I can absolutely confirm that there were not thermal pads, just paste, and as I indicated in an earlier post, not applied correctly. WAY too much paste and most of it was not on the chip or heatpipe contacts.
The whole thing is 100% conservative (since there's no appreciable energy → mass conversion taking place inside a MacBook Pro, not even if you're watching a science fiction movie, where violation of conservation laws seems to occur routinely). So every joule of electrical energy that is converted to heat in one of the chips has to be removed from the point at which that conversion takes place.
However, since there are in some sense two paths (one via the internal heat sink and the airflow driven by the fans and the other via thermal conduction through the case), the analysis is more complicated. Basically, it's possible to shift the balance of thermal conductivity to favor the path through the forced air or vice versa. It's thus conceivable that by changing the distribution of heat-sink compound you could keep the same net cooling but divert more heat to the air-flow.
But I'm dubious that this is a likely explanation for a cooler case. More likely is that net heat flow to the case has been impaired and the temperature of the semiconductor devices has gone up. A net loss for the device's longevity.
Obviously, software and firmware updates are not going to change the physical aspects of the system. However, thermal management overall is heavily mediated by software. This goes for power management of the various chipsets as well as fan management, so overall it's definitely possible to get changes in the thermal characteristics via software and firmware updates.
Some of us think we've seen this happen with the 10.4.6 system software and / or the 1.0 firmware update. I observed a distinct improvement (decrease) in the case temperature when operating on external power after the updates.
"Well, I recently opened my MBP down to the logic board to take a look at these chips and what did I find....
** poor application of the thermal compound. There appear to have been some heat damage in the surrounding silicone wafer as well (a bit of a bubbling up)"
Jean Cyril, interesting post. Given the numerous flaws I am reading about, and experiencing myself, the factory in China is evidently having some quality control issues. I think Apple has rushed these systems into production too quickly.
As you say, the rear vents can be blocked very easily. I'm sure this is a big contributor to many overheating problems.
@ Randall - J-C clearly stated: "there was tons of excess compound leaking out aroung the silicone border of all three chips and very little on the chips themselves." He did not say "excessive heat-sink compound was impeding heat flow" as you suggest. He said it wasn't where is was supposed to be. Really Randall, you are quite the contrarian, aren't you?
Vinbot
I run MenuMeters on mine and so I always see what the CPUs are doing. Heat is definitely a function of CPU activity. My gut on this issue is that the firmware update did nothing. I know people that swear by the firmware update will not like that comment but that is my gut. Mine if fine as long as I am not doing anything but as soon as I run something like parallels the heat shoots up fast. The only way to know for sure that some do and some do not have this problem is to drive the CPUs in the same way and with the same airflow around the machine for the same amount of time and have some accurate way of measuring temp. We have no way of doing any of those things so My gut is that we all have the same problem and it's just perception of severity of the problem that differs.
Thanks for interest/defense, but don't sweat it. I've built enough computers to know what I'm talking about and don't need justification/acceptance from others.
Concerning the widgets, one thing I've done is to unload them all completely so they don't even appear on activity monitor. Not sure that does anything, but its interesting that the widgets are loaded even when not running.
Theoretically sound, but you need to account for the empirical observations. In essence you are saying--the heat has to go somewhere--agreed. And yes, there are essentially two directions, but recall that the material in those two directions is different, as is the surface area, and heat conductivity.
My point was that the scenario you suggest had already occured in the case of my MBP, not necessarily all others, because of the unbelievably shoddy job application of the thermal compound. I wish I had taken a picture, but there was virtually no compound on the chips and a lot of it on the outside. Given the set height of the spacers between the logic board and heatpipe, I am pretty certain that the heat transfer contact was very inefficient for at least one chip because of the silicone swelling TOWARD the heat pipe side.
Proper thermal conductivity does NOT make the MBP hotter on the TOP of the unit as evidenced by those whose laptops are appropriately cool, and more importanly, because I was almost burning my hands before.
implicit in your statement "Improving thermal conductivity between the source, primarily the CPU, ICH, MCH and GPU chips, and the external environment will make the outside of the case hotter, not cooler. If you actually succeeded in making your MacBook cooler, you've probably shortened its life by raising the temperature at the source of the heat, the semiconductor junctions in the key chips" is the assumption that I don't know what I am doing, have never done this before, and must have applied the AS5 incorrectly.
I am sure you did not intend it that way, but that's how you came off to me. Fact, I have been building computers, albeit PCs for over 15 years and have worked on a buch of laptops, some with more circulation constraints than the MBP. Fact, the evidence on the performance benefits of AS5 over properly applied thermal compound are well documented. In this case the job done at the factory was crap!
Your post suggests that heat could be going away from the pipe and into the silicone and other end of the chip. I would argue that is what was going on before. The silicone swelling was towards the heat pipe not away from it, indicating that the trapped heat was inside the chip, not outside. Moreover the top of the laptop was burning.
Bottom line I agree with your argument, but submit that it had already happened, and I fixed it. Of course, time will tell, but since I applied AS5, I have continued to use my laptop as I normally do except that I have turned off sleep and kept it on constantly--something I did not do previously. If you are correct, then there should have been some heat build over that time period (over 4 days now) or some failure indication.
