Intel Core M lets new MacBook go light and fanless, but with sacrifices

1567911

Comments

  • Reply 161 of 202
    snookiesnookie Posts: 139member
    "(much as the successful Microsoft Surface)."

    Amusing. The Surface is far from successful. This is really well known and obvious. Microsoft has lost millions on the Surface. It's both a poor quality tablet and a poor quality laptop. There are not many apps available and returns are high because people hate the interface. Not dislike but hate it. No surprise since Microsoft designed it for their needs instead of their customers.
  • Reply 162 of 202
    Quote:

    Originally Posted by wizard69 View Post





    Using that logic they should just get an Iphone.





    By the way I understand what you are saying, but people value their time differently and a machine that is seen as slow offends their sensibilities. I have an old LINUX machine down in the cellar right now that works fine and likely will become a CNC controller soon, but man is that machine slow!!! Given an iPad or my new MBP I'd grab them before using that old machine to send an E-mail.



    You're correct. Using my logic that would be the case and increasingly IS the case. Many more iOS users can and are using iPhones for general purpose computing. Professors complain all the time about students who bang out term papers on their iPhones. The 6 Plus is powerful enough (and has a large enough screen) to be used in place of a notebook or tablet computer for many others.

     

    BTW, I hear ?Watch is having yield issues. Can you CNC me one in your Tony Stark lab? ?Watch Collection-Milanese Loop, please. :D 

  • Reply 163 of 202
    nhtnht Posts: 4,522member
    Quote:

    Originally Posted by SolipsismY View Post



    Proof that the currently shipping A8X has a 4.5W TDP?

     

    There's no proof given that Apple does not post those kinds of numbers but the K1 is 5-8W depending on device (5W for tablets) and most of the estimates are that the A8X are around 4.5W.  If it's less it's probably not much less.  The Core-M is a very impressive offering by Intel.  

     

    The primary downside is the cost in comparison to ARM and not performance per watt.

  • Reply 164 of 202
    solipsismysolipsismy Posts: 5,099member
    nht wrote: »
    There's no proof given that Apple does not post those kinds of numbers but the K1 is 5-8W depending on device (5W for tablets) and most of the estimates are that the A8X are around 4.5W.  If it's less it's probably not much less.  The Core-M is a very impressive offering by Intel.  

    The primary downside is the cost in comparison to ARM and not performance per watt.

    OK, if the assumption is the iPad Air 2's A8X is 4.5W then what is the entirety of the HW pushing in terms of power? I assume that if there is an educated guess that the other component's power usages has either been systematically detailed or estimated in order to figure the TDP of only the A8X. How much is the iPad Air 2's display and backlit using if only the A8X is using 4.5W?
  • Reply 165 of 202
    nhtnht Posts: 4,522member
    Quote:

    Originally Posted by frantisek View Post

     

    ASUS is 1.4 GHz but Lenovo 1.1 GHz. If you do simple math results are roughly equal.


     

    And the top end Macbook is 1.3 Ghz.  If you do the simple math the top end Macbook will perform more like the ASUS than the Lenovo.

     

    It should bench in pretty close the to 2015 MBA and the 2012 13" MBPr.

  • Reply 166 of 202
    nhtnht Posts: 4,522member
    Quote:

    Originally Posted by SolipsismY View Post



    OK, if the assumption is the iPad Air 2's A8X is 4.5W then what is the entirety of the HW pushing in terms of power? I assume that if there is an educated guess that the other component's power usages has either been systematically detailed or estimated in order to figure the TDP of only the A8X. How much is the iPad Air 2's display and backlit using if only the A8X is using 4.5W?



    Why are you asking me an not readiing Anandtech or Ars to see why that's their estimate?  Do your own research.

     

    And why do you care anyway?  If it's below 4.5W its not going to be half.  That the A8X is faster and lower power than the K1 is believable.  That it's faster and half the power assumes some magic blue pixie dust being sprinkled at TSMC.

     

    A fanless Macbook with mostly Core i5 performance indicates that there's zero reason for Apple to go ARM for the Macs and abandon its current X86 software base, the ability to run Windows software and commonality with the MBP, iMac and Mac Pro processors which must remain x86 for performance reasons unless someone has some viable alternative to Core i7 and Xeon.

  • Reply 167 of 202
    knowitallknowitall Posts: 1,648member
    So, Apples A8X has a higher geekbench 3 score.
    That seems to be end of the line for Intel.
  • Reply 168 of 202
    knowitallknowitall Posts: 1,648member
    nht wrote: »

    Why are you asking me an not readiing Anandtech or Ars to see why that's their estimate?  Do your own research.

    And why do you care anyway?  If it's below 4.5W its not going to be half.  That the A8X is faster and lower power than the K1 is believable.  That it's faster and half the power assumes some magic blue pixie dust being sprinkled at TSMC.

    A fanless Macbook with mostly Core i5 performance indicates that there's zero reason for Apple to go ARM for the Macs and abandon its current X86 software base, the ability to run Windows software and commonality with the MBP, iMac and Mac Pro processors which must remain x86 for performance reasons unless someone has some viable alternative to Core i7 and Xeon.

    A viable alternative exists when Apple introduces its A9 processor on 14nm (or even below) this year.
    It also exists when a 4 or 6 core A8 is used.
  • Reply 169 of 202
    solipsismysolipsismy Posts: 5,099member
    nht wrote: »

    Why are you asking me an not readiing Anandtech or Ars to see why that's their estimate?  Do your own research.

    And why do you care anyway?  If it's below 4.5W its not going to be half.  That the A8X is faster and lower power than the K1 is believable.  That it's faster and half the power assumes some magic blue pixie dust being sprinkled at TSMC.

    A fanless Macbook with mostly Core i5 performance indicates that there's zero reason for Apple to go ARM for the Macs and abandon its current X86 software base, the ability to run Windows software and commonality with the MBP, iMac and Mac Pro processors which must remain x86 for performance reasons unless someone has some viable alternative to Core i7 and Xeon.

    1) I originally asked [@]staticx57[/@] to support his claim, not you, but since you jumped in to defend that claim then the onus also fails to you to support it.

    2) Why is it my responsibility to support your claim? The only reason I can imagine for one to get defensive when asked to support something they state as an unwavering and absolute fact is when they've made it up.
  • Reply 170 of 202
    robertcrobertc Posts: 118member
    Quote:
    Originally Posted by frantisek View Post

     

     

    ASUS is 1.4 GHz but Lenovo 1.1 GHz. If you do simple math results are roughly equal.


    No.

     

    Just to fill you in, Intel has 3 common variants of Core M already on the market:

     

    5Y10:

    - Base frequency: 800 MHz

    - Max turbo frequency: 2.0 GHz

    - Graphics base frequency: 100 MHz

    - Graphics max frequency: 800 MHz

     

    5Y70 (MacBook entry model):

    - Base frequency: 1.1 GHz

    - Max turbo frequency: 2.6 GHz

    - Graphics base frequency: 100 MHz

    - Graphics max frequency: 850 MHz

     

    5Y71 (MacBook top model):

    - Base frequency: 1.2 GHz

    - Max turbo frequency: 2.9 GHz

    - Graphics base frequency: 300 MHz

    - Graphics max frequency: 900 MHz

     

    The Dell Venue 11 Pro (5Y10 version) scores 2180 single-core and 4506 multi-core. The Dell Venue 11 Pro with Core M is a fanless tablet in a mostly plastic chassis:

     

     

    The Lenovo Yoga 3 Pro is a laptop that uses a fan:

     

     

    The ASUS T300 Chi has an aluminum chassis and no fan:

     

     

    If you can see what I'm getting at, no matter how you look at it, the implementation of Core M on the Lenovo is quite poor. I doubt Apple will have that type of trouble with implementation.

  • Reply 171 of 202
    robertcrobertc Posts: 118member
    Quote:
    Originally Posted by knowitall View Post





    A viable alternative exists when Apple introduces its A9 processor on 14nm (or even below) this year.

    It also exists when a 4 or 6 core A8 is used.

     

    14 nm =/= 14 nm. Intel is the only company with a true 14 nm process. Samsung's "14 nm" process is not a true 14 nm process, the same goes for TSMC's 16 nm. There will not be anything below Intel's 14 nm for an A9 to use this year.

     

    Intel will also be launching Skylake-Y (Core M 2014) in Q3 of this year, which has a 4W TDP and offers even higher performance than Broadwell-Y (Core M 2015). It will also likely be paired with DDR4 RAM.

  • Reply 172 of 202
    nhtnht Posts: 4,522member
    Quote:

    Originally Posted by SolipsismY View Post



    1) I originally asked @staticx57 to support his claim, not you, but since you jumped in to defend that claim then the onus also fails to you to support it.



    2) Why is it my responsibility to support your claim? The only reason I can imagine for one to get defensive when asked to support something they state as an unwavering and absolute fact is when they've made it up.

     

    There's no need to defend the claim since its in the right ballpark when assessing the relative merits of A8X vs CoreM.  If you want to be a pedantic ass then he should rephrase his statement as "Huh? The Core M scores better and uses about the same power as the A8x all while running a much more powerful and versatile OS".

     

    ?Its your responsibility because you always ask for unavailable "proof".  The consensus is around 4.5W.  It is presented as an estimate and not as fast (as that's the best guess at the moment.  if you think it's wrong YOU show evidence that it's wrong.

  • Reply 173 of 202
    robertcrobertc Posts: 118member
    Quote:

    Originally Posted by knowitall View Post



    So, Apples A8X has a higher geekbench 3 score.

    That seems to be end of the line for Intel.

    No it doesn't, like I've explained in my earlier posts, the poor implementation on the Levono is not a reference for Core M performance.

     

    I've also explained that Geekbench 3 does not properly compare scores between ARM and x86. Nor does Geekbench compare overall / graphics performance. I don't really understand the obsession with Geekbench, my only assumption is it being one of the few benchmarks on both ARM and x86.

  • Reply 174 of 202
    bugsnwbugsnw Posts: 717member

    Maybe it's just reading about Steve Jobs and Apple over a lifetime and admiring their taste and ability to engineer amazing products. Here's where they are going....

     

    The fewer buttons they have to include on an item, the better.

    The easier and more intuitive the software, the better

    The thinner and lighter, the better

    The cooler, the better (as in Fonzie cool)

    The more Apple designed stuff inside the product, the better

     

    So sure, if I had a time machine and went forward 20 years, I'd expect to see Apple CPUs in everything. The laptops would have zero ports because wireless technologies have rendered them obsolete. Laptops would be insanely thin and light, made out of new composites. And maybe......just maybe.... the iMac wouldn't have a chin.

     

    The keyboard would still be the same. I don't think Apple can get around that.

  • Reply 175 of 202
    solipsismysolipsismy Posts: 5,099member
    nht wrote: »
    There's no need to defend the claim since its in the right ballpark when assessing the relative merits of A8X vs CoreM.  If you want to be a pedantic ass then he should rephrase his statement as "Huh? The Core M scores better and uses about the same power as the A8x all while running a much more powerful and versatile OS".

    I haven't even begun to be a pedant on this topic. I merely asked for any evidence of the TDP of the A8X as I've seen none. I've seen info on the total draw of the iPad but nothing that speculates on what the A8X uses. You made a steadfast claim so I inquired. Why you're getting defensive over what should be an easy question for you to answer based on your previous comments.
    ?Its your responsibility because you always ask for unavailable "proof".

    If it's unavailable then you shouldn't make the claim, nor should you get upset when someone asks you for the same pool of data you used to come to that understanding.
    The consensus is around 4.5W.  It is presented as an estimate and not as fast (as that's the best guess at the moment.

    Gotcha, just like the consensus of the Tea Party is Obama is a secret Nazi Muslim born in Kenya. :rolleyes:
    if you think it's wrong YOU show evidence that it's wrong.

    1) I've never once stated it was wrong, but if I did I would absofuckinglutely show how I arrived at that hypothesis, theory or fact, whatever the case my be, and probably without you even asking.

    2) I'm now stating your comments are wrong to claim an opinion as fact that can't be questioned or tested.
  • Reply 176 of 202
    Quote:
    Originally Posted by SolipsismY View Post





    Wow! That's impressive considering they launched the same year and the tray price of the Core-i7 is more than $100 higher.



    I didn't realized Core M-5Y71 doesn't offer virtualization. I'm curious how that will affect running Fusion or Parallels.



    This page seems so indicate that it supports virtualization, or am I reading it incorrectly? (no sarcasm, genuine curiosity)

    http://ark.intel.com/products/84672/Intel-Core-M-5Y71-Processor-4M-Cache-up-to-2_90-GHz

  • Reply 177 of 202
    solipsismysolipsismy Posts: 5,099member

    This page seems so indicate that it supports virtualization, or am I reading it incorrectly? (no sarcasm, genuine curiosity)
    http://ark.intel.com/products/84672/Intel-Core-M-5Y71-Processor-4M-Cache-up-to-2_90-GHz

    If Intel's site says it then I have to assume CPU Monkey is incorrect. Nice catch.
  • Reply 178 of 202
    knowitallknowitall Posts: 1,648member
    robertc wrote: »
    No it doesn't, like I've explained in my earlier posts, the poor implementation on the Levono is not a reference for Core M performance.

    I've also explained that Geekbench 3 does not properly compare scores between ARM and x86. Nor does Geekbench compare overall / graphics performance. I don't really understand the obsession with Geekbench, my only assumption is it being one of the few benchmarks on both ARM and x86.

    Yes it does. I've (just) read your comments, but other than stating geekbench 3 isn't suited as a benchmark, no argument can be found.
    Geekbench 3 is suited as a cross platform benchmark and the numbers speak for themselves: A8X 1.5 GHz 4488 and 1.1 GHz Core M 4267.
    Corrected for the GHz Core M is faster but only slightly so and I suspect the low clock rate is for a reason (Core M dissipates a lot it seems).
  • Reply 179 of 202
    nhtnht Posts: 4,522member
    Quote:
    Originally Posted by SolipsismY View Post



    I haven't even begun to be a pedant on this topic. I merely asked for any evidence of the TDP of the A8X as I've seen none.

     

    No, you asked for "proof" and when it was stated by me that no proof was available but the consensus was 4.5W you started asking for some kind of detailed analysis. 

     

    If you disagree with the number find another.  More importantly it's a distraction from his primary point which is that they TDPs are very similar.  Not that the A8X is exactly 4.5W TDP.

     

    Quote:

    I've seen info on the total draw of the iPad but nothing that speculates on what the A8X uses. You made a steadfast claim so I inquired. 


     

    I made no steadfast claim on the TDP of the A8X.  This is a lie.  There seems to be a trend in the way you "debate". 

     

    My "steadfast claim" is that "most of the estimates are that the A8X are around 4.5W".

     

    Quote:
    Why you're getting defensive over what should be an easy question for you to answer based on your previous comments.

     

    Because there is never sufficient evidence for you and it is annoying. 

     

    Quote:
     If it's unavailable then you shouldn't make the claim, nor should you get upset when someone asks you for the same pool of data you used to come to that understanding.

     

    The evidence that "most of the estimates are that the A8X are around 4.5W" is clearly available just by googling.  Your inability to discern the difference between "estimate" and "fact" is not really my problem.

     

    As to whether 4.5W is a reasonable estimate the Tegra numbers shows it's probably in the right ballpark.  4.2W?  Sure.  3.9W?  Probably near the low end of the reasonable range of estimates but why not.  2.25W?  No.  How those guys came up with 4.5W I haven't bothered looking.  Nor am I going to.  When you search for A8X TDP the number that comes up is some variant of 4.5W (4.5W, ~4.5W, <4.5W).

     

    If I stated that TDP of the A8X was said to be 5-8W you would have a valid reason to bitch.  Not that you seem to need a valid reason.

     

    Quote:

    1) I've never once stated it was wrong, but if I did I would absofuckinglutely show how I arrived at that hypothesis, theory or fact, whatever the case my be, and probably without you even asking.



    2) I'm now stating your comments are wrong to claim an opinion as fact that can't be questioned or tested.


     

    Oh bullshit.  The point which you keep dodging is that the A8X and the Core M is AROUND THE SAME TDP.  Not one smidgen of evidence indicates otherwise whereas all other current high performance tablet ARM processors with published specs are in that 4-10W range.  

     

    Apple is in the high end of the performance numbers but the low end of the TDP spectrum.  Core-M is in the same ballpark with X86 compatibility and a significantly higher unit cost.  4.5W TDP for almost Core i5 performance is really impressive technology from Intel.  

     

    Presumably if Apple teamed with Intel and got the A8X down to 14nm it would be better than Core-M on a performance per watt basis rather then in the same ballpark but even assuming that you aren't guaranteed that it would scale to the desired Core i7 performance level required in the MBP and iMacs.

  • Reply 180 of 202
    solipsismysolipsismy Posts: 5,099member
    nht wrote: »
    No, you asked for "proof" and when it was stated by me that no proof was available but the consensus was 4.5W you started asking for some kind of detailed analysis.

    I did ask for proof, but not from you. When you decided to jump in with Magic 8 Ball-esque statements I said "OK" and reworded my queries to ask where you received these "assumptions" and how these "educated guesses" were factored. This is where you got your panties in a bind, and despite my repeatedly asking you to name a source other than your watercolor gossip you haven't given a single source that has detailed these "ballpark" figures. Why should I have to validate your claims?
    Because there is never sufficient evidence for you and it is annoying.

    Sure there is. There is sufficient evidence for me that you put your foot in your mouth, metaphorically speaking, of course. I can prove this, too, by simply pointing out your previous comments, including your specific response to my inquiry into your defensiveness which you claim there is "there is never sufficient evidence for [me] and it is annoying" after telling me there is no proof, only some laughable consensus.
    The point which you keep dodging is that the A8X and the Core M is AROUND THE SAME TDP.  Not one smidgen of evidence indicates otherwise whereas all other current high performance tablet ARM processors with published specs are in that 4-10W range.

    And there is not one smidgen of evidence that indicates a flying spaghetti monster is impossible. If I were trying to disprove its existence of FSM I would not be able to do so, but since I'm only trying to ascertain how you received your affirmation that the FSM is real you should be able to do so with something that's a little better than, "Well the consensus of everyone at the Church of the FSM says so. Why would you need to know more. It's all nine. Gah!"
Sign In or Register to comment.