• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Q9550 Core 1 running much hotter than others, bad thermal paste application?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

jarekb

New Member
Joined
Dec 23, 2018
Hi all, hoping I can find out if this is normal for that cpu or if there's something wrong with my cooling setup.

I'm just starting to overclock a very old system (built in 2008) which has been running stable at 2.8GHz (8.5x333) for the whole time.

I did some initial benchmarks to get a baseline set of temperatures
sOtlXiC.png

I then replaced the decade old thermal paste with Thermal Grizzly Kryonaut along with replacing a few broken 120mm fans in my case (have 3 blowing in, 2 blowing out, and a gigantic 200mm blowing out as well).

2CLNgqG.png

Then I started to overclock the system (a process I'm still in the midst of in finding a stable setup) 3.187GHz (8.5x375)
xInTmoR.png

So given that
- The core temps did not change that much after applying new thermal paste
- The old temps had Core 1 running 12+ C higher then Core 2, with Core 3 and 4 always being within 2-3C

Is this running normal or does the above indicate a bad thermal paste application (both from my old setup to the new one)?
Is the lack of change between old thermal paste and new application (before OC) expected or again a sign of bad application?
 
That's normal.

My Q9650 running a ThermalRight Ultra 120 Extreme, a decent air cooler, does the same thing.

My first core runs an average 10*F higher than the other three at idle and a good 20*F higher under maximum load.
 
You don't always see that much spread but it's not abnormal either.
 
Its been awhile since I ran mine, but I think I was only looking at a 5-6c spread. I see a 3c spread on my x5690, and a 10c spread on my 3770k, which does nothing for my ocd.
 
Thanks for the replies all. I asked the same question in the overclockers discord and got a similar response. One member mentioned that it could be caused by air bubbles or gaps in the solder used to connect the cpu die to the IHS, but at this point I'm not interesting in delidding my cpu to drop the temps.

I ended up overclocking this cpu to 3.6GHz and ran a mix of Prime95 and OCCT stress tests for 12 hours and temps did not go above 71C. So I think the thermal paste application is working fine and I won't need to reapply it.
 
Temperature variance is normal, but over 10C is a ton. LIkely something doing under the hood... that said, you're happy where you are at, so no harm no foul it seems.
 
The 10+C difference was there before I reapplied the thermal paste and started to OC. So given that this has lasted 10 years, I'm not too concerned with the difference. Even if the extra OC temps degrade the cpu, I'm only aiming to get another 1-2 years out of this system.

Side note, temps from my 10hr Prime95 Blend run
Core 1 71C max 64C avg
Core 2 61C max 53C avg
Core 3 65C max 58C avg
Core 4 64C max 57C avg

I'll still fine tune the voltages (nb/dram/fsb termination) to see if i can drop the cpu voltages a bit more while being stable at 3.6GHz, so maybe the temps will end up being stable under 70C to reduce the stress on the components.
 
I traded my Q9550 to a member here for his X3360 because the board he had wouldn't run it. But if I can recall, temps more uniform on the Xeon, I think maybe a 3-4c spread. And it was the same thing when I moved to 1366, I ran a 965EE and then a 970 hex. After that I moved to a Xeon x5690, where I saw alot nicer temps, and a faster overclock. In return, I was limited to running the ram a little slower, but I use tight timings so its not really an issue.. As for degrading your cpu, that probably wont happen.
 
Totally a plus one on this reply. The Q9450/Q9550/Q9650's make for a very robust processor.

Even when taxed to the limit in a heavy overclock, failures are very rare.
Depends on the voltage used.
You can still kill one in a heartbeat.
 
Back