Q6600 Heating: Is This A Problem?

[spectre88]

Just put together this system the other day:

Intel Q6600 (NOT OVERCLOCKED YET)
nForce 750i Motherboard
Xigmatek HTD-S1283 after market cooling
Using Arctic Silver 5 (1 thin line horizontally across the center of the processor but enough to make contact with the pipes)

Anyway the problem is after running RealTemp I am finding that there is a 10 degree Celsius difference in different parts of the processor.

TJ Max 53 - 53 - 63 - 65
Core Temp 42C - 42C - 32C - 30C
Min Temp 41C - 42C - 31C - 30C
Max Temp 45C - 45C - 35C - 32C
Screen Shot of my RealTemp Window

By the way these numbers where taken while the system was idle. I was told it is not good having such variance in temperature. Likewise I was also told that the temperature is a little high for a processor that is not overclocked (and I am looking overclock in about a week). Any ideas what is going on? Should I even worry about it?

 

[Idontcare]

This temp delta across the cores is on the higher end of what most folks report. Personally I owned a G0 that had a 7C delta.

There are numerous diseases that give rise to this identical symptom. Ranging from improper TIM application to improper HSF installation to concave/convex HSF and/or IHS surfaces, etc.

From your post it sounds like you are reasonably confident that the TIM application is not to blame, and those Xigmatechs are pretty darn flat from what I've read.

So the cultprit in your case could be that you need to lap your IHS.

In my case I had to lap both my Tuniq and the IHS to get that 7C delta down to 1C delta.

There are even more reasons the temps can be different, but they are beyond the control of the end-user and they include poor adhesion of the IHS to the underlying die and the very real possibility that the core's themselves are simply burning more or less power relative to each other (process variations giving rise to differing leakages in the parts of the chip that matters for these things).