Overclock.net banner
1 - 7 of 7 Posts

·
Registered
Joined
·
322 Posts
Discussion Starter · #1 ·
During stress test I noticed my 5900x core 0 is almost 15 degrees lower than other cores. Is this normal or bad waterblock mount?
Font Line Display device Electric blue Electronic device
 

·
Registered
Joined
·
5,974 Posts
It really depends on a few things:

1) Your preferred core: Ryzen processors prioritize cores to give work to based on their "potential". You can find this info out for yourself using ryzen master (just google it). I would also keep an eye on core usage when you look at temperature.

2) Bad waterblock mount: it could very well be a bad mount, but your temps are super good across the board for a 5900x.

3) Bad IHS mount: this isn't really anything you can change short of RMA. Sometimes, the IHS mounts are wonky from the factory. Most you can do is RMA.

your pic really REALLY does not provide any information that can help us narrow it down for you.
 

·
Still kinda lost
Joined
·
3,515 Posts
EDIT: NVM
 

·
Iconoclast
Joined
·
31,560 Posts
During stress test I noticed my 5900x core 0 is almost 15 degrees lower than other cores. Is this normal or bad waterblock mount? View attachment 2525506
It's normal for cores to have different leakage values, different scheduling priorities, and different boost clocks. Cores at the edge of the die also tend to run cooler because they are only directly next to to one other core. Different CCDs can also have very different properties.

Regardless, you'll want to run a high all-core load and look at maximum temps of each core. Up to about a 10C differential within the same CCD is pretty normal and even a 15C differential between the hottest and coldest cores on a dual CCD part usually isn't indicative of a problem.

If you think there is a problem, you can try remounting the cooler or lapping the IHS, but differentials in this range are usually due to variance in the silicon itself.

Anyway, more information is needed before any potential issue can be adequately identified, let alone remedied.

3) Bad IHS mount: this isn't really anything you can change short of RMA. Sometimes, the IHS mounts are wonky from the factory. Most you can do is RMA.
Solder can be reflowed, or the part can be delidded.
 

·
Registered
Joined
·
322 Posts
Discussion Starter · #5 · (Edited)
It's normal for cores to have different leakage values, different scheduling priorities, and different boost clocks. Cores at the edge of the die also tend to run cooler because they are only directly next to to one other core. Different CCDs can also have very different properties.

Regardless, you'll want to run a high all-core load and look at maximum temps of each core. Up to about a 10C differential within the same CCD is pretty normal and even a 15C differential between the hottest and coldest cores on a dual CCD part usually isn't indicative of a problem.

If you think there is a problem, you can try remounting the cooler or lapping the IHS, but differentials in this range are usually due to variance in the silicon itself.

Anyway, more information is needed before any potential issue can be adequately identified, let alone remedied.



Solder can be reflowed, or the part can be delidded.
Ok thanks I wont worry to much about it. Remounting my monoblock takes about three hours so I will hopefully be leaving it for a while!
that ekwb monoblock for the aorus x570 does not have the greatest mounting system
 

·
Registered
Joined
·
5,974 Posts
It's normal for cores to have different leakage values, different scheduling priorities, and different boost clocks. Cores at the edge of the die also tend to run cooler because they are only directly next to to one other core. Different CCDs can also have very different properties.

Regardless, you'll want to run a high all-core load and look at maximum temps of each core. Up to about a 10C differential within the same CCD is pretty normal and even a 15C differential between the hottest and coldest cores on a dual CCD part usually isn't indicative of a problem.

If you think there is a problem, you can try remounting the cooler or lapping the IHS, but differentials in this range are usually due to variance in the silicon itself.

Anyway, more information is needed before any potential issue can be adequately identified, let alone remedied.



Solder can be reflowed, or the part can be delidded.
Yeah, absolutely, I guess I should have specified that it's not typically worth doing especially with soldered on IHS, especially when still under warranty.
 

·
Registered
Joined
·
798 Posts
Is this normal
It is. Your 1st core (Core 0) seem to have no adjacent active core from the right side (i.e. there's a dummy core instead), hence no thermal influence therefrom.
 
1 - 7 of 7 Posts
Top