Tested: Thermal conditions in the 2018 i9 MacBook Pro dramatically hampering performance

124678

Comments

  • Reply 61 of 158
    Does anyone know if the fans were maxed out? As in know the Max fan speed and if they were at max the entire time once up to speed.
  • Reply 62 of 158
    aknabiaknabi Posts: 114member
    zeudev said:
    Hi Mike,

    Do you have any interest in running another benchmark with just the cpu engaged?  For some of the software developers in your audience, they will only care how well it performs when only the cpu is kicking out heat.
    We're working on it.
    Thanks... indeed it would be great to if there was an "Xcode build test" that ran a big project build over and over... or here's a funny one... download the Android AOSP for Mac and run that full build maxed out for cores. It's the sort of stuff developers use.
  • Reply 63 of 158
    mike54mike54 Posts: 229member
    Could you please mention what was your room temp when you tested this. 
    And can you test this at temps >90F (32C).

    If its throttling due to heat in an air conditioned environment its going to be much worse in higher room temperatures. Knowing room temp is important.



  • Reply 64 of 158
    sflocalsflocal Posts: 4,149member
    I’m sure the iMac Pro also has thermal issues.  All of Apple’s designers/QC testers must be assigned to iPhone & IPad...
    Oh, you're "sure"?  Please cite proof of this nonsense you're spouting off and trying to pass as something valid.

    There are no thermal issues with the iMac Pro.  Give it a rest.
    macxpress
  • Reply 65 of 158
    nicwnicw Posts: 2member
    It’s not “form over function” ... it’s FETISH over function. 

    Apple tells us the iOS user interface won’t translate to the desktop environment. I don’t believe they are genuine in this; they just haven’t got a working model yet. But what I do think is true is that the hardware design philosophy for a hand held device will not translate to a desktop device. 

    Or to make it stupidly clear, thinness has become a problem for the Mac. 

    Stepping back for a second:

    Jobs like the rest of us was swept up in the early success of iOS. But after half a decade of missteps in desktop computing, it’s time to reassess the management reforms he introduced shortly before his untimely death.  There have been too many missteps in the desktop space. Give the desktop line back to the engineers. 
  • Reply 66 of 158
    sflocalsflocal Posts: 4,149member
    jdw said:
    Mike Wuerthele, as an AppleInsider reader, I wish to say that I truly appreciated the testing and work that went into your excellent article.  It reports the facts of the matter and helps would-be MBP buyers be better informed.  Bravo!

    My own personal hope is that the tech media's reporting of this issue can prod Apple positively to THINK DIFFERENT when it comes to the thinness and approaches to minimalism in their notebook line.  Some say that Mac enthusiasts would not like it if Apple decided to use a thicker (and therefore better cooled) notebook design, but I strongly disagree.  My 2015 15" MBP is thicker than the late 2016 and newer models, yet my 2015 is quite thin to my eyes.  The MBP has long been about ports and performance.  The MacBook exists to strip all that away to achieve a thin and light design.  People upset by a thicker (yet so much better) MBP need only consider a MacBook.  And if Apple would kindly refresh the MacBook Air, that could be a second option for Mac portable buyers who prefer thinness and throttling over a thicker and cooler enclosure.  With 3 separate portable lines, surely Apple can come up with something insanely great for those who love the existing notebook line and also for The Rest of Us.
    This was very much a team effort. Five staffers worked on this piece to some extent over the course of the day.

    Appreciate the kind words.
    I will praise when good work is seen.   This was a good piece, and your staff should be commended.  

    This was the first MacBook Pro I got really excited about in a while.  I bought the late 2017 model to replace the sudden death of my 2009 iMac at the office.  I've been quite happy with it but there was always something missing.

    I would not have bought the 2018 model after I saw the lack of LPDDR4.  I definitely will not consider this laptop whatsoever until the thermal issues are resolved I suppose in 2019.  It's a shame actually because that i9 could really have been a screamer.  I suspect Apple had to come out with something serious after Intel being so behind of their CPU timelines.  Still... Intel really should be ashamed of themselves for being so late to the game constantly, and Apple really should have done better QC and made some frank decisions about the realities of a hot CPU in a super-thin chassis.  

    Will be interesting to see how this all goes the next few months.  Will Apple suddenly (and quietly) make changes to the laptops to address this?  
    muthuk_vanalingam
  • Reply 67 of 158
    I think it's time to ditch Intel and pack a pair of A11 bionic in a MacBook Pro. 12 cores sucking just the power of two iphones. that would be interesting product to use. Also, what's with Adobe and not optimising premiere for mac... it's just absurd. Most of creatives that pay for adobe suite use macs.
  • Reply 68 of 158
    asciiascii Posts: 5,774member
    sflocal said:
    Will be interesting to see how this all goes the next few months.  Will Apple suddenly (and quietly) make changes to the laptops to address this?  
    All they have to do is stop offering the i9 upgrade. The i7 (which is the default anyway) is fine.
    EddyD
  • Reply 69 of 158
    People are jumping to conclusions too quickly I think.
    These tests are using the available system that is not really the last and Apple may have already some changes in Mojave that revert the situation (better fans utilisation, better multicores....).

    I tested my MacPro 12 cores 2013 (last model, top) under Mojave and what I get are quite different scores respect to what you get with 10.13. Especially for multicores:
    the numbers are: 3563 for single core (just a little over the standard for this MacPro model, not significant), 30984 for multicores (instead of 26458 of the standard), this is significant.

    So, has someone done the same test on the new Macbook Pro (i7 and i9) using Mojave where clearly something has changed for multicores?

    Again, these real use (not for everyone) tests are using heavily the GPU (OpenCL etc.); as already pointed out, the thermal conditions will be different if you use the computation power of the i9 without the GPU, or not? So are you interested only in graphic performance or computation only with CPU (like Xcode compiling)?
    Results and expectations may be different.

    cgWerks
  • Reply 70 of 158
    orthorimorthorim Posts: 141member
    martinmr said:
    I think it's time to ditch Intel and pack a pair of A11 bionic in a MacBook Pro. 12 cores sucking just the power of two iphones. that would be interesting product to use. Also, what's with Adobe and not optimising premiere for mac... it's just absurd. Most of creatives that pay for adobe suite use macs.
    A11 TDP: 5W
    i7 TDP: 45W

    So you could actually stick 9 A11s in a Macbook Pro. 54 cores anyone??

    I mean I'd probably go for a nice power savings and only use 4 A11s, that'd still be 24 cores. 

    It's going to happen, only a matter of time. I would say it will happen very, very soon.Like 1-2 years max. The ARM chips are on par with x86 while consuming way less power. They're definitely already working on it. I imagine they'll come out with a special A-series processor that's optimized for laptop use. I believe the current phone processors have some super-low power cores, maybe a laptop doesn't need as many as those. 
  • Reply 71 of 158
    sflocalsflocal Posts: 4,149member
    nicw said:
    It’s not “form over function” ... it’s FETISH over function. 

    Apple tells us the iOS user interface won’t translate to the desktop environment. I don’t believe they are genuine in this; they just haven’t got a working model yet. But what I do think is true is that the hardware design philosophy for a hand held device will not translate to a desktop device. 

    Or to make it stupidly clear, thinness has become a problem for the Mac. 

    Stepping back for a second:

    Jobs like the rest of us was swept up in the early success of iOS. But after half a decade of missteps in desktop computing, it’s time to reassess the management reforms he introduced shortly before his untimely death.  There have been too many missteps in the desktop space. Give the desktop line back to the engineers. 
    *yawn*

    eh?!
  • Reply 72 of 158
    frantisekfrantisek Posts: 323member
    DuhSesame said:
    ahobbit said:
    The real question is, how could this have gotten past Apple QC?
    if simple tests as the ones done by Lee, AI and others reveal a potential throttling issue where the i9 is unable to maintain the advertised base frequency, even ending up slower in some cases than last year's i7 model, how could Apple not have known about this?

    Does it mean, Apple cares more about specs on paper than real world performance for pro users?

    Didn't Apple bring in-house many pro users to have them help design the next Mac Pro?
    Why were these people not involved in the MacBook Pro i9 testing?  Wouldn't that seem obvious?
    A wasted opportunity?  
    A waste of in-house resources?

    Apple more and more seems poorly managed...
    For all the money and resources they have, they should produce better results - if it is true that they still care about pro users, as they claim.

    Either they don't actually care as much as they claim - or cannot do any better than that.
    Both are very troubling if you are a pro user.
    Again, short-time bursts.  i9 will make things like opening applications or other small actions faster, but not long-term computing.

    The only thing that can provide maximum turbo boost for an i9 can only be those huge gaming laptops.
    Apple got trapped or being visctim of customers desire for powerfull MBP with 32 GB of memory. As is written above, some will get burst benefits and even some benefits from 2 extra cores but for long term benefits it will need eGPU I guess.I can offload some but not all CPU tasks.
  • Reply 73 of 158
    entropysentropys Posts: 1,158member
    orthorim said:
    martinmr said:
    I think it's time to ditch Intel and pack a pair of A11 bionic in a MacBook Pro. 12 cores sucking just the power of two iphones. that would be interesting product to use. Also, what's with Adobe and not optimising premiere for mac... it's just absurd. Most of creatives that pay for adobe suite use macs.
    A11 TDP: 5W
    i7 TDP: 45W

    So you could actually stick 9 A11s in a Macbook Pro. 54 cores anyone??

    I mean I'd probably go for a nice power savings and only use 4 A11s, that'd still be 24 cores. 

    It's going to happen, only a matter of time. I would say it will happen very, very soon.Like 1-2 years max. The ARM chips are on par with x86 while consuming way less power. They're definitely already working on it. I imagine they'll come out with a special A-series processor that's optimized for laptop use. I believe the current phone processors have some super-low power cores, maybe a laptop doesn't need as many as those. 
    It's possible.  It wouldn't be a bunch of A12s though, it would be a specifically designed arm based chip. A Mac1 it could be called.  That sid, if Apple went to the effort of doing a laptop and desktop class chip though, why not design an x86 stripped of all the legacy sludge that intel or AMD have to support, focussed on only being capable of running mac OS? it would be more efficient and cooler, and likely faster.  
    orthorim
  • Reply 74 of 158
    Mike WuertheleMike Wuerthele Posts: 3,108administrator
    elpopo64 said:
    People are jumping to conclusions too quickly I think.
    These tests are using the available system that is not really the last and Apple may have already some changes in Mojave that revert the situation (better fans utilisation, better multicores....).

    I tested my MacPro 12 cores 2013 (last model, top) under Mojave and what I get are quite different scores respect to what you get with 10.13. Especially for multicores:
    the numbers are: 3563 for single core (just a little over the standard for this MacPro model, not significant), 30984 for multicores (instead of 26458 of the standard), this is significant.

    So, has someone done the same test on the new Macbook Pro (i7 and i9) using Mojave where clearly something has changed for multicores?

    Again, these real use (not for everyone) tests are using heavily the GPU (OpenCL etc.); as already pointed out, the thermal conditions will be different if you use the computation power of the i9 without the GPU, or not? So are you interested only in graphic performance or computation only with CPU (like Xcode compiling)?
    Results and expectations may be different.

    Nothing changes in Mojave in these machines at present.
    cgWerks
  • Reply 75 of 158
    Mike WuertheleMike Wuerthele Posts: 3,108administrator

    mike54 said:
    Could you please mention what was your room temp when you tested this. 
    And can you test this at temps >90F (32C).

    If its throttling due to heat in an air conditioned environment its going to be much worse in higher room temperatures. Knowing room temp is important.



    21C. Guarantee its worse in a warmer environment.
  • Reply 76 of 158
    Mike WuertheleMike Wuerthele Posts: 3,108administrator
    For those anoyed with heating issue, just disable Turbo Boost. You can do it using another app like Turbo Boost Switcher or so. On redding there are even some posts reporting better performance with Turbo Boost disabled on certain tasks due to this issue.
    We started looking into this last night. With any luck, we'll get some numbers today on it.
  • Reply 77 of 158
    friedmudfriedmud Posts: 149member
    xiao-zhi said:
    This is not a new or unexpected problem. 

    The top top end processor for MBP or similar thin form factor notebooks is seldom the optimum performance solution in real terms due to thermal throttling and power consumption.

    Typically, these processors only provide short-term burst-mode performance advantages before they overheat and throttle

    Usually the best overall performance will be obtained with the middle processor choice combined with maximum memory.

    You can test the above to validate.

    This is not true of Macbook Pros in the past!  Let's not start making generalizations here!  I have the 2017 15" MBP with the top-end 3.1GHz i7-7920HQ that can turbo to 4.1 GHz... and under full load (using all 4 cores) it sits right around 3.7 GHz.

    Here's my issue with the current set of tests - we don't know what the instruction intensity looks like for these applications.  Intel's procs are REALLY good at throttling down if there is even the slightest bit of slack in instruction intensity.  It will show as dips in the frequency - but it's not because of thermal throttling... it's because the processor has a moment to "breathe"... so it does.  Absolutely the tiniest stall in the instruction pipeline (due to waiting on file-io or any other resource) will instantly bring the clock speed crashing down.

    We need a better benchmark here.  Can someone with one of these new MBPs copy this program into a file called thermal_test.C :

    #include <iostream>
    int main()
    {
      double stuff = 1.2;
      int junk = 2;

      while (true)
      {
        stuff *= 1.3;
        junk *= 2;
      }

      std::cout << stuff << std::endl;
      std::cout << junk << std::endl;

      return 0;
    }

    (A couple of notes about this program: It's designed to use both the floating point and integer units of the processor to really heat it up.  Also: the printing is just there to keep the compiler from optimizing the loop away)

    Then compile it using "clang++ thermal_test.C -o thermal_test"... that will generate a binary called "thermal_test"... then run it in 6 terminals simultaneously using: "./thermal_test".

    When I do this with 4 terminals on my current MBP (again with the 3.1GHz i7-7920HQ that can turbo to 4.1 GHz) here is what I get using Intel Power Gadget:



    As you can see - it goes right up to 3.7 GHz and is basically pegged there.  The temperature rises towards 100 C... but the fans kick in and keep it from ever thermal throttling.

    We need to run this (or something like it) on the new MBPs!
    muthuk_vanalingam
  • Reply 78 of 158
    Mike WuertheleMike Wuerthele Posts: 3,108administrator
    friedmud said:
    xiao-zhi said:
    This is not a new or unexpected problem. 

    The top top end processor for MBP or similar thin form factor notebooks is seldom the optimum performance solution in real terms due to thermal throttling and power consumption.

    Typically, these processors only provide short-term burst-mode performance advantages before they overheat and throttle

    Usually the best overall performance will be obtained with the middle processor choice combined with maximum memory.

    You can test the above to validate.

    This is not true of Macbook Pros in the past!  Let's not start making generalizations here!  I have the 2017 15" MBP with the top-end 3.1GHz i7-7920HQ that can turbo to 4.1 GHz... and under full load (using all 4 cores) it sits right around 3.7 GHz.

    Here's my issue with the current set of tests - we don't know what the instruction intensity looks like for these applications.  Intel's procs are REALLY good at throttling down if there is even the slightest bit of slack in instruction intensity.  It will show as dips in the frequency - but it's not because of thermal throttling... it's because the processor has a moment to "breathe"... so it does.  Absolutely the tiniest stall in the instruction pipeline (due to waiting on file-io or any other resource) will instantly bring the clock speed crashing down.

    We need a better benchmark here.  Can someone with one of these new MBPs copy this program into a file called thermal_test.C :

    #include <iostream>
    int main()
    {
      double stuff = 1.2;
      int junk = 2;

      while (true)
      {
        stuff *= 1.3;
        junk *= 2;
      }

      std::cout << stuff << std::endl;
      std::cout << junk << std::endl;

      return 0;
    }

    (A couple of notes about this program: It's designed to use both the floating point and integer units of the processor to really heat it up.  Also: the printing is just there to keep the compiler from optimizing the loop away)

    Then compile it using "clang++ thermal_test.C -o thermal_test"... that will generate a binary called "thermal_test"... then run it in 6 terminals simultaneously using: "./thermal_test".

    When I do this with 4 terminals on my current MBP (again with the 3.1GHz i7-7920HQ that can turbo to 4.1 GHz) here is what I get using Intel Power Gadget:



    As you can see - it goes right up to 3.7 GHz and is basically pegged there.  The temperature rises towards 100 C... but the fans kick in and keep it from ever thermal throttling.

    We need to run this (or something like it) on the new MBPs!
    This is why we used Cinebench. We know that there are no breaks in instruction intensity.
    prismatics
  • Reply 79 of 158
    orthorimorthorim Posts: 141member
    entropys said:
    It's possible.  It wouldn't be a bunch of A12s though, it would be a specifically designed arm based chip. A Mac1 it could be called.  That sid, if Apple went to the effort of doing a laptop and desktop class chip though, why not design an x86 stripped of all the legacy sludge that intel or AMD have to support, focussed on only being capable of running mac OS? it would be more efficient and cooler, and likely faster.  
    Agree they'd certainly develop a custom chip, even if it's just minor modifications to remove stuff that's specific to a mobile device. Single thread performance might also suffer somewhat, but nowadays most CPU hungry tasks are parallelized already. Even XCode is parallelized now! It's using all my 4 cores for Swift (and is absurdly slow but that's another topic... main reason I am looking for a lot of CPU horsepower for my next upgrade)

    However, x86 won't happen, that's too much of a mess. Engineering wise, once you have a mess, it's much more difficult to un-mess it than to use something that's not a mess. They could recompile the entire MacOS to a new processor architecture easily. Applications would have to be emulated in the beginning but in theory all that's needed is a new compile run in XCode and they're good to go. 
  • Reply 80 of 158
    friedmudfriedmud Posts: 149member
    For those anoyed with heating issue, just disable Turbo Boost. You can do it using another app like Turbo Boost Switcher or so. On redding there are even some posts reporting better performance with Turbo Boost disabled on certain tasks due to this issue.
    We started looking into this last night. With any luck, we'll get some numbers today on it.
    Mike: can you guys please try my simple program above?  We need more controlled testing on this than hitting "Run" in a rendering application.

    This is important to me because my team is getting ready to buy 5 of these new machines.  We do massively parallel scientific computing and use every single ounce of capability out of our laptops workstations... so if this thing really is thermal throttling then I need to know about it!
    muthuk_vanalingam
Sign In or Register to comment.