Dual-Core Pentium 4 Slated For 2Q 2005

Quantrell writes "Today is the first full day of the International Solid-State Circuits Conference in San Francisco, and Intel has announced that dual-core Pentium 4s are coming in the second quarter, one in the Extreme Edition line (no surprise there), and also the Smithfield Pentium 4 800 series, which is the next so-called consumer desktop line. No word on pricing, yet."

Comments

hehe

[essreenim]

looks like I was right - VIIV is dual pentium V (5)'s not a pentium 64. Or am I wrong : (

Re:hehe

[lhaeh]

I believe they have 64 bit instructions as well.

Intel was quiet about implementing that since its an AMD tech.

Re:hehe

[stupidfoo]

I don't know, but we won't find out until 2Q, whenever that is.

Re:hehe

[Anonymous Coward]

2Q = 2nd quarter of the year = April,May,June

Re:hehe

[Skye16]

Should be Q2, not 2Q. That's what grandparent was referring to.

What kind of sockets will there take?

[lhaeh]

I don't remember reading about anything new, but I can't see these working on cuttent chipsets.

Re:What kind of sockets will there take?

[bhtooefr]

I do see dual-Xeons working on an i875, though, so I wouldn't be shocked at all if there's an LGA775 dual-core CPU.

Re:What kind of sockets will there take?

[David Ziegler]

That must be because you didn't read the article:

The Pentium Processor Extreme Edition will be combined with a new chipset named the Intel® 955X Express chipset, formerly codenamed "Glenwood," that includes features such as Intel® High Definition Audio, PCI-Express and faster dual-channel DDR-2 memory.

Intel will also couple its mainstream "Smithfield" processor with two new chipsets named the Intel® 945G Express chipset and Intel® 945P Express chipset, both previously code-named "Lakeport" in the second quarter of the year.

And:

And before you ask, the new dual-cores require new chipsets, so these CPUs will not run on existing systems. AMD fans, meanwhile, can still hope that their dual-core parts will run in Socket 939.

Re:What kind of sockets will there take?

[Zocalo]

I wouldn't be shocked at all if there's an LGA775 dual-core CPU.

The question then becomes one of how well the existing motherboards would cope of course. The ideal would be for it to be entirely transparent and the dual cores are handled by the CPU in a similar manner to hyperthreading. If that's not possible then we'd be looking at a BIOS update at least, and even then it might not be possible to maximise the benefit of dual cores with out a motherboard designed for the purpose.

In any case, with AMD in a similar situation with its own upcoming dual core CPUs, it's going to interesting to see how the two companies approach it. There's going to be some unhappy customers if one company manages to enable upgrades to dual cores on current motherboards and the other doesn't, that's for sure...

Re:What kind of sockets will there take?

[Elminst]

Perhaps if you RTFA, you'd find out;

And before you ask, the new dual-cores require new chipsets, so these CPUs will not run on existing systems.

but wait, this is slashdot. Where posting comes before reading.

Re:What kind of sockets will there take?

[Anonymous Coward]

"That must be because you didn't read the article:"

Maybe it is because he has already read Dual Core Intel Processors Sooner Than Expected article?.

"AnandTech reports that Intel's Smithfield processors are going to get here sooner than they originally predicted; most likely within the next few months. Apparently, the Intel roadmaps reveal that the launch dates for next generation desktop chipsets, 2MB L2 Prescotts and Dual Core Smithfield processors (operating at 3.2GHz per core) are almost upon us - way ahead of the original Q4'05 roadmap estimates. Hopefully, that means Intel will actually start shipping the new technology instead of waiting four months after the announcement for retail products."

Re:What kind of sockets will there take?

[mako1138]

AMD fans, meanwhile, can still hope that their dual-core parts will run in Socket 939.

Kevin McGrath (AMD's chief architect of x86-64) gave a talk about dual cores at my school last month. I asked him if 939 would support dual cores, and he said it would, though he didn't have a timetable. He also reiterated that we'll be seeing dual cores coming on all product lines.

Part of the reason AMD can do this, I think, is their discipline in keeping a consistent power envelope, so the motherboard and heatsink manufacturers don't have to scramble to support a new incredibly hot processor. I anticipate that Smithfield will require massive cooling.

The other reason is the memory controller is built onto the processor, as opposed to Intel's traditional arrangement of it being on the North Bridge of the chipset. Thus no change of chipset is needed (in theory).

Re:What kind of sockets will there take?

[corngrower]

I think you'll be able to screw them into a standard light bulb socket. I don't know how much light they'll give off though. Maybe quite a bit for a brief instant as you flip the switch on.

Re:What kind of sockets will there take?

[lhaeh]

It says new chipsets, not new sockets. There is the posability that it will use LGA775.

but wait, this is slashdot. Where posting comes before reading.
Like reading my question? :)

Re:What kind of sockets will there take?

[nadadogg]

Well, the main article that this discussion is attached to claims that it will require an all-new chipset, so I wouldn't get your hopes up too high.

Re:What kind of sockets will there take?

[ip_fired]

There is a good article on AMDZone that talks about how AMD is doing it's dual-core. They have put 2 cpus each with their own cache but they share the on-chip memory controller and the hypertransport links. According to AMD, there is only a 10% loss in performance by using the shared components.

Not bad considering you only have to have one socket on your motherboard to accomodate a dual processor system now. And it will even work in current motherboards, using the same 939 socket.

Re:What kind of sockets will there take?

[bhtooefr]

Well, I didn't say I wanted one ;-)

Tell me when Intel gets a dual-core P-M to the desktop, or even when AMD gets their dual-core A64 out, and then I'll listen.

Re:What kind of sockets will there take?

[nadadogg]

I would probably hop back over to intel if they were to get a dual-core to the market before AMD does. I don't do too much gaming, but I do, however, do a lot of video conversion/etc, and that requires some more horsepower than my barton 2500 can put out.

Re:What kind of sockets will there take?

[Elminst]

The answer is still the same;
You CANNOT use it in your existing system. You HAVE to buy new mobo even if it is the same socket.

Re:What kind of sockets will there take?

[lhaeh]

There are other considerations. What about my water cooling system, will I need a new water block?

About time...

[Anonymous Coward]

Twice the inefficiency!

Re:About time...

[garcia]

Yeah, I wasn't so concerned about the price of the chips themselves but in the increase of your electric bill.

Either the machines will be sucking so much power that your lights will dim while running RC5-72 or your AC will have to be cranking in order to keep the room cool.

Re:About time...

[JudahGabriel]

Twice the inefficiency! (Score:5, Funny)

While the parent is modded up as funny, what was said isn't too far from the truth, at least from a software perspective.

There's a saying that goes: Joe gives a little, John takes a little. In this case, the more hardware improves, the more inefficient software becomes. This process is initiated as developers start ignoring minutia and low-level details of software development, use high-level frameworks, and stop investing time implementing efficient algorithms.

(yay I'm unbanned from /.)

Re:About time...

[lateralus_1024]

"Twice the inefficiency!"
Yes, but you forgot to note at 4x the TCO, give Intel Corp. all their dues.

Re:About time...

[eric_brissette]

...or your AC will have to be cranking in order to keep the room cool.

You should be fine so long as AC and the computer aren't on the same circuit in your home.

Re:About time...

[mrwonton]

I don't know about AC, but I live in Michigan, and plan on putting one of these in every room. Central heat? Naaah, distributed dual core pentiums.

Re:About time...

[gfody]

yea we should all go back to reinventing the wheel each time, cause' those high level frameworks we use are so inefficient.

Re:About time...

[HungWeiLo]

...but if the AC is using the dual-core computer, then it can first pr0st twice as fast!

Lack of bandwidth?

[ajiva]

While dual processors is great and all, I'd rather see double the memory bandwidth then double the processing power. In the case of Intel processors (especially duals) memory bandwidth is severly lacking, and while DDR-2 should help a bit, I don't expect to be that impressed with the new dual cores.

Re:Lack of bandwidth?

[lhaeh]

More cache would help too. It would be a good idea if they somehow got the processors to share cache. If they are both working on the same thing it would mean a big performance increase.

Re:Lack of bandwidth?

[yamla]

I'm not sure the extra overhead would be worth it. That'd be a heck of a lot of synchronisation to prevent conflicts.

Re:Lack of bandwidth?

[Monkelectric]

That's the real problem with hyperthreading. Running 2 processes at once (in the sense that it does, anyways) blows up your cache. You really need 2 seperate caches or a muhc much larger one.

Re:Lack of bandwidth?

[Jeff DeMaagd]

In the past, dual CPU Intels had a slower bus because of transmission line issues, a multi-drop bus is harder to make work properly than a point to point bus. But given that there isn't a distance between CPUs, and that the CPU bus is prossibly wired together on-die, that transmission line issue drops back down to a point-to-point bus.

Re:Lack of bandwidth?

[Anonymous Coward]

Rewritten with proper grammar and spelling:

While dual processors are great and all, I'd rather see double the memory bandwidth than double the processing power. In the case of Intel processors (especially duals) memory bandwidth is severely lacking, and while DDR-2 should help a bit, I don't expect to be that impressed with the new dual cores.

Re:Lack of bandwidth?

[fitten]

Depends what the threads are doing...

Re:Lack of bandwidth?

[exley]

In that case, you'll have to talk to memory, chipset, and board people. The memory interface is still in the northbridge/MCH (memory controller hub) on Intel-based architectures, as opposed to AMD's x86-64 offerings which have memory controllers integrated into the processor. Also, when it comes to memory, bandwidth isn't the only issue; latency is also critical.

Re:Lack of bandwidth?

[exley]

They can have processors share cache if they want to; there are disadvantages to this, though. Cache is relied upon heavily, which means lots of requests going to the cache from the CPU core. With two or more cores trying to access the same cache, you're going to have collision issues. With two separate caches, it's not hard to have them both maintain the same information and make sure that things are kept in proper order due to cache coherency protocols.

With separate caches, each CPU has its own resource, and no matter what, they are still working on the same thing -- keep in mind that cache is merely a reflection of what's happening out in main memory.

Re:Lack of bandwidth?

[superpulpsicle]

The thing I don't understand from Intel or AMD is why are they putting so much effort into designing and manufacture so many chipsets and cache changes for minimal performance increase.

Whatever happen to just simple gradual mhz improvement. 30mhz to 40mhz to 100mhz to 500mhz works for any consumer. Both companies are just a marketing mess now.

Re:Lack of bandwidth?

[Jeff DeMaagd]

Their ability to scale frequency is diminishing, lately, the ability to lay down more transistors meant bigger caches, which often has less impact than a second core would. Don't forget that nearly every CPU manufacturer is going dual core, not just Intel and AMD, it is just that they are relatively late into the game now.

I don't think it is too out of line to expect that programmers are going to start considering better multitreaded design. There are limits to what can be done, but for most software, that limit hasn't been approached, in my opinion.

Re:Lack of bandwidth?

[SkyWalk423]

Sounds like you need to retake physics. You think they just gave up on the clock speed race because they got bored with it??

Re:Lack of bandwidth?

[Kupek]

Most CMPs designs have give them a seperate L1, but shared L2.

Re:Lack of bandwidth?

[ArbitraryConstant]

Or do what IBM does and have the processor pass hints to the OS about how well the processes are coexisting together.

Re:Lack of bandwidth?

[timeOday]

That's the real problem with hyperthreading.... You really need 2 seperate caches or a muhc much larger one.

Didn't they already try that? The P4 Extreme Edition. It hardly helped at all.

Re:Lack of bandwidth?

[corngrower]

In some apps, it turns out, that multithreading doesn't help much might be expected. When one really analyzes what's happening, one often finds that one thread locks a resource that's needed by the other threads, so effectively only one thread ends up running at a time because of record/resource locking. In fact a well written, non threaded code can frequently execute faster because they don't spend half their time making calls for acquiring and releasing locks on resources.

But I'm sure you're correct in that software designers will start paying more attention to how they can make their software run more efficiently on multi-core processors.

How many developers out there have yet to realize that size optimization is speed optimization? Reduce your data and code size and the processor can spend more of its time executing from cache, and less time fetching from that relatively slow main memory. Also avoid deeply nesting subroutines within major loops.

Re:Lack of bandwidth?

[Anonymous Coward]

More cache would help too. It would be a good idea if they somehow got the processors to share cache. If they are both working on the same thing it would mean a big performance increase.

POWER5 has a shared L2 cache for its cores. It's probably a little too much for work for the first generation. Maybe Intel will do it with the second generation.

Re:Lack of bandwidth?

[InvalidError]

What do you suppose MP rigs have been doing so far?

SMP and CMP can be more than an order of magnitude more efficient at managing thread synchronization simply because internal chip bandwidth and buses are far less expensive than package/socket/PCB pins and bandwidth. Also, propagation delays across a monolithic chip are far less than they would be across IO-pads + package + socket + PCB + etc.

Multi-core synchronization can be an order of magnitude more efficient than multi-chip can be because multi-core is not exposed to nearly as many unknowns and physical/economical/electrical limitations.

Re:Lack of bandwidth?

[SacredNaCl]

In the case of Intel processors (especially duals) memory bandwidth is severly lacking

They do have 800 FSB with 64 bit support in the Xeon's now, which makes them quite attractive dual processor machines provided you have the budget, and DDR2 support as well...
The days of the 400/533 FSB for the Xeons while everything else goes passing by are over. Of course, they don't yet have the 1066 of some of the extreme editions.

That isn't what gives me pause though. I've been considering a dual xeon purchase using either a Tyan or an Iwill MB. I would like to see how these dual cores compare to two physical processors and if the Xeon line is going to get a major upgrade along with it, thus ending my purchases future proofing.

On the other hand, if they do come out with a dual core Xeon, and it needs a new chipset... I'll be able to buy the system I want to buy now for bargain basement prices as people upgrade, and even without an upgrade path it may still be a fastastic machine. I just hate having to wait though, I already waited for 800FSB and dual PCI-E (though only the tyan is truly 16x/16x the Iwill is really an 8x/8x config).

I need a better magic 8 ball.

Re:Lack of bandwidth?

[Your Average Joe]

You can have double the bandwidth, just buy a multi processor AMD system...

Re:Lack of bandwidth?

[haxor.dk]

Theres ample RAM bus band width (no pun intended). It's RAM latency where the improvement is needed.

Re:Lack of bandwidth?

[phoenix_rizzen]

Intel needs to go back to the drawing board and re-design their SMP implementation. There's nothing worse than adding CPUs to a system, only to reduce the amount of memory throughput available to each CPU.

With Intel's SMP setup, total system memory bandwidth is a constant, regardless of the number of CPUs in the system. As you add CPUs, the amount of bandwidth available per CPU decreases.

Throw a couple of CPU cores into each CPU socket and you lower the available memory bandwidth to each core even more.

With AMD's SMP setup, total memory bandwidth per CPU is a constant. As you add CPUs to a system, though, total system memory bandwidth increases.

Not sure how this would work with multiple cores in on CPU, though. Would there be a separate memory controller per CPU socket, or per CPU core??

Re:Lack of bandwidth?

[runderwo]

If the scheduler is smart enough to keep threads that belong to the same process on the same CPU, this effect is minimized since threads share the same pages (and thus cache entries). Scheduling threads from one particular process across multiple CPUs is not a bad idea when the CPUs are otherwise idle, but it's a terrible idea when the CPUs are busy, if one is to expect HT to lend any advantage.

Re:Lack of bandwidth?

[pipingguy]

That's a good comment I think. Parallel processing and bigger words (8>16>32>64) are the way to go, but then software has to be re-written/modified to use 64 bits and multi-threading.

Since current computers are more than fast enough for what, 90% of most users' needs, will there be a demand for dual Opterons once the price comes down?

Would a true "dual system" require dedicated, seperate hard drives?

Re:Lack of bandwidth?

[tod_miller]

Whatever happen to just simple gradual mhz improvement. 30mhz to 40mhz to 100mhz to 500mhz works for any consumer.

They were pWnd by physics, they used all kinds of optical tricks to shrink down the details of thier wafers, but then they realised that when it got too small, it got really hot, then the electrons started to orgy, going into each others homes, having car key parties... typical behaviour for your average electron if they can get away with it.

And now have to just sell two computers as one... think of it like SLI for your processor.

Don't worry about marketting to Joe Pubichair, PC World reps know what to do in any case: Hi Joe, this is the one we make the biggest markup from, so it must be good! ;-)

Well...

[Blue-Footed Boobie]

I hope AMD isn't too far behind Intel on this one...

Looks like the "Who is Winning the CPU War" line just shifted again.

Re:Well...

[essreenim]

No, they have a dual core lineup codenamed 'Egypt' comming out.

Re:Well...

[BlueThunderArmy]

Looks like the "Who is Winning the CPU War" line just shifted again.

No, I think Apple's still on top of that.

Re:Well...

[Anonymous Coward]

HAHAHHAHAHAHHAHAHAHAHAH
thats cute

and technically it would be Apple anyways, it would be IBM, apple doesnt do shit when it comes to chips.

Re:Well...

[mako1138]

Not really. Intel has been playing catchup all this time, first with 64-bit and now with dual core. Opteron was built from the ground up to support more than one core, which is the beauty of it.

Here's a long discussion on the current dual core situation on Ace's Hardware. (They use a lot of codenames. "Smithfield" I think is what this /. story refers to. "Yonah" is somewhere in the future.)

Re:Well...

[Leroy_Brown242]

Well, now with two cores that are near half as efficient, I think they are just about to catch up. :)

Re:Well...

[Vampyre_Dark]

Only if you hold the charts upside down.

Re:Well...

[Blue-Footed Boobie]

Yeah, I was aware of AMDs inplementation.

But, paper-launches aside, I want to see who actually has a chip "in-stores" for purchase first.

Re:Well...

[leoxx]

You mispelled IBM.

Re:Well...

[Anonymous Coward]

the PPC spec was co-designed by not just IBM & Moto, but the AIM group (Apple, IBM, & Moto).
So yes, Apple does contribute to chip design.

Re:Well...

[Anonymous Coward]

Yeah, cause everyone knows that RISC is better than CISC.

Right?

Re:Well...

[ArbitraryConstant]

"I hope AMD isn't too far behind Intel on this one..."

Looks like a few months.

"Looks like the "Who is Winning the CPU War" line just shifted again."

It would be more impressive if these weren't P4 cores. Heat dissipation and throttling will be a problem. Prescotts already throttle under high load in normal thermal conditions, and now there's two of them. AMD has comitted to keep their dual-core processors under 100 watts total.

Re:Well...

[Anonymous Coward]

apple doesnt make cpu's.

Re:Well...

[drinkypoo]

Is there any citation that describes how much expertise apple actually lent to that? My assumption is that apple provided a demand for chips and input on how they would like the instruction set to look, and very little else. IBM and Motorola are known for making their own silicon. Apple isn't.

Re:Well...

[pipingguy]

Does this mean that if I bought a dual Opteron system (with dual cores each, thus adding up to 4 processors) I'd have to buy XP Server (or whatever it's called) rather than just XP Pro?

Re:Well...

[mako1138]

No, you don't. =) Most software makers will license by counting the number of sockets, Microsoft included. But others will count the number of cores. See this.

There's no XP Server; the existing products are Win2K Server and Win2K3 Server Standard.

Well Intel I got to hand it to you . . .

[OverlordQ]

you may come out with dual-cores before AMD, but since your dual-core is a kludge on top of a kludge, I'm guessing AMD will beat you again.

Re:Well Intel I got to hand it to you . . .

[tesmako]

A kludge winning out in the end sure would be consistent with x86 history.

Re:Well Intel I got to hand it to you . . .

[fitten]

I'm not sure where people keep getting the idea that dual core Intel parts are "kludges". They are kludges in the same vein as having two processors share the same bus, the way that the Xeons (P2 and up) have all shared the same bus. There's not a lot of magic there. It's pretty easy to do in the scheme of things but it isn't necessarily as efficient as other, more complicated ways of doing it. Basically, some Northbridge logic has moved into the package and onto the die to allow arbitration of the external bus to the dual core chip. That's about as simple as you can get and exactly mirrors the current (and past) Xeon multi-processor bus architecture.

Re:Well Intel I got to hand it to you . . .

[ackthpt]

A kludge winning out in the end sure would be consistent with x86 history.

As underscored by the following strategy:

and also the Smithfield Pentium 4 800 series, which is the next so-called consumer desktop line.

Doesn't seem that long ago that people at Intel were saying absurd things like, 'consumers will never need 64 bits' or 'consumers will never need dual core'.

Hell, look out the window at El Camino Real and tell me how many of those consumers crusing up and down that road need those 4WD vehicles. Yet consumers buy them in droves. Consumers want, you don't offer, you surrender a market. Seems they've learned not to underestimate what consumers want (which often has little to do with what they need.)

Re:Well Intel I got to hand it to you . . .

[Gr8Apes]

They are kludges compared to the much better internal architectures of the AMD dual core chips, or the potentially even better Power5 chips.

Re:Well Intel I got to hand it to you . . .

[mapmaker]

your dual-core is a kludge on top of a kludge

Maybe Intel should call them dual-kludge processors instead of dual-core?

Re:Well Intel I got to hand it to you . . .

[Anonymous Coward]

You're a pathetic little shit. If I were you, I'd inhale cyanide gas.

Re:Well Intel I got to hand it to you . . .

[fitten]

Well... some years ago, these "kludges" were considered state of the art, not shoehorned solutions that barely scrape by.

Perhaps the better way to phrase it is:

AMD has a better dual-core architecture than Intel.

The use of the word "kludge" is very subjective. I would agree with you that AMD's multiprocessing architecture (HyperTransport - AMD's dual-core architecture is just their multiprocessing architecture on one die) is superior to Intels current architecture (shared bus), but I do not agree that Intel's architecture is a "kludge" any more than any older (or obsolete) technology becomes a "kludge" when something newer and/or better comes along. Perhaps those who are using the term "kludge" actually mean to use "obsolete".

Re:Well Intel I got to hand it to you . . .

[Gr8Apes]

I would phrase it more kindly, except Intel wasn't first. They weren't even second or third. IOW, they were caught with their pants down and "kludged" together something so they wouldn't be caught in dead last with nothing to show other than 10 year old technology (P6 single core).

A dual core shows that they're at least attempting to play in the same ballpark, no matter how badly executed their dual core solution is. (130W - YOUCH!)

What about P5?

[pranay]

What about P5? I will need it to play my copy of Duke Nukem Forever!

Re:What about P5?

[KyleJacobson]

Well this might be your first problem with Duke Nukem Forever. http://www.vgcats.com/comics/images/040815.jpg

Awsome!

[irokitt]

Now I can fry two eggs at once!

Re:Awsome!

[Anonymous Coward]

Actually, now you can have green eggs and ham :)

Re:Awsome!

[Vampyre_Dark]

Or make waffles with your laptop. MMMMM QWERTY flavour!

Re:Awsome!

[gnuman99]

Now I can fry two eggs at once!

No, that what Itaniums are for.

Processor ^2

[CKnight]

This sounds like HT on steroids. Speaking of which, is the HT technology run from the core? If so you can now have 1 processor acting as 4.

Re:Processor ^2

[mako1138]

Yeah, HT is built into the processor. It'll never give you as much of a performance increase as a true separate physical processor though.

This sounds like HT on steroids.

HT is Hyper-Threading. You're running some threads with the bits of the processor that aren't in use. Marketing has cleverly made HT synonymous with dual processing. Dual core is true dual processing, with two full sets of ALUs, FPUs and whatnot.

Re:Processor ^2

[Zemplar]

"This sounds like HT on steroids. Speaking of which, is the HT technology run from the core? If so you can now have 1 processor acting as 4."

Too bad Intel already used the 'Quad Pumped' moniker for their FSB!

Re:Processor ^2

[questionlp]

I believe the consumer version of the dual-core Pentium 4 will not have Hyper-Threading enabled [1] whereas the more expensive Xeon version will. Part of it is due to market segmentation and power consumption.

[1]: http://www.anandtech.com/printarticle.aspx?i=2329 (page 3)

Re:Processor ^2

[ArbitraryConstant]

It's really nothing like HT...

HT is very difficult to use meaningfully. Performance gains are trivial unless the two processes behave very well together. Dual cores should actually be able to be used meaningfully together.

Re:Processor ^2

[jammindice]

All windows xp users can only use 2cpu's, does this mean i can only run my "Quad Pumped" Intel on my Gentoo Box?

seems like a shameful waste of hardware to me.

--

Re:Processor ^2

[InvalidError]

HT might not be as good as a second core but keep in mind that most people run a fundamentally (mo)no-tasking working environment - their CPUs are 99% idle or running spyware/virus/NAV 99% of the time.

HT is nice for people who want to maximize throughput, not so great for people who want peak performance for fundamentally single-threaded code. For the extra ~6W my P4 PC uses while HT is enabled and running a pair of SETIs, instead of completing one WU per nearly 3h I get through a pair of WUs in roughly 4h, a >50% throughput gain. My P4 is also the only PC I have which consistently plays videos smoothly, the others (P3-1G, A64-3000+) look like they drop a frame every now and then.

So, when I look at the P4 as a media-oriented CPU designed to deliver the smoothest media playback possible on a PC, HT is plenty good enough and quite possibly less likely to hurt than the upcoming apparently half-baked Intel CMP.

In any case, I got my P4 last year, my A64 laptop this year so both AMD and Intel have 3-4 years to sort their stuff out until I let them tempt me again.

Re:Processor ^2

[DudemanX]

Actualy Windows XP is smart enough to recognize a HT CPU as one real CPU and one virtual CPU. That's why you can use HT on XP Home(which only supports one CPU). So I imagine that at the very least XP Pro would see a dual core CPU with HT as two real and two virtual CPUs and work just fine like that.

What I'm interested in seeing is how XP Home will support dual cores. Even if it can't yet tell the difference between a dual core CPU and two physical chips , I'm sure MS will end up patching it or something. I can't see Intel releasing a dual core "mainstream" CPU that isn't fully supported by MS's "mainstream" OS.

Re:Processor ^2

[mako1138]

For the extra ~6W my P4 PC uses while HT is enabled and running a pair of SETIs, instead of completing one WU per nearly 3h I get through a pair of WUs in roughly 4h, a >50% throughput gain.

Nice, nice. That's an impressive gain. I've read that multithreaded-app users benefit more from HT than multitaskers do.

Re:Processor ^2

[InvalidError]

Apps written specifically for HT should be able to use the cache more intelligently - actually, one of Intel's favourite/recommended uses for HT is cache preloading so the processing thread will generate fewer misses.

Since HT's goal is to help keep the pipelines filled when one thread stalls, I am guessing SETI must be a somewhat missy piece of code and that must be why a pair of them on an HT/SMT CPU works well.

Question

[ICECommander]

Does anyone know how likely the forecast is? i.e. has Intel started production?

Re:Question

[ICECommander]

I mean past 'initial production' because that can mean anything.

Re:Question

[bbrack]

since 90nm fab times should be on the order of 4-5 months, I'd assume Intel is running these in fairly high volumes (and has been for a month or so)

Re:Question

[pair-O-scissors]

2Q '05 isn't that far in future. Doesn't anybody know how these things perform? I would assume that after designing the cores they would have an initial batch to test. Somebody's GOTTA have one of these alongwith the chipset to check out the performance before mass producing. Also would it be too much to expect speed step from the new dual cores?

Pricing...

[riptide_dot]

"No word on pricing, yet."

Is that kind of like saying "if you have to ask, you can't afford it"?

P4 EE - $989

Gotta go; I have to sell a kidney or three to afford this thing...

Re:Pricing...

[Preeminence]

Hm... the question is, will the price include the walk-in meat locker you'll need to cool it?

Re:Pricing...

[fitten]

Actually, suspected pricing was released a few weeks ago, if you follow the rumor sites. Supposedly, dual-core 2.8 Smithfields will be about $250USD and going up from there.

Re:Pricing...

[M0nkfish]

"No word on pricing, yet."

Oooh! Oooh! I have a word.

Expensive.

House heating

[kff322]

Joy! Now I can save energy by heating my house using this new processor! Even better, my forced air system will also serve as a great cooling system!!!!!

Cores.

[Anonymous Coward]

Motorola/IBM have had multi-core PPCs demonstrated ever since the very, very first G4 lab units all those years ago.

Yet no one has ever productized a multi-core PPC. (Unless you count the Cell, which you probably shouldn't.)

Why is this?

Is there something about multicore technology which caused IBM/Motorola to decide it was not worth the bother of putting in a box and selling?

Inversely, is there something about multicore technology that makes Intel think we'd actually start caring about the P4 again once it's included?

Re:Cores.

[Wesley Felter]

POWER4 and POWER5 are multi-core PowerPCs. Currently there are not any multi-core low-end PowerPCs because they simply cost more to make than the customers are willing to pay. But costs are going down all the time.

Re:Cores.

[Anonymous Coward]

Power5 much?
Dual to 8 way.
See
http://www.theinquirer.net/?article=121 45

Re:Cores.

[TheKidWho]

The upcoming 970MX is dual core.

Re:Cores.

[Genady]

IIs there something about multicore technology which caused IBM/Motorola to decide it was not worth the bother of putting in a box and selling?

In a word yes. There wasn't a buyer. Until Apple discovered that hell must freeze over before it can get a G5 into a laptop.

Check out the new Dual Core G4's with nice bus attachments.

Inversely, is there something about multicore technology that makes Intel think we'd actually start caring about the P4 again once it's included?

Yes. It's this brick wall we're starting to butt up against in relation to processor speed and heat dissipation. If you can't keep making things faster you're going to have to parallelize and get a boost by doing things across more processors.

Re:Cores.

[Anonymous Coward]

It was going pretty well till a boy, his mom and her austrian boyfriend tore the place to shit.

Re:Cores.

[EpsCylonB]

Software has to be written to take advantage of dual cores. Now that the ability to scale the clock of a CPU is slowing down it seems like the best way to increase performance is to finally force multithreaded applications on the consumer. You only need the OS to be dual CPU aware to see a benefit however, one core for the OS and one for apps.

Re:Cores.

[T'hain Esh Kelch]

How about the POWER processor? Im pretty damn sure its multicore, and thats a PPC.

Re:Cores.

[badmammajamma]

Ask yourself why AMD never released a hyperthreading chip when they supposedly developed this technology long before Intel. I'm guessing it's because it doesn't really do jack shit except increase cost and complexity. It's a marketing gimmick.

Cooling

[kff322]

On Pentium 4 Box in small print "984093274983 ton heat sink NOT included"

No need for heaters next winter

[GatesGhost]

with the dual cores, you'll have a roaring bonfire in only a few seconds after turning it on. heehee. cause pentiums run hot.

Re:No need for heaters next winter

[Grey_14]

thanks for clarifying at the end there, you know, just in case anyone missed it.

It's arithmetic

[panth0r]

If 0 = 0, 2 x 0 = 0...

Ridiculous

[deusexcrottsma]

The notion of having a dual core processor is ridiculous. I mean, with just one processor I'm perpetually afraid that my computer will attain self-awareness, and with 350 watts or so at its disposal there's no telling what kind of havoc it'll wreak upon my tender organic tissues. Now with two processors in my computer box there's no telling what kind of trouble I'll be exposed to. Why don't you just embed a .357 Magnum revolver into the chipset? Processor designers are deranged.

Re:Ridiculous

[Anonymous Coward]

The notion of having a dual core processor is ridiculous. I mean, with just one processor I'm perpetually afraid that my computer will attain self-awareness, and with 350 watts or so at its disposal there's no telling what kind of havoc it'll wreak upon my tender organic tissues. Now with two processors in my computer box there's no telling what kind of trouble I'll be exposed to.

Then again, maybe they'll fight amongst themselves and never realize that it's you who keep them in slavery-- decoding porn and rendering endless demons in stupid little mazes.

Re:Ridiculous

[fred fleenblat]

Humans have self-awareness with only 25 watts.

How does it feel to know that your computer may have a consciousness 14 times more powerful than your own?

Re:Ridiculous

[Anonymous Coward]

When everyone started talking about kitchen appliances having microprocessors, I never thought they ment as heating elements.

Tee hee

Re:Ridiculous

[wondafucka]

It's just like having multiple wives. With a dual core, they just fight for supremacy. It's once you have the triple core, that they start to take sides.

Re:Ridiculous

[adachan]

Hope you dont plan on running multiple CPUs with only 350 watts....

And how about heat?

[Anonymous Coward]

The real question is: How much heat are these puppies going to generate?

Or is Intel, ahem, blowing smoke here? :)

So what?

[angrist]

I want to know when I can buy a toaster over with "Intel inside"

Re:So what?

[Elminst]

Just cut a slot in the top of your case... the current P-4's are plenty hot enough to toast bread.
you'll still have to flip it tho to get the other side.. ;)

Re:So what?

[corngrower]

This chip would be a good fit fore one of those newfangled two slice toasters. I think we may have to wait somewhat longer for for a chip to power those four slice toasters, especially if it's to interact with the operator like the one on Red Dwarf.

... questions ...

[ninjagin]

So, will it still use socket 478, and when do we see moboards with the new accompanying chipsets and DDR2?

I'm a little leery of getting excited about having more juice squeezed out of the P4 line, and maybe it's because I'm not entirely clued into the extent of the benefits gained from dual-core P4s. Are they doing this just to gain time before they introduce a new architecture?

I'm looking to build a new AMD-based system this summer, even if they are a little later-to-the-dual-core-table. As far as I can tell, this news doesn't present any substantive reason for me to change that plan.

Can someone more knowledgeable help me get some perspective on this?

Re:... questions ...

[Wesley Felter]

So, will it still use socket 478, and when do we see moboards with the new accompanying chipsets and DDR2?

No, it will still use socket 775, and the new chipsets will come out at the same time as the processor.

I'm a little leery of getting excited about having more juice squeezed out of the P4 line, and maybe it's because I'm not entirely clued into the extent of the benefits gained from dual-core P4s. Are they doing this just to gain time before they introduce a new architecture?

A dual-core P4 is cheaper than a dual Xeon. Or to look at it the other way, a dual-core P4 has more performance than an equally-priced single-core P4.

I agree that the Opteron is the way to go.

Re:... questions ...

[Anonymous Coward]

The article stated that it would not fit in existing Intel sockets. Get ready to buy a new motherboard.

The AMD solution will work with the current 939 socket.

Re:... questions ...

[djohnsto]

It will not use socket 478, but it may use LGA775. Dual-core P4's will let you execute 2 simultaneous threads at about 1.5-1.8X speed they would run on a single core P4 (given the same clockspeed). Single-threaded apps will not see a performance improvement (although you could run 2 single threaded apps and get an aggregate improvement). These will probably also be 64-bit enabled.

If you want dual-core, I would imagine Intel's will be cheaper than AMD's at intro. The Smithfield processor is in their performance mainstream segment (i.e. same as current Pentium 4 - not Xeon). AFAIK, AMD will intro dual-core with their Opteron line. Not sure when it hits the Athlon FX / Athlon 64 line.

Re:... questions ...

[merlin_jim]

a dual core architecture is two processors on the same chip, which should theoretically double the computing power.

And in true multithreaded apps that might even be possible. But the sad reality is that the number of instructions you can run per second in the real world is related more to how many bits you can move in and out of the processor than the actual speed of the processor.

Most instructions can be executed in a handful of clocks. adding AND multiplying can generally be done in the same clock. On some procs you can add in ZERO clocks (that's how the add+multiply in one clock trick is accomplished)

So for maximum performance, you need a memory bus that can keep up with that demand. But we're seeing multi-GHz processors on sub-GHz memory buses.

Screw dual core. Give me dual-bus!!!!!!!

Re:Name

[RaguMS]

Which independant tests will confirm is slower than a broken 286.

If by that you mean "Faster than a single-core P4, but not quite as fast as expected", then you're probably right.

Cell details comming

[the_sidewinder]

"Stay tuned to Ars, as our own Hannibal is on the scene, and will be bringing us more news--including details of IBM's Cell processor--later today and in the coming days." I can't wait

Re:Cell details comming

[Anonymous Coward]

Prepare to be disappointed. Cell is not all the community thinks it is. The hype that has been allowed to brew up over cell is a failure of IBM's marketing machine, because generally nothing is quite as damaging to a product as high hopes dashed upon the rocks of mediocre performance. But then again, that didn't seem to be the case with Sony and the "movie theatre quality" graphics of the PS2. But once again, I say, prepare to be disappointed with cell. It is not the revolution you have been lead to believe.

Reading around though

[Anonymous Coward]

AMD has been providing working real dual-core samples to partners for months, whilst dual-core Intel processors are apparently in short supply.

This smells of Intel running to get there first before AMD, so they aren't second again with a technology.

HyperThreading is disabled in the Smithfield dual-core product too, so expect a mere 50% overall performance increase at the same clock speed (2.8GHz, 3GHz, 3.2GHz soon afterwards) for Intel. AMD stand to gain more from dual-cores, as they have no HyperThreading equivalent at the moment, and AMD have said that dual 2.4GHz will be possible, that's two 4000+ rated processors, probably overall performance of 6000-7000+. That's a bit better than the 5000+ performance from a dual core Smithfield.

Dual core AMD will likely perform a lot higher than dual core Intel therefore.

Re:Reading around though

[Kupek]

HyperThreading is disabled in the Smithfield dual-core product too, so expect a mere 50% overall performance increase at the same clock speed (2.8GHz, 3GHz, 3.2GHz soon afterwards) for Intel.

I'm not sure what you mean, so I might be wrong. My understanding of what you said is that with HT enabled, you can expect a 50% performance increase. That is unfortunately not true. I'm part of a research group that's using P4s with HT, and the most realistic speedup you get is under 10%.

The problem with the chip is that eventhough two threads can have instructions executed simultaneously, they share everything. For example, if you run two floating point intensive applicaitons at the same time, they're both using the same floating point unit (units? I can't remember), which makes them basically sequential. Squeezing performance out of these is not easy.

I have some experimental numbers in front of me where on a 4 way SMP where each processor is a P4 with HT. The difference with a particular application is that going from one thread to four makes a difference from 16.0s to about 5.3s. But then going from four to eight (using all HT contexts in the machine) makes a difference of 5.3s to 5.1s.

Re:Reading around though

[trompete]

Did you happen to figure out where the extra .2 seconds was made up inside the code? Wait-states?

Re:Reading around though

[Kupek]

I really don't know. I can make lots of guesses, but they're just that, really. Keep in mind these numbers are two days old, and I took yesterday off. :)

I also will probably never find out, since architecture isn't really my bag; I need these numbers to compare against a different implementation of the same application using some of my stuff. As long as I can get the performance of my stuff to approach those numbers I reported, I'm happy.

Re:Reading around though

[ArbitraryConstant]

If you have two branchy threads that are very easy on memory I can imagine big performance gains on the order of 50%. That's not very common though.

Re:Reading around though

[Kupek]

No, that will actually be just as bad as having two threads doing lots of floating point calculations. Codes with lots of branches make heavy use of the integer units. The most gain to be had is probably by having one thread that is compute bound and the other is I/O bound.

Re:Reading around though

[getch();]

I wish everyone would stop calculating dual core performance by just adding up speed ratings. The increase in performance from adding another core is ZERO for any single-threaded application. Likewise, the benefit for multi-threaded apps depends mostly on how much parallelism was extracted from the application.

If we're talking about an application that exhibits speedup that's near the absolute limit (2x speedup from adding an extra thread), then we can start talking about whether AMD or Intel is faster. The problem is, almost no desktop applications work that way. For example, you'll see about zero benefit for most games.

AMD's dual core solution may well be better than Intel's. The problem is that the parent post doesn't explain why. It hardly even presents any facts at all.

Re:heating solution

[Leroy_Brown242]

I don't think this is something that can be chanced. I suggest buying these computers by the rack!

But honey, they're to heat the house!

Re:Name

[Vampyre_Dark]

FLAMEBAIT? Damn... the dual core p4 already caught fire!

130 Watts!!

[leathered]

The jokes about the heat these puppies will pump out couldn't be more appropriate. An article at Tom's states that the Smithfield core has a thermal design power of 130W making it by far the hottest x86 CPU ever seen.

In contrast, AMD's dual core offering will offer no increase in TDP over their present single core designs.

Re:130 Watts!!

[Jeff DeMaagd]

Isn't 130 still less than the power needed by the latest ATI and nVidia graphics boards?

Re:130 Watts!!

[AFCArchvile]

Thermal design power and electrical power (V * I) are two different things. Still, that means the motherboard needs to provide whatever current level that the dual-P4 needs to operate correctly, and that means every component in there must be able to run correctly. This is in an age when a significant amount of computer crashes are caused by faulty or inconsistent power supplies or components (remember the "bad caps" issue with Abit boards?).

Re:130 Watts!!

[Jeff DeMaagd]

In contrast, AMD's dual core offering will offer no increase in TDP over their present single core designs.

Wait a minute. The very article you linked said this:
Competitor AMD so far has not released TDP specifications for its dual-core processors, which initially will be available only for servers.

Unless AMD has released this info in the last two weeks, then there's nothing to go on to make such a claim.

Re:130 Watts!!

[eRacer1]

Isn't 130 still less than the power needed by the latest ATI and nVidia graphics boards? The GeForce 6800 Ultra card "only" uses 110W according to this article. If the core was 90-nm like Prescott and the latest Athlon 64s instead of 130-nm the power requirements would likely be even lower.

Re:130 Watts!!

[dago]

Also note that Intel's thermal design power correspond to "typical usage" (whatever that means) and thet AMD's TDP is absolute max (see this)

Re:130 Watts!!

[dago]

Let's google it : 'amd dual-core TDP

So that's 95W max TDP for the fastest AMD against 130W 'typical' TDP for intel's. That's abour 50%-100% more depending on the exact configuration & performances.

Good explanation of how this will actually help?

[ewanrg]

For all the articles on here about the new dual cores, I have seen little that explains how those will actually help the user.

I mean, I assume that unless Windows is rewritten to take advantage of dual cores that you won't see much performance increase. And I assume that just getting OS support won't be enough for applications to really see much improvement either.

SO unless you're a reasonable l33t linux dude/dudette, or I've missed the boat (also possible I'm sure), where do I see the advantage of this system?

---

More craziness here too :-)

Pentium numerology

[missing000]

'nuff said.

Re:Intel is a major polluter

[cianduffy]

You get this whinging around every major industrial plant, and indeed every Intel plant

They're well within the EPA (Ireland) limits at their Irish plants. We still have whingers. Including one guy who insisted the yellow colour of the steam at night was in fact sulphur and not just the uplighters they have...

THIS IS GREAT NEWS!

[Anonymous Coward]

I can't wait for the dual-core Pentium 4 ... TV-sized monumental tower-like boxes for the fans to fit in! BOOST THE BUS, DOUBLE THE DECIBELS! :-D

Dupe?

[Anonymous Coward]

Previous story: NASA Proposes Warming Mars

Re:Good explanation of how this will actually help

[Chirs]

It works exactly the same as an SMP system. Any OS capable of handling SMP will be able to handle this, including WinXP Pro (but not "Home").

As for app support, any time you're doing a task that is parallelizable, you may be able to benefit.

If you are running two totally different processes at once, then you get immediate benefits. (And immediate subtle bugs, if the processes share resources and weren't properly written for SMP).

If you are running a single multi-threaded app, you get immediate benefits. (And immediate subtle bugs, if the app wasn't properly written for SMP).

If you only run a single app, and that app has only a single thread, then you will not gain much at all.

New Case Design - Not mentioned

[dsginter]

Nere's a pic of the new form factor that they will be using for the case.

Re:New Case Design - Not mentioned

[bhtooefr]

No, I think this is more like it (and that's the P4 model - the Xeon model is even bigger). I mean, bread can only take so much heat, and the heatsinks to get the rest away would be bigger than the toaster ;-)

mad max processor(s)

[Anonymous Coward]

Kick 'er in the guts, billy. Can't wait to install this in, er' I mean on, my laptop with a cool blower and vents. It sounds like some monstrosity of dual 32-bit processor(s) bigger than a palm pilot and guaranteed to make all my geeky friends drool. What colors does it come in?

What about x86_64?

[ajaf]

Is Intel planning on developing a processor with that arquitecture to compete with actual AMD64? If not, I think they're missing one point before they go with anything else.

Re:What about x86_64?

[JustNiz]

The EM64T extension is equivalant to AMD64.
I'd be suprised if these didn't have it, as it seems to be standard on all new Pentium 4's now.

Dual-core Extrme edition with HT

[JustNiz]

It will be interesting to see how much this baby will cost.

It will also be interesting to see what clock sped it runs at, and how much heat it throws out.

Intel Heat issues

[Reapman]

Can't wait to see what kind of numbers this produces... I'd like to get a unit for converting some legacy music / movies into a digital format so will be interesting to see how much of this turns out to be hype / fact.

Got a question though. sorry if this is somewhat offtopic... as well I know this is /. so Intel == Bad, but since most of the comments I already read on this post were about the intense heat generated by Intel processors I have a question...

I currently run an intel 3.2 Extreme Edition (no... I did'nt buy it, I won it, I'm not that rich or crazy) that runs on average about 42 Celcius degrees... no the fan isn't stock, it's a $40 fan / heat sink I bought, but it isn't an amazing fan unit or anything like that... the last AMD I had was a 700, so I'm pretty "outa da loop" on AMD's, but it ran around 60 degrees Celcius on average and died in under a year. 42C doesn't seem that bad to me.

Does anyone have some articles showing how cool AMD's run? This isn't a troll or anything I'm seriously curious, as I am thinking of buying a secondary system but don't want to overheat my house.

Re:Intel Heat issues

[Quattro Vezina]

If it helps, my Athlon XP 2800+ runs at about 40C on average, and my heatsink/fan is about the same price as yours. I only have one case fan besides my processor fan, however, so if you have more (or less), that could skew things a little.

Re:Intel Heat issues

[Anonymous Coward]

My Althon 64 3000+ (90nm) typically runs at 37-38C for stock speeds and at around 42-43C if I overclock it by about 25%. I could see spikes close to 50C when I overclock by that much but not when I run at the speed it is rated for.

Re:Intel Heat issues

[AFCArchvile]

my 2500+ Barton is running at 42 right now, so you're doing fine. My general rule of thumb is that you want to be below 50 C, and 60 C is in the range where the life could be shortened with sustained exposure.

Re:Intel Heat issues

[basics]

My AMD64 3200+ idles at about 32C and hits maybe 38C under load. I have never run anything to put the computer under load, but after playing WoW for a while, the highest it has been is 38.

This is with the stock heat sink/fan, and I do not keep the room particulary cold.

Re:Intel Heat issues

[ArbitraryConstant]

The 130 nm AMD parts are pretty bad (not Intel bad, but not great), but the 90 nm parts are actually very good. They're rare at the moment, and they're not the highest speed grades, but they're very reasonable for high performance desktop processors.

Re:Intel Heat issues

[corngrower]

That's about where mine's at too. Actually at the stock speed, idleing, its about 40 C. Overclocked to 3000+ (2200Mhz) it idles at about 44 C. It will rise several degrees after it's been busy for awhile.

Severe Incredibleness!

[ackthpt]

Twice the inefficiency!

Intel Pentium IV now with RSN Dual Core Technology!

And we'll have an operating system which takes advantage of it, as long as it's called Linux, right? Because Windows is still shuffling around with getting the 64 bit version done, yet.

NASA & Intel

[Anonymous Coward]

In a related news, a NASA spokesperson said the space agency is planning to buy a bunch of these Dual-Core Pentium 4 to speed ud by a great factor the warming of Mars's atmosphere.

Re:NASA & Intel

[Duhavid]

They should implement a 128 bit file system instead. And get some water over there.

Re:Good explanation of how this will actually help

[Wudbaer]

The problem is not so much the OS (both Linux and XP should be able to make use of multiple CPUs/cores well enough) but the applications. If your main apps that really need that much CPU power cannot make good use of multiple cores it doesn't matter too much if the OS does.

So then

[mcc]

Is there any advantage whatsoever to having a dual-core processor over just having two single-core processors?

Re:So then

[Jeff DeMaagd]

Not much. But dual processor has previously been the domain of workstations, servers and Macs. Soon it looks like dual core will bring dual processing to the consumer x86 PC level.

Also, for Intel, I think having a single physical CPU die means that the FSB can be higher. With previous Intel designs, the dual processor chips had a slower FSB because it is a multi-drop bus, being harder to make work right than a point-to-point bus.

Re:So then

[Richard_at_work]

Motherboard realestate. Laying out an SMP motherboard is expensive, and uses more materials and components than a dualcore motherboard will, thats it really.

Re:So then

[Chirs]

Well, you can put a freaking fast bus between them, and you save on motherboard space.

On the downside, you end up sharing bandwidth to anything off-die (pci, memory, etc.)

With a two single-core opterons, each one can each have their own full-speed bus to memory (and can access each other's memory at a slight latency penalty).

Re:So then

[Kupek]

It's cheaper.

The chip makers have always been trying to squeeze more performance out of a chip. For the past ten years or so, upping the clock speed has been the easiest way to do that. Upping the clock speed is getting harder and the gains are smaller. So what they're trying to do is explore other ways of increasing the performance from a single chip. Right now, dual core looks like a better way of using the transistors on the chip.

Question of OS Software compatibility.

[PhaxMohdem]

My question regarding these processors is this.. Will MS windows pick up a dual core CPU as two individual processors as it would in an dually box, and if so would you be required to run MS XP Pro instead of home, (seing as Home only supports 1 CPU and Pro supports 2).

Also on that same note, if you have a dual core proccessor with hyper threading, creating the illusion of 4 cpu's would this be able to run under MS windows Pro or would it need a "server" edition version of software which supports 4 CPU's... (Same for a dual Processor box with 2 dual core chips in it.)

Any one who can shed some light on this for me would become my personal idol. Thanks.

Re:Question of OS Software compatibility.

[wild_berry]

Please don't worship me, I'm a little geek like you. But I do have two Prestonia Xeons in my workstation in the office. XP Pro recognises 4 processors and schedules work accordingly: I don't have a "server" version of Windows.

Having heard Microsoft say that a dual-core processor will be dealt with as one chip in terms of licensing, it may be possible to use XP Home as the OS. However, I expect that anyone who pays enough for a dual-core proc will get XP Pro with their machine.

Re:Question of OS Software compatibility.

[ePhil_One]

Also on that same note, if you have a dual core proccessor with hyper threading, creating the illusion of 4 cpu's would this be able to run under MS windows Pro or would it need a "server" edition version of software which supports 4 CPU's

Windows 2000 SP4 and Windows XP Pro both run fine on a Dual Xeon P4 w/ HT enabled. Task Manager sees 4 CPU's as expected. Pre-SP4 systems might complain because they are unaware of Hyperthreading, but I think MS had not really gone into the overkill mode that highlight XP. Prior to XP they were pretty trusting of folks, license limits were managed via trust. In other words is an 11th client tried to connect it worked, instead of rejecting you with a nasty message about how you need to upgrade to server edition.

Re:Question of OS Software compatibility.

[Anonymous Coward]

Again, Microsoft has stated that they will treat one CPU as a single CPU for licensing purposes, no matter how many cores it has.

http://www.eweek.com/article2/0,1759,1680142,00.as p

Re:Question of OS Software compatibility.

[JustNiz]

You run windows? ewwww.... you deserve every limitation you've paid for.

Is there any way..

[jvagner]

..to emulate multi-core CPU functionality using discrete boxes? Could it happen that these extra boxes we have laying around could eventually just act as extra-cores?

Re:Is there any way..

[CommieOverlord]

That's essentially what clusters are (in rough form). They emulate the larger SMP computers. And with an OS like Mosix you can run a single OS across the cluster, giving the impression of using one large machine with many processors.

But, no, you're not going to be able to use extra old computers to make your workstation run faster.

Re:Is there any way..

[ArbitraryConstant]

You can emulate multi-core functionality on a single CPU, and I suspect that would be faster. Communication in SMP machines happens at the speed of main memory (many gigabytes per second) with a latency of substantially less than a microsecond. On dual-core chips, I suspect a lot of that can happen in the latency of L2 cache, which is in the tens of nanoseconds.

Between different boxes, the best you can do with consumer hardware is gigabit ethernet, which will give you a latency of under a millisecond if you're lucky, and under a hundred megabytes per second of bandwidth.

It's not worth it unless they don't have to work closely together.

If you really want to optimize your app for SMP before dual cores are available, just get an SMP machine from AMD or Intel. That's the closest you'll get, though I suspect that would be significantly faster than dual-cores.

Re:Is there any way..

[Kupek]

No. That's a cluster, which is a different beast. Communication between nodes in a cluster requires message passing, usually over ethernet. Dual core is closer to SMPs, which means shared memory. Communicating via shared memory is faster.

Think about it this way. Let's say you and a friend are working on a math problem. An SMP is you both have your own sheet of paper, but you can read and write on his sheet of paper, and he can do the same to yours. A dual core is you both share the same piece of paper. A cluster if you both have seperate sheets of paper, and you can only communicate over the phone.

Re:Intel is a major polluter

[Anonymous Coward]

Ah, if only we could all brush away accusations about major wrongdoing so easily.

In the case of Intel, the air and water have been tested and indeed they are polluting.

And regulation? Surely you jest.

The EPA in the USA is controlled by an executive from one of the worst polluting companies in the USA. Under the Bush regime, pollution is considered perfectly all right.

Just as it is in Texas.

Dual-core P4 EE w/ HT - best product name ever.

[eric_brissette]

Honestly. The Dual Core Intel Pentium Processor Extreme Edition with Hyper-Threading Technology.

Am I the only one that thinks that sounds funny? Like someone took a steaming PR doodoo into the buzzword generator at Intel?

Personally, I think I'll hold off until they release the Dual Core Intel Pentium Processor Fusion Edition Titanium Pack PRO with Spastic-Threading Nano-Techno-Giga-Awesome Technology.

Re:Dual-core P4 EE w/ HT - best product name ever.

[WelcomeToTheFallout]

At which point the surface of the earth explodes in a ball of fire; shards of molten, hyper-threaded, silicon flying through the air at ludicrous speeds.

Re:Dual-core P4 EE w/ HT - best product name ever.

[Sardak]

And, if you shorten it down, you get the P4 DCEEHT (pronounced "deceit").

Re:Good explanation of how this will actually help

[gnuLNX]

Really OS support won't matter much unless it is on servers. It is the applications that will have to be re-written to take advantage of it. And I would venture to guess that either current applications will be re-written or they will shortly find there way to the grave as new companies produce similar software that does take this into account.

IMHO I believe that a paradigm (hate that word) shft is about to take place in programming. I would love to see rocessors hit a brick wall and good programmers being worth their weight in gold again! There was a fine art to code optimzation that has all but been lost...but I see it coming back with a vengance...writing parallel code will be pretty tough for a lot of coders.

Re:Good explanation of how this will actually help

[Wordsmith]

Hey, that makes me wonder ...

MS has already said they'll treat dual-cores as single CPUs for licensing purposes. It might be reasonable to assume dual-cores will bring SMP into the mainstream, and within a year be pretty standard in new mid-to-high-end systems.

So will there be a point pretty soon when MS unlocks "Home" -- maybe via a windows update -- to properly use these dual-cores? If it's willing to consider them single CPUs for licensing, hopefully it's also not going to maintain its artificial crippling of these systems for the home users likely to eventually buy them.

I wouldn't be surprised to see an update that allows for SMP with dual-core systems, but that still locks out SMP on traditional SMP machines.

What about Windows licensing issues?

[DrVomact]

AFAIK, normal Windows 2000 won't let you run more than 2 processors on any one box. But if you activate Hyperthreading on a dual core chip, you effectively have 4 processors--2 of which Windows will refuse to use. (I found this out when I got a dual Xeon Workstation at my job, but the doofi in IT wouldn't give me a Windows Server license because I had "no need".) Could this be the reason why Hyperthreading is turned off on the low-end dual cores? Because the average Joe won't want to spring for a Windows Server license to get the full potential out of his new box?

Re:What about Windows licensing issues?

[Anonymous Coward]

Microsoft has stated before that they only count the actual CPU. Dual Cores doesn't matter. And HT shouldn't matter at all either, since it's still one CPU trying to act like there's two.

Re:What about Windows licensing issues?

[dabraun]

Windows XP recognizes the difference between hyperthreading and seperate processors. I have XP Pro running on a dual P4 with HT enabled - that is, task manager shows four processors. This works fine. I'm not sure if the code to tell the difference was ever backported to Win2K (which shipped before HT existed.)

Odds are that they will do the same thing for multi-core (i.e. not count each core but rather each chip - I am sure that Intel and AMD will provide a way to tell the difference just as they do with HT.)

With HT there's actually more to it than just the lisencing issue - XP treats HT processors differently - it knows that if proc 1 and 2 are really the same chip / proc 3 and 4 are the other chip that given two threads it should prefer to run them on seperate physical processors when possible. I am not sure if issues like this would apply to multi core but they might since the two cores will likely still share some things.

Re:What about Windows licensing issues?

[Humorously_Inept]

With regards to Win2K, my understanding is that it doesn't support HT, or more specifically that it's not HT-aware. If you enable HT on your Win2K system it will see two processors and it will try to schedule to them the same way it would to two physical processors. Naturally, that's not ideal. I can't find the link at the moment and I don't remember if it came from Intel or Microsoft but you should avoid HT on Win2K. Couple this with the fact that these new chips won't support HyperThreading and your problems are solved on two fronts. If you want to use HyperThreading effectively, for whatever small benefit it really provides, I think you need to be using WinXP.

Re:What about Windows licensing issues?

[coldlatte]

sounds like turn off the hyper threading for W2k systems, leave it on for 2003 (.Net)

http://download.microsoft.com/download/e/2/3/e23 9d d00-9ae1-41a9-bb3f-c01f658d16b1/hyperthreading.doc
(word format)
Quick quote form the document--
Although Windows 2000 is compatible with Hyper-Threading Technology, we expect customers will get the best performance from Hyper-Threading Technology using Windows .NET Server. This is because the Windows .NET Server Family is engineered to take full advantage of the logical processors created by Hyper-Threading Technolog

Re:Intel is a major polluter

[plinius]

Oh, so just because you don't like these people means they're wrong?

As for EPA limits, GW Bush, who stole the elections of 2000 and 2004, has quietly nullified all of the environmental regulations put in place over the past 20 years.

And we never heard about it on the evening news, because the US media is run by rightwingers.

Re:Intel is a major polluter

[cianduffy]

Did I say I meant the EPA in the US Just because the US government uses an acronym doesn't mean that others can't. Their IRISH plants meant IRISH EPA limits, is what I said.

All This Talk...

[eno2001]

...about which chip is more energy efficient or which one runs hotter, but where is the truth? I am thinking of upgrading one of my RedHat 9 GNU/Linux systems to Fedora Core 3 (maybe the 64-bit distro) and I'm having a hard time trying to figure out which CPU to go with. Intel Xeon or P4 vs. AMD Opteron or Dual Athlon 64s. One of my main considerations is energy efficiency, but... I don't want an efficient PC at the cost of a slower PC. My past experience with AMD has been pretty dismal, but maybe they've changed. I had a DX4 120 AMD clone chip back in the 90s and it had all kinds of problems. I later got a K5 chip which was rated at 233MHz, but seemed slower than my Pentium 200 MMX. So these days, I hear people rave about AMD chips, the 64 bit chips from AMD and Asus mobos. I'm tempted, but... is it going to be more of the same? Is that 2.5 GHz AMD Athlon 64 going to really perform as well as a 2.5 GHz Intel P4? If I go dual (which I tend to prefer) will a Dual Athlon 64 system really rock compared to a dual Xeon? It's nuts!

Re:All This Talk...

[michaelwigle]

I feel your pain on your experience with the old AMD chips. It's true, back in the day I wouldn't have touched them with a 12' memory stick. Every AMD chip I ever ran into did weird glitchy things. But truth be told, after the K5 series AMD started to really pull their stuff together. When they released the Athlon for the first time it was a complete turn around on reliability and performance and they have held to that ever since. I have to tip my hat to the folks at AMD because they started way behind but through actual hard work and a great product they've managed to be a real challenger to Intel. IMHO, go for the AMD chip for a power machine, it's not only an equivalent to better performer but most likely will save you a few bucks at the same time. A win-win situation in my books. :)

Re:All This Talk...

[Metapsyborg]

Is that 2.5 GHz AMD Athlon 64 going to really perform as well as a 2.5 GHz Intel P4?

No, that 2.5 GHz AMD chip is going to perform as well as a 3.5-3.8 GHz Intel chip (or higher).

I've never had a problem with my AMD64 machine, and I run it for long periods of time under a decent amount of stress (and heat); I wasn't in the PC market when the earlier chips were around though so I didn't have a chance to become bitter towards AMD.

NASA could use these

[danbond_98]

Considering the heat one of these will kick out, NASA could use a few of those to warm mars. Surely far simpler than all these global warming gas plans plus they could donate the space cycles to something worth while.

Well, the problem is....

[Duncan3]

One point twenty-one Jiggawatts!!!???

I Hate This!!

[Nom du Keyboard]

Notably, hyperthreading will be relegated to the domain of the Extreme Edition dual-core line for now, as Smithfield's hyperthreading is turned off.

So HT is in there, and they turn it off. I pay for a chip that has it (cost of a chip is rather proportional to its size), and they turn the d@mn thing off.

It's got to be a Marketing thing, rather than an Engineering thing, which is why I want the entire Marketing Department dumped into the same hole the Lawyers are going into.

A$$hole$ truly says it!

Re:I Hate This!!

[TheRealStubot]

It's got to be a Marketing thing, rather than an Engineering thing, which is why I want the entire Marketing Department dumped into the same hole the Lawyers are going into.

That's why they'll be the first against the wall when the revolution comes!

Pricing

[qtothemax]

No word on pricing yet

This news bit had been posted on anandtech a bit ago, and seems decently reliable and realistic. 2.8ghz for $241 isn't bad at all, pricing is right between today's prices for a 3.2 and 3.4. I personally though am waiting for AMD's dual cores which will supposedly work on my current motherboard, though it looks like at first the only dual core will be an FX processor, with the insane price that goes with that.

Re:Pricing

[corngrower]

I haven't read much recently from AMD about what's happening with their schedule is for releasing their dual-core chips. I think the dual core Opteron was supposed to get released about the same time as Intel's dual core chip, and that the dual core amd64 was to be released about a half year later. I wonder if AMD is waiting for Intel's release before releasing their chip. That certainly would take the wind out of Intel's sails. I had heard that they had produced working dual core chips several months ago.

While Intel is certainly not down and out, I think they'll be finding it tough going this year until they manage to re-take the lead on top end x64 chips. AMD will be able to charge nice prices for its top of the line chips for awhile yet.

What do people here know about purchasing plans for IT managers for the coming year? Are they looking to get into 64 bit computing, or are they planning on waiting awhile yet until Intel gets their chips in order?

Re:Intel is a major polluter

[plinius]

True true, but my post was about an American company perpetrating something in America which the American EPA will never do anything about because its run by a corporate polluter.

And that is thanks to little boy Bush, who stole the elections of 2000 and 2004 and whose Republican operatives committed felonies in order to put him in office, for which there is ample hard evidence.

Learn How Bush Rigged the 2004 Elections

ACE?

[Prince Vegeta SSJ4]

wow ACE really is THE place for the helpful hardware man.

Need a dual core, ddr-2 compatible 64 bit processor with that skil saw?

Re:ACE?

[Anonymous Coward]

[aceShardware.com]

Please return to elementary school and learn how to read. Reading is an essential skill that will serve you well for life!

Re:And how about FR?

[cyfer2000]

And what kind of flame retardant they will put in their epoxy and PWB? Are they poisonous or potentially carcinogenic?

Re:Good explanation of how this will actually help

[mdielmann]

If you only run a single app, and that app has only a single thread, then you will not gain much at all.

Well, yeah, but I haven't used DOS for years!

But seriously. I'm currently running one Office app, one CAD program, one development environment, 3 web browser instances, an IM client, and a text editor, not to mention all the various other things Windows and I need to have running in the background.

Re:Good explanation of how this will actually help

[Jace of Fuse!]

If you are running two totally different processes at once, then you get immediate benefits. (And immediate subtle bugs, if the processes share resources and weren't properly written for SMP).

So now all those users infected with Spyware won't notice as much slowdown! Har har har.

(Sorry, I just had to.)

Self fulfilling prophecy

[Anonymous Coward]

If you don't build multi-core cpus then no one is going to write applications to exploit what's not there, are they?

I think when T. J. Watson, Sr. made his observation that the world wasn't going to need more then 4 or 5 computers, he based it on the fact that there were almost no applications written for computers at the time.

Re:Intel is a major polluter

[EXrider]

And we never heard about it on the evening news, because the US media is run by rightwingers.

HAHAHAHAHAHAHA! Now that's funny shit. Fox News, I'll give you that one, but all US media? Riiiight.

Could someone please tell me . . . .

[Anonymous Coward]

why Intel named their new processor after a ham.

Sour grapes make the best whine

[glrotate]

So much angst in you people, it really makes me smile.

Hyperthreading is Overrated

[turgid]

Just look at the scaling graphs here for parallel bzip2. Note the almost linear scaling on proper SMP and NUMA architectures vs. the embarrassing curve on intel's old-fashioned bus architecture with dual pentium 4 xeons with hyperthreading. Also notice the high clock frequency on the intel processors compared with the performace achieved on the "slower" ones...

So you're a neo-fascist then, I suppose?

[plinius]

Because what other type of person would cheer that his party rigged the elections and thereby attacked the core of democracy? FYI, Democrats were also complicit in rigging the elections of 2004.

Naw, I'm a paleo-fascist.

[glrotate]

Old skool baby. You should get help. Seeing boogeymen around every corner isn't normal or healthy.

Re:Naw, I'm a paleo-fascist.

[plinius]

Evidence doesn't matter to you.
Facts don't matter to you.
I should have known you're just a teenage brat.

Re:Intel is a major polluter

[plinius]

It's been proven, if facts matter to you. FAIR has shown an overwhelming rightwing bias. But I suppose you don't care for facts, you prefer to live a superficial life.

Chip War?

[Sundroid]

On the day Intel introduced its dual-core Pentium 4, IBM and Toshiba showed off their baby, Cell microprocessor, the so-called "Supercomputer on a Chip", which Sony intends to use for its PlayStation 3. But a Reuters news article raises this interesting point: "If history is any lesson, Cell is by no means guaranteed to encroach on the most successful microprocessor technology to date, the so-called x86 architecture that is the mainstay of the PC world and the profit center for both Intel and AMD...Intel itself once aimed at pushing the x86 technology to the side with a chip called Itanium...After a cool market reception, the Itanium project drifted away from those grand expectations. Today, Itanium remains a niche product marketed primarily at the relatively limited segment of supercomputers and high-end servers."

Where's the dual core Xeon?

[supremebob]

Shouldn't they release the dual core Xeon before the dual core "Extreme Edition" P4? Business and server apps are better designed to handle multi-processing than Half Life 2 would.

Re:Intel is a major polluter

[Anonymous Coward]

If you really insist on putting an extra G in "whining", I suggest that you spell it "whigning." In English, a G is almost never silent after an N, whereas there are plenty of words where it is silent before an N: reign, impugn, sign, align. "Whinging" looks like it should rhyme with "ringing."

Are the mods smoking crack today?

[lhaeh]

How is that a troll?

Actually, IBM/Toshiba/Sony are the leaders ...

[Evil Schmoo]

Reuters reports that the real news coming out of this conference is the I/S/T consortium's Cell microprocessor is ready to hit the market. The Cell will be what powers the Playstation 3.

Re:Good explanation of how this will actually help

[runderwo]

If you are running two totally different processes at once, then you get immediate benefits.

If the processes are dependent on memory bandwidth, you will have trouble as they will be constantly fighting over the cache. For pure computation, perhaps, but nothing that depends on locality of reference for best results.

Kidneys for sale...

[RandyOo]

Funny you should mention that: I just happened across this site, where a blog entry joking about selling your kidneys to buy a Hush "Quiet PC" ended up with a bunch of unfortunately serious replies.

Too Little Too Late.

[Anonymous Coward]

Yeah...
Genuine Intel - now with Hyper-Cooling, and Symmetric-Multi-Tasking.

Sorry Intel - next time, im buying a Macintosh. They're actually designed to work with dual-core.

Re:Good explanation of how this will actually help

[pipingguy]

If you are running two totally different processes at once, then you get immediate benefits.

Doesn't this require that you tell the system to use each processor independantly?