Review Repasting a laptop CPU - quite the significant difference even in "light" load like Geekbench 6

[tamz_msc]

Just repasted my Dell Inspiron, which is a little over two years old. It has been subject to quite significant use - from gaming to work. Lately, it would thermal throttle having HWiNFO open to the monitoring tab doing nothing, even though there was no actual slowdown doing everyday tasks and navigating Windows.

So I decide it's time to repaste. Got hold of a tube of Gelid GC Extreme and all the other stuff and got working. It was a bit intimidating at first, having never serviced a laptop to this extent. But Dell provides extensive service manuals with straightforward instructions, so the process was a breeze. Trivia -- noticed that the CPU die is even smaller than the TU117M GPU die - pictures on the web make Tiger Lake U look larger than it actually is.

So anyway here are the results running Geekbench 6.1. No, don't ask me for Cinebench runs as I consider it irrelevant for my use case.

Dell Inc. Inspiron 14 5410 vs Dell Inc. Inspiron 14 5410 - Geekbench

 

[A///]

concise writeup. none of that excess blog drivel people love to lore about. curious what your results would be trying some of the leading brands that don't need a bake in period or don't need changing every few years. may as well be the ilk of rubbing two toothpicks together but who knows. hope the weather isn't too bad in your part of the world, lad. cheers

 

[dullard]

Did you clean out the dirt, dust, and other gunk in there at the same time? I'm just curious as to how you'd partition the gains between better paste and potentially better cooling from cleaning out the laptop.

 

[tamz_msc]

Did you clean out the dirt, dust, and other gunk in there at the same time? I'm just curious as to how you'd partition the gains between better paste and potentially better cooling from cleaning out the laptop.

That is a consideration, but like I mentioned, the temperatures would spike to near or at the throttling point even when sitting idle on the desktop. Under such situations the fan would barely spin up. Now the idle spikes are much more under control.

Also, when I removed the heatsink, the old thermal paste, even though not completely dry, barely covered the dies. So I would guess that it's more due to the suboptimal coverage of thermal paste rather than dust and dirt that caused the high temperatures. I'm not discounting the possible effects of dust buildup though.

 

[A///]

in my experience and laziness the dust needs to have matted over a good bit for a divisive fluctation to happen. most modern fans can spin up more to compensate for dust build up to a certain amount before their ability to cool diminishes due to build up. maybe if tamz lived inside a wool processing plant he would experience more buildup that would affect it, but with laptops and most computers dry crusty paste will be the troublemaker. I like to reapply once a year.

 

[yottabit]

I repasted my HP Omen 15 laptop and saw a ridiculous difference. I didn’t benchmark, but the fan noise was getting out of control even for doing normal desktop tasks and since then its been more or less silent except under heavy loads. I agree it’s probably more due to with inadequate factory application than anything else