Huge core temps difference, need help

[wsarahan]

Hi guys

I have a doubt sbout core temps difference

I have here a 3770k @4.5 not delided with H100i cooler

Did the small test in prime, the one that stress the cpu in the max and i saw a huge core temp differences between the cores

Take a look:

Core0 - 63C

Core1 - 74C

Core2 - 76C

Core3 - 71C

Is this normal? I just remounted the cooler and reaplyed the TIM, a MX4 and nothing changed

What should i do?

Another question, this temps are normal to a non delided 3770 @4.5 with 1.23vcore with h100i installed?

Thanks guys

 

[Saylick]

When you removed your heatsink before you reapplied your TIM, was the contact patch evenly distributed? In other words, if you were to remove your heatsink right now, would you have uneven TIM distribution on the heatsink?

 

[wsarahan]

When you removed your heatsink before you reapplied your TIM, was the contact patch evenly distributed? In other words, if you were to remove your heatsink right now, would you have uneven TIM distribution on the heatsink?

Yeah i think so, now that applyed again same thing happened, same difference in cores

 

[ShintaiDK]

The temps look relatively very normal. Core 2 being highest temp due to its physical location. And core 0 benefitting from being next to the unused IGP that acts like dark silicon, hence cooling.

The good old picture:

 

[wsarahan]

The temps look relatively very normal. Core 2 being highest temp due to its physical location. And core 0 benefitting from being next to the unused IGP that acts like dark silicon, hence cooling.

So should i not worry about it?

And the temps for the 4.5 non delided 3770k is it normal? It`s stable but i changed my h100 to a new h100i and i just runnned a couple of minutes to see the performance, about 5 minutes in prime selecting the more stress test i got this temps

 

[ShintaiDK]

So should i not worry about it?

No.

 

[wsarahan]

No.

Thanks for the quick help man

I`m just a little worried because the temps, i mean the high temps considering i`m using a WC and only with 1.23vcore at 3770k

 

[wsarahan]

No.

But with IBT my temps are much higher than prime

See if it`s normal

Core0 - 73

Core1 - 82

Core2 - 85

Core3 - 79

I know that Ivy is a hot chip when you not delide it, but considering i`m using a H100i this temps are not so hot?

Thanks

 

[ShintaiDK]

Its hard to say since every overclock and every chip is individual. Also your chip can have a large gap between the IHS and die.

The only thing I can say is that your original temperature difference between cores is perfectly normal and logical.

 

[cytg111]

I'd say that your temps look very satisfying. Unless you are going into throttle territory, dont worry about them, intel knows what they are doing, those temp numbers might as well be the number of smurfs running around throwing electrons at each other for fun..
In other words, it doesnt matter.

 

[wsarahan]

I'd say that your temps look very satisfying. Unless you are going into throttle territory, dont worry about them, intel knows what they are doing, those temp numbers might as well be the number of smurfs running around throwing electrons at each other for fun..
In other words, it doesnt matter.

Really thanks, I`ll let like this so, but i won`t lie, i expected more with H100, mid 70`s or something like that

I bought a Gelid Extreme TIM, will wait arrive and put this one to see, using MX-4 now pea method

 

[ShintaiDK]

Remember the random draw at chips. No 2 chips are equal.

 

[aigomorla]

but considering i`m using a H100i this temps are not so hot?

Thanks

Do not look at your temps compared to other people.

no 2 cpu's are ever identical unless u happen to get lucky and get a full try bin on the same node.
(LOLZ.. i got lucky once and got 2 identical chips... the ID number was off by 1 on each of them.. meaning it was the next bin when the cpu was made.) <--- that is the ONLY time in my numerous CPU killing days have i had 2 cpu's being near identical.

Dont look at other watercooling temps and expect yours to be close.
Unless its another AIO, running the exact same tests as you, in the exact same ambient temperature with the exact same... (ect) you get the point...

This is why i dont even bother looking at tests which dont show the delta between the load temps and the ambient temps... to me its like someone showing u a piece of candy going look candy without telling u what type of candy it is..

without knowing the op's ambient temps inside his case.. the air temp running inside his radiator.. and his coolant temp thats circulating inside the AIO... and finally the FLOW at which the coolant is being curculated...

its truely impossible to tell him his temps are high or within range.

 

[tweakboy]

If you lap the cpu and block you will get even temps all round.

But I dont suggest sanding it ,,,,,,,