Sign in with
Sign up | Sign in
Your question
Closed

ARM Vs. x86: The Secret Behind Intel Atom's Efficiency

Last response: in Reviews comments
Share
a b Î Nvidia
a b å Intel
December 24, 2012 8:06:37 PM

Excellent! Was wondering about this for some time. Also made the mistake of thinking Intel was behind in the mobile space... Well done Toms.
Score
17
December 24, 2012 8:33:39 PM

I'll be very interested to read the Cortex A15 follow up. From what I gather, if compared on the same lithography the A15 core is much larger than the A9, which likely means more power, all else being equal. It brings performance up to and sometimes over the prior generation Atom, but I wonder what power requirement sacrifices were made, if any.

I'm thinking in the coming years, Intel vs ARM will become a more interesting battle than Intel vs AMD.
Score
3
Related resources
Can't find your answer ? Ask !
December 24, 2012 8:42:08 PM

@tipoo, we're not going to hang our hat on it just yet (until we run the numbers ourselves), but A15 runs hot, which is what we hint at in our article.
Score
3
December 24, 2012 8:43:21 PM

tipoo I'm thinking in the coming years, Intel vs ARM will become a more interesting battle than Intel vs AMD.


I was until I saw the numbers. Intel spent $8.4 billion in 2011 ($6.6 billion in 2010 and $5.7 billion in 2009) on R&D - http://www.intc.com/intelAR2011/business/research/ - while ARM isn't worth $1b. It may take a few years but Intel are seriously massive, they'll soon be the go to guys for mobile. Plus (according to rumours) with Haswell focusing on power saving, it could be a big leap forward.
Score
11
December 24, 2012 8:45:38 PM

AlanDang@tipoo, we're not going to hang our hat on it just yet (until we run the numbers ourselves), but A15 runs hot, which is what we hint at in our article.


I'm guessing the same thing. So far we've only seen it in a tablet (Nexus 10), and even that with its 10 inch tablet sized battery didn't last particularly long. ARM has the distribution advantage right now, but I think once Intel gets its foot in the door it will be the 900lb gorilla in this market as well.
Score
5
Anonymous
December 24, 2012 8:57:22 PM

ARM isn't just ARM holdings, it's nVidia, Samsung, and Qualcomm just to name a few of the heavy hitters. And it should also be noted that even if your SoC is better, if the OEM integrating it is incompetent, it won't matter. I'm certain more Surface RT devices have been sold compared to the Acer W500 because it had better availability, a stronger marketing campaign, and overall is a far more solid device. Don't miss the forest for the trees.
Score
0
December 24, 2012 8:59:09 PM

blubbeyI was until I saw the numbers. Intel spent $8.4 billion in 2011 ($6.6 billion in 2010 and $5.7 billion in 2009) on R&D - http://www.intc.com/intelAR2011/business/research/ - while ARM isn't worth $1b. It may take a few years but Intel are seriously massive, they'll soon be the go to guys for mobile. Plus (according to rumours) with Haswell focusing on power saving, it could be a big leap forward.


Didn't Qualcomm alone overtake AMD as a chipmaker? The thing about ARM is that anyone can get a licence for the ISA and build a custom core around it, like Krait.
Score
4
December 24, 2012 8:59:29 PM

And the Chromebook with an A15. We're looking at several other SoC's but what's empowering is how simple math can help you look at efficiency.

But it's Christmas Eve, and sometimes there are more important things than running benchmarks and soldering wires to SoCs. ;) 
Score
14
December 24, 2012 9:14:14 PM

Measurements taken when running Windows. Arm and Intel chips both require different coding styles to make them preform. The way windows has been coded favours the Intel Arch. Intel chips like to lean heavy on their cache, Arm chips prefer code to do as much processing on it's data as it can using all it's registers before moving on.

Bit like comparing a 4x4 and a sports car in a muddy field.

Score
-5
December 24, 2012 9:23:25 PM

nvidia tegra 3 is a piece of overhyped garbage.
Score
-3
December 24, 2012 9:47:47 PM

of all the ARM types, u took one of the weakest one? I like to see the numbers vs Qualcomm krait and Apple's A6
Score
16
December 24, 2012 9:51:53 PM

down with arm! if i ever do use a tablet (wont be anytime in the foreseeable future) i want to be able to run x86 programs.
Score
-5
December 24, 2012 10:32:41 PM

just another article showing how much of a piece of crap tegra is.
Score
8
December 24, 2012 10:55:57 PM

I agree that it will be interesting to see ARM based tablets/phones to battle with upcoming Intel Atom cpu's in tablets (and phones?). Intel's really stepping up their game and it'll just force ARM to step up theirs.

We're already at quad-core ARM SoC's running at up to 1.7ghz. My understanding is that they won't be able to run too much faster anymore and will instead have to do more work per cycle. I think that's where Atom has the advantage in this case, which is why they can run slower but still accomplish the same amount of work with less power (but not faster).

If only this would translate into the Desktop/Laptop sector. We haven't had huge performance gains in the longest time. I miss the old days of Athlon 64...
Score
3
Anonymous
December 24, 2012 11:06:41 PM

Once again, Intel manages to sort of tie with a middle-of-the-pack last-generation ARM CPU, well done.

Tegra 3 is junk and everybody knows it, bring on the Samsung Chromebook or a comparable tablet running an A15, and run Linux on them. Or rather than just waiting for Tom's to do it, check out the benchmarks on Phoronix that show the A15 eating the Atom for breakfast...
Score
8
December 24, 2012 11:30:19 PM

@jessica. Phoronix tested the older Atom which we note has the same name but nowhere near the same performance or energy efficiency. Also keep in mind that Phoronix didn't actually try to run their benchmarks off the battery. A15 is fast, but the performance isn't free.

The advertised battery life on a Chromebook is very telling. It gets 6.5 hours of web browsing with a 30 watt-hour battery, so it's drawing 4.62W under that workload. This Atom tablet is drawing closer to 3.11W under an average web browsing workload and the Tegra 3 is drawing about 3.9W. We're looking at Snapdragon next and then A15, but I think everyone's going to be surprised once they look at the actual numbers when you're running apps other than Phoronix's benchmarks.
Score
5
Anonymous
December 24, 2012 11:35:29 PM

Alan: But the new mobile Atom is slower than the old netbook Atom, which is why you never see them in the same benchmark. Those power savings weren't free, and yet the A15 still destroys either Atom.
Score
2
December 25, 2012 12:35:36 AM

Geekbench would argue differently.

Atom Z530 (the "Z-class" Atom Phoronix tested).
http://browser.primatelabs.com/geekbench2/search?utf8=%...

Atom D525 (the fastest netbook Atom Phoronix tested, which beats Exynos Dual in some applications)
http://browser.primatelabs.com/geekbench2/search?utf8=%...

http://www.phoronix.com/scan.php?page=article&item=sams...

Atom Z2760
http://browser.primatelabs.com/geekbench2/search?utf8=%...

The Atom D525 had a TDP of 15W in 2010. The Atom Z2760 has the same performance with a TDP of
Score
7
December 25, 2012 12:37:05 AM

less than 2W. The Exynos has a TDP of 4W.

If the A15 is 2x the power consumption, is it consistently 2x as fast?
Score
8
Anonymous
December 25, 2012 12:51:04 AM

Jessica... you couldn't be more wrong. Good grief!
Score
-4
December 25, 2012 2:01:26 AM

What is the price of an Intel SoC compared to ARM with comparable perfomance? And take into account Rockchip, MediaTek and all the chinese manufacturers that are having a field day with ARM's licensing policies.

What is happening with Android support? Windows is going to be less relevant on a tablet oriented enviroment and OEMs love the FREE nature of Android.

On 2W processing power I think Android benchmarks are more relevant.
Score
2
December 25, 2012 2:17:07 AM

Quote:
the following numbers come from Intel's team, not the Tom's Hardware lab


Until Tom's is able to run there own test's I don't trust the numbers. Not that I don't trust Intel but we've seen in the past how ridiculously easy it is to tilt benchmarks in your favor. Just remember some creative editing would make it look like AMD's CPUs kill Intel's, which clearly they don't.
Score
3
December 25, 2012 2:36:32 AM

similar articles on Toms and Anandtech, praising Intel over ARM ?
I am usually not into conspiracy theories. But this looks suspicious. Sorry, Chris et al, but i smell a rat.
Score
-2
December 25, 2012 2:42:36 AM

More and more it's looking like Jim Keller is pretty much THE memory controller guru. Considering the lower-than-optimal memory bandwidth modern AMD processors have (especially the APUs) and that the memory controller was the primary source of performance in the Athlon 64, it's no surprise that AMD's re-hired this guy.
Score
1
December 25, 2012 3:07:33 AM

mayankleoboy1similar articles on Toms and Anandtech, praising Intel over ARM ?I am usually not into conspiracy theories. But this looks suspicious. Sorry, Chris et al, but i smell a rat.


This was bound to happen, since I don't think Intel would call different report groups in different times. That would say 'special treatment' to some sites ;D

Anyway, interesting numbers, to say the least. Although, Tegra 3 is getting long in the tooth. Krait or Exynos 5 or even Apple's A6 would have been better to test against.

Also, remember these things (tablets and phones) spend most of their life not using 100% cpu number-crunching goodness, just power saving features and a few cycles to load Facebook, haha.

With that in mind, the least power used while on max load, the better, since a few seconds less won't make a difference. Battery life does make the difference here. Lowest idle and fast P state changes is what Intel should aim to, here.

Cheers!
Score
3
Anonymous
December 25, 2012 3:19:44 AM

Great a muscle car and a volkswagon use little gas at idle, but run them at full load, and it is a different story! Sure, Intel has done much to get its power hungry CPUs sipping little power at idle, with power
gating, but start up all those grains and see how they compare to ARM. I think this article is more an example of Intel's marking than an actual power usage comparison! TRUE power usage comparisons have both power usage at idle and power usage at full load, so if I don't use my phone Intel has great power gating, but the real test is when the phone is in full use mode, and there is no power plug near by! Is Intel going to provide me with the extra battery for free!
Score
1
Anonymous
December 25, 2012 3:27:54 AM

Correction Marketing not Marking, for IntelLowPowerAtStandStillBUT POST!
Score
0
Anonymous
December 25, 2012 4:04:09 AM

But standby power/time is important. How does atom/win8 handle always on receiving emails and notifications, etc?
Score
0
December 25, 2012 4:18:34 AM

mayankleoboy1similar articles on Toms and Anandtech, praising Intel over ARM ?I am usually not into conspiracy theories. But this looks suspicious. Sorry, Chris et al, but i smell a rat.

You should know us better than that, friend. All of the disclosures are there, and we did our own cross-analysis based on the numbers we have in-house. The fact of the matter is that we don't have the equipment that Intel does to get as granular as they're getting. We never trust benchmark results blindly, but we've done as much due diligence as possible to make sure the data we're presenting here is as accurate as possible.
Hope everyone's having a Merry Christmas. Back to prepping the little guy's first tricycle!
Best,
Chris
Score
7
December 25, 2012 4:20:45 AM

Thanks for enjoying another article, sponsored by "Inel".
Score
-6
December 25, 2012 6:00:04 AM

x86 is an architecture with long years of r&d an billions to spend on it. On the other hand ARM is a tech that is getting some attention now
Score
-1
December 25, 2012 7:27:31 AM

I just watched a 2 hour movie (720p mvk) using my Samsung ATIV XE-500T (Atom Z2760) through the miniHDMI to my 47" TV. The battery still shows 80% after the movie was done, and all this with no stutters, no heat, no fan noise. It's just awesome as a portable secondary device.
Score
2
December 25, 2012 10:26:14 AM

Quote:
Even at idle, though, the Acer W510 uses less power than Microsoft's Surface. In this scenario, the CPU cores aren't cranking away, but the graphics core is still refreshing the screen and reading from memory. This constant reading taxes the memory controller, and is one reason why the Atom maintains low power consumption.


IIRC, win8 had redesigned this to ensure that at idle, screen wouldn't refresh at all. So GPU would use lesser power. I dont know if that was limited to x86 arch only, or was for all x86 and ARM too.
Score
-1
December 25, 2012 2:45:04 PM

This is interesting but one can't rule out that the lower efficiency of the ARM tablet is chalked up to "Windows on Arm" and not the SOC.
Score
-1
December 25, 2012 3:02:36 PM

@mayankleboy1. That's right, but there's two parts to the problem.

Historically, if nothing was changing, the CPU would still send the GPU the same image over and over and the GPU would process that image over and over before putting it into the framebuffer for the panel to read. For panel self refresh, the idea is to put RAM on the panel itself and then have the panel loop from that extra bit of RAM, allowing the CPU and GPU to sleep. Cool, but it adds cost to the panel.

In a hybrid version of that, you can let the CPU/GPU sleep, but the panel is still looping memory from the framebuffer instead of calculating everything. This cuts down on the cost of the panel because you can use the regular system RAM instead of having extra circuitry on the panel, and you can still put the CPU and GPU in low power states. However, since the RAM is still part of the normal system RAM, you still need the normal system memory controller running.

@hixbot. That's 100% true. If you look at Transformer Prime it has similar battery life to the Surface with a smaller battery. That mans it's more efficient. By our quick math, Tegra3+Android = Atom+Windows 8. Part of the Android's increased performance comes from better 4+1 use, and part of it come from the fact that Android is a more stripped down OS without the same real multitasking that Windows 8 has, and panel differences.

All in all, I think the messge that Chris and I wanted to put out was that a year ago, no one thought Intel could come close to ARM. At the end of 2012, we can see an actual shipping Win8 tablet beating an actual shipping Win8RT tablet, bot which were released at the end of October. That's a pretty impressive feat.
Score
3
December 25, 2012 7:21:17 PM

Oh! I get it, performance be damned, power consumption is the only metric that matters.
Tom's has hammered away at the power thing so as to skew all testing toward INTEL.
Go back look at the hundreds of reviews, see that when INTEL started using smaller scale how Tom's jumped on the bandwagon. I want AMD to succeed so we have competition. I want other companies to prosper as well. This goes for the stupid assumption that every Tom's reader has thousands to spend.
All this crap about how a game won't play at super high res is not serving the readers. We don't have $5000 to spend on a system that will be obsolete in a few months. Your marathons are so up the a$$ of advertisers it is sad. I want to be a Fan, but it is time to get real.
Score
-2
December 26, 2012 12:32:24 AM

Thanks for the insight into the mobile / tablet space in terms of Intel / ARM offerings ... quite enjoyed this.

Cheers.
Score
2
December 26, 2012 1:15:21 AM

If i'm going to buy a "tier1" tablet and smartphone, it wont have a 32nm Intel chip, or a 40nm ARM chip.
Score
0
December 26, 2012 8:15:53 AM

blubbeyI was until I saw the numbers. Intel spent $8.4 billion in 2011 ($6.6 billion in 2010 and $5.7 billion in 2009) on R&D - http://www.intc.com/intelAR2011/business/research/ - while ARM isn't worth $1b. It may take a few years but Intel are seriously massive, they'll soon be the go to guys for mobile. Plus (according to rumours) with Haswell focusing on power saving, it could be a big leap forward.


Leap forward eh? What about, cough, price?
Do you know that even if Intel grabs all the market, minus Samsung and Apple, it's still "lolwhat" amount of money to them?

There is no way this Juggernaut used to price fixing and what not can get away with ARMs margins.
So they need to beat ARMs with times better performance + power consumptions (and they are VERY far away form that) so that they can start asking at least 75-150$+ for their chips. Uhm, I don't see this happening at all, not now, nor in the foreseeable future
Score
-1
December 26, 2012 8:19:23 AM

PS
So Intel spending money to enter market, which isn't at all financially interesting. Why?
I guess because of fear, that one day all those non-PC thingies will finally kill PC.
Score
-1
December 26, 2012 12:20:10 PM

kartuLeap forward eh? What about, cough, price?Do you know that even if Intel grabs all the market, minus Samsung and Apple, it's still "lolwhat" amount of money to them?There is no way this Juggernaut used to price fixing and what not can get away with ARMs margins.So they need to beat ARMs with times better performance + power consumptions (and they are VERY far away form that) so that they can start asking at least 75-150$+ for their chips. Uhm, I don't see this happening at all, not now, nor in the foreseeable future


Which is why I said 'it may take a few years'. You underestimate what tens of billions worth of research over the next 5 years can do.
Score
0
December 26, 2012 2:40:17 PM

You fail to acknowledge an important point: WinRT does not support use of Tegra's 5th, low-power companion-core. I think you noted that Android has "better" support for the companion core, but my understanding is that the companion core doesn't work at all under WinRT. That is obviously a huge downside for the Tegra which depends on that to not completely suck as far as power usage goes.

Of course, that raises the question of why the hell Microsoft chose the Tegra3 for the Surface. I can only assume Nvidia is one step away from giving away Tegra3 SoCs for free, as there are other mature ARM solutions on the market they could have tapped, most notably the dual- and quad-core Qualcomm Snapdragon S4 variants. Either of which are superior to Tegra3 in terms of performance and power efficiency.

Also, it has to be noted that Clover Trail's GPU is really, really awful. It kills the idea of running legacy x86 games on the tablet, except for maybe Solitaire and Minesweeper.
Score
1
December 26, 2012 2:56:32 PM

I think that this chip war will also take place on the software side.
Fortunately, if this Atom SoC puts intel on par with ARM in the tablet space, Android will have had all this time to try to put on par with Windows on app availability...
Probably, content providers prefer Android over Windows because of being a closed ecosystem, while content consumers want whatever is easier for them... while others will love the x86 "openness".
Score
0
December 26, 2012 6:03:19 PM

jessica sanchez de la hoyaOnce again, Intel manages to sort of tie with a middle-of-the-pack last-generation ARM CPU, well done.Tegra 3 is junk and everybody knows it, bring on the Samsung Chromebook or a comparable tablet running an A15, and run Linux on them. Or rather than just waiting for Tom's to do it, check out the benchmarks on Phoronix that show the A15 eating the Atom for breakfast...


"-last generation ARM CPU..." All that statement shows is that you're biased or misinformed. The Tegra 4 isn't out yet so the T3 is current generation tech. What is funny is how you didn't point out Phoronix's use of an actual last gen Intel CPU in their benchmarks. Funny how, to you, it matters for one but not the other...
Score
-1
December 26, 2012 7:50:35 PM

i hope this leads to a slue of media pc's like we saw 2012-11.......
Score
0
December 26, 2012 11:34:45 PM

There is a surprise. Intel chips are more power efficient and outperform in single thread applications. I am starting to see a pattern here.
Score
2
December 27, 2012 2:00:36 AM

Intel's cherry picked benchmarks piss me off. Tegra is a POS. Let's see the krait comparisons!
Score
0
December 27, 2012 2:27:45 AM

dozermanIntel's cherry picked benchmarks piss me off. Tegra is a POS. Let's see the krait comparisons!


Keep in mind that Surface has a lot of buzz in the market right now and the RT has already been released with the Tegra 3. Windows is going to release an Intel version here very soon so it makes a lot of sense that Intel targeted the Tegra 3 for it's comparison. In this case it's not "cherry picked" it's Intel making a case as to why you should by the Atom powered Surface as opposed to the Tegra version.

Not to mention this article was focused on power consumption not processing power...
Score
0
December 27, 2012 7:28:47 AM

This just confirms what most already knew, Arm by insulting Intel have awoken the giant. History wise it reminds me of the Amd days... one big question remains - Will history repeat itself but this time with another Cpu company that starts with an A or will it go the other way?

(Those figures in the article shows that even with the x86 die overhead its very possible to make it very efficient with the right R&D staff - So much for the Arm fanbois claims that Intel couldn't make this happen with a x86 cpu a couple of years back... Amazing how fast people/company's can be proven wrong when givin the competition the right motivation)
Score
0
!