Intel looks to build ultra-efficient mobile chips Apple 'can't ignore'

245

Comments

  • Reply 21 of 100
    wizard69wizard69 Posts: 13,377member
    drdoppio wrote: »
    Performance-wise, including battery life, Medfield is already competitive with the best ARM chips. It seems that the major obstacle before a wider adoption of Intel's mobile chips is price. This was discussed with more details in the ....

    A better test would be a comparison of how long the platform can run crushing the web. Honestly Apple can get far better performance by turning up the clock rate and widening the path to memory. Is that the right thing to do on a cell phone? The real test of a cell phone is being able to make that call to your lover after an 8 hour day of use and maybe a call to your wife also.

    Frankly this is the Android problem but expressed in hardware. A phone that drains the battery quickly is of little practicle use.
  • Reply 22 of 100
    wizard69wizard69 Posts: 13,377member
    Wishful thinking if you ask me. Intel has yet to demonstrate a willingness to do custom silicon. This is make or break for Intel, if they can't do custom they can't win.

    The problem is simple engineering these days involves putting functionality on silicon not a PC board. The greater the functionality that can be built into a custom chip the greater the advantage for the users of ARM architecture. In the Apple 2 days and the early Macs, Apple got significant advantage from clever design on the printed circuit board. Today there is little on a circuit board beyond the processor, RAM and flash. Each iteration Apple needs and will find ways to integrate more functionality on chip. Silicon is this generations printed circuit board.
    aizmov wrote: »
    Next year the Atom SoC will be far ahead of ARM not only in performance but also power efficiency. I won't mind my next iPad or iPhone having an Atom inside of them.
  • Reply 23 of 100
    wizard69wizard69 Posts: 13,377member
    2012 has barely gotten started, so I think it is a stretch to declare Intels success.
    ksec wrote: »
    Then there is the cost equation and many other trade off.
    2012 Medfiled, is performance / watt just about comparable with current ARM design. Although Intel Wins in performance, ARM still wins with lower power usage. But then it is bigger, and hotter.
    iI isn't clear at all that Intel wins with its 2012 technology. There are many ARM chips to be released real soon that may offer a real advantage over ATOM.
    2013 with 22nm ULP, it should be a lot better for idle performance but then you have 28nm LP version of Cortex A15 big.LITTLE SoC, which could means performance and Idle power wise should keep the advantage to ARM.
    64 bit is what will push ARM into the future. I don't even consider Cortex A15 a huge advantage. It is notable that Apple smartly concentrated on GPU performance instead of CPU performance in the new iPad. Intel is waging old battles in a dramatically changed battle field. The CPU is only a minor part of the equation.
    2014 Intel will be releasing their Mobile Atom on 14nm first before their Mainstream CPU parts. Which means Intel will again played the game with Nodes advantage, and this time it could well be two nodes ahead as TSMC aren't even reading 20nm parts til late 2014.
    Intel is always about two nodes ahead. All this indicates is that Intel screwed up really badly and is running hard to gain traction in a market where they have little success. The reality is this, ARM processors take up less space on the die thus at ever node far more auxiliary functionality can be integrated into the SoC.
    Although Samsung might be capable of holding up.
    So if all things goes well 2014 will truly be the year that Intel is truly equal or better then ARM camp in terms of performance and power.
    The problem here is that ARM and the foundries are not standing still. Nor have they played all of their cards.
    But then how much better? According to Intel Roadmap, the consequence of quickly shifting Atom through nodes is that there aren't much changes planned to its architecture. Which means its performance / Mhz will pretty much be the same. For higher performance it will have to scale to higher clock speed. The Cortex A15 is, from a high level Point of view,a faster and more powerful Out of Order architecture then Atom. But Atom has Hyperthreading, 64 Bit ( performance advantage on x86 side only ), and better software optimization. So Lets give Intel a 20% advantage here for Smaller Nodes, Higher Frequency and may be other tricks they have.
    Idle Power will still be hard to beat, the LITTLE Cortex A7 was designed with Ultra Low Power in Mind. I would doubt Intel could win, But lets call it equal with a node advantage from Intel.
    Cost -
    So Nvidia would have to add R&D per Unit , Unit Price Per Wafer and their Margin ~ 20% to their price of SoC.
    Intel owns their Fab, so they get BOTH the Fabs Margin and their Final Sale Margin. Intel could properly make a SoC that has same performance and priced the same as Nvidia but still gets 3x% profits margin.
    Now , Unlike other players Apple only make SoC for themselves. Apple makes NO profits on the SoC. Just for the simple Numbers, Apple could make a same cost SoC as Nvidia but with 20% more transistors. And Hence Faster performance.
    Now for Intel to grab Apple's SoC business, they must provide equal or better performance x86 SoC for the cheaper price. Intel would then have to make chips that is good enough ( Post 2014 ) and with less then 15% Profits Margin. Much lower then what they are used to get with Desktop CPU 50%.
    And then even so, the cost of the current Apple SoC is less then $30, even intel could provide something equal or better at $25. Would Apple even be bothered with the $5 dollars on the iPhone per unit sold but to lose control of their own SoC?

    I really need to hit the sack but I have to suggest this, Intel has messed up Sandy Bridge E and Ivy Bridge pretty badly. Do not believe that Intels goals for 2014 are written in stone.

    Also do not forget about AMD, they could easily couple an ARM CPU with an AMD GPU. More than a year ago they where using ARM A9 cores at Global Foundries to test process development. With the right amount of integration beyond the GPU, AMD could have many a customer knocking at their door. Beyond all of that AMD has fusion processors in various forms. Intel simply hasn't demonstrated the chops to build low end hardware.
  • Reply 24 of 100
    bullheadbullhead Posts: 493member

    Quote:

    Originally Posted by DrDoppio View Post





    Performance-wise, including battery life, Medfield is already competitive with the best ARM chips. It seems that the major obstacle before a wider adoption of Intel's mobile chips is price. This was discussed with more details in the Lava Xolo thread on AnandTech.

    image


     


    interesting how you do not also look at the battery life for some context...http://www.anandtech.com/show/5770/lava-xolo-x900-review-the-first-intel-medfield-phone/6


     


    great, _this_ years Intel chip runs twice the clock than _last_ years iPhone to get "faster" performance.  But it will run out of power _way_ before the iPhone will, for example in web browsing the intel phone runs out of power over twice as fast as the iPhone 4S.  I am all for better performance, but not when it is going to kill the battery.  Intel has a lot of work to do to make power efficient processors.  The MHz war is meaningless in the mobile space.

  • Reply 25 of 100
    red oakred oak Posts: 931member
    ksec wrote: »
    Then there is the cost equation and many other trade off.
    2012 Medfiled, is performance / watt just about comparable with current ARM design. Although Intel Wins in performance, ARM still wins with lower power usage. But then it is bigger, and hotter.
    2013 with 22nm ULP, it should be a lot better for idle performance but then you have 28nm LP version of Cortex A15 big.LITTLE SoC, which could means performance and Idle power wise should keep the advantage to ARM.
    2014 Intel will be releasing their Mobile Atom on 14nm first before their Mainstream CPU parts. Which means Intel will again played the game with Nodes advantage, and this time it could well be two nodes ahead as TSMC aren't even reading 20nm parts til late 2014. Although Samsung might be capable of holding up.
    So if all things goes well 2014 will truly be the year that Intel is truly equal or better then ARM camp in terms of performance and power.
    But then how much better? According to Intel Roadmap, the consequence of quickly shifting Atom through nodes is that there aren't much changes planned to its architecture. Which means its performance / Mhz will pretty much be the same. For higher performance it will have to scale to higher clock speed. The Cortex A15 is, from a high level Point of view,a faster and more powerful Out of Order architecture then Atom. But Atom has Hyperthreading, 64 Bit ( performance advantage on x86 side only ), and better software optimization. So Lets give Intel a 20% advantage here for Smaller Nodes, Higher Frequency and may be other tricks they have.
    Idle Power will still be hard to beat, the LITTLE Cortex A7 was designed with Ultra Low Power in Mind. I would doubt Intel could win, But lets call it equal with a node advantage from Intel.
    Cost -
    So Nvidia would have to add R&D per Unit , Unit Price Per Wafer and their Margin ~ 20% to their price of SoC.
    Intel owns their Fab, so they get BOTH the Fabs Margin and their Final Sale Margin. Intel could properly make a SoC that has same performance and priced the same as Nvidia but still gets 3x% profits margin.
    Now , Unlike other players Apple only make SoC for themselves. Apple makes NO profits on the SoC. Just for the simple Numbers, Apple could make a same cost SoC as Nvidia but with 20% more transistors. And Hence Faster performance.
    Now for Intel to grab Apple's SoC business, they must provide equal or better performance x86 SoC for the cheaper price. Intel would then have to make chips that is good enough ( Post 2014 ) and with less then 15% Profits Margin. Much lower then what they are used to get with Desktop CPU 50%.
    And then even so, the cost of the current Apple SoC is less then $30, even intel could provide something equal or better at $25. Would Apple even be bothered with the $5
    dollars on the iPhone per unit sold but to lose control of their own SoC?

    Well thought out. At end of day, there is value for Apple for owning the SoC and not putting it in the hands of Intel. Just look at what Intel is trying to do to Apple with this Ultrabook campaign

    Can you imagine Apple going back to Intel at this point for iOS. Not me. No way
  • Reply 26 of 100
    drdoppiodrdoppio Posts: 1,132member
    Regarding the comments about battery life of Intel's Medfield chip in the Lava Xolo phone, anandtech's benchmarks indeed show about half of iPhone's battery life. It could be assumed that Apple can increase the clock speed and match Intel's performance, while Intel could lower the clock speed and match iPhone's battery life. Some trade-off has to be made, and I'm not sure there's a clear winner in this situation... It should also be noted that iPhone is well ahead of the bulk of Android phones regarding battery life, which allows for a more beneficial comparison of Intel's chip with ARM on the same software platform.
  • Reply 27 of 100
    waybacmacwaybacmac Posts: 309member

    Quote:

    Originally Posted by Entropys View Post


    But will Apple be able to customise the chips for optimisation with its software, like they currently do with ARM?  IF so, why not if it has advantages?  I think we are a ways of that though.



    BINGO! Entropys, you have hit the proverbial nail on the head. Apple customizes it's chips on the devices where customization pays off the best--iPhones and iPads. To that end Apple acquired two semiconductor design firms: P.A. Semi and Intrinsity. Both specialize in ARM chips. To date, Apple hasn't seen the need to use customized chips on the Mac line. They've been able (usually) to get Intel's best and newest before the other computer manufacturers. That may may change with this UltraBook business. If P.A. Semi and Intrinsity's talents can be applied to Intel chips, or if Apple acquires someone with those talents, then maybe custom Intel chips could be forthcoming. Personally I doubt it. I don't think Apple cares if it is or isn't THE predominant PC manufacturer. As long as the mainline Macs are considered "best of their breed" in quality and make money, Apple is good with that. Then there is the "Air" line. Here Apple sits in the catbird seat, it can stay with Intel or go to one of their own design whichever has the performance advantage (performance being the balance of speed, temperature, battery life, cost, etc.). So when Intel CEO Paul Otellini says he wants to build better chips for the iPad, I hope he understands he's not going up against ARM chips but Apple's.

  • Reply 28 of 100
    majjomajjo Posts: 574member

    Quote:

    Originally Posted by DrDoppio View Post



    Regarding the comments about battery life of Intel's Medfield chip in the Lava Xolo phone, anandtech's benchmarks indeed show about half of iPhone's battery life. It could be assumed that Apple can increase the clock speed and match Intel's performance, while Intel could lower the clock speed and match iPhone's battery life. Some trade-off has to be made, and I'm not sure there's a clear winner in this situation... It should also be noted that iPhone is well ahead of the bulk of Android phones regarding battery life, which allows for a more beneficial comparison of Intel's chip with ARM on the same software platform.


    we have to compare apples to apples though (pun unintended). because the xolo is running android, we cant directly compare to the iphone in terms of battery life, because we dont know how much of the difference is due to the os and how much is due to the soc.


     


    while xolo's results are not earthshattering, it does show that intel can go toe to toe with arm on battery life. and imo it effectively kills the claim that x86 will never be as power efficient as arm. i say that intelms outlook on mobile is alot brighter than it was a year ago.


     


    will apple use intel socs? i dont know, but i can imagine apple abandoning all the investments they made into designing their own arm chips

  • Reply 29 of 100
    tallest skiltallest skil Posts: 43,399member

    Quote:

    Originally Posted by zunx View Post

    Intel, I have been telling you this for years! Intel should go to 4 nm, nanotechnology and even beyond right now. And now means now. Now you see the writing on the wall. Good riddance!


     


    As much as I share your belief that they should jump straight to Skymont, that's physically impossible and reeks of a the naïveté of youth.

  • Reply 30 of 100
    drblankdrblank Posts: 3,383member


    I think Apple, for the Mac side of the business, is going to continually use Intel chips, and just go after the portion of the computer where they need to put some technology investment which is the battery and the SSD.  Graphics has three or four guys working on later generation graphics chipsets for the laptop/desktop market.  But the battery for laptops is important as well as the SSD since those two areas are expensive parts of the design that needs attention that they can do with or without Intel. 


     


    Apple partnered with Intel on the Thunderbolt design, so Apple and Intel can play together as long as Intel wants to.  It's just the processors for the mobile devices Intel didn't latch on quick enough.   Yeah, Intel can make a faster processor at the expense of battery life, but battery life is equally important.  In the processor arena, everyone is playing leap frog.  What Intel has now is going to be upstaged by someone else, back and forth and ultimately, if the Intel chips can't give long battery life, then it almost doesn't matter how fast the processor is for these mobile devices.  People aren't doing the same type of work on a mobile device as they are no a workstation.

  • Reply 31 of 100
    drdoppiodrdoppio Posts: 1,132member
    majjo wrote: »
    we have to compare apples to apples though (pun unintended). because the xolo is running android, we cant directly compare to the iphone in terms of battery life, because we dont know how much of the difference is due to the os and how much is due to the soc.

    while xolo's results are not earthshattering, it does show that intel can go toe to toe with arm on battery life. and imo it effectively kills the claim that x86 will never be as power efficient as arm. i say that intelms outlook on mobile is alot brighter than it was a year ago.

    will apple use intel socs? i dont know, but i can imagine apple abandoning all the investments they made into designing their own arm chips

    I agree, and others here made compelling points that Apple isn't likely to abandon its investments in ARM either. I believe Intel has a better chance to get its processor adopted by manufactures using Android and WP. In the case of Android, it's quite remarkable that most existing apps will work without modification, but that's a topic for another forum...
  • Reply 32 of 100
    gazoobeegazoobee Posts: 3,754member

    Quote:

    Originally Posted by AppleInsider View Post



    "Our job is to ensure our silicon is so compelling, in terms of running the Mac better or being a better iPad device, that as they make those decisions they can't ignore us," Otellini said.


     


    He's absolutely right about that being their job, it's the only rational way forward for them.  Whether they succeed or not is another question entirely. 


     


    It's so funny though (IMO) how over the years intel has gone from a position of "not sure if they care" to make enough chips for Apple computers, through deciding that "they might be interested," past "Apple is great!" and "we're partners!," and now after the breakup, is all "well, maybe they will come back if we are really good."  


     


    It's like the history of a dating relationship.  If Apple continues to reject them, the next logical position for intel is "Well, we never liked them anyway, so we're seeing other companies now."

  • Reply 33 of 100
    gazoobeegazoobee Posts: 3,754member

    Quote:

    Originally Posted by Shameer Mulji View Post


    @Nvidia2008


     


    If you read the review on Anandtech, regarding Intel's latest Medfield mobile SOC, it performed very well in terms of CPU performance, battery life and camera performance.  The only area where it lagged was GPU and that's only because Medfield is single-core design.  Later this year, Intel is releasing a dual-core CPU / GPU variant of Medfield plus the fact they will be fabbing 22nm mobile SOC's next year and 14nm version in 2014, so in terms of manufacturing they're well ahead of everyone in the ARM camp.  With the introduction of Medfield, Intel is finally back in the game and as long as they execute their mobile roadmap the way they have their PC roadmap, they will be a major player in the mobile device sector within two years. 



     


    No offence, but this sounds like marketing speak.  


     


    With terms like "Intel is finally back in the game," and so forth it reads like it was lifted right out of an intel promotional pamphlet or stock report.  I read stuff like this about Microsofts latest exciting projects all the time (they've put them out for years), and it doesn't necessarily mean they are actually doing anything or becoming competitive.  


     


    You will notice that both Apple and ARM never put out statements like this and only announce what they've done, not what they will be doing or how great they are going to be in the future etc. 

  • Reply 34 of 100
    ankleskaterankleskater Posts: 1,287member

    Quote:

    Originally Posted by ChiA View Post


    With hindsight, Intel made a strategic mistake in 2006 when it sold off its ARM XScale division. 


     



     


    So true.

  • Reply 35 of 100
    gazoobeegazoobee Posts: 3,754member

    Quote:

    Originally Posted by zunx View Post


    Intel, I have been telling you this for years! Intel should go to 4 nm, nanotechnology and even beyond right now. And now means now. Now you see the writing on the wall. Good riddance!



    4nm is physically impossible.  I'm guessing this is an autocorrect error? 

  • Reply 36 of 100
    ankleskaterankleskater Posts: 1,287member

    Quote:

    Originally Posted by Gazoobee View Post


     

    You will notice that both Apple and ARM never put out statements like this and only announce what they've done, not what they will be doing or how great they are going to be in the future etc. 


     


    Good point.


     


    At the same time, hasn't Intel been tipping its hand forever? Isn't it their traditional strategy to publicize the names and architecture of nextgen processors 1-2 years before launch?

  • Reply 37 of 100

    Quote:

    Originally Posted by nvidia2008 View Post

    We have got to the stage where PCs are so powerful but so ghastly to use, let's face it. 2012 is indeed the turning point in our modern era.


     


    I don't find PCs "ghastly to use".  Quite the opposite, in fact.

  • Reply 38 of 100
    ankleskaterankleskater Posts: 1,287member

    Quote:

    Originally Posted by Entropys View Post


    But will Apple be able to customise the chips for optimisation with its software, like they currently do with ARM?  IF so, why not if it has advantages?  I think we are a ways of that though.



     


    Good question. Also a good question for ARM's other licensees too. This is not just a tech battle between Intel and ARM. It's a business model battle.

  • Reply 39 of 100
    tallest skiltallest skil Posts: 43,399member

    Quote:

    Originally Posted by Gazoobee View Post

    4nm is physically impossible.  I'm guessing this is an autocorrect error? 


     


    That was said when they hit 1000nm.


     


    I realize we're hitting quantum limits soon, but 4nm should be possible.

  • Reply 40 of 100
    ankleskaterankleskater Posts: 1,287member

    Quote:

    Originally Posted by I am a Zither Zather Zuzz View Post


     


    I don't find PCs "ghastly to use".  Quite the opposite, in fact.



     


    I agree. I still prefer my MacBook for browsing in many instances.

Sign In or Register to comment.