I'm upgrading my Hard drive tomorrow and was curious if it's worth it to reapply some Arctic Silver 5 on the CPU / any other chips. Any suggestions?
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.
Worth it to reapply Arctic Silver 5?
- Thread starter oplix
- Start date
- Sort by reaction score
I'm sure it would help with your temps. However, since i doubt your overclocking your mbp or anything like that, it probably won't make all that big of a difference.
In my gaming computer, i just switched the thermal paste on my i7 930 from the stock paste that came with my aftermarket cooler, to arctic silver and my temps at load dropped almost 25 degrees centigrade. Arctic silver is pretty awesome.
In my gaming computer, i just switched the thermal paste on my i7 930 from the stock paste that came with my aftermarket cooler, to arctic silver and my temps at load dropped almost 25 degrees centigrade. Arctic silver is pretty awesome.
There have been a few threads about this on here. Try using MRoogle for "thermal paste" and words like that.
From what I've seen, it makes some difference, a few degrees C at idle. Not as much as some "OMG Apple puts on so much paste" people would have you believe.
I think that the fan control algorithm is designed to keep the CPU at a pre-determined full load setpoint (eg 95 C). If so improving the thermal conduction will keep the CPU at 95 C but using fewer fan rpm - so no real gain. Unless you install some fan control firmware as well.
But if you want to do it & have the skills, why not go for it?
I would be very interested in some controlled before-and-after temperature tests, not just of the CPU but of the heatsink, enclosure etc. running the same full load program eg Cinebench or "yes > /dev/null" x 2.
Have fun!
From what I've seen, it makes some difference, a few degrees C at idle. Not as much as some "OMG Apple puts on so much paste" people would have you believe.
I think that the fan control algorithm is designed to keep the CPU at a pre-determined full load setpoint (eg 95 C). If so improving the thermal conduction will keep the CPU at 95 C but using fewer fan rpm - so no real gain. Unless you install some fan control firmware as well.
But if you want to do it & have the skills, why not go for it?
I would be very interested in some controlled before-and-after temperature tests, not just of the CPU but of the heatsink, enclosure etc. running the same full load program eg Cinebench or "yes > /dev/null" x 2.
Have fun!