Intel Lynnfield CPU Bests Nehalem In Performance/Watt

Vigile writes "Not many people have debated that Intel's Nehalem architecture is the fastest available for consumer desktop computers since it was released last year, but quite a few have complained about the cost of the platform. Intel just released new Lynnfield-based processors under both the Core i7 and Core i5 names and tests are showing the new CPUs beating Nehalem in both performance-per-watt and performance-per-dollar tests to a startling degree. And while raw performance probably still goes to the Nehalem-based Core i7 CPUs, the lower prices of motherboards and memory for Lynnfield processors will likely more than make up for it." Update: 09/08 14:03 GMT by T : There are more eye-wateringly exhaustive examinations of the new chips all over the Web; here's HotHardware's version, and Tom's Hardware's.

Comments

arm

[RiotingPacifist]

How do these compare to "popular" arm chips? Ideally ones powerful enough to run netbooks not just phones.

Re:arm

[Kjella]

If you imagine ARM as a women's flyweight newbie and Lynnfield as the men's heavyweight world champion in boxing, you got a pretty good idea how that match will play out. Not nearly the same class and the results are as expected.

Re:arm

[Sockatume]

How is a raven like a writing desk? They're entirely different kinds of products. They're not even the same kind of computer (RISC versus CISC), but even setting that aside, you'd be better off comparing an ARM CPU to one of Intel's low-end Atom offerings.

Re:arm

[FlyingBishop]

The troll is not that far off. However, a car analogy is probably better:

The ARM chip is a Prius hybrid.

The Intel Core series is a 15-year-old diesel schoolbus.

Re:arm

[afidel]

More like a diesel locomotive, more absolute power burned but significantly better MIPS/Watt. If the ARM architecture gave better MIPS/Watt then you'd see supercomputers based on it, you don't.

Re:arm

[Anonymous Coward]

They're entirely different kinds of products.

They're both Von Neumann architectures. If my application is parallelizable and not written in assembly, I can easily switch it from running on n Lynnfield CPUs to running on m ARM CPUs. In that case, only performance-per-watt and performance-per-dollar matter.

Re:arm

[Anonymous Coward]

GP is obviously asking about perf/$ and perf/watt, where ARM has always been king. The question was almost certainly rhetorical, nearly trollish, given the obvious result.

In other news, every Intel processor since the P3 has been a RISC execution core with an x86 translation layer (and all competing processors have had similar translation layers, even for "RISC" ISAs, for about the same period of time).

So, to summarize:

"How does a Cortex A9 compare to a i7-870?"
"Poe wrote on both"

Re:arm

[zokier]

I'd be interested to know how would a highly clocked, 8 core, ARM based CPU made at 45nm process and fitted to be compatible with some current tech (PCIe 16x for GPU and DDR3 memory) compare to anything offered at desktop market now. Of course there is the little problem with no actual software optimized for that kind of architecture.

Re:arm

[644bd346996]

There are plenty of ARM processors with a great MIPS/W rating. Just not a great FLOPS/W rating, which is what keeps them out of supercomputers.

Re:arm

[FlyingBishop]

A bus still tends to be better than a hybrid in passenger miles per gallon. I wasn't disputing the superiority of server CPUs for use in servers.

Re:arm

[TheLink]

> They're both Von Neumann architectures.

And an alien will find there's little difference between you and a rat. Come back to planet earth sometime.

Heck both are not even made of dark matter- that would be the most interesting point if the alien is made of dark matter (which apparently makes up >90% of the matter in the known universe, according to many scientists).

Re:arm

[Anonymous Coward]

> They're both Von Neumann architectures.

And an alien will find there's little difference between you and a rat. Come back to planet earth sometime.

We're getting well off topic here, but surely 'how does a human compare to a rat?' is a valid question. Especially for an alien I would imagine.

Re:arm

[Kjella]

What is it now, the secret Linux on ARM "this will crush Intel any day now" idiot brigade doing the modding? I guess I'm in for another dose of negative karma on this post, but wtf.... we're talking about 2-500$ quad-core CPUs with a 95W TDP. There's not an ARM processor in sight that is even remotely competing in this class. Maybe if this was about some low-end Atoms the question could have made a little bit of sense, but now it's just to laugh at. And the ARM fanbois with no humor I guess.

Re:arm

[Mister Whirly]

I would imagine that if it was a great drop in cost compared to performance that they would already be making them. If we are having this discussion now, engineers have had it years before and decided to pass.

Re:arm

[TheLink]

I agree that it's a valid question.

My point is using the AC's logic, one might even answer "They're both Von Neumann architectures" to your question too. Which isn't a very useful IMO.

As for an alien, it may depend a lot on the alien :). Who knows, maybe with some aliens it might be like us comparing a small snowflake with a bigger one, before the snowflakes both vaporize in a furnace - esp if their timescales are much longer than ours...

Re:arm

[Anonymous Coward]

Go ahead. Scoff at the ARM architecture based on aliens and dark matter arguments and mod that shit up to +5 insightful. Meanwhile I'll be running my company on ARM and beat yours at the marketplace.

Re:arm

[MightyDrunken]

Heck both are not even made of dark matter- that would be the most interesting point if the alien is made of dark matter (which apparently makes up >90% of the matter in the known universe, according to many scientists).

Well that could answer the Fermi paradox.

Re:arm

[TheRaven64]

The popular ARM chips are single-core Cortex A8 running at between 600MHz - 1GHz. They perform slightly better than (single-core) Atom clock-for-clock on most workloads, and slightly worse on a few. The next generation chips that are just starting to hit the market are based on the Cortex A9, which does a bit better clock-for-clock and scales up to 4 cores per die. ARM chips also typically have the memory and flash controllers, GPU, and a DSP on die. Something like the OMAP3530 consumes around 250mW in real use or around 15mW when playing back MP3s on the DSP. They are typically limited to around 1GB of RAM, with only about 256MB being available in package-on-package configurations (i.e. not requiring a more expensive motherboard).

In short, they compare like apples and oranges. In terms of performance per watt, the ARM chip most likely wins by an order of magnitude - more if you include the DSP. In terms of absolute performance, the i7 wins by at least an order of magnitude.

Re:arm

[RightSaidFred99]

It (the ARM) would get crushed. It's not designed for that scenario. ARM is old. It's good at low power and decent performance, but it's certainly not uber technology.

Re:arm

[GauteL]

Could someone please stop modding this down? It really is not a troll and the post is essentially true.

Comparing a quad core high-performance desktop processor with a lightweight low power mobile/embedded processor would be completely pointless. An argument which the parent actually made with a little bit of humour.

The quad core desktop processor would absolutely trash the ARM in performance, but would result in a phone that would (literally) burn a whole in your pocket for 5 minutes until it ran out of battery.

It really is like comparing apples to oranges. Both types of processors are completely unsuitable for the main uses of the other.

Besides I can think of at least one use for a female flyweight newbie for which I'd never consider a male heavyweight champion.

Re:arm

[Anonymous Coward]

No doubt.

The Fermi paradox rocks my socks!

Re:arm

[Anonymous Coward]

So ARM makes for good cocksuckers?

Re:arm

[Anonymous Coward]

and your mom sucks! go figure!

Re:arm

[Anonymous Coward]

If my application is parallelizable and not written in assembly, I can easily switch it from running on n Lynnfield CPUs to running on m ARM CPUs.

Okay Mr. Smarty pants, please parallelize one of the following problems:

1. Linear programming: Maximize a linear function subject to linear inequality constraints
2. Context Free Grammar Membership: Given a context-free grammar and a string, can that string be generated by that grammar?
3. Lexicographically First Depth First Search Ordering: Given a graph with fixed ordered adjacency lists, and nodes u and v, is vertex u visited before vertex v in a depth-first search induced by the order of the adjacency lists?

Hint:They're all P-Complete, so if you can parallelize one of them then you've proven P=NC. (Note: that's NC not NP.)

Re:arm

[TheRaven64]

If you built a supercomputer out of ARM cores, with the same total power consumption as an existing system, it would have a much higher integer performance total than any existing supercomputer, and probably not do too badly on the floating point front (especially if you used of-the-shelf ARM chips and included the on-die DSP and GPU in the total). Each chip, however, would have its private memory, and moving data between them would be very slow. In terms of the sort of problems supercomputers are typically used for, its performance would be very poor. That said, you could put 64 600MHz ARM cores each with its own 512MB flash and 256MB of RAM on package on a board and still be well under the power consumption of a typical Intel or AMD chip, which might make sense for web server type workloads.

Re:arm

[Anonymous Coward]

just to clarify that's millions instructions per second and fuck loads per second?

Why are they still calling it i7?

[themysteryman73]

Perhaps it's nearly time to upgrade my aging Athlon X2 5600+...

One thing that I find interesting about this is that Intel decided to still call it "i7" when it apparently doesn't fit into the Nehalem-based i7 motherboards. As the article correctly pointed out - why not call it "i6" to prevent consumer confusion?

Re:Why are they still calling it i7?

[noundi]

Perhaps it's nearly time to upgrade my aging Athlon X2 5600+... One thing that I find interesting about this is that Intel decided to still call it "i7" when it apparently doesn't fit into the Nehalem-based i7 motherboards. As the article correctly pointed out - why not call it "i6" to prevent consumer confusion?

My processor goes to i11.

Re:Why are they still calling it i7?

[Anonymous Coward]

why don't you just go to i10 make that faster

Re:Why are they still calling it i7?

[Anonymous Coward]

no, but see, his goes to i11.....

Re:Why are they still calling it i7?

[tylerni7]

Lame, my processor is restricted to the reals.

Re:Why are they still calling it i7?

[Mister Whirly]

Can I raise a practical question at this point? Will your processor be able to run Stonehenge v1.8 tonight?

Re:Why are they still calling it i7?

[bluesatin]

The marketing department spokesperson was quoted as saying 'i10'nd to name things the same as previous products, to reduce the amount of brands we have to promote'.

Re:Why are they still calling it i7?

[DragonWriter]

One thing that I find interesting about this is that Intel decided to still call it "i7" when it apparently doesn't fit into the Nehalem-based i7 motherboards. As the article correctly pointed out - why not call it "i6" to prevent consumer confusion?

Intel has never been particularly keen on matching processor branding to sockets; while its probably frustrating to enthusiasts who'd like to be able to rely on the branding rather than several different axes of variation within the same brand name while building systems, its hardly new with the Core i7 brand, and there is really no reason to expect (as opposed to "want") it to change.

Re:Why are they still calling it i7?

[lordmetroid]

On the other hand my professor can count to 100

Re:Why are they still calling it i7?

[LeperPuppet]

The i7 label is apparently for chips with hyperthreading and the i5 label is for those without hyperthreading. The naming scheme is idiotic since you now have socket 1366 Nehalem i7s and newer socket 1156 Lynnfield i7s. Since Intel wants to have two different sockets for different markets (high end enthusiast and low/mid range buyer) why not actually clearly label the corresponding products? Maybe they think customer confusion will somehow lead to increased sales or greater sales of processors with higher profit margins? Maybe they're just too lazy to try to explain to their customers what hyperthreading is and why they should pay extra for it.

Re:Why are they still calling it i7?

[toddestan]

As the article correctly pointed out - why not call it "i6" to prevent consumer confusion?

Probably because Intel loves to confuse consumers? For extra fun, try to figure out the pattern for what chips have virtualization and for which chips don't.

Lack of focus

[BadAnalogyGuy]

I've begun to feel that Intel is lacking focus in their chip lineup. While it makes sense that they have different series for different markets, within those lineups they have too many disparate chips that just cloud the water.

Atom Z vs Atom N is one such case. The Atom is supposed to be their embedded processor series, but they just can't shake off the PC market yoke and focus solely on embedded customers.

They have server CPUs, desktop CPUs, mobile CPUs, and embedded CPUs. But within each segment there are just too many choices that make it difficult to understand the whole picture without true data analysis like this article.

Re:Lack of focus

[afidel]

It's called binning and price discrimination. One is a technical/economic tool to maximize profits based on the non-perfect nature of chip manufacture, the other is the capitalists favorite tool to extract the maximum profit possible out of the consumer.

Re:Lack of focus

[fuzzyfuzzyfungus]

One wonders if this is a lack of focus, or if they aren't hugely interested in having the whole picture be understood(at least by end users).

If, for instance, they are assuming that the first line of selection will be done by the (presumably informed and competent) OEM, the amount of choice the customer has to deal with is considerably reduced. You pretty much just choose the device you want, and then pick from a very limited set of chips available in that device. Server configuration is still a bit complex; but so are server applications, so server customers can suck it up.

Aside from enthusiasts who just like following the stuff, the relevant picture isn't all that complex. If you want a netbook, you get Atom, game over. If you want an ultralight, you get some cut-down ULV version of the core2. If you get a laptop or a normal desktop, you get a core2. If you get an enthusiast model or workstation, you get a quad core. Within each device type, there are only really a few clock speed/cache size options to choose from(and, unless you are doing Serious Computing or gaming, it hardly matters which one you choose).

Re:Lack of focus

[afidel]

Wow, how is explaining the market dynamics of chip manufacturing a troll?!? Was it the use of the word capitalist? You do know that it's a perfectly valid economics term, right?

Re:Lack of focus

[Kjella]

They have to keep up the idea that there's competition by having many different brands offering many different options. Truth is that in many markets it's now a grand choice of Intel, Intel and Intel. Then you usually make more money keeping your customers confused and selling outdated or low-end processors to high-end prices than making it all very obvious. AMD is struggling badly to carve out any sort of niche where they can get a premium, delivering value processors isn't making them enough money to do what's necessary neither in R&D nor in process development. Intel on the other hand is pounding away at systems-on-a-chip, SSDs, higher-end graphics and really moving towards the Intel computer with your choice of Dell, HP or Compaq sticker. You can tell nVidia fears that future too.

Re:Lack of focus

[L4t3r4lu5]

Was it the use of the word capitalist? You do know that it's a perfectly valid economics term, right?

Only when followed by the word "pig!" and uttered in an Eastern European accent.

Re:Lack of focus

[commodore64_love]

the relevant picture isn't all that complex. If you want a netbook, you get Atom, game over. If you want an ultralight, you get some cut-down ULV version of the core2. If you get a laptop or a normal desktop, you get a core2. If you get an enthusiast model or workstation, you get a quad core.

What was that you were saying about "not confusing"? Jeez.

Okay I have a Pentium 4. I want to upgrade. What's the modern-day equivalent of a Pentium 5? I see Atom notebooks for sale for $300 but those seem to have less power than what I have now, so do I get a Core2 CPU instead, or is that considered too old? Am I suppose to get an i7 processor? Or perhaps an i8? Maybe I should look at AMD instead?

This confusion reminds me of what happened with Apple with they had Performas, Quadras, and Centris Macintoshes. It was a mess of models that had no logical consistency, left the consumer confused, the company strung-out supporting too many brands, and almost drove Apple to bankruptcy in 1995 (as happened to Atari and Commodore the year before).

Re:Lack of focus

[bemymonkey]

So you got a Mac because of the lack of choice available at each price point?

That makes a whole lot of sense...

Re:Lack of focus

[vlm]

Wow, how is explaining the market dynamics of chip manufacturing a troll?!? Was it the use of the word capitalist?

Ahh you left out the rest of the phrase "capitalists favorite tool to extract the maximum profit".

Everyone knows the actual favorite capitalist tools to extract profit, in order, are:

1) Form a Monopoly (Microsoft)

2) Form an Oligopoly / Cartel (OPEC, to some extent Intel/AMD)

3) Form a confuse-opoly to eliminate the free market by making it incomprehensible (cell phone pricing, and to a lesser extent, CPU pricing)

4) Government intervention via patents, regulations, govt contracts, special taxes against your competitors

5) Marketing using sex appeal (well, maybe not the capitalists favorite tool, but it's my favorite)

6) Global movement of capital and jobs via world trade while preventing populations from moving along (due to culture / language / govt issues)

Blah blah blah, somewhere around favorite tool number 25796274 there is yours, "price discrimination"

So that is what made it vaguely "troll-y"

Re:Lack of focus

[MadKeithV]

You're new here aren't you? The /. way involves:
3. ????
4. Profit!.

Re:Lack of focus

[MadKeithV]

The only people confused are the tech-savvy people who care. The rest of the consumers are just happy with the performance of the computer / laptop they bought at the price-point they could afford, blissfully ignorant of the differences in power, cores, and battery / energy use.
I have an Atom-based netbook. I didn't know it was an atom when I bought it. I knew it was only â199 for an ultraportable that would allow me to read PDFs, read and write office documents (in OpenOffice), and surf the web.

Re:Lack of focus

[ThePhilips]

I've begun to feel that Intel is lacking focus in their chip lineup.

Yes. Because they can afford it.

Now fab-less AMD remains in remote second position and can't really compete against the Chipzilla.

That gives a perfect chance to Intel to further fragment the market to maximize the profits.

Re:Lack of focus

[skolima]

Actually, if you're using a Pentium 4, Intel Atom is also an upgrade. I recently went from Pentium 4 HT 540 (Prescott, 3.2GHz) to Atom 330 (1.6GHz, dual core, HT, D945GCLF2 motherboard) on my home media server and noticed a performance increase (of about 20% with the things I cared to measure, but never the less - an increase) while the power usage dropped to 20% of what it used to be.

Re:Lack of focus

[coolsteve]

You're new here aren't you? The /. way involves: 3. ???? 4. Profit!.

No one on /. these days remember that meme. I know. I've tried using it before, and with little success...

Re:Lack of focus

[Anonymous Coward]

Depends. Coming from a P4, you'd have to be happy with anything, but the part you probably would is something like the E7400, which is a dual core Core 2 part running at 2.8GHz per core. It's plenty fast enough for most normal tasks, but it's FSB runs at 1066mhz, so there's no point pairing it with DDR3. For a little more money, the Q8400 gets you a quad core system at 2.66ghz per core, but a faster FSB of 1333mhz.

In direct answer to your question: the modern day equivalent to a P4 is a Core 2 Duo.
The modern equivalent to a professional multiprocessor workstation is an i7.
The Core 2 Quad fits in somewhere between - a part for hobbyists and video editors, for the most part.
The Atom is a low power part - my understanding is that it's based on the Celeron version of the Core 2 (which, in turn, is an redesign based on the Pentium M, which was in turn more or less based on the Pentium 2), but optimised for power efficiency. As far as I can work out, they're a little less fast than an equivalently clocked Pentium 4, but much, much more efficient. That said, a 1.6ghz P4 was...pretty slow.

While it's a bit confusing, it doesn't take that long to work out, and frankly, most people will buy whatever Dell sells them - ie. a Celeron with Vista Home Basic.

Re:Lack of focus

[DaMattster]

An educated consumer is a dangerous consumer. You want to be able to trump up a product to sell it. But, you can't trump up a product if your buyer is an expert.

Re:Lack of focus

[epine]

An educated consumer is a dangerous consumer. You want to be able to trump up a product to sell it. But, you can't trump up a product if your buyer is an expert.

Yes, but theory of the beneficent invisible hand is based on voluntary transactions between informed parties. It seems to me that most American businessmen are capitalist, except for than annoying bit about informed consumers. Instead we have a giant industry to manufacture overwhelmed consumers who purchase on color preference.

You comment about trumping up a product is valid as an inference from how the system actually works, but has no point of contact to the philosophical system (capitalism) that we use to justify having structured the system in this way.

What you end up is the profit motive shorn of the invisible hand responsible--in the philosophical sense--for weeding out inefficient producers.

In this case, the informed consumer has good cause to conclude we're being bamboozled by too much manufactured choice. The video card industry is the epitome of this. Models come and go faster than one can establish a reference point from reliable review.

On another note, the Toms Hardware article (a fairly decent one) made the usual blunder comparing the three channel i7 configuration to various two channel alternatives.

With three channels you can load more RAM into your system before filling the second pair of slots on each channel (which results in a small RAM performance loss). The number I would like to see is six sticks in three channels vs eight sticks in two channels. The other benefit of the third channel, which they neglected to point out, is that it increases your maximum RAM build by 50%.

I'm using R fairly heavily these days. Great statistics language, but doesn't cope especially well with serializing data off disk. You get better results with your entire data set loaded into memory. I don't have to image a very large data set before 24GB starts to seem attractive.

Re:Lack of focus

[apoc.famine]

You aren't understanding the parent.
 
You take your $800, go to Dell's website, and select $700 worth of computer, and pay $100 tax and shipping. And then you get a free printer with 6 pages worth of ink in it.
 
You aren't Intel's customer. Dell is. As the GP said, "Aside from enthusiasts who just like following the stuff, the relevant picture isn't all that complex." You fall into that "enthusiast" category, by even thinking about P5 or AMD.
 
For Dell, it's a no-brainer. You find a decent chip you can buy in lots of 100k, and put them in systems. For Dell's customers, it's a no-brainer. You take your money, and buy something with it.

Re:Lack of focus

[cffrost]

You put yourself on several peoples' Freaks list. That's begging for stalker-mods.

Re:Lack of focus

[lordmetroid]

What the hell are you trying to say, I can not understand you for all the numbers and all the question marks.

Re:Lack of focus

[commodore64_love]

>>>if you're using a Pentium 4, Intel Atom is also an upgrade.

False.
P4 == 9,726 MIPS at 3.2 GHz - 3.039 MIPS/MHz
Atom == 3300 MIPS at 1.6 GHz - 2.06 MIPS/MHz

Pedantry note

[Kupfernigk]

I think the submitter means "disagreed", or "argued" not "debated". I expect that in the early stages quite a lot of people debated the subject, but when the results become clear they stopped arguing and there was a general agreement

Yes, I know it's pedantry, but some of us like to live in a world where different words mean different things that make a useful distinction. And now, please, do get off my lawn before my dog comes and pees on your shoes.

Re:Pedantry note

[LordKronos]

but when the results become clear they stopped arguing and there was a general agreement ...
some of us like to live in a world where different words mean different things

Are you referring to things like the difference between future tense and past tense? :-)

P.S. Please don't bother trying to pick apart any wording that I use in this post. I don't really give a crap about minor stuff like that, but if YOU are going to be a pedant, try not to be so hypocritical about it.

What are Intel's naming department on?

[Lemming Mark]

I don't understand how Core was an improvement on Pentium. Pentium was iconic and a household name (which is pretty difficult to achieve in such a low-level field, especially as Intel typically doesn't sell direct to consumers). Core is boring and misleading. For instance, Core 2 Duo ... whuh?! Doesn't sound that impressive but definitely sounds muddled.

Now there's this i7 and i5 business. Maybe I'm just old but I preferred when "Pentium n" is the new processor and probably better than my "Pentium n-1". I can understand they may have wanted to avoid the Sexium but at least that would be distinctive. Core is about as boring as traditional IBM naming.

Their hardware is excellent these days. They went through some doldrums but generally seemed to sort themselves out pretty effectively and come out with ace stuff. Their Linux support is usually great too. Maybe one of these will be my new PC...

Re:What are Intel's naming department on?

[NotQuiteReal]

I can understand they may have wanted to avoid the Sexium but at least that would be distinctive. Oh, I don't know, I've heard that Sex Sells.

Besides, if they had stuck that naming, we would probably have the Orgasium by now, and who wouldn't buy that!

Re:What are Intel's naming department on?

[a09bdb811a]

Just be thankful they didn't come up with something completely rotten, like, I dunno, "Phenom".

Re:What are Intel's naming department on?

[Sockatume]

"Centrino". They're not just selling a CPU these days, they're selling a platform. Thus, they are marketing a platform too. The fact that CPUs are still named at all is for the benefit of enthusiasts.

Re:What are Intel's naming department on?

[nyctopterus]

I think it has to do with the marketing idea that you concentrate on your primary brand--in this case "Intel". You de-emphasise your sub-brands by giving them dull generic sounding names. I suspect this was the idea behind Apple changing the Powerbook and iBook brands to MacBook Pro and MacBook respectively. Emphasise the "Mac" umbrella brand.

I think it's dumb, but there you go.

Re:What are Intel's naming department on?

[Anonymous Coward]

While I agree their current naming is stupid, Pentium was the next in line after the 80486 (ie. 80586, "Pent", 5, 586, get it?). Calling current chips 586's doesn't make any sense either.

Re:What are Intel's naming department on?

[Anonymous Coward]

When the Pentium 4 came out, it performed worse then the Pentium 3. Towards the middle or end of the Pentium 4's life was the only time AMD clearly trounced Intel. The Pentium brand was largely ruined.

Intel made an attempt to re-image itself and shack off any stigma associated with it's old, we're-the-biggest-so-we-don't-care-if-our-chips-suck selves, changing it's "Intel Inside" logo to "Intel Leap Ahead," dropping Pentium, etc.

Not sure that the new stuff is necessarily better, but I believe that was the reasoning behind the change.

Re:What are Intel's naming department on?

[commodore64_love]

>>>The fact that CPUs are still named at all is for the benefit of enthusiasts.

False. If you bothered to learn your history, you'd know the reason why CPUs have names instead of numbers is because the courts ruled companies cannot trademark numbers. Thus the 80586 became the Pentium and that tradition has continued to today. They cannot just go back to calling them 80986 because of legal reasons.

Re:What are Intel's naming department on?

[mikael]

I can understand they may have wanted to avoid the Sexium but at least that would be distinctive. Core is about as boring as traditional IBM naming.

They could have used 'hexium' or 'hexagonium' but the first name would have sounded like the CPU had a curse on it, and the second would lead users to think their data would disappear.

Re:What are Intel's naming department on?

[baka_vic]

The later Pentium 4 procs were just pushing higher and higher speeds with no regards to the incredible TDP they were now producing - basically it was the epitome of a hot and expensive processor which lost to AMD at that time. The Itanium sinking also doesn't help.

The Core stuff is from a different architecture lineage, and I guess they want to tell people that these processors were something different from the hot P4s.

But you're right, the Core naming is retarded - Core and Core 2, are different families, the solo, duo and quad, are the number of cores...people get confused as to what the '2' means - "is it 2 cores?" etc.

Re:What are Intel's naming department on?

[pohl]

These things happen. Sometimes words, and sometimes even letters, carry hype all by themselves. You wake up one day, and a capital letter X is all the rage. Apple buys NeXT, then you have MacOS X. Some agile methodology gurus want to sell some books and they invent eXtreme Programming. Microsoft, with a marketing department full of ironic hipsters from Seattle, decide that would make an awesome name for Windows XP too. X is everywhere. Fast forward a few years, and the X is out, and the word Core is in. Live on the bleeding edge of the RedHat-derivitive universe with Fedora Core. Apple APIs abound: Core Data, Core Animation, Core Image. The megahertz race gives way to multicore. X sounds cool. All by itself, it is one phoneme away from something that evokes coitus. Core is hip, central, musical: hardcore, grindcore, metalcore. Even the term for the captured state of an abnormally terminated computer program sounds cool: core dump.

Re:What are Intel's naming department on?

[Lemming Mark]

Wait, that's a phenomenal name!

Re:What are Intel's naming department on?

[Anonymous Coward]

Sure they can, but then so can everyone else :D

Re:What are Intel's naming department on?

[koxkoxkox]

No, not the Sexium, more like the Hexium or something like that (yeah, I know it was for the joke, but still ...)

Re:What are Intel's naming department on?

[moosesocks]

Maybe I'm just old but I preferred when "Pentium n" is the new processor and probably better than my "Pentium n-1".

Not to troll, but the Pentium 4 line was a distinct step backwards from the Pentium III in many respects.

Yes, the Pentium 4 was the successor to the Pentium III. However, it wasn't particularly good in terms of performance, power consumption, or cost. AMD clearly had the upper-hand during that period of time.

Of course, Intel, AMD, ATI, and nVidia *all* need to clean up their product lines so that they make sense to consumers!

Re:What are Intel's naming department on?

[Chaos Incarnate]

Core only supports x86, Core 2 x86-64. Dumb scheme, but that's what it mean.s

Re:What are Intel's naming department on?

[Lemming Mark]

Maybe I'm just old but I preferred when "Pentium n" is the new processor and probably better than my "Pentium n-1".

Not to troll, but the Pentium 4 line was a distinct step backwards from the Pentium III in many respects.

Of course, that's not trollish at all, as it's simply true! The Pentium 4s were still, in some sense, "better" as they (eventually! The first Pentium 4s were slower than PIII in some benchmarks, I think) exceeded the raw performance of the PIII. But in terms of microarchitecture, the Pentium 4 was a bit of a dead end. A shame in a way, as it did seem like Intel were thinking toward the future in some respects with that design and then instead ended up wasting a lot of time. The Pentium M, which begat the Core etc was much closer to the PIII than the Pentium 4 was to either of them.

Some good tech from he Pentium 4 has made its way into other Intel chips, though, so it's not a dead loss. It was a disappointing period, though.

Re:What are Intel's naming department on?

[ceoyoyo]

Little problem with your chronology - Apple has been naming their, uh, core APIs with "core" since at least the beginning of OS X, around 1999. Long before the rest of the industry thought it was cool.

Re:What are Intel's naming department on?

[guyminuslife]

"I'm an Apple fanboy, I was doing whatever it is you're doing before it was cool."

Re:What are Intel's naming department on?

[ceoyoyo]

Aw, muffin. Are you jealous? You know, you could just get a Mac if you want one so badly.

The only thing worse than a "fanboy" is a "fanboy" in denial.

Re:What are Intel's naming department on?

[pohl]

True, but not terribly relevant to the trend. CoreFoundation may be that old, but CoreFoo didn't become a leitmotif until Tiger.

Re:What are Intel's naming department on?

[TheRaven64]

CoreGraphics is also that old. There are several other Core* frameworks; they were the C APIs used to implement Cocoa (Objective-C) and Carbon (Classic-style C). When Carbon was deprecated, Core expanded to include any relatively low-level Cocoa APIs. As the grandparent said, the chronology is way off. For example, Microsoft was using things-with-X-in-the-name as a buzzword from around '95 onward (DirectX, ActiveX, XBox, and so on). Apple were using Core in their APIs since 2000.

Re:What are Intel's naming department on?

[Anonymous Coward]

No, the only thing worse than a fanboy is your pathetic attempt to rebut. Go back to grade school with that weak shit.

Re:What are Intel's naming department on?

[pohl]

I'm not sure that bringing up CoreGraphics undermines anything, given that the marketing name for it was "Quartz". You're right that I left out a lot of uses of X. I'm sure one could come up with more.

Re:What are Intel's naming department on?

[Anonymous Coward]

Linux support usually great .. however over at Phoronix, Lynnfield gets its arse handed to it a couple of times by a Phenom II X3 710. Perhaps a little early for Lynnfield on Linux ;)

Phoronix Lynnfield review

Re:What are Intel's naming department on?

[toddestan]

They cannot just go back to calling them 80986 because of legal reasons.

But what if they decided to name the processor the EightOhNineEightSix(tm)?

Re:What are Intel's naming department on?

[toddestan]

How come everyone seems to have forgotten that Intel is still making new Pentium processors? For example:
http://ark.intel.com/Product.aspx?id=42805

Interestingly, they have dropped all that III IV Dual-Core, etc. nonsense, and they are now just "Intel(R) Pentium(R) Processors" again.

Apparently they are budget processors of some kind, which are positioned between the Celerons and the Core 2's. At least I think so - I'm finding Intel's product line rather confusing at the moment too.

AES benchmarks

[a09bdb811a]

These chips have some kind of AES acceleration, called AES-NI.

Are there any benchmarks of this? I use dm-crypt on Linux w/ AES-128 and the throughput is pretty low, about 60MB/sec tops, not as fast as the disk itself.

Re:AES benchmarks

[Neil Hodges]

The VIA Nano has had AES, SHA-1, and SHA-256 acceleration since its inception.

Re:AES benchmarks

[Anonymous Coward]

No, Nehalem/Lynnfield does NOT have AES-NI. Westmere will, but only in 2010. As a heavy 1Gbit+ encrypted network user, 3x AES speedup is enough for me to postpone my purchases until then.

Re:AES benchmarks

[cfalcon]

But do you trust hardware you can't examine to do your encryption for you? This is a bigger problem for the hard drives where we have to trust a controller to have the data stored at some encryption standard (and can't analyze the data ourselves to check that the encryption was applied properly), but a hardware implementation could, for instance, cache your keys somewhere not nearly volatile enough.

Re:AES benchmarks

[Anonymous Coward]

But by the time you get AES-NI, it won't be NI any more because you will have it.
NI stands for New Instruction. Once it's out in the wild, it looses the NI postfix.

Re:AES benchmarks

[Anonymous Coward]

I call FUD. AES-NI instructions will do mathematical operations and will be documented in the development manuals just like any other instruction. And you can examine the output in your debugger like any other program. You could as well ask "But you do trust hardware you can't examine to do your addition for you?"

Re:AES benchmarks

[ceoyoyo]

A regular processor running a software encryption algorithm could cache your keys, data, whatever somewhere not nearly volatile enough.

Nehalem vs. Nehalem

[Demetrius Berman]

Lynnfield is a Nehalem processor, just as the earlier Bloomfield is a Nehalem processor, hence the title to this article makes no sense. The difference is in socket (LGA 1156 vs. LGA 1366), and intended market ... with a couple design differences as well.

Re:Nehalem vs. Nehalem

[Andy Dodd]

By definition, if it isn't a Nehalem die, it's not Nehalem, even if it's just a "tock" variant (die shrunk - see Intel's "tick/tock" roadmap) of Nehalem it's still a different chip design.

In this case, the CPU has significant design differences from a Nehalem CPU. There's a lot more than just removing some pins from the package. The CPU had to be changed significantly (one DDR channel removed, QPI replaced with DMI) in order to allow those pins to be removed.

The removal of QPI in favor of DMI (much slower but simpler/cheaper) is a *significant* difference.

Re:Nehalem vs. Nehalem

[Anonymous Coward]

It has differences from the LGA 1366 desktop and server parts. However both the new Lynnfield and also upcoming mobile parts (Clarksfield?)and the earlier LGA1366 desktop and server chips (Bloomfield being the desktop part, the server chip name eludes me at the moment) are all based on the Nehalem architecture. Westmere will be the die shrink on Nehalem with various parts for different markets, then the next architecture will be Sandy Bridge.

Re:Nehalem vs. Nehalem

[zokier]

There isn't a single Nehalem die/chip. Nehalem refers to the general architecture on which Lynnfield, Bloomfield etc chips are based on.

Re:Nehalem vs. Nehalem

[kill-1]

Exactly. I also don't see why the memory for Lynnfield CPUs should be cheaper as the submitter claims.

Re:Nehalem vs. Nehalem

[Anonymous Coward]

i7 has DMI between the PCIe bridge (Tylersburg) and the "ICH". Since (IIRC) the PCIe bridge is in the CPU on Lynnfield, having DMI matches i7. You have to keep in mind, i7 is a 3-chip solution, and Lynnfield is 2-chip.

Re:Nehalem vs. Nehalem

[Wesley Felter]

Two DIMMs are cheaper than three.

Re:Nehalem vs. Nehalem

[LordKronos]

Unlike some old designs years ago where the memory had to be installed in pairs, Nehalem's triple channel system is designed so that it can also operate in single or dual channel mode. So the number of DIMMs isn't the issue. You can get away with a single DIMM if you wan't. So, installing 2 DIMMs on a Lynnfield system should be no less expensive than installing those 2 DIMMs on a Bloomfield system. The only thing Lynnfield will do to improve the cost is
1) Decrease the cost of the motherboard/chipset/cpu itself (less transistors, pinouts, and traces needed for 2 channels vs. 3 channels).
2) Drive down the cost of the more expensive DDR3 memory as it becomes more commonly used.

Re:Nehalem vs. Nehalem

[cheezedawg]

The removal of QPI in favor of DMI (much slower but simpler/cheaper) is a *significant* difference.

Funny. QPI was not removed or replaced by DMI in any sense of the word. In fact, DMI has existed for at least the last 5 generations of chipsets from Intel. On last year's Nehalem release, QPI connected the CPU with their "IOH" called Tylersburg. Tylersburg was then connected to the ICH with DMI. It was a 3 chip solution. With Lynnfield, the CPU and IOH have been combined to a single chip. The QPI connection still exists, its just internal to the (now combined) chip. Almost everything that used to be north of DMI is still north of it (including PCIe graphics, memory, etc), and everything that used to be south of it is still south (like USB and SATA).

Nevermind Performance per Watt

[aaaaaaargh!]

Okay I know it's important for big server farms, but personally speaking I'm not interested in performance per watt at all. I'm only interested in one thing: Which processor/motherboard/graphics card/OS combination gives me the biggest bang for the bucks for my gaming, compilation, and simulation needs?

Re:Nevermind Performance per Watt

[remmelt]

Bang for the price-of-admittance buck or bang for the total-cost-of-ownership buck?

See, not only server farms need to pay their electricity bills. A modest system can be built in the under 50W range, where gamer systems don't have the 1000W PSU for nothing. There is a huge difference at the end of the month.

Re:Nevermind Performance per Watt

[zokier]

Best bang for buck CPU's are in the sub-$100 segment, maybe some dual-core Athlon, especially if you take the whole system cost into account. Lynnfield is designed for much bigger bang, at much higher cost, with lots of new tech (new socket, "new" memory(ddr3) etc).

Instead of asking best bang/buck, ask for largest bang at certain price point for cpu-mb-ram -combo. Also upgradability/longevity of the platform should be considered. With the new lynnfields you have large probability to have a viable upgrade path in the future, but with s775 its not as certain. Imho choosing CPU has never been more difficult than what it is now with i5 out. On the other hand, luckily most CPU's are powerful enough for average desktop usage. While some other CPU could be 10% faster than other in certain benchmark, in real life usage the differences usually aren't that clear, and performance should be adequate in most situations.

Re:Nevermind Performance per Watt

[the_other_chewey]

Okay I know it's important for big server farms, but personally speaking I'm not interested in performance per watt at all. I'm only interested in one thing: Which processor/motherboard/graphics card/OS combination gives me the biggest bang for the bucks for my gaming, compilation, and simulation needs?

So you don't pay for your electricity consumption?
On the not-so-long term, energy costs outweigh hardware costs.

Re:Nevermind Performance per Watt

[Anonymous Coward]

I don't have to worry about that here in the basement, because my mom always pays the electricity bill and she always brings pizzas and coke down here when I knock on the door. On my 30th birthday, I even got a cake from her!

Re:Nevermind Performance per Watt

[oakgrove]

Indeed, the system I'm posting this on now has a Core2Duo e7400 proc at 2.8 GHz, 4 GB of RAM, a 500 GB HDD, and integrated nVidia graphics. While running Ubuntu 9.10, with compiz effects on, according to my Kill-A-Watt meter, it pulls a cool 46 watts at the plug just browsing around. I can't complain.

Re:Nevermind Performance per Watt

[Belial6]

I calculated the payback of my last round of upgrades (two years ago) since the new systems used noticeably less power than the old systems, as well as suspend working on the new systems. My computer: 6 months to recoup the entire cost in electricity bill savings.
My wife's computer: 3 months to recoup the entire cost in electricity bill savings.
My son's computer: 4 months to recoup the entire cost in electricity bill savings.

Re:Nevermind Performance per Watt

[drsmithy]

A modest system can be built in the under 50W range, where gamer systems don't have the 1000W PSU for nothing.Mostly they do. You need an insanely powerful system to come even close to needing a 1kW PSU. A quite capable gaming system will comfortably function with a PSU of half that capacity.

Re:Nevermind Performance per Watt

[toddestan]

I pay 8 cents a kilowatt-hour. My current computer idles at around 100W. If I was to reduce my computer's power use by 50W, that would save me about $35 a year in electricity in a year. And those numbers are assuming I leave the computer on 24/7, which I don't. I actually only have it on about 1/3 of the time, so that cuts the savings down to about $12 a year.

So basically, I don't see how energy costs outweigh the hardware costs. Certainly lower power is better, all other things being equal - but if the lower power parts are more expensive then I'm probably better off just eating the additional electricity costs. Perhaps the real lesson is to turn the computer off when you're not using it to cut energy costs?

Obvious

[leromarinvit]

Translation: High end CPU sucks power and is expensive. News at 11.

Re:Obvious

[Anonymous Coward]

Yeah, but its cheaper and sucks less power than even higher end CPU's.

Re:Obvious

[leromarinvit]

Which (Intel) CPUs would you describe as being even higher end than i7 Nehalems? Obviously I was referring to them, not the Lynnfield ones.

Off topic, but vaguely related

[DoofusOfDeath]

Since we're talking about different Intel chip lines...

I've been laptop shopping, and I've seen two different kinds mobile processor names: P* vs. T*. For example, P8600 and T9600.

Could someone explain to me the significance of 'T' vs. 'P'?

Re:Off topic, but vaguely related

[Anonymous Coward]

There's several differences. I'm not a pro at this, but I did some research before buying my laptop.
You can look at ark.intel.com for details, but one big thing I noticed is that the P processors are 25W and T are 35W (Should mean longer battery life for P).
I also think that P is for Penryn, and that this is a new architecture somehow.

Re:Off topic, but vaguely related

[hoffmanbike]

P is newer with a 1066mhz FSB and roughly comparable clock and cahce to the T series. The T series has a 800MHz FSB. use this to find the true differences: http://ark.intel.com/

Re:Off topic, but vaguely related

[Skuto]

Both are Penryn. P vs T is indeed the power efficiency.

Re:Off topic, but vaguely related

[Anonymous Coward]

grep for the model number at http://en.wikipedia.org/wiki/List_of_Intel_Core_2_microprocessors

Re:Off topic, but vaguely related

[ciroknight]

The "T" vs "P" originally came from the name of the socket the processor fit in, but quickly became corrupted by Intel marketeering. The modern guideline is that the "T" chips tend to be Mobile Core 2 Duo ("Merom") which are 65-nm chips and run hotter, and that the "P" chips are all Mobile Core 2 ("Penryn") which are 45nm and run 10W+ cooler. The latter, obviously, get significantly better battery life and come with all of the bells-and-whistles enabled by default (VT-x, EIST, faster FSB at 1066 MT/s, etc).

If you're going for the most battery life you can squeeze out of the machine, it's worth the $50 upgrade. Otherwise you likely won't notice a difference in overall performance unless you're a heavy movie editor or exceptionally heavy gamer.

LGA 1366 dead now?

[Ultronator]

Does this mean they're done making core i7, socket LGA 1366 cpus? Oh well, so much for my hopes of someday upgrading mine. LGA 1366 was a nice 9-10 month run while it lasted.

Re:LGA 1366 dead now?

[Anonymous Coward]

You should be OK for a while, Westmeres ("core I9", 6 cores) will also use socket 1366 when those come out in a year or so. They aren't abandoning 1366, but it's targeted to high end CPUs, while the lower count sockets are for midrange.

Re:LGA 1366 dead now?

[afidel]

No, Gulftown is slated for socket 1366 next year. Of course it's list price is $1500 so whether you will want to upgrade to it is questionable.

Re:LGA 1366 dead now?

[zokier]

I think LGA 1366 will live on as the high-end platform it always has been. LGA 1156 will be middle-end, and 775 in low-end and legacy.

Re:LGA 1366 dead now?

[pete-wilko]

Ya... well I outlaid on a core i7-920 setup 3 months ago, justifying the outlay cost as I thought there would be a decent lifespan in the architecture, so should be easy to upgrade for a couple of years on that platform. D'oh!

Re:LGA 1366 dead now?

[Ultronator]

Glad to hear my machine still has options for gaming in the future.

Re:LGA 1366 dead now?

[LordKronos]

http://anandtech.com/cpuchipsets/showdoc.aspx?i=3634&p=7

Intel told me something interesting when I was out in LA earlier this summer: it takes at least 3 cores to fully saturate Lynnfield's dual-channel DDR3-1333 memory bus. That's three cores all working on memory bandwidth intensive threads at the same time. That's a pretty stiff requirement. In the vast, vast majority of situations Lynnfield's dual channel DDR3 memory controller won't hurt it.

Move up to 6 or 8 core designs and a third memory channel is necessary, and that's why we'll see those processors debut exclusively on LGA-1366 platforms. In fact, X58 motherboards will only need a BIOS update to work with the 6-core 32nm Gulftown processor next year. P55 looks like it'll be limited to four cores and below.

Re:LGA 1366 dead now?

[LordKronos]

Yeah, I know what you mean. I even did my research and knew about the LGA-1156 socket coming down the line, but my understanding was that it would only be for the core i5 chips. I had no idea that the i7 line was going to switch too. However, at least we have >4 core models to look forward to sometime in the future. However, those will probably maintain a premium price. I was originally thinking that in 3 years there would be a lowend (but significantly faster that today's lowend) core i7 I could throw on there for $100 or so. Now I think there's a good chance that won't be the case.

Re:LGA 1366 dead now?

[pete-wilko]

yeah - in a sense i think the 920 was the problem, as you got pretty good bang for your buck (on 1136), almost too much when compared price performance against the 940/50. had heard rumors that it would be discontinued in the new lineup in the second half of this year... turns out this was the announcement they'd planned. but yeah, i was exactly the same as yourself.

More in-depth coverage here

[MojoKid]

Power consumption, H.264 encoding, file compression and image manipulation tested here, as well as Intel's on-chip PCI-Express links in multi-GPU setups: http://hothardware.com/Articles/Intel-Core-i5-and-i7-Processors-and-P55-Chipset/

Fishy numbers?

[NameIsDavid]

Something seems strange with these numbers. My i7 920 system, overclocked to 3.2GHz, draws 95W at idle (monitor excluded). This is based on the APC utility that monitors my UPS unit into which my computer is plugged. This is with 6GB of DDR3-1600 RAM and a silent ATI 4670 card. Now, my GPU draws much less than the test system. However, the 60W difference between Nehalem and Lynnfield seems odd since that would means that my system would drop to 35W idle with Lynnfield!

Tests

[zokier]

Lots and lots of tests and bechmarks. Looking good.

Intel 'Lynnfield' Core i5 750 and Core i7 870 Performance Testing Introduction :: TweakTown
Intel Core i5 and Core i7: Lynnfield CPUs reviewed - Intel, Core i5, Core i-750, Core i7, Core i7-860, Core i7-870, Lynnfield, Bloomfield, AMD Phenom II X4 - PC Games Hardware
Core i5 750 - Core i7 860 and 870 processor review
HEXUS.net - Review :: Intel Lynnfield Core i5 750, Core i7 860 and Core i7 870 CPU review: bombarding the mid-range : Page - 1/12
Legion Hardware
Intel Core i5 750 & i7 870 Review - Page 1 - The Next Nehalem-based CPU lineup
PC Perspective - Intel Lynnfield Core i7-870 and Core i5-750 Processor Review
Introduction - Intel Lynnfield Core i5 and Core i7 Processors | [H]ard|OCP
In Theory: How Does Lynnfield's On-Die PCI Express Affect Gaming? : Introduction - Review Tom's Hardware
AnandTech: Intel's Core i7 870 & i5 750, Lynnfield: Harder, Better, Faster Stronger[/QUOTE]
Intel Core i5 750 Core i7 870 Review - Overclockers Club
Techgage - Intel Core i7-870 & i5-750 - Nehalem for the Mainstream
Core i5-750 and Core i7-870 Processors Review | Hardware Secrets
Intel Core i5 750 Processor Review - TechSpot News
Intel Core i5 And Core i7: Intel?s Mainstream Magnum Opus : Introduction - Review Tom's Hardware
Intel Lynnfield Core i5-750 & Core i7-870 Processor Review
Intel's Core i5-750 and Core i7-870 processors - The Tech Report - Page 1
bit-tech.net | Review - Intel Core i5 and Core i7 Lynnfield review
bit-tech.net | Feature - Intel Lynnfield: Details and Architecture
Intel Core i5, Core i7 800 Processors and P55 Express - HotHardware
Intel Core i5-750 Processor BX80605I5750 | Intel Core i5-750,BX80605I5750,Lynnfield,LGA1156,CPU,Proocessor, Intel Core i5-750 Lynnfield LGA1156 CPU Benchmark Performance Test Processor Review | Benchmark Reviews Performance Tests
Intel Core i7 870/Core i5 750/P55 Express chipset Review :: Introduction :: Motherboards.org

I mind Performance per Watt

[fnj]

How nice for you not have a care in the world for energy consumption. As one who actually has utility bills to pay, my electric bill is around 20 cents per kwh, or $1752 per kw-year. If I can save 200 watts of continuous consumption, that's $350.40 less per year for me to pay. Also, I don't play games, so I don't give a flying fuck for graphics performance after the first couple of notches. The same goes for cars. I could drive a car that goes 200 mph and 0 to 60 in 3 seconds, but mpg is important to me. I get 44 mpg average.

Re:I mind Performance per Watt

[eabrek]

How about a processor with 0 power consumption?
Of course, it is 0 performance...

Infinite performance per watt!

Performance per watt is a useless measure. People should use lowest power in a desired performance envelope. But that is not a single number you can use in a sound bite...

Re:I mind Performance per Watt

[edmudama]

Depends heavily on where you live.

Residential power here is $0.055 per kWh with no hourly, daily or seasonal restrictions or pricing... roughly 25% of what you pay. There are almost no cases where paying extra for a green technology locally will recover the costs.

Re:I mind Performance per Watt

[Anonymous Coward]

A concept that seems to be lost on most people is that you can care regardless of the price.

Re:I mind Performance per Watt

[The End Of Days]

A concept that seems to be lost on you is that people don't have to care.

Re:I mind Performance per Watt

[fnj]

Did you flunk basic math? Zero divided by zero is undefined, not infinity.

Re:I mind Performance per Watt

[Anonymous Coward]

can != have to

I just happen to prefer people whose values aren't limited to the contents of their wallets. Once your basic needs are covert you can afford to look a little bit further than that. So you might also let your opinion about, for instance, the merits of green technology be influenced by other factors than just the price you pay.

Re:I mind Performance per Watt

[Anonymous Coward]

covert -> covered

Re:I mind Performance per Watt

[eabrek]

Wow, relax people... troll? fail at math?

The point is, optimizing for performance per watt is the same as minimizing performance. Everything that adds performance costs power. Anything that reduces power with no effect on performance is going to get done in the high performance design...

Re:I mind Performance per Watt

[Anonymous Coward]

here energy is dirt cheap
and since it is so cheap it is included in my rent, please dont tell the landlord about ours 6 pc !

Re:I mind Performance per Watt

[fnj]

This is just wrong. Optimizing for performance per watt is not minimizing performance. 2.1 times the performance for 2 times the watts is optimizing for performance per watt.

Re:I mind Performance per Watt

[eabrek]

The problem is, you cannot process more instructions in less time without using more power. A 210% speed up will use at least 210% more power.

The rule of thumb in CPU design is 3 power for 1 performance - that is simple voltage/frequency scaling. If your architectural idea is worse than that, start over. Most teams today are using a 2 for 1 rule (1% perf will cost 2% power). That is why many-core (which promises 1 for 1) is so attractive to managers and non-technical people...

Re:Fishy numbers?

[0123456]

However, the 60W difference between Nehalem and Lynnfield seems odd since that would means that my system would drop to 35W idle with Lynnfield!

Isn't the claimed 60W difference under load? The idle figures I see in the Tom's Hardware article only show about 30W difference.

However, from what I've read of the article, the new CPUs will pretty much shut down cores which aren't being used and lose an entire north-bridge chip that currently takes about 20W by itself.

CPU performance articles are quaint

[Junks Jerzey]

While I hardly think 640K is enough for anyone, this story strikes me as an odd curiosity, certainly not something worthy of the Slashdot front page. In the age of netbooks, the iPhone, and notebook computing, does the ultimate pinnacle of performance even matter any more? Even with desktops, I just bought a $600 Dell that's so far beyond anything I can throw at it (with the usual exception of those few extraordinarily demanding GPU-bound games that need $400 video cards just to scrape by), that CPU performance is no longer on my radar. And it's not even an i7; it's the last revision of the Core 2 Duo.

Re:CPU performance articles are quaint

[TheRaven64]

That's why the comparison is performance per Watt, not performance, although good job posting before reading to the end of the headline. Performance per Watt is still very much relevant. Even if the chip is fast enough already, a new version that has the same performance in half the power means more battery life.

who cares if you're not talking server chips?

[Anonymous Coward]

Honestly, when you build a computer how many people are going? "Well this chip provides more performance per watt than that one, thus I'll buy it." All that matters to the person getting the pc 99% of the time is performance per dollar spent. In a datacenter power matters, at home.. not so much.

so?

[toby]

Why compare 2 Intel products? Where's the comparison with AMD, or - in a perfect world - low-power, high-threads SPARC?

Intel == destructive monopoly, quit playing into their hands. Up next: Worthless comparisons of Vista and W7...

Re:so?

[Microlith]

There already have been comparisons against AMD, and AMD remains competitive with Intel's low end. Also, no one cares about the latest high-threads SPARC because you can't get a modern one for less than several thousand dollars. If you're in line for one of those you're looking at the XEON line which outright smokes AMD's Opterons, and probably running software that won't run on SPARC.

Re:so?

[DaMattster]

I would say AMD and Intel are neck and neck performance-wise. They both alternate years of being in the lead. Intel does not really have a monopoly. There are really only two manufacturers in the CPU market.

Re:so?

[moosesocks]

No. That's not quite right.

AMD is competitive with Intel's mainstream line.

Intel currently has the upper hand for the ultra-high-end, and ultra-low-end (ie. Atom). AMD's mainstream line is about on par with Intel, and has a considerable cost advantage.

Re:so?

[0123456]

I would say AMD and Intel are neck and neck performance-wise.

If by 'neck and neck' you mean 'often slaughtered by a CPU running 500MHz slower', yes. Which wouldn't be a problem for AMD per se if AMD's slower CPUs were significantly cheaper to build than Intel's; most of the revenue is in the mid-range even if much of the profit is at the high end.

They both alternate years of being in the lead.

AMD get ahead when Intel screw up. Currently Intel are not screwing up and don't seem likely to screw up for some time to come.

There are really only two manufacturers in the CPU market.

That'll be news to ARM (though I suppose that if you're pedantic enough you could claim that they don't actually _manufacture_ CPUs).

Re:so?

[0123456]

AMD's mainstream line is about on par with Intel, and has a considerable cost advantage.

I'd say it's more than AMD's high-end is currently on par with Intel's mainstream line, forcing AMD to sell for a much lower price than Intel's high-end to compete.

Antitrust

[toby]

"Truth is that in many markets it's now a grand choice of Intel, Intel and Intel."

Which helps explain why AMD has an antitrust suit against Intel.

More Reviews

[tab_b]

Here are a few more reviews for today: The Tech Report, Phoronix, AnandTech, X-bit labs, and Benchmark Reviews. It's all enough to make your eyes bleed. There's a list for the Core i7 870 at 0x6877.com

What about low end processors?

[DaMattster]

This is slightly off-topic but I really like the newer Celeron processors. It is almost an overclocker's dream as you have a fair amount of room to play with. The new Celeron E3000 series based on the Wolfdale architecture looks especially promising and has 1mb of cache. I bought the Celeron E3300 and I am planning to pump it to 3GHZ. My guess is that a simple increase of 500MHZ should be interesting enough without putting undue stress on it. From there, I'd like to see how it compares with Dual Core Xeon. Oh, it will still get smoked by the Core i7 and its ilk but it will be powerful enough to run a FreeBSD-based webserver and more. I have a Celeron 560 laptop with 2GB and it cruises.

Anandtech's version

[klui]

http://anandtech.com/printarticle.aspx?i=3634

Half of US power is generated by coal

[symbolset]

We all breathe. The fact that power consumption is coming down on mainstream PCs is of huge benefit to people generally. It also means more technology uptake in places where power is less available or more expensive, and that's good for the economy. Across a hundred million units bringing power consumption down by 10 Watts is a Gigawatt, or roughly the power delivered by The last US nuclear power plant to come online. Watts matter at home too.

Dog Bites Man

[Trutane]

A next-generation chip bests its predecessor. Yawn.

--

Even those who arrange and design shrubberies are under considerable economic stress in this period in history.

It matters to me.....

[BLKMGK]

I have a 4ghz C2D (overclocked). My x.264 encoding of BD rips takes as long as 18 hours - I keep the quality high. I am looking for more cores and more clock speed if I can get it in order to lower encoding time. I already know more cores will benefit me since a slightly slower 4 core I've "drag tested" has beaten my C2D by a couple of hours but runs hot as hell. I've been looking to move to an I-7 920 clocked to 4ghz but if I can get one of these 800 series CPUs for less, overclock it to nearly the same performance level as a 920, and use less power I'm VERY interested.
.
Since the two CPUs I'm considering use different damned sockets these tests are important to me.
.
Compiling code is another app where these CPU can come in handy. I frequently compile a fairly complex app in Linux (still more machines!) on both a C2D and a Atom 330 - the C2D STOMPS the Atom as you might imagine. An I-7 with additional cores would likewise stomp the C2D and if my compiled HTPC app could benefit I'd consider it but the Atom is good enough (thank you VDPAU!). The more complex the project the more potential time saved compiling obviously. I'm betting that some people do this activity often for a living? Distcc farm perhaps? Rendering maybe? Weirdly the Atom shows up as 4 cores in Ubuntu and I can actually task all 4, I guess the 330 got Hyperthreading? The C2D still wins that race though no matter what I do. (duh) The Atom is overclocked to over 2.1ghz too BTW :-)
.
So yeah - CPU performance for some folks is still a big deal. Your $600 Dell would be hating life for hours on end processing data like my desktop does. For gaming and lighter weight apps I'd agree that such power isn't needed. I own a netbook and a few other laptops that suit me fine day to day. My NAS all run underclocked Celeron as well. the right tool for the right job, if your Dell does the job be happy and enjoy it but please do not be so narrow minded as to not understand others need a different tool for their activities...

Morons abound.

[ciroknight]

You're both right, which is why the argument is stupid. There's the Nehalem microarchitecture which refers to the chip's construction, and there's the actual Nehalem chip, which was the first chip built with the Nehalem architecture.

It's like arguing whether or not you can call the USS Greenville a "Los Angeles boat".

Re:Morons abound.

[LordKronos]

No, they aren't both right. Nehalem is an arcitecture and nothing more. There is no Nehalem chip.

http://www.intel.com/technology/architecture-silicon/next-gen/

Intel® Microarchitecture, Codenamed Nehalem

Intel's latest-generation microarchitecture, first exemplified as the Intel® Coreâ i7 processor, represents the next step in faster, multi-core technology that intelligently maximized performance to match your workload

It was first exemplified as the Core i7, and as we know the Core i7 debuted with 3 models...the 920, 940, and 965 (extreme), all 3 of which were Bloomfield designs. While I have no doubt some websites probably incorrectly reported it as the "Nehalem chip" before its release, that's all it was....incorrect reporting.

Performance per dollar is a stupid metric

[JustNiz]

Performance per dollar is a stupid metric.
A crappy old pentium 3 from ebay wins every time, but is that really what you want to buy and use?

I don't think that's entirely true

[Chuck Chunder]

He said "Aside from enthusiasts who just like following the stuff".

I might have a passing interest in the progress of processor technology but I'm actually specifically interested at the moment because I'm on the look out for a new laptop. For me that means looking at a Toshiba, a Lenovo (and Dell, Fujitsu etc) and trying to make a judgment about what advantages and disadvantages each has and comparing them to see what makes the most sense for me.

I've found Intel's processor designations mind boggling. They do have a pretty good comparison tool though, which can help clear things up once you've narrowed it down to a few options.

Re:I don't think that's entirely true

[apoc.famine]

But there again you're seeking information far outside that which 90% of the people buying a new computer look for.
 
Intel doesn't give a shit about how easy it is for you or anyone else to compare processors. It truly doesn't matter.
 
Dell will buy 100k processors of some generic type. It doesn't matter if they are the same, worse or better than what the're selling now. They'll slap them into a system, mark the entire thing up, and slap the specs on their website.
 
Joe Average will come along, pick his price range, and buy that computer.
 
That's it.
 
Anything you or I or anyone else wants to know about that chip doesn't matter. It makes no difference to Intel or Dell. Those of us seeking information are a cork bobbing in the ocean of "I have this much to spend on a computer. Hey - look! There's one which matches that price. I'm going to buy it!"
 
Computers aren't sold on specs to 95% of customers. They're sold on form-factor and pricepoint. Intel can have a dozen lines and it doesn't matter for the most part how well they perform. Nobody, for the most part, cares.

Re:Fishy numbers?

[drsmithy]

My i7 920 system, overclocked to 3.2GHz, draws 95W at idle (monitor excluded). This is based on the APC utility that monitors my UPS unit into which my computer is plugged.

That seems extremely high for an idle system. You should check it with something like a Kill-a-watt.

Are you sure you don't have a few instances of Prime95 or something running in the background ? :)

Re:Fishy numbers?

[pete-wilko]

I've got a similar setup (and hence am now eating bricks), the anandtech article I think sheds some light on this: http://anandtech.com/printarticle.aspx?i=3634 - basically this new line-up actually has some smarts in it to properly power down non-utilized cores, whilst being able to operate within the thermal profile to run individual cores at very high speeds. Sounds pretty nifty, for us basically we'll have to wait and see how affordable the gulfstrem line of processors coming out next year is.

Long story short - 1366 is now good for 4+ cores, anything else and this new offering takes the cake (the gulfstream coming out next year is meant to be 4+ cores... so maybe there's hope for the platform not being obsolete yet...)