Posted on 04/27/2010 2:41:44 AM PDT by Swordmaker
"Laptop design is largely governed by heat management. There are a few key components inside a laptop that run quite hot. The CPU and GPU are perhaps the most important of these, as they focus huge energy output into a tiny square of silicon. One way that notebook designers have faced this challenge is through the use of metal laptop bodies," Zara Baxter and John Gillooly report for PC Advisor.
"This works to a degree, but the unfortunate side effect is that when things run really hot the entire laptop body heats up to an uncomfortable level. This can quite easily shift a laptop from being pleasantly warm in one's lap to something that is dangerous to one's health," Baxter and Gillooly report. "Such situations are why product marketers made a concerted effort in the early half of this decade to shift the naming of such devices from laptop to notebook."
Baxter and Gillooly report, "During testing of the latest Core i7-620M based 17in Macbook Pro we noticed that there were problems running certain tests in our benchmark suite. The score being spat out for the Photoshop tests - fourth in a suite of six test applications - in particular was quite low, and we wondered whether it was down to heat issues. When the test was then successfully run as a standalone test with the Macbook sitting on its side, unibody base exposed to the air, we suspected that the Core i7 was struggling within the Aluminium shell of the Macbook."
MacDailyNews Note: Just a bit of clarification: The MacBook Pro was being put through benchmark tests which naturally push the hardware to its extremes in order to measure maximum results. During normal use, these extremes might never be reached. (Unless you're trying to watch an Adobe Flash video. )
Baxter and Gillooly continue, "To test our suspicions further we booted into Windows (hooray for Boot Camp) and watched what happened when the CPU was loaded to full... We then switched to Maxon's Cinebench 11.5. This is a 3D rendering benchmark that is used to test multithreading in CPUs, and loads up all cores with rendering tasks. During this test the Core i7 spiked at 95 degrees Celsius, tantalisingly close to the boiling point of water."
MacDailyNews Note: The temperature noted is under Windows, not Mac OS X, running a benchmark app designed to max out the processor, and measures the processor surface temperature, not the MacBook's external case temperature.
Baxter and Gillooly continue, "We repeated the Cinebench test in OS X, and, as with the Windows version, the CPU temperature climbed precipitously high - topping out at 90 degrees Celsius. The underside heat sensors were only registering 39 degrees when this happened, even though the underside near the CPU was almost too hot to touch."
MacDailyNews Note: 39° C equals 102.2° F.
Baxter and Gillooly continue, "Worryingly the heat buildup in the CPU doesn't register on the enclosure sensors. This is despite the chassis getting hot to the touch, and the heat buildup being registered on all the hardware-based sensors in the Macbook Pro."
MacDailyNews Take: What the hell is worrying about that? It's proof positive of Apple's excellent hardware design. The part that the user can touch gets warm, but not dangerously so, while the processor when abnormally stressed to maximum levels hit temperatures that do not exceed Intel's thermal spec for the MacBook Pro's Intel Core i7 Mobile Processor (I7-620M) of 105° C.
Baxter and Gillooly continue, "To test just how much an influence cramming the Core i7 into the unibody Macbook Pro has we re-ran the tests on a Fujitsu Lifebook SH 760. This uses the same Core i7-620M CPU as the Macbook, but is designed with a copper heatsink that vents out the left side of its plastic shell. The CPU started out with an idle temperature of 40 Degrees. After 3 consecutive Cinebench runs the maximum CPU temperature seen was 81 Degrees, a full 20 below that experienced with the Macbook. It was also cool to the touch."
MacDailyNews Take: Meaningless. Baxter and Gillooly's own tests prove that Apple's 17-inch MacBook Pro Core i7 processor does not exceed Intel's thermal spec of 105° C even when running a test designed to maximize processor load. Furthermore, the screenshot Baxter and Gillooly show clearly indicate a "FUJITSU PC" running an "Intel Core i5." Not an Intel Core i7. So, of course, it's running cooler. PC Advisor's propaganda piece is falling apart at the seams.
Baxter and Gillooly continue, "Apple sits the Core i7 at the top end of its Macbook Pro Range. From our testing in both Windows and OS X it seems that while the CPU is powerful, the heat output associated with it running at full load is definitely a cause for concern. In this case the fantastic looks of the unibody Aluminium design are let down by the sheer amount of heat buildup experienced."
MacDailyNews Take: Bullshit. The MBP's i7 heat output is within Intel's own safe operating temperatures and the MacBook Pro's case seems to be executing one of its main missions, to dissipate a processor's heat before it reaches the user, in excellent fashion, as Baxter's and Gillooly's own tests prove.
Baxter and Gillooly continue, "The generally cool styling of the Macbook Pro just doesn't seem too capable when put up against the sheer output of Intel's Core i7 processor. This is reinforced by the Fujitsu Lifebook running 20 degrees cooler in the same tests with the same CPU."
MacDailyNews Take: Again, bullshit. Baxter's and Gillooly's own screenshot clearly says "Core i5," not "Core i7," for the Fujitsu POS. It "reinforces" nothing but their own incompetence and/or duplicity.
Full article here.
MacDailyNews Take: To recap: The MacBook Pro's Core i7 processer never exceeds Intel's thermal spec limit. The MacBook Pro's case gets warm under maximum and abnormal processor stress, up to 3.6° F above normal human body temperature, or 98.6° F. OMG, alert the media!
Enjoy the overblown hyperbole this piece of propaganda is sure to generate among the less-than-critical thinkers throughout the blogosphere and mainstream tech media.
At the very least, Apple ought to demand a prominent retraction and apology.
Comparing PC with i5 to Apple's i7... Oranges to Apples...
AND reports processor SURFACE temperature as though it were temperature of laptop case!!! NOT!
If you want on or off the Mac Ping List, Freepmail me.
Kudos to the guys at Mac Daily News.
So much for PC Advisor’s reputation, assuming they had one prior to this nonsense.
Am I understanding this correctly? The unibody was designed to be a huge heat sink?
Why don’t they start comparing bricks to marshmallows?
Precisely: heat sink, heat spreader, and heat dissipator.
The "unibody" (solid) aluminium case works much better than a case that is interrupted by a battery access opening/door -- and works far better than a plastic case -- for thermal management.
But it also replaces any type of heat sink directly on or near the processor? Is that a good idea? I’d think that better ventilation and fan controls would be part of the design also. Was there something wrong with actually sinking the CPU as part of the design?
I even created an iterative graphic modeling system (on an Apple//, no less) that treated heat flow like electrical current flow, using Ohm's Law, Watt's Law, and the solid state formula for resistance, R = ρ L / A ...
When it comes to effective and efficient industrial (thermal) design, Apple has their act together...
Leonard210, thermal management design is the art of designing the complete path of the flow of heat from its source (the active devices on a chip) to the "outside world", where it can be "dumped". The goal is to minimize resistance to heat flow at every step along the way.
No matter how well you heat sink the chip, if you can't get that heat all the way out into the outside air, you have a failed thermal design. A plastic case is a big resistor at the end of the chain; a metal case is a much smaller resistor.
And a solid metal case (uninterrupted by holes that decrease the cross-sectional area of the heat path) is the lowest practical resistance.
User-replaceable batteries (with access ports punched through the case) are a big limitation on good thermal design. Folks who insist on them must learn to live with lower computing power.
Apple optimizes every element of their designs. That's why I sit here and laugh at the bozos who think it is smart to just stuff cheap components into a box -- and call it a "PC"...
“This silly report could have been written by a high-school C- student.”
Or a guy who only has a passing acquaintance with thermal design from the 90’s. Like me. :-)
Thanks for the info, however, I’m learning from you guys.
Excellent explanation for the ill-informed around here. Er, that would be me. :-)
The MacBook Pro has heat sinks, heat pipes, and two fans for the processor and the dedicated graphics chip. Even with that, a lot of heat in any notebook also gets into the case.
Cool. I just haven’t kept up with heat dissipation technology. I kind of dosed off after they started putting liquid cooling systems in.
So Mac Daily is saying you shouldn’t run non Apple softwware?
Shouldn't Apple disclose that if you run High Performance non Apple Stuff you might get burned?
It's part of Apple's heat management on the laptops that are made with the aluminum unibody. Aluminum is a good conductor of heat... and the fan system is designed to blow around the unibody and extract the heat from it.
No, it is not a replacement... it is incorporated with those as well. The design has fans and incorporates fan controls. It's one of the ways that Apple keeps their laptops the quietest around.
No, but if you do run Windows7, Vista, or XP, you had better be sure you have loaded the optimized drivers Apple has provided so that it will control the fans as best it can.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.