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

[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

[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

[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

[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

[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

[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.

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?

[tylerni7]

Lame, my processor is restricted to the reals.

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

[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

[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.

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.

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?

[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]

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.

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.

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

[zokier]

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

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.

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'?

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!

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.

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...

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