Of course they throttle at high enough temps. When did I ever say they don't?
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Ramifications of the MBP's very high temps (or, a means to this topics end)
- Thread starter LurkinTuff
- Start date
- Sort by reaction score
When I had an overheating problem in my Early-2011 MBP it was actually a problem, would get to 98c after 5 seconds of being on, with nothing running.
Compared to a bench model at the store, which got to 60c after 2 minutes of being on...
So yes, some do, but most don't.
Compared to a bench model at the store, which got to 60c after 2 minutes of being on...
So yes, some do, but most don't.
If it is throttling then it is overheating.
They are all throttling.
That's like advertising that the car can do 205mph but it is actually electronically limited to 155 because the tires installed at the factory will blow up at 205mph.
No, it throttles to prevent overheating. If it overheats, it shuts down. The problem is not how Apple portables handle heat. The problem is some people's expectations. Every computer has operating limits that you can't exceed. Those who are continuously gaming or running other apps that pushes an MBP to the limits should have chosen a computer better suited to their particular needs. No computer made by any company is expected to meet the needs of every user. It's up to the user to pick the right machine for the workload they intend to put on it. If you buy a MBP expecting to treat it as a high-end gaming machine, you made the wrong choice.
Even a car that can reach 205mph is not designed to run at that speed continuously. If that's your intention, your expectations are wrong.
One comment on the throttling. I looked up two reviews (links below) of the early 2011 15'' base and high end models.
Both machines showed throttling in a benchmark scenario that puts extreme load on both the GPU and CPU at the same time. This is a highly artificial situation that puts the system under extreme stress. Still the MBPs could handle it without showing signs of failure.
In a more realistic test, using Battlefield Bad Company 2, the high end model (2.2 GHz) remained stable at 90C, without CPU throttling. The base model apparently showed some throttling in some games, but it remained unclear whether it was due to high temperatures or due to an insufficient power supply. Note also that this is maybe not completely unexpected, since the early 2011 base model has a very weak CPU, compared to the other components.
In my opinion this shows that the MBP can operate at it's full potential under most conditions. There is no fundamental design problem, but there are some limits since this is a laptop with limited amount of space for all components.
Here are the two tests I'm referring to:
http://www.notebookcheck.net/Review...-GHz-quad-core-glare-type-screen.50344.0.html
http://www.notebookcheck.net/Review...1-2-0-GHz-Quad-Core-Matte-Screen.49689.0.html
Both machines showed throttling in a benchmark scenario that puts extreme load on both the GPU and CPU at the same time. This is a highly artificial situation that puts the system under extreme stress. Still the MBPs could handle it without showing signs of failure.
In a more realistic test, using Battlefield Bad Company 2, the high end model (2.2 GHz) remained stable at 90C, without CPU throttling. The base model apparently showed some throttling in some games, but it remained unclear whether it was due to high temperatures or due to an insufficient power supply. Note also that this is maybe not completely unexpected, since the early 2011 base model has a very weak CPU, compared to the other components.
In my opinion this shows that the MBP can operate at it's full potential under most conditions. There is no fundamental design problem, but there are some limits since this is a laptop with limited amount of space for all components.
Here are the two tests I'm referring to:
http://www.notebookcheck.net/Review...-GHz-quad-core-glare-type-screen.50344.0.html
http://www.notebookcheck.net/Review...1-2-0-GHz-Quad-Core-Matte-Screen.49689.0.html
Heat wears out electronics more quickly.
Keep temps low.
The lower they remain, the longer the CPU will run.
All things being equal, out of the box, used for general purposes (email, browsing, word processing) and graphic design (photoshop, dreamweaver, etc) -- anything that doesn't use the CPU 24/7 (e.g. 3D rendering, video production) will ensure the lifespan of this model remain ~5 years.
I'm more worried about the AC adapter prematurely failing. It uses 85W on a machine that needs over 90W. My AC adapter gets burning hot to the touch when I'm using my MBP. The adapter is being pushed to the limit and, yeah, the blistering hot temperature makes it a fire hazard.
Since these computers cost a lot of money, it is fair to say we want to get every bit of use out of these things, and it is fair to say we expect a $2000+ device to be made with quality and care.
Unless you plan on using your computer for decades, this is absolutely meaningless. Too many people are confused about the practical implications of the heat non-issue because of people like you who don't understand what they're talking about.
The problem is that it is a 'pro' machine, you can easily bump up against the thermal limits while encoding. If you're implying that the 'pro' part isn't really pro then I would agree with you.
As for the other part you understand my point, do you not?
I road race 600cc sportbikes, I can leave one stock engine wise and pretty much hammer it for thousands, if not tens of thousands of miles, without blowing it up.
I have read somewhere that you should keep the CPU about 15*C under the Tjunction.
I also know that its not just absolute heat, but drastic changes in heat that really stress the chip.
Without trying to stress my CPU, (and with no little to no dust on the fans/heatsinks/vents) during normal usage my temps rise and fall from about 50*C to 97*C. By normal use, I mean running LightRoom 3 editing and adjusting 300Mb tiffs on an external display.
If Im just editing, my temps are about 70*, whereas if Im adjusting a large file, they get up into the high 90's. The highest Ive seen is 97*, however, it is completely usual to see temps of 95*.
I also know that its not just absolute heat, but drastic changes in heat that really stress the chip.
Without trying to stress my CPU, (and with no little to no dust on the fans/heatsinks/vents) during normal usage my temps rise and fall from about 50*C to 97*C. By normal use, I mean running LightRoom 3 editing and adjusting 300Mb tiffs on an external display.
If Im just editing, my temps are about 70*, whereas if Im adjusting a large file, they get up into the high 90's. The highest Ive seen is 97*, however, it is completely usual to see temps of 95*.
Yes, "Pro" is a marketing term. If a doctor, who is a professional, uses a Mac mini, does that make it a "Pro?" If a truck driver uses a Mac Pro, does that mean it's no longer a "Pro?"
Yes, I understand your point, but my point is that a device that is built with operational limits is rarely designed to operate at those limits for extended periods of time.
Do you have any evidence of this? Encoding should only stress the CPU, not the GPU. The reports about throttling that I have read only observe it when they artificially put extreme stress on both CPU and GPU.
Besides, this is still a freaking laptop. If the only purpose of your computer is to encode videos for extended periods of time, you bought the wrong computer. Although it's probably still doing a decent job.
Unless you plan on using your computer for decades? Really? Ive heard/seen many MBPs fail, and macbook pros have not been around for decades. By fail, I mean the logic board fails, at which point its practically totalled because of repair cost. Are you saying that lower peak and average temperatures (lets say ~8*C) wouldn't have meant a significantly longer lifespan?
Are you aware that people bake their dead MBP's back to life, in order to resolder a connection thats become brittle/broken from excessive heat/ thermal cycles?
That's great. Unless you can clearly attribute these failures directly to "heat", it's not relevant.
What does "significant" mean? If a lower average temperature adds 10 years on to the end of the 50-year lifespan of the electronics in your machine, why does it even matter? If a higher average temperature subtracts 10 years from a 50-year total lifespan, why does it matter? Computers last a typical person 3-5 years.
No, I am not aware of this. In fact, I have never seen "baking" a MBP logic board (i.e. reflowing the solder) suggested in 5 years of forum membership and more than a decade of repairing and servicing Apple machines.
But hey, you might be right.
Also, have you seen the amount of TIM applied to the CPU/GPU on these things? Given the case design and the amount of heat these chips produce, properly applying TIM is, at the very least, a good idea. If you want a machine thats a reliable as it is costly, keeping the temps just a few degrees cooler can go a long way. And by long way I mean an extra year or two or three.
----------
So you dont think that these failures (where reflowing the solder helps) are caused by excessive heat?
What do you suggest the cause is?
----------
So you dont think that these failures (where reflowing the solder helps) are caused by excessive heat?
What do you suggest the cause is?
Again, I ask you: what is the point of extending the life of a computer by 2 or 3 years when it's already going to last for 20+ if you wanted it to? It's completely pointless.
First you could point out one of these cases you're talking about in which solder reflow actually solved a problem. I have seen zero evidence that this is widespread. If you are going to make the case that the "high temperatures" experienced when the machine is under load is causing actual problems with the solder on the logic board, please provide evidence. Isolated issues with electronic components out of production runs numbering into the millions are to be expected, which is probably what you're referring to.
So, youve been repairing apple machines for a decade. You claim that the lifespan of a MBP is closer to 50 years than 5.
In all your years of loyal service to Apple machines, were the only notebooks with any damage or failure that could possibly be attributed to excessive temps the 40 and 50 year old notebooks?
In all your years of loyal service to Apple machines, were the only notebooks with any damage or failure that could possibly be attributed to excessive temps the 40 and 50 year old notebooks?
If you want a Mac or any other computer to last, it can last decades.
I have not repaired a 50-year old notebook, because 50-year old notebooks do not exist. Please ask intelligent questions, or don't ask questions at all.
Still waiting for one example of solder reflow that solved a problem as well.
Edit: I googled "solder reflow macbook pro" and found numerous instances of reflowing the solder "fixing" something. However, I have not found any evidence that sustained heat was the cause of the problem that reflow fixed. I accept that reflowing the solder with a heat gun or reflow station will work, but still do not accept that sustained operating temperatures are directly attributable to a widespread, specific issue.
Last edited:
Yeah thats called being sarcastic.
My point is you're saying that the lifespan of a MBP is closer to several decades than several years. Im sure it will last quite a while if I rarely use it and never stress it. However, I do use it, and it reaches into the high nineties during normal use on a regular basis.
I am neither abusing nor neglecting my MBP in order to reach these temps. You can project all you want, but the fact is the oldest MBP is... like 5 or 6 years old?
The fact is this, no matter what: Higher temps and larger differences in ambient temp vs chip temp and larger differences between idle and load temps always equal less time before failure.
Running at 5* under the Tjunction on a regular basis is not conducive to a long lived computer. And this happens without neglect or abuse.
read some of the comments...
http://russell.heistuman.com/2010/04/27/cooking-the-books-or-baking-my-macbook-pro-logic-board/
And 5 or 6-year old MBPs still work perfectly fine.
That's not the disagreement. The disagreement is about whether or not the decreased MTBF is going to affect your machine before you buy a new one, to which the answer is almost certainly "no."
If the computer is having to actively limit performance to prevent itself from overheating, something is wrong. Apple sold me a 2.2ghz quad core cpu, if it runs at 1.2ghz under load... what is that?
I should be able to push any computer to 100% and have it run at the performance level I paid for. People paid for a 2ghz cpu not one that runs at 1ghz when the speed is actually needed.
I don't expect the computer to last forever but if it throttles or shuts off, it means the computer is at risk of immediate damage. That is not acceptable.
I agree with most of the comments here.
My current machine, a fully maxed out 2.5GHz i7 15", was an AppleCare replacement for my custom configured Mid-2010 2.4GHz i5 MacBook Pro, with 8GB, 128GB SSD, and Antiglare.
If i had the choice, I would have kept my old unit. This one runs much hotter, when connected to my 27" Cinema Display, the fans are almost on all the time, and the unit feels very hot to the touch. On my old unit, the fans were never audible, unless I was using my web development tools, or playing a game.
On this new one, I also can only get about 4 hours of battery life, with low brightness, word open, and light safari browsing. My old machine got 7.5-8.
Also, I don't mean to "thread jack", but I took the RAM from my old machine and put it in my new one. It's 1067MHz DDR3, I know this machine is supposed to have at least 1333Mhz. Would this have an adverse affect on my battery life, and thus explain the lower battery run time I am experiencing? iStat pro reports I have 9 charge cycles with 100% health, meanwhile my old machine had 98% and over 200 charge cycles on the battery, so I don't think this has anything to do with the battery itself. The graphics switching was enabled in both machines.
My current machine, a fully maxed out 2.5GHz i7 15", was an AppleCare replacement for my custom configured Mid-2010 2.4GHz i5 MacBook Pro, with 8GB, 128GB SSD, and Antiglare.
If i had the choice, I would have kept my old unit. This one runs much hotter, when connected to my 27" Cinema Display, the fans are almost on all the time, and the unit feels very hot to the touch. On my old unit, the fans were never audible, unless I was using my web development tools, or playing a game.
On this new one, I also can only get about 4 hours of battery life, with low brightness, word open, and light safari browsing. My old machine got 7.5-8.
Also, I don't mean to "thread jack", but I took the RAM from my old machine and put it in my new one. It's 1067MHz DDR3, I know this machine is supposed to have at least 1333Mhz. Would this have an adverse affect on my battery life, and thus explain the lower battery run time I am experiencing? iStat pro reports I have 9 charge cycles with 100% health, meanwhile my old machine had 98% and over 200 charge cycles on the battery, so I don't think this has anything to do with the battery itself. The graphics switching was enabled in both machines.
Take it up with Intel.
In most scenarios throttling is limited to reducing the turbo multiplier i.e you are still running at 2.2ghz when under heavy CPU load, just not 2.8ghz at maximum turbo the whole time. So you are getting the speed that you pay for almost all the time, if you want to be belligerent.
Of course, there is a scenario where your chip hits the temp limit of 100 deg C (roughly, can't be bothered to look it up) where it would then enter a harsher throttle state below stock speed. This state only usually arises if vents are blocked, heatsink filled with dust to prevent damage.
You can download software to run under boot camp to monitor throttle states if you are interested in factual data.
It's the Intel chip that throttles performance or shuts down, not the computer.
Name one notebook computer that can run at maximum limits continuously without issues. They're not designed to run at 100% continuously. If you expect that, you bought the wrong computer. You shouldn't be buying a computer that needs to run at 100% to handle your normal workload. You should buy one that can handle that workload at something like 60%.
So you'd rather it didn't throttle or shut off and simply burn up? Again, if you don't like the throttling or shutting down to prevent damage, talk to Intel, not Apple.
I agree with this. Most CPU specs are produced in absolutely ideal conditions, with more than adequate cooling, and no dust whatsoever. (Not to mention thermal paste that is applied correctly, unlike every MacBook Pro I've ever been inside of) Just as with battery life, mileage varies, and is never really what is actually quoted. Maybe if the MacBook pro was 3" thick, with triple the fans, large heatsinks, and a power adapter that produces double the wattage you might see close to the promised performance. However, I don't think anyone would quite like the idea of carrying around a thick deadweight "laptop". Look at the top of the line Alienware laptops for example, (Although it's a bad one) they are basically what I just described. Very thick, lots of fans for cooling, unbelievably heavy, and they use quite a bit of power. I'd take my MBP over one of those any day.
There are actually some upcoming changes on the intel end that could make a cooler laptop workable. It's just that popular opinion is that Apple will thin it down more like the macbook air. If you look at the weight comparison, the 13" air is 2.96 lb whereas the pro is 4.5 lb. The ODD and HDD should come out to at least half a pound saved there with the port reduction possibly cutting a couple ounces. The 15" which is often what people owned previous to the air, comes in at 5.6 lb. A 15" air would still be light, but you'd notice the difference from the ones out currently.
Intel is aiming to get their mainstream laptop cpus close to what the Air is at now in tdp around 2013. That would be a huge step toward quieter machines that aren't forced to throttle down assuming the enclosure remains the same. The problem is I think Apple wants the line to look like the macbook air does currently even though you could still retain significant weight reductions without going to the ultra thin design.
Just my thoughts on it...