[SOLVED] I ran a 5600x that even a Scythe Fuma 2 could barely cool.

[ProfSeaGoat]

I no longer own this CPU, but I tried asking on Reddit about this and just got hit with why would you upgrade to a 5800X? calling it user error and some other nonsense.
Essentially my 5600X would shut off the PC after hitting 95c This is my original thread and in the thread you can see I go back and forth with a few members and a journey happens.

https://forums.overclockers.co.uk/threads/do-you-play-games-that-melt-your-system-state-your-gpu-and-temps-for-the-whole-system.18948373/page-2

In the end the 5800X is easier to cool than the previous 5600X and This I think is a bad chip, likely the integrated heat sink is not quite in contact with the actual die or something.
I upgraded everything I could cooling wise and even still, the 5600X failed to run at acceptable temperatures.

I am hoping for a clear, rational discussion here, thank you.

 

[drea.drechsler]

...
likely the integrated heat sink is not quite in contact with the actual die or something.
...

Sounds like a likely assembly defect...it may be rare but not unheard of.

I've been curious about the process for assembly of AMD CPU's. I imagine they apply low-temp solder in a paste form on the top of the die, bond the heat spreader in place then oven bake the chip to reflow the solder paste. It's easy to see where an occasional inadequate dispensing of solder paste could get past an inattentive inspector before the IHS goes on. Or maybe an interrupted bake cycle didn't fully reflow the paste. Or maybe some even more esoteric reason: like contamination of the under surface of the IHS prevented proper solder bonding to get proper thermal transfer.

 

[drea.drechsler]

...
likely the integrated heat sink is not quite in contact with the actual die or something.
...

Sounds like a likely assembly defect...it may be rare but not unheard of.

I've been curious about the process for assembly of AMD CPU's. I imagine they apply low-temp solder in a paste form on the top of the die, bond the heat spreader in place then oven bake the chip to reflow the solder paste. It's easy to see where an occasional inadequate dispensing of solder paste could get past an inattentive inspector before the IHS goes on. Or maybe an interrupted bake cycle didn't fully reflow the paste. Or maybe some even more esoteric reason: like contamination of the under surface of the IHS prevented proper solder bonding to get proper thermal transfer.

 

[Phaaze88]

If that's the case, that 5600X should've been eligible for a warranty replacement from AMD.

 

[ProfSeaGoat]

If that's the case, that 5600X should've been eligible for a warranty replacement from AMD.

Sadly I bought the 5600X second hand.

 

[drea.drechsler]

Sadly I bought the 5600X second hand.

Sad indeed!

Looks like someone passed on their problem child...someone who could more easily have gone the warranty route to get replacement.

If I had it I'd have filed a claim with AMD anyway. It's worth a try.

 

[mdd1963]

that's curiously bizarre, as the 5800X samples are actually known for running a tad on the very warm side!

Perhaps a previous owner half- attempted a delid with a heat gun, and partially ruined the heat spreader/die contact interface solder, or, it could have been a rare factory dud as suggested... (If the rubbery seal around the edges looks factory/similar to that of the 5800X, perhaps just a dud.)