Core temp difference in 11-12gen Intel CPU for P-cores - is THAT typical?! Let me know pls

[webbmaster]

When looking at each core - when one core 3 start hitting max temp - whole CPU package in HWinfo said this MAXimum - when all other cores are still pretty ok.
This is proper for this 11-12gen Intel?

Strange that one core is 68 and other 98 - in multicore test Cinebench 23 for example. Could you share your differences?
PS. This is for i9-12900HK laptop with Liquid metal pad - so it must be evenly on CPU, not as paste could be?!

 

[Falkentyne]

When looking at each core - when one core 3 start hitting max temp - whole CPU package in HWinfo said this MAXimum - when all other cores are still pretty ok.
This is proper for this 11-12gen Intel?

Strange that one core is 68 and other 98 - in multicore test Cinebench 23 for example. Could you share your differences?
PS. This is for i9-12900HK laptop with Liquid metal pad - so it must be evenly on CPU, not as paste could be?!

You have a contact pressure issue. What is your cooler you are using with this?

Your odd numbered cores are all the hottest cores, with core 3 and 5 the hottest.
Those are the same hottest cores on 8 P core chips, just with 2 more cores. The problem is your delta is too extreme. Coldest to hottest should not exceed 12C.
Check the mounting contact pressure and make sure there is an even spread.

This will require that you remove the liquid metal pad, use regular thermal paste and apply a medium sized (about 3 grains of rice sized total) drop in the middle of the IHS, then attach the cooler head-on. Use partial turns in an "X" shape pattern (one turn top left, one turn bottom right, one turn top right, etc etc) evenly until it is fully tightened.

Then immediately remove it (the same partial turns btw!) the opposite way, remove it directly straight up, then take a good picture of the CPU surface and the heatsink/heat block, with the paste spread.
Someone here will be able to tell you if you have a contact issue from doing this.

 

[webbmaster]

You have a contact pressure issue. What is your cooler you are using with this?

Your odd numbered cores are all the hottest cores, with core 3 and 5 the hottest.
Those are the same hottest cores on 8 P core chips, just with 2 more cores. The problem is your delta is too extreme. Coldest to hottest should not exceed 12C.
Check the mounting contact pressure and make sure there is an even spread.

This will require that you remove the liquid metal pad, use regular thermal paste and apply a medium sized (about 3 grains of rice sized total) drop in the middle of the IHS, then attach the cooler head-on. Use partial turns in an "X" shape pattern (one turn top left, one turn bottom right, one turn top right, etc etc) evenly until it is fully tightened.

Then immediately remove it (the same partial turns btw!) the opposite way, remove it directly straight up, then take a good picture of the CPU surface and the heatsink/heat block, with the paste spread.
Someone here will be able to tell you if you have a contact issue from doing this.

it is MSI GE76 Raider laptop, and Liquid Metal Pad they called and fans are same mostly, standard for Msi.. So to take it off quite screwy whole thing...will try to make a ticket for them...

 

[joemama]

Seems like a TIM or cooler contact uniformity issue

 

[Falkentyne]

it is MSI GE76 Raider laptop, and Liquid Metal Pad they called and fans are same mostly, standard for Msi.. So to take it off quite screwy whole thing...will try to make a ticket for them...

Oh, the standard convex MSI heatsink at work again.

Had to sand one of those on my GT73VR to get the 4 cores delta down to 4C.
Trick to sanding those is to remember that "less is more". You only want to sand just enough to make the block flat--and not a single 0.2 millimeter more than that.
Then when you repaste and remount, you should make sure you apply partial turns in an X shaped directional pattern for each screw position.

What some people do also is they remove the C-clip under the springs to very slightly increase contact pressure. You don't get much, maybe like an extra a 0.1mm worth of screw turning pressure available, since it's then the screw washer stop itself that limits you. But I don't know if MSI is still using that type of spring loaded screws on the 12th gen throttlebooks.