Intel Unveils New Chips to Battle AMD

An anonymous reader writes "Reuters is reporting that chip giant Intel hopes to get back on track in their continued market share war with AMD when they unveil a new line of chips at their upcoming twice-annual developers forum. From the article: 'AMD, once content to mimic Intel's advances, has set the technological pace in recent years with innovations such as putting two processing cores in a single chip -- moves that have helped it gobble market share from its much-larger rival.'"

Comments

Which innovation?

[hyc]

Of course, IBM had multicores years ago, so AMD wasn't really the innovator on that front.

Re:Which innovation?

[cfx666]

>Of course, IBM had multicores years ago,
So did SUN with their UltraSparc platform. But for the consumer market this really was something new.

Cfx

Re:Which innovation?

[vitya404]

Well, I do think Fokker was a great guy, but whatever you say, the Wright brothers (or Whitehead) were the first to fly with a self-propelled plane. I could enumerate many other examples. It is great to have something that anybody can buy, but it is rarely the first step.

Re:Which innovation?

[Sique]

They just claim to have been flying first with a self propelled plane. Alberto Santos-Dumont was the first to show the large public himself flying in a self propelled airplane :) (and there is still this odd picture allegedly taken in 1902 showing him in one of the early constructions).

Re:Which innovation?

[dunstan]

No, IBM were first with the dual core Power. Sun have now leapfrogged ahead with Niagara, which not only has 8 cores but has four threads per core, so the OS sees a single processor as a 32 way system.

Re:Which innovation?

[Bert64]

But that's multiple physical processors on a single board, that's no more sophisticated than a dual processor motherboard.

Re:Which innovation?

[SillyNickName4me]

Hello moderators, it rather looks like the reply to my post is correct and this is indeed 2 seperate CPUs, and not a dual core CPU.

Re:Which innovation?

[vitya404]

Yeah, you're right in a way. Still, I would rather believe anybody that they have detonated a nuke. They don't have to show it to me, that it works. If I see the bomb, and they tell me that it works, that's fairly enough... :). By the way, some people around did this "do you beleive it?" game for some decades :(

Re:Which innovation?

[quad4b]

I get pretty tired of the Intel / AMD war. There is no innovation in these chips, just iterative improvement. IBM's Cell processor, although it may not succeed, is the first interesting architecture which has some weight behind it, to appear in years.

Intel, AMD, who's better? Who cares.

Re:Which innovation?

[samkass]

There was large amounts of photographic evidence of the Wright brothers' accomplishments, some of which was lost in the Ohio floods early in the 1900's, and some of which survives today. Needless to say, NO ONE is documented to have flown out of ground effect, nor make a coordinated turn, until the Wright brothers demonstrated their plane publicly in France. By 1906 when Santos-Dumont made his little hop, the Wright brothers were flying for 20-30 minutes at a time at heights of 100 feet before spectators from the US Army as well as others in his town.

The Wright brothers didn't demonstrate publicly because they were in it for more than a hobby. Not being an independently wealthy tinkerer, they wanted to make their living making airplanes, and realized that they had the only viable design anyone had come up with, so not trusting the patent system, held out until they could secure agreements with various military organizations. They were engineers more than scientists.

Much of the "evidence" of earlier flight, including claims that Ader flew in the late 1800's, was concocted to try to overturn the Wright brothers' patents on their system of differing the angle of attack of the two wings in order to bank the plane. (Almost no one had banked planes before, either... most others were still thinking of planes like ships that would use the rudder to steer, which at those speeds every pilot now knows would lead to a stall.) Newspaper reports from before the patent battle clearly admit the Wright brothers unique invention, while those after the patent battle try to find almost anyone else to assign the invention to. As most know, though, the Wright brothers won every patent battle they faced and the only "evidence" of earlier flight lies in retellings of myths on sites like wikipedia.

Re:Which innovation?

[ErikZ]

That sound interesting. Does anyone know how well it works?

It sounds fast, but it could act like a beuwulf cluster of 386s.

Re:Which innovation?

[PsychoSid]

It works OK, but only has one FPU per-chip. The next iteration will fix this apparently.

Re:Which innovation?

[justthinkit]

And why do you arbitrarily pick the thing that the Wright brothers achieved as the first significant step? The Wright brothers, even if you accept their disputed claim, were merely the first to fly with a self-propelled plane. But surely a far greater achievement was that of by the glider pioneers who preceded them by decades - Otto Lilienthal and the like - who actually proved that a heavier-than-air machine could fly, and built up the basic understanding of wings and aerodynamics without which the Wright brothers would never have had a frame to strap their engine into?

The progression of gliding begins with a child holding a piece of paper above their head, then jumping. After that it is only about flight duration.

Wright innovations like realizing that the propeller was like the wing, and designing both for maximum lift (unlike everyone else), using a wind tunnel (unlike anyone else) to a level of propeller efficiency comparable to propellers made 100 years later IS remarkable.

As is their work with engines. I believe it was about 5 horsepower but about as heavy as a modern automotive engine. And, as mentioned earlier, their work with control services, introducing one that no one else had ever used before.

In short, comparing gliders to airplanes is like comparing the typewriter to the computer.

Re:Which innovation?

[Sique]

The main accomplishment of the Wright Brothers was the steering. The principle of flying 'heavier than air' was shown to be sound before (Lilienthal et.al.), and the idea to have the plane being self propelled was obvious. It was just a matter of time until the gas engines were light and powerful enough.
But it was the Wright's analysis of the bird flight, and the realisation that you have to have bendable wings and tail/front flaps to get to a controlled flight, that was really new. Ironically it was this idea that was published in the patent application of 1904, which enabled the other flight pioneers to get their planes ready until 1910.

Re:Which innovation?

[somersault]

I think perhaps all those millions of people out there with existing x86 hardware, OS, and (most importantly) software, care

Re:Which innovation?

[SeeMyNuts!]

Sun's MAJC processor was dual core VLIW and was around in the late 1990s, but I don't think that CPU saw much action outside of some graphics cards. It looks like Power 4 went dual core around 2000. Regardless, multi-core CPUs have been around nearly, if not longer than, 10 years, so neither AMD nor Intel are even close in 'innovating' this technology as portrayed in the article summary.

Re:Which innovation?

[default luser]

There is nothing "innovative" about Cell. Cell is basically multiple vector processors on a chip, and is a very predictable path for Sony after the release of the Playstation 2.

Each Cell SPE is simply a highly-optimized vector unit with 128-bit registers. It is capable of operating on 4 32-bit operands per cycle, just like SSE2/3 and Altivec. The difference is, eache SPE runs an independent task, while the Altivec / SSE units execute vector instructions in parellel with normal operations. However, the SPE is cut down: it has no branch prediction hardware or out-of-order execution, and depends on the main processor filling and emptying its Load /Store memory.

If you think the speed makes it innovative, think again: neither the Cell SPE nor its predecessor, the Emotion Engine, are IEEE754 compliant for 32-bit floating-point operations (for speed reasons). Cell can do IEEE754 compliant 64-bit floating point, but at an estimated speed hit of 10x, which makes it just "competitive" with existing solutions.

Sony / IBM actually inflate the performance numbers of the SPE, advertising it as 25.6 GFLOPS. But this doesn't take into account that the two pipelines of the SPE are NOT flexible, and can only perform certain types of instructions. The "Even" pipe can do arithmetic, and the "Odd" pipe can do Load / Store / Permute / Branch. Thus, the maximum arithmetic thoroughput per SPE is cut in half, to 12.8 billion arithmetic operations per second, and the double-percision performance is just 1.28 billion arithmetic instructions per second.

It's a nice idea for a media processor, but the complexity guarantees it will have a hard time finding buyers, and programmers for the Playstation 3 will be slow on optimization.

I mean, really, can you really break down a game into more than a few concurrent tasks without going crazy trying to synchronize it all? In addition, how many of those independent tasks can be designed with few or no branches? In your average code, branches make up about 20% of instructions. With an 18-cycle penalty per-branch, you'd have to keep those SPE branches under 1% of all instructions to avoid a serious performance loss.

Re:Which innovation?

[owlstead]

Basically, I don't think too many game creators will program the cell vectors directly. They'll just use the correct API's to make use of all that power. Let the 3D, audio and physics engine worry about the cells...

http://news.com.com/PlayStation+3+chip+goes+easy+o n+developers/2100-1043_3-5476933.html

Don't know if the above link says this, but I googled it up, and lets hope it is right.

Re:Which innovation?

[CrowScape]

Actually, change "decades" to "centuries." http://www.uh.edu/engines/epi3.htm So the Wright brothers were the ones who moved flight out of the Middle-Ages.

Intel giving in to the pressure

[grindcorefan]

Does anyone really wonder why Intel's announcement are getting so much press coverage lately?

Well, I don't wonder. It's all looking like good old IBM vs. Amdahl again. Surprising though that Intel seems to think they need to resort to FUD already. Perhaps they really think the heat is on.

Re:Intel giving in to the pressure

[Nuffsaid]

No doubt, the heat is on their processors!

Re:Intel giving in to the pressure

[vhogemann]

Does anyone really wonder why Intel's announcement are getting so much press coverage lately?

Probably because everyone wants to know what Apple will be using on their computers a year from now ;-)

Innovative dick comparison

[Opportunist]

Where's the innovation? And I'm not talking about AMD, Intel is just as guilty for equaling innovation with "make that damn thing run faster". Instead of shifting gear, they just basically upped the engine speed. 100 MHz, 600 MHz, 1 GHz, 4 GHz... now that the ceiling is more or less reached and enough waste heat is generated to heat a medium sized home, they change the measurement. Instead of length, we compare circumference. One core, 2 cores, 4 cores, 8...

Where is that innovation?

Re:Innovative dick comparison

[cfx666]

The innovation is that the good old software you are running does not automatically profit from cpu upgrades any more. So you need some new which is a good thing for me, cause Im a software developer.

Cfx

Re:Innovative dick comparison

[10Ghz]

And I'm not talking about AMD, Intel is just as guilty for equaling innovation with "make that damn thing run faster".

If they made their processors slower, then they would be "innovating"? What do you want processors to do, really? EVERYBODY wants their CPU to be as fast as possible. If you could choose between two identical CPU's, but one of them were twice as fast as the other, which one would you choose? the slower one? I doubt it. So why are you then whining as if making CPU's faster is a bad thing, since everybody wants faster CPU's? What benefit would there be in having slow processors?

And they have been doing pretty interesting things in order to make it faster. Pentium Pro with the on-die cache, SIMD, multithreading etc. etc.. Hell, even Cell with it's SPU's was designed the way it is, so it would be as fast as possible. But according to you, that's not innovcation?

ow that the ceiling is more or less reached and enough waste heat is generated to heat a medium sized home, they change the measurement. Instead of length, we compare circumference. One core, 2 cores, 4 cores, 8.a

Uh, they are still comparing performance, the means to get performance has just been changed that's all. They are NOT adding cores for the sake of adding cores. They are adding cores in order to increase performance.

But since you apparently think that making CPU's faster is not the way to go, why not share ith us what YOU want processors to do?

Re:Innovative dick comparison

[smash]

I see your point, but really, who cares?

Up till now, pushing Mhz has likely been the cheapest way of ramping up speed.

At the end of the day, I couldn't give a crap whether or not my CPU that performs X teraflops does it by running a "dumb" core at extreme speed, or runs a really complex core at slow speed.

Now is the time to get into the nitty gritty of making chips more efficient, now we've exploited the cheap and easy ways...

smash

Re:Innovative dick comparison

[gormanly]

Well, there's been tons of innovaton at Intel. Even just looking at the CPU side, between the speeds you list:

100 MHz (1994): DX4 (P24C), Pentium (P54 version) - both, AFAICR were 0.6 um processes, and the DX4 had a 33 MHz bus and the P100 had a 50 MHz bus. I can't remember which was released first though. 600 MHz (Summer 1999): Pentium III (Katmai), the first rev of Pentium III, which was a new revision of the P6 core used in the PPro and PII chips. It had a new instruction set, SSE, and 512MB (external) L2 cache and a 100 MHz bus. Like the Pentium II, it also had Intel's MMX instructions for 64-bit SIMD integer operations. 1 GHz (Spring 2000): Still a Pentium III, though now with 133 MHz FSB and smaller (256MB), on-die L2 cache. No real changes from the 600 MHz version, but then it's only 2/3 faster again - and Intel were working on the Netburst architecture for the Pentium 4 and had somewhat taken their eye off the ball at this point. 4 GHz does not exist. Currently P4EE is at 3.73 GHz, but the clock speed race is over.

Intel gambled on Netburst, which was designed to get faster rapidly, and scale all the way from the 1.4 GHz at launch to 6 or 7 by now. Yes, they lost, but that doesn't mean that they weren't innovative - it's just that their process teechnology couldn't keep up, and failed to meet predictions. That's not the CPU designers' fault.

The earlier processors did scale fantastically well (486 16->120 MHz; P6 150->1400 MHz) but they hit an unexpected brick wall this time, so they've gone around it with clever scheduling and power management, and doing dual core versions of what is essentially a new rev of the P6. There's plenty of innovation in that chip too...

Also, remember that during the same timeframe, they've invented and developed the PCI, PCI Express and Universal Serial Bus(es). Pretty innovative, really, IMHO.

And yes, I'm typing this on an Athlon 64 and all 3 of my home PCs are AMD-powered.

Re:Innovative dick comparison

[khellendros1984]

I'm guessing that the gp is referring to picking a single aspect of processor performance, and equating that single aspect directly to performance of the machine, i.e. "1200Mhz will always perform faster than 1000Mhz". No manufacturer says this (that I know of), but so many consumers take it as a given, when the only number quoted to them concerning a processor is clock speed.

That isn't to say that I agree with the grandparent, though. Intel's Pentium-m processors are pretty nifty...lower power usage, high performance (compared to a P4).

So, while it may not be some amazing quantum leap, I'd say that Intel is showing plenty of innovation, at least from the standpoint of the consumer market.

Re:Innovative dick comparison

[Opportunist]

The question remains, though, if more throttle is more speed, too, or just more smoke? Or rather, if that smoke generated is warranted by the speed increase. More horsepower only equals more speed if you manage to get it onto the ground, if your traction sucks because you try to run metal tires instead of rubber, you will generate a lot of heat, noise and some quite cool FX, but you won't pick up more speed.

To get out of bad analogies, of course more MHz == more speed. The question is, though, if there are not more fruitful ways to pump speed out of your CPU. With some really innovative features added, you can actually maybe generate more power output instead of generating just more waste heat.

For example the invention of branch prediction and multiple pipelines were something that DID increase performance considerably, even though the MHz didn't. So was adding the math coprocessor to the core. Or the increase in cache ram. All very good ways that speeded up calculation speed considerably, without increasing waste heat.

I blame the clueless customer. He only looks at "Wow, XX GHz!" and buys the crap, not caring if it's the equivalent of running his engine in first gear at 15.000 rpm. Remember that inane "2500+" labeling by AMD? A result thereof.

Re:Innovative dick comparison

[big_gibbon]

How about making them run cooler and more efficiently, and hence quieter since they wont need 3 fans pointing at them? The best chip that intel have made in recent times was the Pentium M, in large because they focussed on the needs of a mobile chip, i.e. power consumption and hence efficiency.

Also take a look at Via's new chips and boards, particularly the Epia series of boards. They're ideal for media centre applications, since the chips run very cool but quickly enough to do the necessary work.

P

Re:Innovative dick comparison

[10Ghz]

I fail to see why this turned in to discussion about clock-speeds. The GP was talking about "making the damn thing faster". That IMO means just what is sounds like: Making the CPU as fast as possible. It does not mean "squeeze as much Mhz from that thing as possible!" CPU's haven't been about Mhz in a long time already.

if there are not more fruitful ways to pump speed out of your CPU.

Yes there are, and Intel and AMD are pursuing them as we speak. But even that isn't good enough for the GP, who is now whiniong when they put more cores in to one die. Is he going to whine about the ever-growing caches next? And this isn't about clock-speed, this is about how fast (as in: how fast they carry out their tasks) the CPU's are. GP was apparently annoyed because Intel and AMD care only about performance. Well, what should they then care about? Should they be trying to make their CPU's as slow as possible? And before you say: "how about power-consumption?". Well, they are caring about that now, because the excessive power-consumption (and the heat that follows it) is killing performance. Now, when they are tackling power-consumption, they are also giving themselves more headroom to grow the performance of their chips in the future.

Re:Innovative dick comparison

[10Ghz]

How about making them run cooler and more efficiently

They are doing just that. And they are doing that because the excessive heat was killing performance. They couldn't make it faster because it was running so hot.

and hence quieter since they wont need 3 fans pointing at them?

My A64 3200+ has exactly ONE fan cooling it.

Also take a look at Via's new chips and boards, particularly the Epia series of boards. They're ideal for media centre applications, since the chips run very cool but quickly enough to do the necessary work.

But wouldn't you just love it if they doubled the performance (increasing it from "crap" to "OK-ish" while keeping the power-consumption at the same level? Of course you would.

Re:Innovative dick comparison

[JPribe]

(sig) Great UHF quote!!!!

Re:Innovative dick comparison

[muhgcee]

At the end of the day, I couldn't give a crap whether or not my CPU that performs X teraflops does it by running a "dumb" core at extreme speed, or runs a really complex core at slow speed.

You might not care at the end of the day, but at the end of the month when your power bill comes in you might care at least a little bit.

Re:Innovative dick comparison

[WolfZombie]

enough waste heat is generated to heat a medium sized home

Maybe it's about time someone came up with a method to use this heat for additional productivity. Maybe the element used in heating water for an expresso machine!

Re:Innovative dick comparison

[Karhgath]

But what happens here is the same as with cars. Ok, bad analogy follows, but oh well. If you put too much HP in your car, you might not even be able to attain the highest speed as other components comes into play: tires, trasmission, aerodynamics, etc. Putting a dual-V12-whatever won't magically make a Mini as fast a a ferrari. HP = MHz and Twin Engine = Core/Dual procs, etc.

Furthermore, why aren't EVERYONE buying ferraris/lambo/porsche/etc? I mean, they want the fastest car possible right? Or why aren't every car makers putting V12 in their cars? Oh yeah, price, gas, normal usage, etc. all come into play. If you only drive to work, a ferrari isn't very useful. Sure, a speed freak/track racer might want to have the ferrari more than a ford, but then again, they are the hardcore gamers of the automotive world.

Intel is kind of selling engines by only advertising the HP in them. That's not all there is, and your mom probably don't need a damn 3.0 ghz for her dell with 256 ram and a slowass hard drive and shitty onboard graphic card. Yeah they have the Celeron, but they still sell them by the speed, for normal people a 3.0 ghz pentium and celeron isn't different, so obviously they'll take the cheapest of the 2.

Re:Innovative dick comparison

[Splab]

The buzzword you are looking for is Extreme Multiprogramming. (Well ok, two words).

Not to be mistaken for extreme programming. It's based on CSP (Communicating Sequential processes) - Occam, c++csp, jcsp etc. support this model originally made for transputer.

Re:Innovative dick comparison

[Nimey]

P-100s had a 66 MHz bus and a 1.5x multiplier. The 50 MHz bus was used only on the P-75 and the rare 486DX-50.

Re:Innovative dick comparison

[somersault]

no, he was talking about Mhz, as you can see from his listing of all the different 'Hz'es.. he's saying that they should innovate rather than just do what they do already, but faster. Having more cores isn't actually that 'innovative' if you take innovative as meaning a new and original idea, but it is useful enough for now. Obviously since the Pentium M runs a lot faster than the P4 at similar clockspeeds, then Intel have eventually realised that !(Mhz>all). AMD could likely increase processor efficiency too, but they're a lot more efficient than Intel to start with..

Also, computers used to run faster when they got hotter, which was a vicious cycle resulting in them burning themselves out. So modern CPUs are designed to run slower the hotter they get. Well that's what I heard :p

Re:Innovative dick comparison

[RyuuzakiTetsuya]

If they made their processors slower, then they would be "innovating"? What do you want processors to do, really?

Lower power consumption? Lower heat output to the point where I don't need a Rube Goldberg Device?

Re:Innovative dick comparison

[Amouth]

heat my house and cook my dinner? i mean why else would i buy a p4 prescott

Re:Innovative dick comparison

[tepples]

But wouldn't you just love it if they doubled the performance (increasing it from "crap" to "OK-ish" while keeping the power-consumption at the same level?

I wold also like a fairy to give me a thousand U.S. dollars every day, but it isn't going to happen any time soon. It will take a lot of time to figure out how to change both processors and the programs that run on them to run faster in the same transistor budget. Remember that compilers become better much slower than transistor budgets increase.

Re:Innovative dick comparison

[repvik]

I really wish my CPUs had the on-die L2 cache you're mentioning there....

Re:Innovative dick comparison

[TubeSteak]

change the measurement. Instead of length, we compare circumference.

Girth
Not circumference, but girth.
That is the word to use

You measure the "girthiness" because it makes you feel more like a man.

Re:Innovative dick comparison

[smash]

Actually, no I wont - I've noticed no real different in my power bill whether my PC is left on 24x7 or not.

smash.

Re:Innovative dick comparison

[rainman_bc]

I notice a bigger difference when using my 22" CRT... Damn monitor heats the whole top floor of my house haha!

Re:Innovative dick comparison

[triffid_98]

Furthermore, why aren't EVERYONE buying ferraris/lambo/porsche/etc? I mean, they want the fastest car possible right? Or why aren't every car makers putting V12 in their cars? Oh yeah, price, gas, normal usage, etc. all come into play. If you only drive to work, a ferrari isn't very useful. Sure, a speed freak/track racer might want to have the ferrari more than a ford, but then again, they are the hardcore gamers of the automotive world.

V12+Mini=death sled. "looks like a fish, moves like a fish, steers like a cow"

Energy efficiency

[Lisandro]

"Multicore gives us the ability to get back on traditional performance growth lines," Intel Chief Technology Officer Justin Rattner told reporters on Monday. "We have become fanatical about energy efficiency. We have to continue to make progress in terms of energy efficiency."

    Does this means these new multicores will fry eggs even faster? I hate it when my meal isn't done in time!

Re:Energy efficiency

[Anne Thwacks]

Does this means these new multicores will fry eggs even faster? I hate it when my meal isn't done in time!

No - Sun manage to get four multithreading cores in their Niagra, and only run at 72 watts with 32 threads. see this

However, with Intel's cores, I expect be able to have a hot dinner faster than you can say Microwave".

Re:Energy efficiency

[asliarun]

[quote] No - Sun manage to get four multithreading cores in their Niagra, and only run at 72 watts with 32 threads. see this However, with Intel's cores, I expect be able to have a hot dinner faster than you can say Microwave". [/quote]

Oh, do you expect to fry an egg with a 31W (TDP) Sossaman, or do you expect to do it with a (TDP not confirmed officially) 80W Woodcrest? I can understand faboi comments like these, but for it to get modded to 4-Interesting is seriously lame.

Re:Energy efficiency

[foobat]

wow, so not only does it cook eggs faster, with it's multiple cores it can be an eggtimer as well! Now THAT's innovation!

Re:Energy efficiency

[DrSkwid]

DRM != Trusted Computing

Re:Energy efficiency

[Ninjy]

Though, you'd expect the sun to be able to fry an egg faster than that. I feel like I misread something somewhere, though...

Re:Energy efficiency

[Sebastopol]

Intel has ditched the p4 architecture and are focusing specifically on low power.

If you actually read technical reviews (anandtech and tomshardware), you will see that the dual core Yonah hovers around 13 W when the USB2.0 is not active. With 4 core on the next process, it wouldn't surprise me if it had an average power of ~10W.

So, the frying eggs joke should be focused at AMD, and not Intel.

Advantages and disadvantages of multicore

[replicant108]

Advantages

        * Proximity of multiple CPU cores on the same die have the advantage that the cache coherency circuitry can operate at a much higher clock rate than is possible if the signals have to travel off-chip, so combining equivalent CPUs on a single die significantly improves the performance of cache snoop operations.
        * Assuming that the die can fit into the package, physically, the multi-core CPU designs require much less Printed Circuit Board (PCB) space than multi-chip SMP designs.
        * A dual-core processor uses slightly less power than two coupled single-core processors, principally because of the increased power required to drive signals external to the chip and because the smaller silicon process geometry allows the cores to operate at lower voltages.
        * In terms of competing technologies for the available silicon die area, multi-core design can make use of proven CPU core library designs and produce a product with lower risk of design error than devising a new wider core design. Also, adding more cache suffers from diminishing returns.

Disadvantages

        * Multi-core processors require operating system (OS) support to make optimal use of the second computing resource.[1] Also, making optimal use of multiprocessing in a desktop context requires application software support.
        * The higher integration of the multi-core chip drives the production yields down and are more difficult to manage thermally than lower density single-chip designs.
        * From an architectural point of view, ultimately, single CPU designs may make better use of the silicon surface area than multiprocessing cores, so a development commitment to this architecture may carry the risk of obsolescence.
        * Scaling efficiency is largely dependent on the application or problem set. For example, applications that require processing large amounts of data with low computer-overhead algorithms may find this architecture has an I/O bottleneck, underutilizing the device.

http://en.wikipedia.org/wiki/Dual-core

Re:Advantages and disadvantages of multicore

[10Ghz]

Multi-core processors require operating system (OS) support to make optimal use of the second computing resource.

And that support exists in just about all OS'es already

Also, making optimal use of multiprocessing in a desktop context requires application software support.

Not really. Your OS is already running several processos in the background. SMP allows for those processes to be evenly distributed to two cores. Also, if you run more than one app at the same time (like we all do, basically), you will benefit.

Re:Advantages and disadvantages of multicore

[Anonymous Coward]

        Multi-core processors require operating system (OS) support to make optimal use of the second computing resource.

And that support exists in just about all OS'es already

I've been talking to groups who are hitting scalability problems, i.e. not getting extra throuput with extra processors, even though they are using multi-threading with the correct threading design patterns. The problem is that while the solutions for that problem exist, they're not supported in the OS yet and are not likely to be. Part of this is because certain vendors are patenting the known solutions without bothering to provide any implementations.

Too Little, Too late?

[zaguar]

I think intel might be too late to the party. The dual-cores of this generation have been bought already, and the upgrade cycle won't start again until 2007-2008. And guess what? Intel lost this battle. My X2 3800+ performs better, OC's better, and is faster than a similar priced Pentium D. And I got it a good few months ago. So did many others.

Intel needs to accept that they lost this war, and focus on the 07-08 season, when the target market is upgrading again.

Also - faster, more efficient chips? Compared to what? The Netburst architecture chips? They put out more heat per watt than my heater, FFS!

Re:Too Little, Too late?

[hyc]

What a sobering thought. "We've invented a space heater that produces computations as an operational byproduct."

Re:Too Little, Too late?

[MichaelSmith]

They put out more heat per watt than my heater, FFS!

Sorry to be the one to tell you this, but all your watts are heat.

Re:Too Little, Too late?

[myspys]

'cause we all know that none of the new macs will be using the dual cores

so, since YOU have already upgraded, you're saying that they've lost ALL dual core sales "available"?

Re:Too Little, Too late?

[Tim C]

They put out more heat per watt than my heater, FFS!

Got any proof of that?

Re:Too Little, Too late?

[10Ghz]

The dual-cores of this generation have been bought already, and the upgrade cycle won't start again until 2007-2008.

Dude, WTF?!? Are you saying that everybody who is going to buy a dual-core processor has already bought one, and next such CPU's wont be sold until 2007-2008?? What if someone decides to upgrade his computer in the summer (for example) to a dual-core machine? By your logic, he does not exist and/or he should wait until 2007/2008 because "that's when the next upgrade-sycle is in, you can't upgrade before that"?

People are upgrading their computer all the time. People are buying new computers all the time.

Re:Too Little, Too late?

[richman555]

Funny, I don't even have a dual core processor yet.

Re:Too Little, Too late?

[kfg]

Are you saying that everybody who is going to buy a dual-core processor has already bought one, and next such CPU's wont be sold until 2007-2008??

No, what he is saying is that the early adopters, those who have legitimate imediate need and those who purchase simply because "it's better" have already bought one and that it won't be a conventional purpose for a couple more years.

The Cycle: A concetration of buying from the early adopters, a slow dribble for awhile, then commodity buying.

Poster is not suggesting that Intel won't sell any dual core chips, merely that they are entering the market just as the market shifts to slow dribble sales. That they've missed the boat on the early buyers who you rely on to return your capital investment.

KFG

Re:Too Little, Too late?

[10Ghz]

No, what he is saying is that the early adopters, those who have legitimate imediate need and those who purchase simply because "it's better" have already bought one and that it won't be a conventional purpose for a couple more years.

There will be plenty of people buying dual-core chips even after the early-adopters have bought theirs. Some people simply didn't have the need/money to upgrade right after dual-cores were released, but they might upgrade a bit later. And new computers are being sold to companies/people all the time.

Of course dual-core CPU's wont be that common when compared to singles, reason being that they are hi-end parts, and hi-end parts always have slower sales than mainstream CPU's.

There are the geeks and early-adopters who must have the fastest CPU possible. And the dual-core CPU that was the fastest CPU last week, might not be the fastest next week.

Re:Too Little, Too late?

[kfg]

And new computers are being sold to companies/people all the time.

I not only did not say anything to deny this, I avered that it was so.

KFG

Re:Too Little, Too late?

[Orgazmus]

If we say that Wh is the heaters output, Wc is the the CPU, and Wi is the input voltage multiplied with the input current:

Wh/Wi > Wc/Wi

So, there you have it

Re:Too Little, Too late?

[Deliveranc3]

No they haven't, if AMD had continued it's tradition of excellence by releasing a $250 4200 and a $200 3800+ then yes it would have been all over for a few years (Those prices might seem unreasonable except Intel pulled them off).

Dual cores should have been a huge generation, but they ended up being only a half generation :(

Re:Too Little, Too late?

[mybecq]

They put out more heat per watt than my heater, FFS!

Unless you can get your heater or CPU to output other than 1 joule every watt, I'm going to have to disagree.

Re:Too Little, Too late?

[Surt]

There's no bi-yearly upgrade cycle that everyone adheres to. Some people have upgraded, some haven't, and cpu sales cycles are actually quite smooth. Look at a graph of intel's or dells quarterly sales. Intel can bring out new tech whenever it wants, and if it is good, it will sell.

http://www.investors.com/editorial/IBDArticles.asp ?artsec=17&issue=20060216

http://quicktake.morningstar.com/Stock/Income10.as p?Country=USA&Symbol=INTC&stocktab=finance&pgid=qt qnnavfinstate

Re:Too Little, Too late?

[ErikZ]

Did you look at AMD's earning's statement? They made a profit, but compared to all the income they brought in, it wasn't much in comparison.

I was also dissapointed at the 800$ 4800X2, but from looking at their statements, it looks like the X2s just cost more to make.

Intel can price their dual core cheaper because they're using the new 62nm fabs. The cores take up less space on a die and therefor cost less to produce.

It's interesting that it took a (multi?)billion dollar fab for Intel to catch up to AMD. Once AMD gets their's up and running, their product will smoke anything Intel has out.

Re:Too Little, Too late?

[Gr8Apes]

On the AMD side, dual-core may become more common than single core, very shortly. The fact that dual-core chips are drop-in replacements for single core chips helps, as vendors don't have to change anything other than their CPU order. In the intel world, you have to have a different motherboard.

Also, I think there's about to be a new wave of dual-core buying, as AMD based DDR motherboards start hitting the shelves. Something about making a fast system faster... and the fact that these motherboards will not accept the current non-DDR CPUs.

I know I've been waiting for this. This should leap-frog anything Intel has in the works for the next year and a half at least, performance-wise. The only question is whether to buy an Opteron or X2, and then upgrade later. Given that single core Opterons are running at commodity prices, that makes the most sense from a budget perspective. But, the 3800 X2's are almost in the same commodity category. (Gaming is not my main pursuit, so an X2 is probably the better solution for me).

Re:Too Little, Too late?

[geekoid]

dude, he is giving complex business advise about a very large company as if the people at that company who make these decisions don't know their market,and look to /. for advice.
What did you expect?

Re:Too Little, Too late?

[TubeSteak]

You got modded Funny, but I've closed the central heating vent to my room and I'm only running a 1.33 GHz AMD T-Bird

My mid-tower puts out a lot of heat. Enough to keep my room warm, even with the door open.

Re:Too Little, Too late?

[Darby]

Given that single core Opterons are running at commodity prices, that makes the most sense from a budget perspective. But, the 3800 X2's are almost in the same commodity category.

The dual core Opterons and the X2's are basically the same chips AFAIK. The Opterons all have 1MB cache per core, whereas only some of the X2's do. The conventional wisdom (which could be wrong) is that the Opterons are better tested and higher quality in general since they are intended for servers. Another frequently stated idea is that due to the above, the Opterons are better overclockers.

Up until recently, a dual core Opteron was cheaper than the equivalent X2, AMD recently raised their Opteron proces and lowered their X2 prices to get them more in line since many people were getting Opterons for their non server computers.

Like me, for example, who ordered an Opteron 175 right before the prices went up ;-)

Now I just wish people would get their act together with better dual core support. That'll allow me to get major speed improvements without spending anything on hardware.

Re:Too Little, Too late?

[Gr8Apes]

The Opteron 1xx series with the 939 socket are essentially the same as the 1MB cache series. However, if I were buying an Opteron, I'd either go for max overclockability (there's a word) in 939 socket FX type configuration, or a 2xx series pure Opteron solution. Nothing like multiple CPUs to make for multi-core heaven. The beauty is, you can start with a 2xx, drop in a second 2xx, and have 2, 3 or 4 cores, depending on which chips you buy (note: the 3 core option appears to work in some configurations, but has issues)

Re:Too Little, Too late?

[mink]

Only?!
Having one as well I can tell you it is probably the hottest AMD chip ever. That was like 5 years ago. Everything I have gotten since has been considerably cooler running. Pick up a cheap old Athlon XP chip (2000 will beat it out speed and heat wise) and welcome to a world where your processor can not cook the motherboard caps.

Pretty light reading, but...

[porkThreeWays]

It's a fluff piece, but there was nothing mentioned in there to make me believe Intel is really doing anything new. All I saw was mention of 4 cores. Are cores the new mhz race? 2 cores is all 99% of people will see benefit from right now. The 4 core race is moot because it's like a race for automakers to produce the first production 16 cylinder family sedan. It's not going to really benefit anyone. Really only a marketing gimmick. I'd rather see Intel clean up their current 2 core chips.

Here's what most consumers need in a computer...
A low latency desktop that can handle about 2-3 running applications with no slowdown that runs cool and doesn't use a lot of power.

Here's what we are getting...
A high latency desktop with fat pipes that run hot, optimized for running 7-8 cpu intensive applications at once, and idles at 200 watts. Because it should take 10+ seconds to open a basic program on an out of box pc.

Re:Pretty light reading, but...

[Tim C]

Because it should take 10+ seconds to open a basic program on an out of box pc.

Most of that will be down to the disks, which is nothing to do with Intel or AMD.

Re:Pretty light reading, but...

[Deliveranc3]

A few months ago released a press release saying that they were going to cut power inefficiency in their chips to 1 millionth their previous levels. We've already seen the chips they produced with this breakthrough.

Net power loss less than 5%...

Intel is not wrht gambling on anymore, too many false rumours... Wait for some real results before believing ANYTHING from them.

Re:Pretty light reading, but...

[Surt]

There's very little intel can do in its chip designs to make the desktop low latency. That's all about app and OS design. Intel is putting a lot of focus into getting their chips to run cool, and if people are really demanding the sort of PC you're describing then surely someone will build one out of the ULV chips intel is going to be supplying.

And on the other hand, I'm thinking about whether or not I can afford to get a dual-processor dual-core in my next box, so I'd love to see them move to quad-core, and so would everyone buying a workstation. Not as large as the home PC market, but not much worse than 10x smaller.

Re:Pretty light reading, but...

[Cyno]

Actually, with a company like Intel pushing 4-cores they may be selling them for well under $1000 and get many other companies, including Microsoft, to further optimize their software for multiprocessing/threading. 4 cores are still 4 cores, even if you don't have a lot of CPU-RAM bandwidth. If you can fill them with enough data they will crunch 2 times faster than 2 cores. Plus any geek like me would love to have 4 physical CPUs in our desktop.

AMD will probably have theirs out before I'll be forced to buy Intel, and maybe it will be socket compatible with my 939, but probably not.

Even if Intel makes a 4-core Netburst/P4-based monstrocity that sits on top their 1066Mhz FSB it will sell to those who can't wait and don't want to pay the premium for Opteron/Dual Xeon or go non-x86. But if they make a 4-core Pentium-M with 2MB of cache per core, even w/o an onboard memory controller it would dominate. AMD would have to have a 4-core chip to compete. If Intel could get those running at 3.8+ Ghz, OMFG! That's what AMD really has to fear: Intel cranking up the clock speed on the Pentium-M or patching on a memory controller. I doubt it will easily scale up that high, but Intel might be able to squeeze something out of 65nm.

wanna compare cpu speeds?

[80 85 83 83 89 33]

if you want to do an easy way to compare amd and intel chips, here is a very simple perfomance check i love to run on every computer i come across: put windows calculator in scientific mode (yes, mathmatica or maple will do factorials in a fraction of the time, but try to post windows scores for comparison purposes....) type in 100,000 hit the n! button ignore the warnings that it will take a long time, don't even bother clicking on "Continue", because the calculation is still going. and report how long it takes to complete a factorial of 100,000 please report what CPU you have **windows XP 64-bit edition is twice as fast as 32-windows** celeron 800MHz (coppermine): 333 seconds (5min 33sec) 1.4GHz celeron (tualatin) does it in 205 seconds P4 3.2Ghz and Athlon 3200+ both do it in about 80 seconds.... my next post will have a long list of scores. from slowest to fastest, there is a difference of 66x...

Re:wanna compare cpu speeds?

[80 85 83 83 89 33]

sorry for the bad formatting, but the lamness filter is killing the proper layout.

factorial times for "100,000!"

look at the two athlons running at 2.0GHZ (3200+ and 2400+) and notice how it is frequency dependant

P4 3.2GHz 81 seconds

athlon XP 3200+ (2.2GHz socket A, barton)81 seconds

Pentium 930 dualcore (3.0GHz) 82 seconds

P4 3.0GHz (laptop) 90 seconds

Pentium 920 dualcore (2.8GHz) 90 seconds

athlon 64 3200+ (2.0GHz socket 939, venice) 91 seconds

athlon XP 2400+ (2.0GHz) 93 seconds

athlon XP 2100+ 106 seconds

athlon XP 2000+ (1.67GHz) 121 seconds

athlon mobile XP 1800+ (1.52GHz) 122 seconds

celeron 2.7 GHz (northwood core) 130 seconds

celeron 1.4GHz (tualatin) 205 seconds

athlon 900 (thunderbird) 228 seconds
(used msconfig to disable everything)

celeron 1.1GHz 253 seconds

celeron 800MHz (win98) 333 seconds (5min 33sec)

celeron 800MHz (XP pro) 373 seconds

PIII 800 (XP pro) 378 seconds (used msconfig to kill all crap running)
474 seconds (lots of junk running)

PIII 450MHz (underclocked coppermine) 490 seconds

PII 333MHz 686 seconds

PII 300MHz 760 SECONDS

P 166MHz 2417 seconds

P 100MHz ~4000 seconds (66 minutes)

P 75MHz 5330 seconds (1:28:50)

Re:wanna compare cpu speeds?

[80 85 83 83 89 33]

sorry for the formatting, here it is easier to understand:

a very simple perfomance check i love to run on every computer i come across:

put windows calculator in scientific mode (yes, mathmatica or maple will do factorials in a fraction of the time, but try to post windows scores for comparison purposes....)

type in 100,000

hit the n! button

ignore the warnings that it will take a long time, don't even bother clicking on "Continue", because the calculation is still going.

and report how long it takes to complete a factorial of 100,000

please report what CPU you have

**64 bit XP is twice as fast

celeron 800MHz (coppermine): 333 seconds (5min 33sec)

1.4GHz celeron (tualatin) does it in 205 seconds

P4 3.2Ghz and Athlon 3200+ both do it in about 80 seconds....

Re:wanna compare cpu speeds?

[Anonymous Coward]

I'll add this to the list:
Opteron 146 (2.0GHz) : 43 seconds

Now I know what is the purpose of 64bit desktop CPUs - extreme Calculator performance!

Re:wanna compare cpu speeds?

[SamBeckett]

The only thing your "test" is testing is the FPU on the processor. The code for factorial is small, so that and all the intermediaries are able to be stored in cache, regardless of the processor. And how exactly are you timing it? Do you sit there and say "1-one thousand, 2-one thousand, ..." ?

Re:wanna compare cpu speeds?

[Kaetemi]

Athlon XP 2400+ 2GHz: 103 seconds

Re:wanna compare cpu speeds?

[SamBeckett]

Oh, and calc.exe says 2.8242294079603478742934215780245e+456573 in 62 seconds. BOOM.

Re:wanna compare cpu speeds?

[Nuffsaid]

Ha! My super-optimized Gentoo x86-64 system, running on a 2 GHz Turion64 CPU, all unnecessary services and processes killed, under optimal condition (downslope, wind from behind, air temp below 10C) uses exactly zero seconds to tell me "Error" in kcalc...

Re:wanna compare cpu speeds?

[Nuffsaid]

OTOH, seriously, win calculator on 32 bit Win2000 under VMWare on the same machine takes exactly 120 seconds, on par with my home Athlon XP 2000+. Impressive, for an emulated machine!

Re:wanna compare cpu speeds?

[kainhander]

Athlon XP 2700 (thoroughbred b) - 87s

Re:wanna compare cpu speeds?

[Jepler]

I don't know what method you're using to compute factorials, or whether the time includes converting to decimal and displaying on a terminal. My programs do not. (actually, the last one does convert to decimal and write to a file, and it still runs quite a bit faster than yours)

My test system is a Sempron 1800MHz 64-bit processor, and I wrote my programs in Python

A naive program which calculates 1 * 2 * ... * 100000 just as written takes 54s CPU time. Another program which uses a "divide and conquer" approach takes 2.6s CPU time. Once again, it's clear that algorithm can make a much bigger difference than CPU speed.

The "divide and conquer" method creates a list of all the numbers to be multiplied. At each stage, it divides this list into two sublists of equal length, and multiplies the pairs to create a new list. If there's an odd item, it's simply placed on the new list. This means that the intermediate products stay smaller longer.

If you *are* timing the conversion and output stage, then you should use a math package that is designed to create decimal output. I used "DecInt", and added a step which converted Python longs to DecInts when they appeared in the intermediate results. The run time is a bit longer, but the result is still calculated and written to a file in just 5.5s CPU time.

Here's the best I can do at getting my code into this post without hitting the fitlers:

def f(n):
nums = range(2, n+1)
while len(nums) > 1:
. print len(nums)
. hl = len(nums)/2
. if len(nums) % 2:
.. a = nums[:hl]
.. b = nums[hl:2*hl]; b.reverse()
.. odd = nums[2*hl]
.. nums = [odd] + [i*j for i, j in zip(a, b)]
. else:
.. a = nums[:hl]
.. b = nums[hl:2*hl]; b.reverse()
.. nums = [i*j for i, j in zip(a, b)]
return nums[0]

Re:wanna compare cpu speeds?

[ErikZ]

Is windows Calc threaded? Will it use both processors in a dual core machine?

Re:wanna compare cpu speeds?

[Dr. Mortimer]

Took somewhere between 60 and 70 seconds on my Pentium M 1.7 ghz (Thinkpad T41p).

Re:wanna compare cpu speeds?

[Mr Z]

You might be great at math and algorithms, but I don't think you did so well at reading and comprehension. Did you miss this little detail in the original post? (The parent of the "big list of scores".)

put windows calculator in scientific mode (yes, mathmatica or maple will do factorials in a fraction of the time, but try to post windows scores for comparison purposes....)

The algorithm or the fact that he computed a factorial doesn't matter. Rather, the GP was defining a specific, fixed workload that anyone running Windows can generate for themselves easily.

--Joe

Re:wanna compare cpu speeds?

[Duhavid]

Calculator Performance Unit (CPU)

Re:wanna compare cpu speeds?

[Lonewolf666]

PIII 800 (XP pro) 378 seconds (used msconfig to kill all crap running)
474 seconds (lots of junk running)
That one is interesting. So XP Pro has enough unnecessary stuff running by default to make it 25% slower??

Re:wanna compare cpu speeds?

[TheLink]

but what's the result?

It's no point producing the wrong result in faster time ;)

Re:wanna compare cpu speeds?

[ShinSugoi]

Athlon XP64 4200 X2 - 38 Seconds.

I'm also running Win64, so I'm not sure how huge a difference that makes.

Re:wanna compare cpu speeds?

[80 85 83 83 89 33]

well, it also shows how much work per clock a chip can do. if you compare two different chip designs that are running at the same frequency, then you can cancel out the frequency units and measure how well each chip is at crunching integers....

Re:wanna compare cpu speeds?

[80 85 83 83 89 33]

And how exactly are you timing it? Do you sit there and say "1-one thousand, 2-one thousand, ..." ? no way would i count them in my head. i use a sundial, it's so much more convenient! except when the damn clouds kick in.

Re:wanna compare cpu speeds?

[Jepler]

It figures---the post with the useless numbers was modded up to 5, but the important one that gave the "methodology" wasn't moderated up past 2, so it went unseen (by me, at any rate). It doesn't look like Windows CALC.EXE even computes the full value of factorials, judging from the way it displays 100! in scientific notiation. This truly is the contender for most worthless benchmark! Though, to bring this back around to increasing performance, I suggest the NOPsort. It's a version of bogosort optimized for overclocked computers. First, NOP is executed, which may sort the array. Then, the order of the array is checked. If the process never completes, just overclock your computer some more. I'll wait for someone else to publish the first NOPsort benchmark, though.

Re:wanna compare cpu speeds?

[Mr Z]

Well, the computed number is worthless, but the workload is well defined and CPU bound. It's not a comprehensive benchmark, but it's probably better than the old Norton SI, since it's likely to not actually fit in cache.

Naming Conventions

[Freaky Spook]

They also need to name their chips better to actually differentiate more simply between their lines.

Telling a customer the difference between a Pentium D, Pentium 4, Pentium 4 EE, Celeron D is hard enough without actually having to know what chips are out and what is offering the best performance for price. It feels a lot like market saturation sometimes.

AMD at least is a little bit simpler to follow.

Re:Naming Conventions

[powerlord]

It also is a taste of wat the market is going to be like with the 6 flavors of Windows Vista we've been hearing them mention :)

Too bad most people don't think they have an alternative ... although considering the push we've been seeing Mac's get in the media, that might not stay true for long :)

If I was a consultant, marketing PC to Mac migration services would probably be a good (and easy/fun) thing to do for a bit.

Re:Too Little, Too late? My Arse!

[W33B]

Intel needs to accept that they lost this war, and focus on the 07-08 season

You just made me want to go out and buy a K6 or a PIII with that dumbass comment, WTF are you on about 07-08 season!?

Upgrading is not a seasonal thing except for the uber geek desperate to get the latest and greatest and I've got news for you...that is deffiantely not thier target audience and they have not yet lost the war!

Dont get me wrong, I'm not an Intel fanboy but I think you are ill informed and probably have your head stuck so far up your front side bus that you are blind to the reality of the situation.

IBM Power 6 @ 6Ghz

[CypherOz]

IBM have Power6 chips running at 6Ghz. IBM have been able to do 4 cores with this new technology.

Refer here

Re:IBM Power 6 @ 6Ghz

[Somegeek]

Unfortunately your link (or the article that it links to) doesn't say anything about 6 GHz. A little googling found some that did however, but they still talk about this as 'in the Lab'. I bet Intel and AMD can get stuff running at high speed 'in the lab' too. All of the stories that I have seen say that the chip will come out at 4-5 GHz, and not for another year.

It's also important to remember that one of the reasons that Intel is walking away from the clock speed race is that AMD showed that it wasn't necessarily the best way to higher performance. My point is that just because the new IBM chip may have four cores and a high clock speed doesn't mean it will be any faster than a chip with AMD's architecture. No one will really know until it's released and compared against whatever else is available at the time.

Link to an article that does mention the 6Ghz Power 6:

http://www.theregister.co.uk/2006/02/07/ibm_power6 _show/

Re:IBM Power 6 @ 6Ghz

[tomstdenis]

IIRC Intel has made 10Ghz transistors already but the trick is to make enough of them (at once) to make a processor.

So it is possible that Intel makes a 4.5Ghz P4 and AMD makes a 3.2Ghz AMD64 but the yield would be so low that you'd be paying nearly 5 digits per processor to recoup costs.

Would you honestly pay [say] $17,000 for a 4.5Ghz processor? Specially when you could get an entire 2P SMP system for that much? (and often less).

Tom

Re:IBM Power 6 @ 6Ghz

[Quila]

IBM had 110 GHz processors in the lab years ago, under a .18 micron process.

It only really matters when the stuff gets to market at a decent price.

Re:IBM Power 6 @ 6Ghz

[cciRRus]

Wow, that's pretty hardcore.

Re:IBM Power 6 @ 6Ghz

[geekoid]

"It's also important to remember that one of the reasons that Intel is walking away from the clock speed race is that AMD showed that it wasn't necessarily the best way to higher performance."
incorrect, Marketing high clock speeds is not nearly as relevent as it was when you had to have the latest next clock speed just to run the latest app. That is no longer true.

Intel has know the a chips power isn't just clock speed, and lest we forget they were driving that way before AMD becames a serious competitor. Back in the days where you only had about an 80% chance that an application would run with an AMD chip at all.

Re:IBM Power 6 @ 6Ghz

[tomstdenis]

Exactly. Though IIRC Intel had a functional ALU at 5Ghz in the labs once. That's a bit more meaningful than a single transistor at 100Ghz. Of course the P4 is a bit of a hack since it has delay stages in the pipeline so it's not really a >3Ghz device.

I have no idea what AMD has (even though I work for them) but I know they're more interested in branching out then up. (hint: quad-core).

AMD has 2.8Ghz dualies out (or at least will) ... damn I gotta get some "work samples" for my gentoo box...

Tom

Re:IBM Power 6 @ 6Ghz

[be-fan]

These days, the big isue isn't transistor switching speed, but wire delay. It's possible to make very high speed ALUs (using dynamic logic, the ALUs in a Pentium 4 run up to 7.6 GHz), but getting instructions to it becomes very difficult without inserting pipeline stages just to allow the signal to propagate on the wires.

Re:Too Little, Too late? My Arse!

[zaguar]

...your head stuck so far up your front side bus...

Ironically, the AMD64 series CPU's have no front side bus. This includes the X2 series. They have a hypertransport bus, which is similar but different. This is one of the premier reasons that the X2/Opterons scale so much better than the Intel equivalents, they do not have a saturated FSB as they have direct HTT links CPU-CPU.

love that headline...

[smash]

Intel aims to get back on track with new chips

As opposed to "Intel aims to get further behind with new chips"?

What the hell else would they be doing??

smash.

Re:love that headline...

[marcushnk]

perhaps just to perpetuate the status quo, by minimizing effort? (and letting their winged marketing monkeys of doom go for broke)

performance increase since 75MHz pentium?

[80 85 83 83 89 33]

P4 3.2GHz or a barton core 3200+ = 81 seconds

first gen pentium 75MHz = 5330 seconds

that is 66 times faster, and six doublings of performance!!!

Competition good... Sloth bad...

[ursabear]

I think the competition has been good for all of us. AMD's strength in the market has kept Intel on its toes, and advances from IBM's Power processors has kept many architectures running pretty well.

I (this is IMHO) believe that Intel has been doing some laurel-resting for a number of years now. I do believe that they will come to bear with better stuff on a gradual basis. My only fear is that Intel will allow itself to do like GM, Ford, AT&T... allow itself to be way too slow to be quick to adapt. I personally would like to see IBM, AMD, and Intel all have truly great, smokin' processors going way into the future - it seems that it would only be good for us in the long run.

Re:Intel has unrealistic expectations

[Anonymous Coward]

If nobody learns multithreaded programming, and it is as important as you think it is, then every microprocessor company (or, rather, their customers) is screwed. This is the way the industry is moving, so the software industry had better keep up.

However, since I'm usually doing more than one thing on a computer at a time, I don't require programs to be multithreaded in order to see the benefit of multicore processors. For instance, I can transcode video while web browsing or watching HD video and still have a snappy user interface. Perhaps most programs aren't multithreaded, but at least I am.

TRANSPUTER

[poptones]

I'm pretty sure transputer predates IBM's multicore POWER. Furthermore, transputer was inherently multi - up to four cores on a die and they could be interconnected easily via into larger arrays.

Re:TRANSPUTER

[Anonymous Coward]

I'm almost certain that Inmos never released a multicore version of the Transputer. You are correct, however, in that the Transputer were designed for parallel processing using high-speed communication links. It was very easy to assemble massive numbers of Transputers into a cohesive computing system. I have a 256-processor sysyem in my basement, right now (it was used for missile simulations and weather modeling, before it was surplused).

On the other hand, Inmos did release a 32-MAC DSP chip, but it wasn't really programmable, you could only change the filter coefficients.

Re:TRANSPUTER

[deadline]

The transputer was not multi-core. It could be connected to other transputers very easily, however. What it did have that made it rather interesting was process management in hardware.

General Rant - no need for multi-threaded aps!

[MadCow42]

>> * Multi-core processors require operating system (OS) support to make optimal use of the second computing resource.[1] Also, making optimal use of multiprocessing in a desktop context requires application software support.

The bolded section continually drives me nuts... NO, you don't have to have multi-threaded applications to get benefits from a multi-CPU system. When was the last time you EVER ran one program on your computer? Take a look at the Task List some day... there are probably 20-30 threads listed there at any one time.

I love my dual-Athlon system because there is NO lag... it never freezes up while starting an application or while one task is doing somethign intensive. I'm free to do surf while burning CD's, or read email while rendering images.

The world of one-program thinking is over, and has been for YEARS. Think SYSTEM when you look at the benefits of multiple CPUs.

MadCow.

Re:General Rant - no need for multi-threaded aps!

[maraist]

However, when I LAME something I see that one of my intel's cores goes to 100% while the other does not, so I DO find that applications need to be written for multi core systems.

I use "grip" and it asks you how many CPUs you have; it will simultaneously schedule that many encodings in parallel.

Technically the fast fouier transform can be parallelized, but it only makes sense with very large datasets, and MP3s break the audio clips into 1/32 second bit-streams, not a whole hell of a lot to transform. I'm sure there are pieces that can be parallelized of the MP3 format.. For one, you could have the encoder simultaneously try two different paths to determine which is smaller w/ less loss of data. You wouldn't be able to just give one frame to one CPU and another frame to another because they are semi-dependent on the ordering and context that's stored in subsequent frames.

In general, I've found that the best multi-CPU loads are in fact separated units of work where there is zero recognition of peer threads (except for state-safe locking/synchronization). While you CAN eek extra performance by writing threaded code, it just costs too much, is TOO damn hard to debug and the majority of the market is still single CPU'd anyway.. Parallelized web / servlet tasks are excellent scaleable units of work. Other things like IO-tasks tend to suck compared to multiplexed IO streams like UNIX "select" and friends.

multithreaded programming

[marco.antonio.costa]

I know I'd love to see great built-in multithreading/concurrency support in the next C++ standard. Java already does MT pretty well, right?

bottom line, people won't see massive improvements in performance with those new multicores until ppl really get the hang on developing multithreaded soft. I better read up on the subject.

Re:multithreaded programming

[thisislee]

If you're only using one program not designed for multithreading, you would see no benefit. If I understand correctly, running two cpu intensive programs at once, you would see a benefit.

Re:multithreaded programming

[Amouth]

"Java already does MT pretty well, right?"

the only thing java does well is eat memory

Re:multithreaded programming

[geekoid]

it is 2006 and people still aren't multi threading? hello? It is very simple people, get with it.
My god, if I was a manager of a project and my programers weren't using multi-threading they had better have a good reason. A Damn good reason.

Re:multithreaded programming

[marco.antonio.costa]

lol.. I don't find it that simple. maybe if "Hello, world!" programs were multithreaded I'd have less trouble getting the hang of it. :-|

Re:Too Little, Too late? My Arse!

[powerlord]

Ironically, the AMD64 series CPU's have no front side bus. This includes the X2 series. They have a hypertransport bus, which is similar but different. This is one of the premier reasons that the X2/Opterons scale so much better than the Intel equivalents, they do not have a saturated FSB as they have direct HTT links CPU-CPU

Isn't the direct CPU-CPU connections and the direct CPU-Memory connections an evolution of the EVA architecture AMD licensed from DEC? Sure would seem like their long range strategy is starting to pay off (and that they had the vision to see where they needed to go to inovate in the PC field).

BTW, The most striking thing I learned I my first day of OS design (a while ago), was how the computer industry keep re-inventing the wheel.

Mainframes seem to lead the way, PCs seem to be a few generations behind, but developing the same features, in the same general order. Handhelds seem set to follow in the path of their PC breathren.

I suppose we should be surprised, but its amusing to think about things we view as innovative in the PC arena and realize how long some of them have been around in the mainframe world.

It may be innovative, but it doesn't seem so

[Slayer]

That's the big problem with Intel's marketing: You are correct, they do produce faster and more powerful products every year (people who say all Intel does is cranking up the GHz have no clue about silicon processing IMHO) but the general public doesn't really get excited.

Back in those days it was exciting to see MHz climb up every year, we got excited about RISC technology (while most of us didn't know what the actual benefits of RISC are) introduced by Pentium and PowerPC. We were thrilled about DVD players and being able to see our digital photo collection just downloaded from our brand new digital camera. And we loved our music collection from the internet^H^H^H^H^H^H^H online music store :-)

You couldn't do any of this with P100s ... remember, it was the geeks who had to have the fastest meanest computer on the block back then.

Flip to present: New technologies are still being invented, although most of it for the purpose of imposing copy right enforcement (HD-DVD, BlueRay, Sony root kit CDs) or to spy on our surfing habits (Spyware hardly existed 10 years ago, viruses were just gaining real momentum) or to hammer us with SPAM (Ten years ago I got none. nada. nichts).

Now tell me why one would actually spend money and buy into this! Try really hard to get excited about this technology!

THAT is what the GP possibly wanted to express. You (the potential customer) see increases in GHz but no real value for yourself.

Re:It may be innovative, but it doesn't seem so

[somersault]

"You couldn't do any of this with P100s"

huh? I'm pretty sure you could do all of that with a Pentium (though maybe nothing else at the same time). I could play MP3s choppily on my 40Mhz 68030..

And I think the main reason that we're not excited about more Ghz anymore is because the CPUs are much faster than the memory we have, and so we dont see the benefits for memory intensive applications..

Re:It may be innovative, but it doesn't seem so

[Slayer]

Well, if you look at current CPU development, they do invent ways of addressing the problem with memory band width: Cache management, fast memory busses, smarter code generation, ...

That's exactly what I meant when I said: "There is a lot of inovation in this field"

But explain me one little fact: When PCs got their first CD burners, everybody started burning music CDs and other stuff. It was assumed: Why would anyone buy a CD burner if you couldn't do that? I have yet to find a computer which can copy DVDs out of the box. What's the point of buying a new computer right now?

Re:It may be innovative, but it doesn't seem so

[somersault]

your CPU isnt modular like your PC, you cant add on your better memory controller without upgrading the mobo. You dont usually have to dump your whole machine just to upgrade one part.

There's nothing wrong with Intel increasing performance by bumping up the FSB etc, until they realise again that they have been barking up the wrong tree for the last few years. Things like cache management etc are more clever. Bumping up the speed is kind of a given though anyway, not exactly innovative, unless it's because you've made a breakthrough in fabrication techniques or transistor design etc

I may have been an AMD fanboy for the last 6 years, but I could learn to like Intel now that they are having to play catchup anyway. I dont like arrogance.

Keeping the options afloat

[the eric conspiracy]

Intel has a serious problem in that they are perceived, and rightly so that to be a technical laggard. They are bleeding market share and their stock price has dived.

As a result Intel is trying to revamp their product line to become more competitive - but to keep from losing customers they are trying to darken the sky with marketing. This will work for a while because Intel has some credibility amassed from its earlier successes.

But if they fail to deliver at least parity with the next round of designs they are going to lose market share as fast as AMD can build Fabs. And right now they are running the risk of the 'Osborne Effect' - promising new product so attractive that the company loses large sales volume on current sales.

So Intel is making some really big bets here. If we get into the same time frame in 2007 with AMD still having a clear technical lead we could see AMD and Intel all of a sudden having a 40/60 split in market share, and a duopoly where once there was a monopoly.

Intel Marketing

[Exter-C]

Although Intel is not alone on this front it has been shown many times in the past that Intel will announce a technology or chip or both and yet it takes 6+ months for those to actually hit the market. We really should just wait until we actually HAVE the products from Intel (and other people) before raving about how fantatic they are.

Re:Intel Marketing

[TubeSteak]

Although Intel is not alone on this front it has been shown many times in the past that Intel will announce a technology or chip or both and yet it takes 6+ months for those to actually hit the market.

Yea, the dreaded soft launch

They'll send out samples to the major hardware sites... but nobody else will be able to get their hands on it for months.

I think they do it mostly to keep people from buying a competing product in the meantime. Ya know: Buy X now, or wait for really flashy cool Y in 6 months. Most purchasing decisions can be put off that long anyway.

Number of apps depends on OS

[jscotta44]

Most Macintosh users are used to using 6-7 application at the same time. Just a minor thing and forgivable since you are probably not a Mac user. Examples of apps constantly running on even 400 MHz G3 iMacs:

- Word processor
- Browser
- E-mail
- MP3 Player
- Photo Manager/Editor
- PDF Viewer
- Calendar application

That is just an example that I know intimately because it is my wife's that I took a look at to she what she was running. She rarely ever closes an application once she has opened it. And she happily runs her day multitasking away with nary a complaint. I asked her if she wanted a new iMac or PowerBook/MacBook Pro. Her response was, " Why? My computer does everything I need." But I do know that, in the case, her ignorance is bliss. I know that is she were to try a new dual-core iMac, she would not question it - especially with the number of applications that she runs.

However, I have to say that I am very happy with the longevity of the Macintosh (and very, very low maintenance) for the typical daily use of non-geeks.

I do agree with you that I would have expected things like applications to open so much faster and not be the cookers that the new silicone has tended to be. Hopefully, all the makers can fix that. I think that the competition between IBM, Freescale, Intel, and AMD are good things and I hope to see it-the competition-to continue. We all benefit.

Now if only Apple would release a Universal OS build that would allow me to run on each of those CPU makers.

Re:Number of apps depends on OS

[SpeckledJim]

Most pc users would have no problem running the list of apps you mention either. I think the comment about running 2 or 3 apps simultaneously refers to those which will grab a lot of the CPU, and be using it constantly, ie rendering apps, video transcoders, etc. The list you made will mostly be sitting idle unless you are actually using one of them.

Re:Number of apps depends on OS

[jscotta44]

Nope, he specifically stated "most consumers". The kind of people you are referring to are probably considered pros or prosumers.

Re:Number of apps depends on OS

[SpeckledJim]

Irrelevant, my point was a pc will handle a list like yours with ease, just as a mac can.

Re:Number of apps depends on OS

[aibrahim]

Well- I think you are both missing each others points.

Mac users typically multitask rather differently than Windows users. I am not talking about technical or power users, just "normal" users.

My experience is that most home Windows users barely multitask at all. 3-4 browser windows (not even tabbed since they are typically using IE.), a word processor and whatever they use for email. If they do something else, like photo management they tend to shut down their other applications. A few more advanced users will run iTunes or Winamp most of the time.

Office users use a bit more, usually either Excel or Powerpoint and a more specialized application, but those special apps are migrating to the web. Also they tend to have heavier email clients- like say Notes.

Add to this the stuff users don't know they are using, like AV software that scans every document before opening.

Windows users routinely quit programs, thus freeing and reallocating memory.

Mac users multitask differently. In fact, I'd argue they multitask in the same way a user accustomed to Linux does typically.

Typically you'll see 3-4 browser windows, each containing multiple tabs, their word processor (usually Word), Adobe Reader or Preview, Quicktime Player, some email program, a calendar application, iPhoto or iTunes and a few trivial apps of interest to the user. Oh, and 3-6 "widgets" which can be considered as browser instances.

They keep them open until something crashes or an Apple update "forces" a reboot.

When they decide to do something more serious, say running iMovie or Garageband, they don't shut down their other applications.

Remember I am not talking about "power" users or professionals. To contrast, I have a different set up running concurrently at this moment:

Shake, Final Cut Pro, Motion, Modo, Lightwave Layout & Hub, Soundtrack Pro, Safari (5 windows ~5 tabs each window), Firefox (1 window, 2 tabs), SubethaEdit (6 docs), Word (2 docs), Excel (no docs), iCal, iPhoto, iTunes (not playing), Mail, iChat AV, Quicktime, Photoshop, Toast, VNC, 3 Terminal windows, Cyberduck (FTP app), Compressor and Qmaster (distributed frame rendering for Shake and Compressor)

Presently I am rendering a flipbook in Shake, wasting time on /. and playing video in Motion. The rest of the stuff is idle for the moment. (Rendering a flipbook takes a good chunk of CPU.)

Typically if I were working on the same tasks on a Windows machine right now I'd be running Avid Xpress Pro, Photoshop and if desperately needed Combustion. I might have Thunderbird or AIM running, if I were expecting to hear from a client.

The upside of a Windows machine is that I wouldn't dare run a browser unless the project demanded it, and thus wouldn't be wasting life posting on /.

Humor aside, its a different pattern of multitasking. People keep a lot of apps open on OS X, to view different output or to change source materials for my Final Cut project.

I would point out that this is an OS issue, not a hardware issue- well not entirely. My Dual G5 has more RAM than Win XP 32bit supports and that helps. 64 bit helps shake, but makes no difference to anything else on my proc list.

If the apps were available for a 64-bit Linux you could achieve this readily. Windows XP 64 doesn't do a good job, and the apps aren't present anyway- but at least it won't crack like Win XP 32 bit will.

Re:Number of apps depends on OS

[AlterTick]

Mac users typically multitask rather differently than Windows users. I am not talking about technical or power users, just "normal" users. My experience is that most home Windows users barely multitask at all. 3-4 browser windows (not even tabbed since they are typically using IE.), a word processor and whatever they use for email. If they do something else, like photo management they tend to shut down their other applications.

I guess it's all dependent on your sample then, because the only Windows user I know who closes other apps before opening new ones is someone running Win98 on a 600mHz machine with 32mb ram. In fact, I regularly encounter people complaining that their computer is too slow who have a dozen apps plus 30-odd browser instances open.

Re:Number of apps depends on OS

[TheRaven64]

You can blame most of that on the maximise button. Where Windows (and the awful UIs that imitate it) have a maximise button, MacOS has a zoom button. The zoom button makes the window big enough to see the entire document (or as much as can fit on screen). The maximise button puts the user back in single tasking mode by taking up all of the screen real estate. The zoom button encourages the user to have multiple windows open, which enhances productivity since it's rare to only need a single resource open at once. This is, perhaps, why multiple monitors are so popular with Windows users - you can maximise two windows at once.

The other thing to blame is the window-is-the-application metaphor on Windows. It is impossible to close all of the windows of an application without it quitting using the Windows metaphor. This made sense back when everyone had 1-4MB of RAM and couldn't manage much more than one app open at once, but it doesn't now. On OS X, you can close all of the windows of an app, but still have it active. This is particularly useful when you want to close one document and open another in the same application. Without this, you need to do things in conceptually the wrong order, or quit and then re-launch the application.

Finally, Windows has no 'Hide Application' functionality. If you want to stop working with an app for a few hours / days you have no way of easily moving it out of the way. You can minimise all of the windows, but then it takes up task bar space. A hide feature preserves the spacial relationship between windows, but doesn't take up any UI space. I find this a much nicer solution to cluttered desktops than virtual desktops (which introduce unnecessary modes into the system), since I can have any combination of applications visible at once easily.

Note that this should not be considered a wholehearted endorsement of the OS X UI. It does a lot of things wrong, it just does fewer things wrong than most other systems.

Re:slightly off-topic

[dtsazza]

Check it out.

Where's the beef?

[Short Circuit]

OK, so I skimmed the article. But I didn't see anything in it detailing the new processors. Just drivel about competition and finances.

The geek in me wants more...

Overhype

[XMilkProject]

So I'm guessing next week we'll have the story about Intel overhyping their product release.

Re:Why woule Apple pick Intel?

[heinousjay]

AMD is not only ahead but being the underdog then that would seem to fit more with Apple culture.

You're mistaking the company for its fans. There are differences here, the main one being that the company itself isn't actually run on pure emotion.

Re:slightly off-topic

[Anonymous Coward]

Um those arent the yonahs(Core duos), these are the yonahs:

http://www.anandtech.com/cpuchipsets/showdoc.aspx? i=2648&p=14

To save you the trouble, "We continue to see that the Core Duo can offer, clock for clock, overall performance identical to that of AMD's Athlon 64 X2 - without the use of an on-die memory controller. The only remaining exception at this point appears to be 3D games, where the Athlon 64 X2 continues to do quite well, most likely due to its on-die memory controller. "

So basically intels laptop chips use less power, and can go head to head against AMD's desktop chips.

In any case Q3 Intel will be releasing the desktop version of the core duo which will up the power envelope slightly, but still be less then the amd,a dn should offer significantly better performance per clock then the core duo.

Pentium M differs quite a bit my friend....

[macavity1977]

You *REALLY* need to look at the Pentium M...

OS: Slackware Linux (Current)
Application: kcalc (Comes with KDE)

These are both ASUS laptops with PC3200 RAM:
2.8 GHz Celeron:     65 secs
1.6 GHz Celeron M: 18.5 secs

This kind of makes you wonder now, doesn't it? It appears that the Pentium M achieves *quite* a bit more per MHz then the Pentium 4.

Aside from that... the calculator in windows is obviously a joke, as the 1.6 GHz machine took 118 secs to do it in WinXP   >_<

Re:Pentium M differs quite a bit my friend....

[TheLink]

"Aside from that... the calculator in windows is obviously a joke, as the 1.6 GHz machine took 118 secs to do it in WinXP "

It depends. What was the result you got from kcalc?

I get 2.8242294079603478742934215780245e+456573 from the win2k calculator

Anyway the windows calculator was built for other purposes.

512MB L2 Cache?

[dunkelfalke]

holy shit, that is really innovative

Re:512MB L2 Cache?

[Amouth]

yea i saw that 512mb & 256mb L2's.. can i have one :)

it is 512kb & 256kb L2's.. and for the record the p3 coppermine (which is the one with 256kb L2) is not jsut another p3 "No real changes from the 600 MHz version" is completely wrong..

clock for clock the p3 coppermine is the fastest proccessor ever designed.. sure it can't do everything that the new stuff can do, but that wasn't what it was ment to do.

the coppermine was a wonderfull design and i would love to see intel bring it back from the dead and make some new high clock cpus with that core - i have heard some reports that that is what they did with the new line "core" but i don't know...

Re:512MB L2 Cache?

[ChronoReverse]

Eh? I like the Coppermine P3's but they certainly aren't the highest performing clock-for-clock x86 cpus. Even within the Pentium 3 family, the Tualatins outperform the Coppermines in that regard. Much of the Pentium M family also outperform the entire P3 family. Likewise with the Athlon families.

Re:512MB L2 Cache?

[Amouth]

the Tualatins is a Coppermine with increased L2 size.. no change in the core other than the L2 addressing. as for the Pentium M's they have a nice power to preformace ratio but from what i have used they are not nearly as fast clock for clock.

Not Just Cell

[DAldredge]

A breakthrough in microprocessing, the UltraSPARC T1 processor features SPARC-based CoolThreads technology and a revolutionary eight cores--each having four threads--for a total of 32 threads that work simultaneously so many tasks are performed in parallel with no waiting. The chip saves energy, while increasing system throughput and employs Sun's radical CMT processor architecture to keep pace with the multi-threaded application environment of the Internet.

Breakthroughs Unique in the UltraSPARC T1 Processor:

• First massively-threaded processor with 32 systems on a 90 nanometer die: akin to a 9.6 Ghz chip, its eight cores at 1.2 Ghz gives Sun a tremendous advantage over the competition;
  
• First to put the rack on a chip: Internal communications tasks right on the chip means less metal to move across for better energy efficiency and higher performance;
  
• First to put four memory controllers on a chip --which routes data between the processing cores and the memory - allowing data to be transferred into the chip as fast as it can be processed;
  
• Only processor to deliver industry-leading performance at lowest power consumption: Traffic flow is increased due to minimal memory latency. Because each individual UltraSPARC T1 core is simpler, it generates less heat-- using less than 70 watts;
  
• From threaded software to threaded chips: Marrying the processor and the system design yields the highest throughput. Get the most out of the world's best multi-threaded software architectures with no recompile. Using Sun Studio 11, the convergence of all three of these innovative technologies (Solaris 10, Java and UltraSPARC T1 with Coolthreads) add up to one innovation;
  
• Best Security: Improves system security while delivering 10x the throughput via crypto acceleration (compared with competitive processors) and runs secure applications faster than competitive processors. UltraSPARC T1 uses lower compute cycles to do the same amount of crypto operations. Processes crypto operations using as little as 5% of CPU capacity compared with Xeon which requires 33% and POWER5+ which uses 50%.(2)

Re:Not Just Cell

[Miguelito]

Man, and I thought some of us AMD fanboys sounded like shills at times. Way to cut and paste Sun ads. :)

I've done some tests on a niagra based box, and I'm not impressed at all. I love that claim that it's the same as a 9.6GHz cpu too.. yeah, I'm sure our engineers will think the same when running their single threaded, serialized apps and processes onthe box too. Nevermind that 31 of the "cpus" will be totally idle.

I think it's stupid for Sun to be pushing it like a 32CPU box.. just as annoying as HyperThreaded Xeons showing up as 2x the cpus is. They're not a full cpu, just count the cores.

Now the dual-core opterons and athlon64s I've played with (and I have an X2 game/dev box at home now) actually perform like I expect. My latest experiences with Intel and Sun multicores have been disappointing to put it lightly. Even FB-DIMMS don't help for running more then one process at a time on the new intels.

Re:Heat Sink... literally!

[evilNomad]

Did you care to look at the new yonah, which uses less power at 100% than the amd counterpart does at idle? I guess not... Sigh fanboys..

Re:slightly off-topic

[Tweekster]

ahh, but Intel doesnt use less power: Intel chip power usage only counts the power of the chip. AMD includes the on die memory controller. So if you add that factor and usage to the intel specs (for comparable features) then AMD is better...

Re:Is windows Calc threaded?

[lupine]

No, I get about 83 seconds on an intel dual core 3ghz.
winxppro - with lots of apps running.

Re:Why woule Apple pick Intel?

[Anonymous Coward]

Think of it this way. Apple wants to transition to the x86 world. Now there is no doubt that Intel is the better partner for this in terms of resources, marketing and stability. *HOWEVER* once they have a stable x86 product line, *THEN* they will IMO look to AMD and discard Intel if Intel falls behind again.

Slower is better to avoid armed conflict

[tepples]

If you could choose between two identical CPU's, but one of them were twice as fast as the other, which one would you choose? the slower one? I doubt it.

Not always. CPUs use electric power and turn it into heat. They get this power largely from fossil fuels such as coal, and access to sources of fossil fuel involves non-trivial politics and in many cases armed conflict. Unless you're using your data center to help heat the building during an appropriate season, it also takes power to move this heat away from the machine. In addition, mobile devices must carry batteries or other fuel with them at all times, and it is in the customer's interest to make the device last as long away from refueling stations as possible.

So if the cost of power and cooling are significant, and a slower CPU can get a given job done in "good enough" time compared to a faster CPU that drains the battery faster, then slower is better. Sure, expanding what jobs a CPU can do is a worthy goal, but by far not all CPUs are used in PCs or servers. The majority of microprocessors are deployed in embedded systems, where the job that the processor does is decided from day one.

Re:Slower is better to avoid armed conflict

[maraist]

Except that all modern CPUs perform throttling and use power-down noops instructions. If you actually cared about a green PC, then you'd shut your screen-saver off, but 99% of people that I know (namely I'm the only one out of 100 people that I know) DON'T shut off their screen saver...

With throttling (especially on laptops) you literally run at a lower speed, and power consumption is quadraticly related to frequency (2x speed is x^2 the power).

Then there's stuff like the video card. I have no idea how power efficient these massive beasts are anymore; they come with their own power supply plug, high powered fan, etc. So unless you know you're never going to game and thus get a REALLY crappy video card, you're probably throwing away more power-heat in your video subsystem than in your CPU at idle and definitely at game time.

Laptops have the luxury of using completely crappy video subsystems (that don't even have their own memory)

But seriously, the best green PC is going to be a laptop, EXCEPT that you're definitely not saving money in that approach. (a $1,000 difference for a semi-comparable system)

Core Duo is NOT a Pentium-D

[doh123]

Core Duo is an actual product, it doesnt just mean dual core.

The Pentium-D is a crappy thrown together dual core P4, not close to the same thing.

if everybody was as efficient as nintendo

[majortom1981]

I think that if programemrs were as efficient at programming as nintendo iss then we wouldn't need such advances. The gamecube ran software almost on par with an xbox 360 with processor specs at almost have the speed. Why can't programemrs just be more efficient at what they do and hwo they use the computers rescources?

Re:if everybody was as efficient as nintendo

[Slashcrap]

I think that if programemrs were as efficient at programming as nintendo iss then we wouldn't need such advances. The gamecube ran software almost on par with an xbox 360 with processor specs at almost have the speed. Why can't programemrs just be more efficient at what they do and hwo they use the computers rescources?

I was going to flame you for your apalling spelling, typing and grammar. But having thought about it, I've decided we should really only flame people for that when it detracts from their message.

Since your message is, "I'm a retarded 12 year old skipping English class to post on Slashdot" there is no need.

Single core to save on software licensing costs?

[tepples]

On the AMD side, dual-core may become more common than single core, very shortly. The fact that dual-core chips are drop-in replacements for single core chips helps, as vendors don't have to change anything other than their CPU order.

Unless the publishers of popular proprietary programs start billing customers per core. Microsoft isn't planning on charging dual-core customers more, but Oracle treats a dual-core CPU as 1.5 billing units.

No news here

[ChrisA90278]

Title: Intel Unveils New Chips to Battle AMD
Summary: Intel will anounce new chip at upcomming event
Real World: At a future event Intel will talk about a chip that will be available some months after the event
What this means: In the future Intel will talk about the future

No news here. Tell me whan I can buy one.

I Really Wonder

[Nom du Keyboard]

I really wonder how many i486 cores -- quite a competent chip for the x86 archtecture, all things considered -- Intel could put onto a die if they decided to do so. And with a modern process technology, how fast they could run.

you already got it...

[YesIAmAScript]

What you list "most consumers need", that's already what Intel is providing. Core Duo is their desktop solution.

These new chips mentioned are server/workstation chips.

Also, I find your latency comments incongruous. Yes, P4 has an overly-long pipeline. But it's not user-perceiveable in terms of latency. It's only reflected in how the processor just doesn't perform as well as might be expected from the processor clocking and transistor count or heat production.

The 10+ seconds thing is more attributable to other factors, if it's even true. Most apps launch very quickly for me and besides, it's more related to the OS and how it uses the disk than the pipeline of the processor.

I'd like to see Intel improve their processors further, but they do seem to be on the right track now and away from NetBurst.

Re:slightly off-topic

[Overly Critical Guy]

I know this is Slashdot where "AMD is better" is the taught mantra, but power consumption tests have already shown Intel's new chips kicking AMD's butt in performance-per-watt.

The Core Duo consumes less power at 100% than the Athlon 64 3800+ X2 does at idle.

Merom is now due this September, and AMD hasn't even switched over to 65nm yet. I know people around here love AMD, but they've always struck me as riding coattails and playing up the "underdog" aspect while copying whatever new thing Intel has done, be it SSE, Viiv, etc.

Re:slightly off-topic - not 64-bit

[Nom du Keyboard]

How do Intel Core Duos rate against Athlon x 2s?

Well, for starters the current Core Duo is a 32-bit only chip. And while 64-bit processing doesn't double your CPU speed or anything like that, there are other improvements in the AMD64 design (more registers, NX bit, etc.) that make for improvements beyond 64-bit integer processing and >4GB address space.

Comparing these two particular processor lines would be a lot like comparing 80286 processors at 16MHz with 800386 processors at the same clock rate. Both might run DOS at the time at a similar speed (IIRC the 286 was actually slightly faster with 16-bit code), but in the end the old architecture rapidly fell away to the new 32-bit processors.

I, for one, would only take a Core Duo system today (including every Intel-based Apple Mac) if it was given to me. I wouldn't spend my own money on one.

Re:slightly off-topic

[danskal]

It's much worse than that! AMD's Thermal Design Power is set higherthan the Maximum power used (worst case scenario), whereas Intel's refers to the Average power. This can translate into a difference of approx 50 watts.

Examples of the actual power consumption differences:
http://www.tomshardware.com/2005/05/09/amd/page20. html

Here is a detailed discussion:
http://www.silentpcreview.com/article169-page3.htm l

Re:Which innovation? - ITANIUM ! away from x86

[acomj]

Intel spent much time money and engineering talent on ITanium. I think it has roots in PA-risc but has a completely new instruction set. It didn't fly, for a variety of reasons I won't go into here.

Itanium didn't work as well as expected, and by all accounts is a failure, but at least intel tried. It has some interesting features for a chip. It could be considered more inovative than grafting 64 bit instructions on a tired old 32 bit instruction set.

Who knows, maybe itanium will make a comback! (there still working on it, Don't hold your breath though)

Re:slightly off-topic

[Tweekster]

Did you not read what I said... What you just said is not a fair comparison... you are comparing Proc + Memory controller to intels Proc only... intel has the memory controller seperate, which uses power, which means when you add those two together (for a comparable chip+features) AMD wins. try to pay attention this time

We need a 4 GHZ or faster CPU

[Frank T. Lofaro Jr.]

We need a 4 GHz or faster CPU. More GHz is the only way to make a CPU intensive non-multithreaded application run faster.

Imagine how fast a 5 or 10 GHz PC will run.

Re:We need a 4 GHZ or faster CPU

[geekoid]

"Imagine how fast a 5 or 10 GHz PC will run."
I would imagine 5 or 10 GHz, respectivly

Re:We need a 4 GHZ or faster CPU

[chawly]

I would instinctively agree with your estimation. And even if we're wrong, it has to be faster than 4GHz ..... I think.

Re:Heat Sink... literally!

[zlogic]

My dad wanted to buy a Pentium 4 PC. I told him not to and recommended AMD. He still wanted to buy an Intel, and bought one. It used a heatsink with size like of a CD (third-party, not Intel), but the fan itself was reasonably quiet. The CPU's temperature rarely rose higher than 45C, which surprised me a lot - I thought it wouldn't get lower than 60C.
And notice how AMD presents their benchmarks:
http://www.amd.com/us-en/Processors/ProductInforma tion/0,,30_118_11599_11605~101503,00.html
An AMD CPU that's 23% faster that Intel's appears to be at least 2.5 times faster. That's because they are showing only the last 85%-125% of the scale.
However I'm still pro-AMD and anti-Intel.

Re:Too Little, Too late? My Arse!

[AlterTick]

Isn't the direct CPU-CPU connections and the direct CPU-Memory connections an evolution of the EVA architecture AMD licensed from DEC? Sure would seem like their long range strategy is starting to pay off (and that they had the vision to see where they needed to go to inovate in the PC field). BTW, The most striking thing I learned I my first day of OS design (a while ago), was how the computer industry keep re-inventing the wheel.

If they licensed it from DEC, I don't think that counts as reinventing the wheel. It's more like finally dragging the x86 architecture kicking and screaming into the age of the wheel.

Re:Heat Sink... literally!

[geekoid]

"However I'm still pro-AMD and anti-Intel."

Because we don't want facts to stand in the way.

Re:slightly off-topic - not 64-bit

[Anubis350]

AFAIK the 64-bit version comes out later this year, it's most likely what Apple is waiting for before updating the pmac line...

Re:Heat Sink... literally!

[dbucowboy]

For me, it's all about reliability... Intel processors have proven to me that they are NOT reliable... I worked in a computer lab were we used only Intel processors and we ALWAYS had problems with them. I've worked with AMD powered PC's with equal or greater workload and I have NEVER had a problem with 1 AMD processor...

Faster? Who cares! I know that my processor isn't going to explode and die... do you?

Formerly FEMPUTER

[DumbSwede]

I thought TRANSPUTER is what FEMPUTER became after trying to live in a MANPUTER's world.

Re:Heat Sink... literally!

[zlogic]

There was a famous video clip made by Tom's hardware in 2002 (or something like that), where they took off the heatsink off an Athlon XP and Pentium 4 while the CPU processing lots of data:
http://www4.tomshardware.com/images/THG_CPU_Coolin g.zip
The AMD melted in 10 seconds, the temperature rising to something like 350 degrees Celcius, while the Pentium 4 lowered its FSB and continued working without any cooling for some time.
However I think the situation would be the opposite today.

Re:Where's the beef? Mod STORY down

[javaxman]

Seriously, the closest this article gets to anything technical are the words "multiple cores" and "enterprise servers". That's it.

What new chips? Are they x86 compatible ? 32-bit ? 64-bit ? Do they have a code name ? Are they going to ship some day ? The article is made more pathetic by the fact that one could actually answer all of those questions, but this story contains virtually no real news for nerds whatsoever.

This story is crap, let's face facts. Slashdot should be protecting it's readers from this kind of marketing drivel, not placing it on the front page where we're tempted to waste our time on it. I want those 2 minutes of my life back, please...

Re:Is windows Calc threaded?

[ErikZ]

I'm going to assume a typo, and that you meant 2Ghz.

Re:slightly off-topic

[NDPTAL85]

Those tests were run using Intel's Pentium D dual core chips which are based on the badly performing Pentium 4 design. The Intel Core Duos are based on the Pentium M and perform much better.

2.824229407960347874293421578e+456573

[80 85 83 83 89 33]

but what's the result? It's no point producing the wrong result in faster time ;)

who knows if this is the correct answer, but, on my celeron 800MHz, with Windows 98SE, calc returned: 2.824229407960347874293421578e+456573

Re:Is windows Calc threaded?

[80 85 83 83 89 33]

actually, his times sound about right, because i recorded somthing similar on another 3.0Ghz dual-core machine

Pentium 930 dualcore (3.0GHz)
82 seconds (XP media center)

Re:Is windows Calc threaded?

[ErikZ]

I can't find a 3Ghz dualcore anywhere. 2Ghz is the max. Where are you seeing these?

Re:slightly off-topic

[Overly Critical Guy]

No, perhaps you should pay attention. Read the benchmarks. The separate memory controller is taken into account. The Core Duo consumes less power at 100% than the AMD at idle, even with AMD's on-die memory controller.

Article with results

[crossmr]

Wasn't there an article that had the results of this new chip showing that it was like 40% faster than thew lead AMD chip? I thought it was on slashdot but this it the only mention I can find of it..anyone happen to have a link to it?