they of course also benefit from better compilers, even the complex predicting OoO chips hedge their bets and include a few SIMD units in the mix to be able to capitalize easily on explicit parallelity. But to be practical; Most languages in popular use today are
Re:Is this really a big deal? (Score:2)
I think that going from 8 to 16 register has made something like a 20% improvement for the x86-64.
So 1) it is a big improvement 2) 8 register is a big problem in practice: you take a 20% hit even with register renaming (in pratice the loss is even greater as the new register ar
Re:Is this really a big deal? (Score:2)
To somewhat offtopic comment on the x86-64 register extension: It is still an obviously good idea. Spill code does not actually have to hit memory as long as there are rename registers left, but having to insert tons of spill code will still beef up the number of instructions
Re:Is this really a big deal? (Score:2)
While I dislike very much the Itanium architecture (PPC, Alpha, ARM Thumb2 mmmh), it is obviously much better than the POS x86 that Intel made.. but removing competition?
I'll stick with x86, thanks!
Re:Is this really a big deal? (Score:2)
Making x86 fast has been perhaps the most incredible feat in the existence of man-kind. And it's also been one of the most damaging. If Intel had never spent a buttload of money and come through with the Pentium and produced a CISC chip that could compete against the RISC chips, because in its core it is a RISC chip, then the world would be walking around with much better performance per watt RISC chips running everything.
Perhaps, we might hav
Re:Is this really a big deal? (Score:3, Insightful)
I'll agree that the later designs have been very nice. The Pentium-M and Pentium-D, along with the AMD designs are all incredible.
You say that the x86 design allows you to to take advantage of system busses in ways that many RISC instruction sets don't, but there's nothing that would stop them from doing so, except there's been little in
Re:Is this really a big deal? (Score:2)
When I stated that I thought they should've gone with a SCSI based replacement, I mean that they would literally have chosen a particular type of SCSI. For example, Consumer-SCSI = SCSI3-160 w/ SCA LVD connector. There are plenty of devices out there that would match that, so there wouldn't have been a need to wait for drives. Just about everything auto-terminates and auto-ids now, so you wouldn't have needed to worry about which SCSI-ID a device had
They've had chips fail before. (Score:2)
But they're a big company. They will overcome such failures.
what is amazing... (Score:3, Interesting)
Re:what is amazing... (Score:2)
hummmmm. (Score:2)
In fact, other than Intel, only chips that had either company supported OS or an OS that was not under control of somebody else , has had a long life. And x86 arch. has outlived its purpose only due to an illegal monopoly (as in unnatural) in MS (and probably Intel as well).
Re:what is amazing... (Score:3, Insightful)
So what this article may actually mean, is that there is no market for _windows_ in Itanium space anymore. Which isn't that suprising, when there is hardly any windows/ia64 applications, what use an empty OS is?
IMO what Itanium needs to become a success, is a
Fortran programmers don't need (or want) Windows (Score:2, Funny)
The Itanium was the only realistic chance we had to get away from the x86 for the forseeable future, and the designers blew it. So sad. Excuse me while I start one of my Leonard Cohen albums, I need something to cheer me up.
Re:Fortran programmers don't need (or want) Window (Score:5, Insightful)
It's not gonna happen. The industry likes migration/upgrade paths, and in 90% of all cases, a design that extends is gonna win over one that outright replaces.
Intel seems to have been unwilling to face that fact, but what they failed to realise is that their monopoly is not big enough to simply force change on people - rather, their move just gave AMD etc. an opportunity to slowly but steadily chip away at that monopoly.
From a market perspective, that's a good thing, of course - but if I was an Intel shareholder, I'd demand that heads roll for this gross mismanagement in the top executive floor.
The x86 market: evolution vs. revolution (Score:2)
Instead, you have companies, l
Re:The x86 market: evolution vs. revolution (Score:2)
The execute an instruction you send an opcode to the chip, with certain parameters. Those opcode were what made up the ISA. (ie: MOV, ADD, LEA) Modern chips take those opcodes, and decode them to another set of more simple opcodes. Those are then executed. AMD calls those internal instruction ROPs; Intel cal
Re:The x86 market: evolution vs. revolution (Score:2)
Re:Fortran programmers don't need (or want) Window (Score:5, Interesting)
PowerPC/POWER is still viable, and IBM may have another go at putting them in consumer machines if an OS that runs on PPC becomes popular in the desktop space.
ARM-derived chips are still going strong. At IDF there was an XScale chip demo'd that ran at 1.25GHz - probably fast enough for 90% of users.
Alpha remains my all time favourite architecture - pure 64-bit, and the PAL code concept is remarkably elegant.
Vista? (Score:2, Interesting)
Re:Vista? (Score:3, Informative)
When it comes to this software, many techies will continue to refer to Windows Vista as "Longhorn", which will no doubt confuse many regular users.
Now instead of having one coherent name known throughout the marketplace (ie. Windows Vista), the name has been fragmented (ie. Longhorn, Windows NT 6.0, etc.).
Re:Vista? (Score:2)
I'm sure they'd love to use the eventual product name as the codename, but you can't rush marketing. The marketing people require full technical specifications, comprehensive feature lists, and functional demos before they can even BEGIN the hard work of naming. They need to go over all the possible synergies with parallel products, and c
Re:Vista? (Score:2)
"I only have Windows Vista, but I also need Longhorn and NT 6.0. I'd better get to the new Microsoft Store in Times Square!"
Re:Vista? (Score:3, Insightful)
You speak of fragmentation too... Do you hear a lot of confusion between what Windows 2000 and NT 5.0 is? Do you hear many call Windows XP as NT 5.1? Windows 2003 Server as NT 5.2?
It's the marketing machine that decides, unless maybe for a percenta
Re:Vista? (Score:3, Interesting)
i doubt it (Score:2)
do you hear anyone calling 98 "memphis" or XP "whistler" nowadays?
Re:Vista? (Score:2)
Vista (being the Longhorn intended for Homes and Small Offices) will likely not support Itanium at all... but don't take my word as gospel on that, I don't know for certain.
But Longhorn Server is most definitely *not* Vista.
Re:Vista? (Score:2)
We're not talking about Vista; precisely what the marketing name of Longhorn Server will be hasn't yet come out of MS. Ship dates for Longhorn Server haven't been set either - I'm betting a fair while after Vista ships; similar to the lag from Windows XP (a desktop os) to Windows Server 2003 being released.
(MS produce Server and Desktop versions of their OS's. XP and Vista are desktop; 2003 is ser
Uhhh, (Score:2)
I think you meant "Pig Iron"
Tee hee
Vista isnt the thing (Score:2, Insightful)
Of course, a few years ago Intel hoped to put Itanium in workstations, but they can hardly have hoped much for that lately. No, Itanium is for servers, and there is Windows Server.
However, internally Windows Server is the same shit as Windows Vista, so if they dont support it in one, they probably dont find it very strategic to support it in the other. And as we all know, Itanium is much more dying
Haha... DRTFA (Score:4, Informative)
I better flame myself before someone else does. This was about "Windows Longhorn Server". Sorry Intel - this must suck big time!
New Design Getting Flushed Away (Score:5, Insightful)
If I remember correctly, the IA64 has 128 general purpose registers and 128 floating point registers. It's a load/store machine and it's pretty close to a RISC arch (really it's an "very long word" instruction set, but lets not get picky).
It was a chance to make a clean break from the old 32-bit legacy chips, however the price was and is too high and AMDs are cheaper and still very powerful.
I really hope this chip doesn't die off. At least with limited support in the new Windows, it will still have a strong server market, but I think a lot of companies are going to be afraid to buy because of running into compatibility problems. I know at where I work, we'd like to have servers that can do anything/general purpose. You put a limit on what the OS can do and then you're afraid of old legacy or propriety software not working correctly
But hey as long as you use Linux, the IA64 is fairly well supported, and it will be better supported in Linux than in Windows!
Sumdog
Re:New Design Getting Flushed Away (Score:3, Informative)
What this allows one to do is just slide the window further down the line before making a function call, or sliding the window around while doing a loop, so you can perform some loop operations without changing the instruction's declared register usage, but rather just by sliding the window.
As the register window rotates around,
Re:New Design Getting Flushed Away (Score:4, Informative)
The Itanium instruction sets allows code to access between 32 and 128 general registers (aka integer registers), at the discretion of the piece of code, and 128 floating point registers. The sliding window design is for the integer registers 32-127 and can indeed be considered as a sliding window on a memory stack. It is up to each piece of code to decide how many registers it wants to use (in increments of 8 ?).
On top of that there is the ability to design a subset of the high registers (registers with an index higher than 32) as rotating. This makes modulo-pipelining worthwile by removing the requirement for register-to-register moves to push things down the (conceptual) pipeline at each iteration of the loop.
Re:New Design Getting Flushed Away (Score:2)
Please explain.
An accurate mapping of the accessible register window would be very clarifying. I just remember myself that there are a number of registers that are always accessible and don't get shifted through the window (like the gp, the sp, and the rp) But I also know that a number of registers also are in the register shift window. I just don't know where the split is, and how large th w
Re:New Design Getting Flushed Away (Score:5, Interesting)
One of the lesser known reasons for Intel's plan to develop and push the Itanium was that it would be a clean break with x86, which means that AMD would not be allowed to make them. Intel would be the only supplier allowed to make the chip. Then they'd get sued for it, and would settle by giving rights to manufacture them to some small company with one fab that's a generation or two behind. AMD would have been stuck with x86, and Intel would have won. (Bear in mind that if the switch had been successful, Itanium would have been adopted long before x86-64 and the Opteron were developed.)
Frankly, I'm glad the Itanium failed. Even though it's a pretty cool chip with an interesting design, I'd rather have Opterons available than not.
Re:New Design Getting Flushed Away (Score:2)
If I remember correctly, the IA64 has 128 general purpose registers and 128 floating point registers. It's a load/store machine and it's pretty close to a RISC arch (really it's an "very long word" instruction set, but lets not get picky).''
And that's where you go wr
Re:New Design Getting Flushed Away (Score:2)
That's another point. It's not like IA64 was the only chance to break from the x86 cruft. I mean, PowerPC is doing very well, and I don't think there's a good reason the world couldn't have switched to Alpha or MIPS, either.
The success of AMD64 shows that people don't _want_ to break away from x86. In the world of closed-source software, backward comp
Re:New Design Getting Flushed Away (Score:2)
Intel spent billions marketing IA64. They gave away lots of expensive hardware to developers. They used their monopoly muscle, and cash, to strongarm all the big industry software and hardware manufacturers into supporting it. They persuaded several companies to abandon their own architectures and bet on IA64 for the future. They ensured that analysts everywhere were united in hailing IA64 as the wave of the future. It COULD NOT have been pushed harder.
It's a measure of just
Re:New Design Getting Flushed Away (Score:2)
The problem is that the IA64 architecture asks compilers to DO THE IMPOSSIBLE.
Shades of Pentium Pro (Score:2, Interesting)
MS was slow to get 32 bit support to the Pentium Pro, and Intel twisted in the wind for a couple years with expensive chips and no support for the mainstream.
Now we have Itanium64 and MS is again (very) late with support, and now saying that the much promised and never yet delivered Longhorn will not give the support to Itanium that it will need.
Maybe Intel ought not to accept MS's promise of support for new chip architectures and look to FOSS for their hot new chip's support for
It doesn't matter what Intel wants. (Score:2)
It'll do no good for Intel if open source OSes support their chips years before Windows does, but relatively few people want to use the non-Windows operating systems.
Re:Shades of Pentium Pro (Score:3, Informative)
Besides, the Pentium Pro CPU core design became the basis for the Pentium II, Pentium III and Celeron CPU's.
Re:Shades of Pentium Pro (Score:2)
Re:Shades of Pentium Pro (Score:2)
If by "Pentium" you mean the original Pentium, prior to the Pentium Pro, could you cite a reference that supports that assertion?
Re:Shades of Pentium Pro (Score:2)
I suppose you think it's just some happy coincidence that Apple is going to start using Intel chips? I think Intel has had enough of Microsoft dictating their success or failure.
Apple will be more than happy to include Intel's new whiz-bang chips they come up with in their products. If the yields aren't that great it doesn't matter because Appl
Xserves? (Score:3, Funny)
Also supported (Score:4, Funny)
Itanium Will Only Be Partly Supported by Longhorn
Excuse me (Score:2)
Sorry, what's positive about that? I guess 99.9% of slashdotters could not care less.
Just pull the plug already (Score:2, Interesting)
Its all the things these newer PentiumV's and PentiumMII's that are coming out, supposed to be. They are VLIW and use very little power, slim, and efficient.
Itanium was supposed to really take off back in 1997 according to all the analysists. How many years is that? Good lord!
HP shot itself in the foot because they had no concept of sunkin investments or sunk costs and demanded everyone use their hogs with full 1 pound heat sinks and a fan that sounds like a jet engine taking off.
To me the
"On the positive side" (Score:3, Insightful)
Sounds like... (Score:2)
Did anyone else (Score:2)
Although stopping short of pulling the plug entirely on Itanium, MS has said ver.sicher.ungsvergleich
MS tears technobabble a new one.
What will Longhorn/ Vista have? (Score:3, Insightful)
Everything I've read on Slashdot and Wired talks to what will not be in Longhorn. What will be in Longhorn that will make it better than XP? More and different vulnerabilities? (Maybe it will ship with a demo .wmv file showing Microsoft executive throwing chairs around offices in response to other MS executives leaving for Google!)
Seriously ... though I'm an Apple user from before Macs were released I've also used every version of Windows - always at work but I've also had every version except XP at home.
With each new Mac OS X release I look forward to what will be in that version - but there's little talk around the water cooler regarding what will actually be in Longhorn/ Vista. Unlike Mac OS releases, which people anticipate because of stuff like Dashboard, iTunes integration, .Mac integration, Spotlight and Automator, all I hear is what Longhorn/ Vista won't have ...
What the Alpha engineers thought about Itanium (Score:3, Informative)
http://www.raytheon-computers.com/ref_docs/alpha_
The rest we know; the Alpha was ditched when HP bought Compaq (who bought DEC earlier), because HP wanted to eliminate any threats to its Itanium bet.
Re:What the Alpha engineers thought about Itanium (Score:2)
Re:What the Alpha engineers thought about Itanium (Score:2)
Re:OS x86? (Score:2)
Re:OS x86? (Score:3, Informative)
Though, Apple have themselves used "Mac OSX Intel" to refer to OSX running on Intel hardware. Thus, I stick with that moniker.
Re:OS x86? (Score:2)
The Apple world seems full of these though, iMac, iPod, AltiVec, OS X, etc, etc, etc.
Re:Heh. (Score:4, Insightful)
Re:Heh. (Score:3, Funny)
Re:Heh. (Score:2, Funny)
Wouldn't it be funny... (Score:5, Interesting)
[drumroll please]
I mean, talk about a soap opera:
Who knows what the moral of this story is?Maybe: Hardware comes and hardware goes, but software is forever?
Re:Wouldn't it be funny... (Score:2)
Windows NT was an outgrowth of OS/2, the one-time joint project between IBM and Microsoft. OS/2 grew to become a well thought-out OS now dead from marketing suicide squads, and NT has become
Just because Cutler came from DEC doesn't mean that NT is somehow a descendent
Re:Wouldn't it be funny... (Score:2)
Dave Cutler wasn't just some guy from DEC. He was one of VMS's main architects. Imagining he applied a lot of what he learned writing VMS to NT just doesn't take that much imagination.
Re:Wouldn't it be funny... (Score:2)
Well, it was. They even shared filesystems, until Microsoft took their ball and went home. It's not "just some guy on
http://www.os2bbs.com/os2news/OS2History.html [os2bbs.com]
"1990 - Th
Re:Wouldn't it be funny... (Score:3, Informative)
Really? Because I lived through that era, and I and a bunch of other folks that did think otherwise. I was a VMS user long before NT ever came out. Coming from a VMS background, it's not hard to see more similarities in the foundation of the
Two points.... (Score:2)
1) Does the fact that DEC successfully sued M$FT for theft of intellectual property mean anything to you? The resolution of that suit [or threat of a suit - I forget the details, and I'm too damned lazy to Google them] was that M$FT would port Windows NT to Alpha. Unfortunately for DEC, NT on Alpha never grabbed much "mindshare", and withered on the vine.
[By the way, Alpha's greatest opportunity was lost when DEC missed the chance to become the supplier of the successor to the 68000-series in the Macinto
Re:Two points.... (Score:2)
Re:Two points.... (Score:2)
Ok, ok, poor choice of analogy.
It was a threat of a lawsuit. They shook hands, signed papers, and became best of buddies afterwords, with DEC/Compaq getting a pittance of a settlement and the rights to resell Microsoft products, also for a pittance.
If it could have been proven that, yes, a whole crapload of Mica was dumped into t
Re:Wouldn't it be funny... (Score:3, Informative)
* Packet-driven I/O subsystem
* Delayed Procedure Calls
* Asynchronous Procedure Calls
* The security model at the lowest level
* The object manager (although it was somewhat non-formal in VMS)
I mean, if you've ever programmed both systems at the kernel level you would be pretty shocked how similar they are. I mean, WNT feels like a more modern VMS with some things new and
Re:Wouldn't it be funny... (Score:2)
There was plenty of cross pollination after The Agreement. It just irks me that people spout off that "nt is really VMS" without considering the whole OS/2 history.
There is a history of various DEC people (besides Dave Cutler) working on OS/2/NT going back to
Re:Wouldn't it be funny... (Score:2)
No, but most of the OS/2 bits are in the upper layer for compatability. They don't make up the kernel and basic heart of the OS. That part is very much VMS-like. The OS/2 bits resides up on top like the Win32 and Win16/DOS and POSIX bits.
I'm sorry that it 'irks' you that NT isn't treated more as an offshoot of OS/2, but plenty of us do consider the OS/2 history and still think of the heart of NT as a VMS offshoot (no matte
Re:Wouldn't it be funny... (Score:2)
Re:Heh. (Score:2)
Re:Heh. (Score:2)
Itanium is a horrible CPU, I have no lost love for this beast. I've been waiting for this to happen and I'm happy!
Re:Heh. (Score:2)
I work on the following platforms on a daily basis:
- x86
- SPARC
- PARISC
- Alpha (yes, they are still out there)
- IBM servers, both PPC and mainframes (not sure about the CPU types)
- Itanium
But I guess that is not enough diversity for you?
Actually, Another Nail in the Coffin for Microsoft (Score:2, Insightful)
But, the Itanium is quite different from earlier x86 chips, because it doesn't make guesses to try to run a series of binary statements in parallel.
Instead, the Itanium provides the facility for running statements in parallel, but it leaves it up to the software to decide when and how to do it. That's a good thing, because a compiler, using the original source code, can do a much better job of parallelizi
Re:Actually, Another Nail in the Coffin for Micros (Score:2)
IIRC, MS has already used BSD code for the TCP/IP stack in NT 3.1. MS bought the code from a third company called Spider that had copied and modified the BSD TCP/IP stack.
Re:Who DIDN'T see this coming? (Score:2)
Microsoft. AMD64 isn't the architecture that needs the "Windows on Windows" emulation in NT 5.2.
Re:Just as well ... (Score:2)
Re:Just as well ... (Score:2)
Re:Contradictory news (Score:2)
2. Rip out anything that doesn't work and doesn't have to work if you just use it for massive databases.
3. Make the press release about how you've optimized your offering for the "workloads" that still work on the system.
4. ????
5. Profit!