Mobile Haswell achieves all day battery life - beginning of the end for ARM?

[blackened23]

http://www.theverge.com/2013/1/7/3847742/intel-fourth-gen-intel-core-processor-battery-life

Let's face it. The only reason ARM is relevant is because despite how hilariously low performance the current parts are, all day battery life is nice.

You can view performance benchmarks of ARM vs x86 in AT's recent review, and as expected ARM is hilariously low performance. But that battery life.

I think there will be a shift in momentum if this claim by intel (which they have made for a long time now) is true. I also think this will be the beginning of the end for ARM competition wise, they can only complete on cost - since it's obvious that they don't have high performance parts.

What do you guys think? Intel seems really adamant that haswell is the absolute best gen to gen efficiency improvement they have ever achieved. What do you guys think? I think there will be a shift in momentum towards fully capable Haswell powered ULV tablets. Now that intel has comparable battery life, aside from cost - ARM is irrelevant.

 

[Red Hawk]

And how big of a battery does mobile Haswell need to have "all day battery life"?

 

[blackened23]

And how big of a battery does mobile Haswell need to have "all day battery life"?

I don't know. I've read that with all other things equal to IVB, that haswell has 3 times the battery life. I'm seeing figures of 14 hours continuous battery life, and all day battery life with power optimizations through the operating system. This will be absolutely a huge deal because 14 hour continuous battery life exceeds that of any current gen ARM tablet.

Like I said - intel seems very serious about Haswell being their best work in the history of intel in terms of efficiency gains. If that is even remotely true, I really see that as a blow to ARM devices in the high end tablet market.

 

[dqniel]

I'll believe it when I see it.

 

[frozentundra123456]

I'll believe it when I see it.

Agreed, but even if the battery life is phenomenal there is the cost issue. An ARM chip costs what, a few dollars, while Haswell costs what, 200.00 or more. Personally, I am so frustrated with the poor performance of my ARM tablet and the annoying Android OS that I would pay the difference, but a lot of people seem satisfied with it.

 

[Charles Kozierok]

If not Haswell, then Broadwell.

I've said it before and I'll say it again -- never count Intel or x86 out.

There will always be a place for cheap toys, and a place for real computers. As long as Intel stays close in the areas of size/weight/batterylife, there will be many people who will prefer their platforms.

 

[NTMBK]

To echo the above, I'll believe it when I see it, and when I see the pricing.

 

[CHADBOGA]

To echo the above, I'll believe it when I see it, and when I see the pricing.

I agree with that.

BTW, Have you learnt your lesson about SemiAccurate's idiot mods yet? :awe:

 

[tweakboy]

Haswell mobile chips in 2013. Desktop chips in 2014. 2013 champ for speed will be Ivy Bridge E 6 to 12 cores + HT. F howard I mean dukes of hazzard

 

[sefsefsefsef]

There is nothing unbelievable about the claim that Haswell will achieve "all day battery life" (whatever that vague statement means). Even my 2011 MBP with a quad-core Sandy Bridge i7 can offer 5-7 hours of moderate use on a single charge. The difference is that my i7 cost 10-20x more than a current ARM chip would cost. Intel does what they do by throwing tons of resources at every problem, so their success is no surprise IMO, but of course the consumer has to ultimately pay for it.

 

[blackened23]

There is nothing unbelievable about the claim that Haswell will achieve "all day battery life" (whatever that vague statement means). Even my 2011 MBP with a quad-core Sandy Bridge i7 can offer 5-7 hours of moderate use on a single charge. The difference is that my i7 cost 10-20x more than a current ARM chip would cost. Intel does what they do by throwing tons of resources at every problem, so their success is no surprise IMO, but of course the consumer has to ultimately pay for it.

Here's the thing. Intel has no interest in competing for extremely low end, low margin parts: they want to provide parts for real computers with real power. If ARM wants the sub 200$ market, good for them.

Thus far, intel has continually been stymied due to battery life - "high end" ARM tablets in the 500-700$ price range have boasted 8-9 hour continuous battery life, so while intel clearly outperformed those parts by a wide margin, battery life was a big drawback.

Battery life will no longer be a big drawback. I think this will be huge for intel in the high end tablet market - If I could look into a crystal ball, I could see a future where intel completely and utterly dominates the high end 600$+ tablet market. Again, ARM is only relevant because of battery life - even the best ARM devices have pitiful performance in comparison to intel.

I think the most humorous aspect of this is that ARM has been quite smug over the past couple of years about how intel "will never catch us" on battery life. To see their hubris catch up to them, and for them to eat their words, will be quite funny indeed.

 

[NTMBK]

I agree with that.

BTW, Have you learnt your lesson about SemiAccurate's idiot mods yet? :awe:

Hah, I was wondering when someone would notice that! I'd already had a couple of minor infractions for posts that were taking the piss a bit, so it wasn't out of the blue. I'll be going back when my vacation's over, anyway.

(The guy did have it coming though.)

 

[Yuriman]

I think you're right in a sense. Haswell and Sky Lake are what make x86 into a power efficient and competitive architecture, but Intel hasn't "won" just yet because there is a huge market for sub-$500 products. It's probably more correct to say that ARM is the death knell for AMD, or at least their CPU division.

I suspect we'll continue to see ARM in phones for a long time, x86 will creep into the more expensive tablets, and we'll see more architecture agnostic programming.

 

[meson2000]

I will also believe it when i see it. Haswell tablets are still a ways off.

Plus, the benchmarks above are very misleading. The benchmark in the graph is heavily dependent on the browser/OS used. The Nexus 7 and Surface RT above use practically the same Tegra 3 chipset (T30L versus T30) but the Surface RT takes nearly half the time to complete the benchmark.

 

[BrightCandle]

Pricing and battery size are pretty relevant. But if I get a similar form factor at that increase in performance then I am happy.

So far it doesn't look like that is imminent, the data Anandtech captured shows only a moderate advantage over ARM in performance at these power levels. Haswell if it delivers this will be a fantastic advancement beyond any Intel has made in recent years by some measure. I just don't see it being 4x the efficiency.

 

[Ventanni]

Unless intel can design and produce haswell and successor chips at $10 a piece, it won't compete with arm where it matters. Sure, it'll compete with arm in performance and eventually battery life, but not in price.

But atom in the other hand...

 

[blackened23]

Unless intel can design and produce haswell and successor chips at $10 a piece, it won't compete with arm where it matters. Sure, it'll compete with arm in performance and eventually battery life, but not in price.

But atom in the other hand...

There are plenty of consumers that do not want bargain bin cheap products. This is precisely why PC makers have been struggling - in prior years they have traditionally competed on price only. Then apple comes along, and turns everything around out of the blue.

There is a lot of demand for real computing devices, there has been quite a paradigm shift recently in the US. Consumers are willing to pay more for better products.

 

[CHADBOGA]

Hah, I was wondering when someone would notice that! I'd already had a couple of minor infractions for posts that were taking the piss a bit, so it wasn't out of the blue. I'll be going back when my vacation's over, anyway.

(The guy did have it coming though.)

Oh undoubtedly, but I thought your banning was ridiculous.

Truly, they are the worst mods of any forum on any subject, I have ever encountered.

 

[Idontcare]

If not Haswell, then Broadwell.

I've said it before and I'll say it again -- never count Intel or x86 out.

There will always be a place for cheap toys, and a place for real computers. As long as Intel stays close in the areas of size/weight/batterylife, there will be many people who will prefer their platforms.

Haswell vs. The World will be an interesting start, not expecting to be surprised either way in how that turns out TBH.

But 14nm is where things will become silly lopsided in Intel's favor IMO.

Regardless the foundry of choice, if you are fabless then you are going to be showing up to a gun fight with a knife when you start fielding your 16nm TSMC or 14XM GloFo parts that are really just hybrid 20nm products designed to finally be competitive with Intel's 22nm...only your products will be competing with Intel's full-blown 14nm legitimate shrink.

Broadwell, Skylake, and Airmont will put Intel a full head and shoulders in front of the competition and there is nothing any of the fabless guys can do about it as their fates are tied to the foundry's roadmaps and those roadmaps are not at all competitive with Intel's come 14nm.

 

[USER8000]

Intel is launching its latest Intel Core processor today at CES 2013. Formerly known as Haswell, the latest Intel chip will bring all-day battery life to the latest ultrabooks. Intel says this will equate to around 9 hours of continuous battery life on new systems. This new chipset is designed for a new ultrabook detachable reference design codenamed North Cape that converts into a 10mm tablet which will run up to 13 hours while docked.
Intel showed off a reference design on stage that has a battery both underneath the keyboard and behind the display on a Core i5 / i7 system. Along with the new reference design, Intel says ultrabooks will require touch input for fourth-generation Core systems. Intel didn't provide exact timings on when systems will arrive with its latest processors, but expect to see them regularly in new Windows 8-powered devices throughout 2013.

I thought the link in the thread had independent benchmarks,comparing battery life,etc with the competition.

Instead, I see what looks like someone regurgitating a PR statement.

So,people are getting overexcited about some marketing blurb from a company??

Are people that bored now??

This thread disappoints.

 

[Charles Kozierok]

Broadwell, Skylake, and Airmont will put Intel a full head and shoulders in front of the competition and there is nothing any of the fabless guys can do about it as their fates are tied to the foundry's roadmaps and those roadmaps are not at all competitive with Intel's come 14nm.

Yep.

That combined with the toy computer fad beginning to fade....

I'm not worried, if I'm Intel.

 

[NTMBK]

Yep.

That combined with the toy computer fad beginning to fade....

I'm not worried, if I'm Intel.

"Toy computer"? No, cheap computer. Same reason that people went for netbooks.

 

[KompuKare]

Surely the one reason ARM will continue to do well is because of what they offer: almost complete control to their licensees to do what they want. Intel has no intention of allowing that. ARM's business model is totally different to Intel's.

Yet, eventually Intel with huge R&D, big pockets and process lead will make their CPUs competitive. We just have to at look how what was possible the worst CPU architecture in the early 1980s eventually caught up and mostly surpassed all the other CISC and RISC vendors. Intel’s secret ingredient may only be money and economics of scale but that they sure have a lot of that: just watch their march to the top position in Semiconductor sales in the wikipedia: http://en.wikipedia.org/wiki/Semiconductor_sales_leaders_by_year

Of course neither Qualcomm or Samsung are going to give up on ARM but since neither has a presence in high-margin areas like servers they are unlikely to be a major threat to Intel there and will hence have less profits to re-invest.

 

[ShintaiDK]

"Toy computer"? No, cheap computer. Same reason that people went for netbooks.

Netbooks are dead and was nothing more than a toy. And tablets(also toy) and smartphones margins are tumbling down. The record profit Samsung and Apple currently got will be something of the past in the future.

 

[ShintaiDK]

Of course neither Qualcomm or Samsung are going to give up on ARM but since neither has a presence in high-margin areas like servers they are unlikely to be a major threat to Intel there and will hence have less profits to re-invest.

Only Qualcomm got a real interest in ARM of the 2.