Thermal Paste Application Methods - Which one is best? - The Workshop

I wonder if in today's CPUs standards the test would be exactly the same considering 7 years ago we didn't had multiple cores, multiple chiplets nor CPUs were so small. Would there be any thermal variation considering CPUs are now more thermal sensitive?

I’ve always suspected this. Only different is too much or too little

ok but how do you measure temps of cores or smaller units of the cpu? I imagine precision thermometers are expensive, the only concern would be bubbles in the thermal paste. spiral is the worst one for this reason, I'm guessing best is just a paper thin layer completely covering the cpu

Win: X

Best: Spiral

Isn't thermal paste non conductive 😅?

its 2023, i still refer this video when changing cpu or aio real great one

That's 10 minutes I'll never get back. Fast forward to the chat at 8 minutes and skip the rest lol

try to deliberately introduce air pockets

X is best ask Elon

god the music was too loud back then

spread past i think

I did the apply-reapply method.

I'd really love an updated version of this with new lab conditions to see if there is minor variance in any of the methods

Wow Luke looks young

you need to find a different barber.

My faith in life has been shattered thanks for the video!

How about no thermal paste :) ?

Long live the DOT METHOD!!!!]
*Thanks, on making the dot amount the size of the top of capacitors as a guide on dot size amount.
Great Tip!!!!!

would you like to include a Brazilian thermal paste? I send you free of charge

Make a updated version

best part of the video was seeing the spread when you removed the radiator cover. 10 minutes is great and all but its all about the long term. Seeing how evenly it spread out afterward is the most important.

I actually learned something.

meth o.d. matters.

Spread technique both on the cooler and cpu (like applying butter on your bread ), with a small ammount of extra paste like a a dot in the center of the cpu and cooler to prevent "airpockets" in the paste.
With my "Assassin III" cooler, the temp is even 3 degreese Celsius under ambient.
Max Temp with an overclocked i7 990x at 5 Ghz with 1.58 voltage arround 76 Celsius at full load with CPU-z.
The brand cooling paste: Kryonaut

Cringe. Get a life

i apply my thermal paste in the style of the ltt logo

You mist one test : What would happen when you put thermal paste on a thermal pad that is mounted on cpu ?

I would love to see this with an AMD CPU, seeing as its dyes are a lot more spread out than Intel ones.

What about the fan speed tho? Was it always on same RPM as the test went by?

It’s not the paste that cool the cpu, that what’s people need to understand. It’s actually just there to optimize the heat transfer.

Never accounted for fan RPM and thermal paste break in time. Fail.

Thanks

this sorta things annoying. a simple 3/4m video turned into a 10m one just for ad revenue

Guys the real question. Why there are no videos on this yet. Intel 12th gen chips are longer now more rectangle than square....

Whats the new best application method for these chips. The pea dot doesnt seem to hit both ends properly.

Oh my god! Who would have thought that adding too much thermal paste would increase the room's ambient temperature by .1 degrees? Crazy!

I always put thermal past with my finger and never have problems. Now thanks to you y know why.

Man I haven't seen a Luke video in ages

I think pea will win

It would've been funnier if it had been 42 degrees.

there are no good applications when you are playing under 39c in summer and your cpu is hitting 100%

I think there should have been more to this test
for example running a high demanding cpu test and the second it ends look at the temp.

Thought paste needed to cure before you see actual results…..so basically a video showing how much paste is too much and too little

I disagree with Luke's conclusion on this video. Clearly the pea-size drop was the best because it resulted in the lowest ambient room temperature at 16.8 degrees celsius.

half a pea better than pea

are you cooler now with that stuff on your finger?

I used the X or line or spread method on CPU's with a large IHS. Prefer the X. Simple reason is that not covering the edges of of a large IHS is much worse than putting on a little too much... and I'm not competing for benchmark scores.
If you put the cooler down at little too much of an angle(not intentionally) when installing it, methods that are not the X or pre-spread have a larger probability of pushing the paste to one side when tightening it. In essence I'm just reiterating the "don't apply too little" message. And on a big IHS that's easier to do by mistake.

God damn this guy's personality is like watching paint dry.