Vent: Thermal paste debate - thick spread or just a pea?

I had a Ryzen 5 3600 run hot for months, tried the pea method and got temps around 85c under load. Then I manually spread a thin layer with a plastic card and dropped to 72c. Anyone else find that the pea method is overrated for certain chips?

3 comments

3 Comments

emma_baker6119d ago

honestly i see this debate all the time and i think people overthink it way too much. like you got a 13 degree drop which is nice and all but did you reseat the cooler or change the mounting pressure at the same time? because that could have been the real fix not the spread method. i've built like 8 PCs with the pea method and never seen more than a 2-3 degree difference no matter what i do. unless you're running a massive overclock or your chip is just wonky it's probably not that serious.

johnson.river19d ago

Wait, did you make sure the cooler was actually making good contact the first time? @emma_baker61 makes a solid point about mounting pressure being the real culprit there. I've messed around with thermal paste on a dozen builds myself, and honestly the pea method works fine 99% of the time. But some chips have weird hotspots or uneven IHS surfaces that need a full spread to cover properly. My old 3600 was the same way, ran hot until I did a thin spread like you did.

simonk9819d ago

Actually, that's a good point about the mounting pressure, but I think there's something specific about the Ryzen 3600's heat spreader. Those chips have a pretty noticeable edge where the die isn't centered under the IHS, so a pea in the middle can leave the hot spots uncovered. I've got a 3600X myself and saw a similar drop when I switched to a thin spread - about 10 degrees. The pea method might work fine for Intel chips or newer Ryzens, but that first gen of Ryzen 3000 seems to benefit from full coverage.