7980XE delidded temps whats normal?

[tps3443]

Nether it was due to the surface of the die mate, I ended up sanding it down as you can see my temps here under 4.5Ghz 1.2v load.

LM doesn't last forever to keep good temps you need to redo it as it will die out on your die when it starts to pit it, normally the 2nd time around will last longer because of the infused LM on the die, I ran my OC for a good 12hour stress test even know my CPU can do 4.8Ghz 1.26V its not stable for my work load.

How much did you lap off of the die? Do you think
you just had a wonky die surface?

I have considered lapping my own die for much lower temps, but it worries me. I’ve seen people have the LM kill the CPU due to not having a protective coating and damages the CPU by seeping to far in to the die.

I think I got pretty lucky with my CPU die, and IHS. It is pretty flat I think. The IHS doesn’t spin like a top on a flat surface. And even before the delid my core to core temps were within 5-9C.

When I first installed my CPU, the core to core temps were about 15-20C apart. I rotated my AIO block 90 degrees, repasted, and reinstalled and it was with 5-10C. And it has been like this ever since even with numerous reinstalls of my processor.

 

[LiveOrDie]

How much did you lap off of the die? Do you think
you just had a wonky die surface?

I have considered lapping my own die for much lower temps, but it worries me. I’ve seen people have the LM kill the CPU due to not having a protective coating and damages the CPU by seeping to far in to the die.

I think I got pretty lucky with my CPU die, and IHS. It is pretty flat I think. The IHS doesn’t spin like a top on a flat surface. And even before the delid my core to core temps were within 5-9C.

When I first installed my CPU, the core to core temps were about 15-20C apart. I rotated my AIO block 90 degrees, repasted, and reinstalled and it was with 5-10C. And it has been like this ever since even with numerous reinstalls of my processor.

I lapped it until it was flat and my die was very uneven and sanding it took a hour or so, as for LM seeping into the die it is possible but not likely the protective coating doesn't stop LM pitting the die, once the LM has formed a coat ontop of the silicon the pitting will stop ether way.

 

[EarthDog]

I lapped it until it was flat and my die was very uneven and sanding it took a hour or so, as for LM seeping into the die it is possible but not likely the protective coating doesn't stop LM pitting the die, once the LM has formed a coat ontop of the silicon the pitting will stop ether way.

in theory you've sanded that off.

Sanding bare die is incredible risky.

 

[tps3443]

in theory you've sanded that off.

Sanding bare die is incredible risky.

I think he had to.His CPU has some seriously wonky core to core temperatures. He has no
choice. If he did it right, it’ll last for years and years. He’s already been running this CPU for a few years anyways right?

 

[LiveOrDie]

in theory you've sanded that off.

Sanding bare die is incredible risky.

LM eats the protective layer over time so really the risk isn't that huge and the silicon thinkness is enough, I got the CPU for half price 2 years ago now so the risk was worth it for me,

 

[yotano211]

Nether it was due to the surface of the die mate, I ended up sanding it down as you can see my temps here under 4.5Ghz 1.2v load.

LM doesn't last forever to keep good temps you need to redo it as it will die out on your die when it starts to pit it, normally the 2nd time around will last longer because of the infused LM on the die, I ran my OC for a good 12hour stress test even know my CPU can do 4.8Ghz 1.26V its not stable for my work load.

I ran an i7 920xm ES, a laptop processor with cool lab liquid metal for over a year and it didn't dry up. I had the same results with a i7 2920xm, another laptop processor for around 10 months, still didn't dry up.
I changed the thermal paste because the next owner didn't want liquid metal applied. People at another forum had liquid metal for over 2 years, it didn't dry up.

 

[Solaris17]

I ran an i7 920xm ES, a laptop processor with cool lab liquid metal for over a year and it didn't dry up. I had the same results with a i7 2920xm, another laptop processor for around 10 months, still didn't dry up.
I changed the thermal paste because the next owner didn't want liquid metal applied. People at another forum had liquid metal for over 2 years, it didn't dry up.

This. Mine is still fine.

 

[tps3443]

I had temperature degradation on my 7980XE. The temps went up about 6-7C on each core. And 4.8Ghz at 1.270V was no longer a viable option, due to the additional heat making it unstable.

I delidded the CPU again, reapplied LM allowed it to soak really well, and now my temps are better than ever.

I am running my new memory kit now.

G.Skill DDR4 4000 CL15 quad channel set.

I run 4,000Mhz at CL14-15-15 1T.

Why debate over tight timmings or high speed, when you can have both!

Anyways, the X299 platform and 7980XE is really a interesting processor. This 3+ year old CPU is really in its prime right now in the later part of 2020. Because high core count CPU’s are become more mainstream. it just makes this processor better with age! It is faster than a overclocked 3950X, and just about anything else too. Time has served this processor better than most!