Thermal test of iPad's A5X chip shows operating temperature of 97 degrees F

1235

Comments

  • Reply 81 of 106
    jnjnjnjnjnjn Posts: 588member
    Quote:
    Originally Posted by jragosta View Post


    First, I don't believe your numbers.



    More importantly, reducing it to 11 nm would also reduce the heat generation - but 11 nm isn't ready either. What's the point of talking about something that isn't available.



    Oh, and btw, some reports suggest that a lot of the heat is coming from the CPU, but a good bit also comes from the battery. If you discharge a battery quickly, it heats up.



    It isnt matter of belief. It's common knowledge that heat dissipation of a chip is directly proportional to its surface (look it up if you don't believe it).

    So if you reduce the feature size from 45nm to - say - 22nm, the chip shrinks in its two dimensions and has a quarter of its original size, so it dissipates four times less. You can calculate the 45nm versus 32nm case for yourself (it should be around 2.8).

    11nm isn't available yet, your right about that.

    But I reacted on the assertion that 45nm versus 32nm wouldn't make much of a difference.

    And that isn't true.

    Samsung and others are able to make SoCs with a feature size of 32nm even 22nm if you look at memory chips.

    So it isn't that strange to make the comparison.



    J.
  • Reply 82 of 106
    myapplelovemyapplelove Posts: 1,515member
    Quote:
    Originally Posted by VanFruniken View Post


    My point is the following.



    If the iPad 3's external housing gets hotter, it may be partly because of better heat transfer from the processor to the external world. Something already mentioned elsewhere, namely heat produced by other parts such as LED backlights, is not directly contributing to peak temperatures in the processor.



    Further, given that the A5X has a metal housing, it may also be evacuating the heat better than the A4's plastic housing.



    Also, opening up the case may not give a representative idea of the processor temperature, because of different heat conduction to the outer shell.



    The only reliable measurement would be to access the on-chip temperature sensors.

    Maybe there would be no FUD articles to publish if it would actually turn out that the actual dye temperature in the iPad3's processor is hardly different from that in the previous models. Would such a realization put CR, the NYT and others to shame? Probably not, they wouldn't even admit their unscientific mistake. They would simply start looking at something else to pick on.



    There's tons of ways of rationalising an issue away, the CPU of the new iPad is apparently getting warmer and warming the iPad more due to its increased demands and the 45nm manufacturing process, that's the most sound conclusion and the most obvious one. Is an iPad not charging while plugged also some shady ploy by cr or ny times against apple? Come on boys, apple wanted to be first in the market with retina display and they are taking a hit with heat issues and charging issues because of the incremental update of the CPU and the added gfx cores on an already very high by today's standards manufacturing process. Apple are to be commended for pushing mainstream tech with retina, but as a rev.1 high tech product the new iPad has its issues apparently.
  • Reply 83 of 106
    addaboxaddabox Posts: 12,665member
    Quote:
    Originally Posted by myapplelove View Post


    There's tons of ways of rationalising an issue away, the CPU of the new iPad is apparently getting warmer and warming the iPad more due to its increased demands and the 45nm manufacturing process, that's the most sound conclusion and the most obvious one. Is an iPad not charging while plugged also some shady ploy by cr or ny times against apple? Come on boys, apple wanted to be first in the market with retina display and they are taking a hit with heat issues and charging issues because of the incremental update of the CPU and the added gfx cores on an already very high by today's standards manufacturing process. Apple are to be commended for pushing mainstream tech with retina, but as a rev.1 high tech product the new iPad has its issues apparently.



    The iPad doesn't have an "issue" with "not charging while it's plugged in" in general, just when its being used at the most power consuming level possible. So that's only an "issue" if you go way out of your way.



    And what about any of this is "rationalizing"? More power hungry parts, gets slightly warmer. The end.
  • Reply 84 of 106
    myapplelovemyapplelove Posts: 1,515member
    Quote:
    Originally Posted by addabox View Post


    The iPad doesn't have an "issue" with "not charging while it's plugged in" in general, just when its being used at the most power consuming level possible. So that's only an "issue" if you go way out of your way.



    And what about any of this is "rationalizing"? More power hungry parts, gets slightly warmer. The end.



    Oh and I thought it had an issue with charging when plugged in, in general, and I was asking myself how do they manage to keep the power without it ever charging when plugged in, some pretty amazing a engineering there...



    Yeah I guess we shouldn't go out of our way then and use power consuming apps, and we should also be checking -how we ll figure it out- which apps are power hungry so we can have our iPads charge too... Boy I'd love to be in aline of work with a critical element to ot, say a pilot and have my iPad plugged and running a demanding app only to see it drain away and not charge.



    What a load of horse manure, the level of talk from some of you guys here are making this site, which is otherwise great in the level of posting by most people, unbearable, ignore list ASAP for me.
  • Reply 85 of 106
    jragostajragosta Posts: 10,473member
    Quote:
    Originally Posted by jnjnjn View Post


    It isnt matter of belief. It's common knowledge that heat dissipation of a chip is directly proportional to its surface (look it up if you don't believe it).

    So if you reduce the feature size from 45nm to - say - 22nm, the chip shrinks in its two dimensions and has a quarter of its original size, so it dissipates four times less. You can calculate the 45nm versus 32nm case for yourself (it should be around 2.8).

    11nm isn't available yet, your right about that.

    But I reacted on the assertion that 45nm versus 32nm wouldn't make much of a difference.

    And that isn't true.

    Samsung and others are able to make SoCs with a feature size of 32nm even 22nm if you look at memory chips.

    So it isn't that strange to make the comparison.



    J.



    I love the way people fall back on 'common knowledge'.



    Please show some evidence that a die shrink from 45 nm to 32 nm would reduce the power usage by 2/3.



    In reality, a die shrink does reduce power usage somewhat, but nowhere near a 2/3 reduction. Have a look at comparable chips with different line widths:



    http://en.wikipedia.org/wiki/List_of...er_dissipation
  • Reply 86 of 106
    pendergastpendergast Posts: 1,358member
    Quote:
    Originally Posted by jragosta View Post


    As explained in a different thread, you are incorrect.



    You should be measuring heat gain while in operation. If the iPad 2 is at 70 degrees when turned off and 80 degrees when in use, that's a 10 degree heat gain from operation. If the iPad 3 is at 70 degrees when turned off and 90 degrees when turned on, that's a 20 degree heat gain from operation - or 100% more.



    Of course, neither one is a significant problem and there are no reports of any problems caused by it. Furthermore, there are at least 4 or 5 different reports of external temperatures and none of them come close to Consumer Reports' figures. Most of them put iPad 3 temperatures in heavy use into the 90's, not 116.



    The post you quoted is not incorrect. It is simply discussing a different type of statistic. You're referring to the percentage increase in heat generated in use compared to the previous model. The other poster is referring to measuring the actual heat at peak.



    The statistic you're using can be very misleading and lead to headlines like "New iPad gets twice as hot as iPad 2".



    The other statistic shows that, regardless of starting temperature, the new iPad runs only a few percentage points hotter than the iPad 2. At the end if the day, isn't that more in line with reality?



    As the old adage goes, there's lies, damn lies, and statistics.
  • Reply 87 of 106
  • Reply 88 of 106
    muppetrymuppetry Posts: 3,331member
    Quote:
    Originally Posted by Pendergast View Post


    The post you quoted is not incorrect. It is simply discussing a different type of statistic. You're referring to the percentage increase in heat generated in use compared to the previous model. The other poster is referring to measuring the actual heat at peak.



    The statistic you're using can be very misleading and lead to headlines like "New iPad gets twice as hot as iPad 2".



    The other statistic shows that, regardless of starting temperature, the new iPad runs only a few percentage points hotter than the iPad 2. At the end if the day, isn't that more in line with reality?



    As the old adage goes, there's lies, damn lies, and statistics.



    A lot of these discussions really do illustrate that even simple physics is a moderately complicated subject, and that just failing to appreciate the difference between steady and unsteady heat flow is enough to derail an argument completely.



    The simplest analyses are often the best, and in this case, it doesn't get much simpler than the observation that the energy content of the new iPad's battery is nearly twice that of the old one, but that it's nominal battery life is the same. That does indeed mean nearly twice the power dissipation, and since virtually all of that energy is dissipated as heat (I'm ignoring emitted light and sound here), and since all heat loss mechanisms depend in some way on differential temperature (relative to the environment), the device, if otherwise physically similar, must run hotter.



    How much hotter depends on the environment, and since the primary cooling mechanisms are going to be conduction and convection, it will be sensitive to factors affecting those processes. I doubt the various tests are controlling those very tightly.
  • Reply 89 of 106
    pine45pine45 Posts: 1member
    I do not have a problem with my ipad 3. I like your comment about getting the heat out My only concern is what kind of cover would be the best if someone is concerned with the heat. I used a leather type cover.

    Another thing is if there is an app to monitor temp?
    Quote:
    Originally Posted by VanFruniken View Post


    My point is the following.



    If the iPad 3's external housing gets hotter, it may be partly because of better heat transfer from the processor to the external world. Something already mentioned elsewhere, namely heat produced by other parts such as LED backlights, is not directly contributing to peak temperatures in the processor.



    Further, given that the A5X has a metal housing, it may also be evacuating the heat better than the A4's plastic housing.



    Also, opening up the case may not give a representative idea of the processor temperature, because of different heat conduction to the outer shell.



    The only reliable measurement would be to access the on-chip temperature sensors.

    Maybe there would be no FUD articles to publish if it would actually turn out that the actual dye temperature in the iPad3's processor is hardly different from that in the previous models. Would such a realization put CR, the NYT and others to shame? Probably not, they wouldn't even admit their unscientific mistake. They would simply start looking at something else to pick on.



  • Reply 90 of 106
    jragostajragosta Posts: 10,473member
    Quote:
    Originally Posted by pine45 View Post


    I do not have a problem with my ipad 3. I like your comment about getting the heat out My only concern is what kind of cover would be the best if someone is concerned with the heat. I used a leather type cover.

    Another thing is if there is an app to monitor temp?



    The best cover would be something like the SmartCover - since the heat is mostly dissipated through the back of the iPad. You don't want a cover that covers the metal part.
  • Reply 91 of 106
    jnjnjnjnjnjn Posts: 588member
    Quote:
    Originally Posted by jragosta View Post


    I love the way people fall back on 'common knowledge'.



    Please show some evidence that a die shrink from 45 nm to 32 nm would reduce the power usage by 2/3.



    In reality, a die shrink does reduce power usage somewhat, but nowhere near a 2/3 reduction. Have a look at comparable chips with different line widths:



    http://en.wikipedia.org/wiki/List_of...er_dissipation



    First of all, I am not 'people'. Second I will rephrase my statement: to the best of my knowledge chip dissipation is directly proportional to its surface area (*). I thought that was common knowledge like Moorse law or the qwadratic surface area to speed relation for wind resistance and the frequency doubling when the feature size of a chip shrinks to half and so on.

    It could be that I am mistaken about the common knowledge, you don't seem to have it you claim.

    I know the chip industry for about 40 years and have seen the trends, and that's the source of my knowledge.

    Further, your wiki link is a disputed article and not in any way clear about dissipation. And most important of all it doesn't have information about the most relevant processor in this discussion: the ARM processor. So you have to come up with something better.



    J.



    (*) take care that I stated that the frequency of the cores and the number of cores (and so on) must be the same for a correct compare between chips. Note that most of the time feature size shrinks and the frequency and transistor count increases (this is also common knowledge), so you have to account for that when you compare actual chips.
  • Reply 92 of 106
    jragostajragosta Posts: 10,473member
    Quote:
    Originally Posted by jnjnjn View Post


    First of all, I am not 'people'. Second I will rephrase my statement: to the best of my knowledge chip dissipation is directly proportional to its surface area (*). I thought that was common knowledge like Moorse law or the qwadratic surface area to speed relation for wind resistance and the frequency doubling when the feature size of a chip shrinks to half and so on.

    It could be that I am mistaken about the common knowledge, you don't seem to have it you claim.

    I know the chip industry for about 40 years and have seen the trends, and that's the source of my knowledge.

    Further, your wiki link is a disputed article and not in any way clear about dissipation. And most important of all it doesn't have information about the most relevant processor in this discussion: the ARM processor. So you have to come up with something better.



    J.



    (*) take care that I stated that the frequency of the cores and the number of cores (and so on) must be the same for a correct compare between chips. Note that most of the time feature size shrinks and the frequency and transistor count increases (this is also common knowledge), so you have to account for that when you compare actual chips.



    Your statement is wrong - even after you rephrased it.



    The A5 and A5x are not the same processor. There are twice as many graphics cores in the A5X.



    Saying that heat dissipation is proportional to chip area is meaningless when you're comparing different chips.
  • Reply 93 of 106
    jnjnjnjnjnjn Posts: 588member
    Quote:
    Originally Posted by jragosta View Post


    Your statement is wrong - even after you rephrased it.



    The A5 and A5x are not the same processor. There are twice as many graphics cores in the A5X.



    Saying that heat dissipation is proportional to chip area is meaningless when you're comparing different chips.



    Reread my posts.



    J.
  • Reply 94 of 106
    addaboxaddabox Posts: 12,665member
    Quote:
    Originally Posted by myapplelove View Post


    Oh and I thought it had an issue with charging when plugged in, in general, and I was asking myself how do they manage to keep the power without it ever charging when plugged in, some pretty amazing a engineering there...



    Yeah I guess we shouldn't go out of our way then and use power consuming apps, and we should also be checking -how we ll figure it out- which apps are power hungry so we can have our iPads charge too... Boy I'd love to be in aline of work with a critical element to ot, say a pilot and have my iPad plugged and running a demanding app only to see it drain away and not charge.



    What a load of horse manure, the level of talk from some of you guys here are making this site, which is otherwise great in the level of posting by most people, unbearable, ignore list ASAP for me.



    Yes, you're probably right. If only more people on these boards simply jumped on whatever grindingly predictable Apple "scandal" happens to be consuming the insect like attention span of teh internet these last 5 minutes or so we could have much more productive conversations. I, too, long for the days when this board was largely comprised of grimly enumerating all the ways Apple sucks. Good times.



    Only a raving fanboy would assume that making a big deal over slightly warmer case temperatures when the device is being driven flat out isn't a huge issue that needs to be dissected ad infinitum.



    I think we can safely assume that any charge against Apple, no matter how slight or unfounded, must be given the utmost attention, because "fanboys" will protest.
  • Reply 95 of 106
    relicrelic Posts: 4,735member
    Quote:
    Originally Posted by addabox View Post


    I think we can safely assume that any charge against Apple, no matter how slight or unfounded, must be given the utmost attention, because "fanboys" will protest.



    Really, have you read some the of the posts around here, the iPad can be spitting out fire from the speaker grill and the, "fanboys" will find every excuse to down play it as much as possible.



    The iPad does get a little on the warm side, especially if you play a game that is heavy on the graphics, I happen to like Shadowgun and Sims 3. I've been using my iPad extensively this past week because I'm stuck in a hospital bed for foreseeable future and I find my self shutting down the iPad between 2 - 3 times a day because it becomes a little to uncomfortable to hold.



    It's not horrible in any sense but I'm also afraid of melting something, it's a new machine. Like I said though I have been using it more then a average person would, do to my situation. I would imagine that most people would probably never get it to the point where they feel they need to shut it down to cool off.
  • Reply 96 of 106
    solipsismxsolipsismx Posts: 19,566member
    Quote:
    Originally Posted by Relic View Post


    I'm also afraid of melting something, it's a new machine.



  • Reply 97 of 106
    addaboxaddabox Posts: 12,665member
    Quote:
    Originally Posted by Relic View Post


    Really, have you read some the of the posts around here, the iPad can be spitting out fire from the speaker grill and the, "fanboys" will find every excuse to down play it as much as possible.



    Tell you what: let's say we wait till Apple ships a product that actually has a problem comparable to spitting fire from the speaker grill, and if anyone defends that I'll join you in being dismayed.



    As it stands, we have a long sorry history of ginned up "gates" that don't amount to much but for their moment in the sun are brandished as proof that fanboys think Apple can do no wrong.



    Myself? I see way more evidence that some people will seize on pretty much anything to portray Apple as coming up short, than I do evidence of these crazed, never-say-die fanboys. Just in general, if you find that kind of thing off-putting, maybe an Apple enthusiast site isn't the best place to hang out? The internet is full of tech sites where being an Apple fan is like declaring yourself a child molester.
  • Reply 98 of 106
    relicrelic Posts: 4,735member
    Quote:
    Originally Posted by SolipsismX View Post






    Oh come off it, I'm not taking any chances, My iPad hasn't been turned off more then 20 minutes since I got it last Thursday, I've played games on it for like 4 hours not stop and there were times I had to prop it up on pillows because it was to uncomfortable to hold. So excuse me if I'm just a little concerned about my new toy. I don't want to be in a situation where I won't have it while I'm in the hospital. It's bad enough I'm fighting this horrible affliction it would be even worse If I didn't have my iPad with me while doing it, your kind of a mean spirited sometimes SolipsismX, oh look those kids are back on you lawn again.
  • Reply 99 of 106
    solipsismxsolipsismx Posts: 19,566member
    Quote:
    Originally Posted by Relic View Post


    Oh come off it, I'm not taking any chances, My iPad hasn't been turned off more then 20 minutes since I got it last Thursday, I've played games on it for like 4 hours not stop and there were times I had to prop it up on pillows because it was to uncomfortable to hold. So excuse me if I'm just a little concerned about my new toy. I don't want to be in a situation where I won't have it while I'm in the hospital. It's bad enough I'm fighting this horrible affliction it would be even worse If I didn't have my iPad with me while doing it, your kind of a mean spirited sometimes SolipsismX, oh look those kids are back on you lawn again.



    I just stupid comments. Saying you're afraid it's going to melt because it's warm to the touch is obnoxious. You sound like a troll who you is claiming they have Apple products so they can keep trying to point out how flawed it is.



    I've been using notebooks for about 15 years now. At times they get uncomfortably warm when sitting on my lap. Never once have I stated that I choose not to use them because I'm afraid of it melting.



    If you aren't trolling and actually think it could melt then replace it because melting is a clear sign that it's a defect. It's also a safety hazard.
  • Reply 100 of 106
    relicrelic Posts: 4,735member
    Quote:
    Originally Posted by addabox View Post


    Tell you what: let's say we wait till Apple ships a product that actually has a problem comparable to spitting fire from the speaker grill, and if anyone defends that I'll join you in being dismayed.



    As it stands, we have a long sorry history of ginned up "gates" that don't amount to much but for their moment in the sun are brandished as proof that fanboys think Apple can do no wrong.



    Myself? I see way more evidence that some people will seize on pretty much anything to portray Apple as coming up short, than I do evidence of these crazed, never-say-die fanboys. Just in general, if you find that kind of thing off-putting, maybe an Apple enthusiast site isn't the best place to hang out? The internet is full of tech sites where being an Apple fan is like declaring yourself a child molester.



    Yea your probably right, Appleinsider has become more of a bitch session as of late. Here we have a new Apple toy to play with and the only threads under the iPad part are all negative. I get caught up into to, it's kind of hard not to when there is nothing else happening. Everytime I start a new thread that asks what new things people are doing no one replys and it dies faster then when it was started. I need to find an Apple site where they just discuss how to be more creative with their tablets. Something like XDA but just for Apple would be cool, oh www.hackint0sh.org , I completely forgot about them, their a fun bunch.
Sign In or Register to comment.