• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Q6600 differences in core temps

Spyke_UK

Spyke_UK

Spyke_UK

Associate
Joined
8 Dec 2008
Posts
882
Location
South Yorkshire
Hi all, this is my 1st post and i have done a search to see if there is a thread already running on this, can't find anything but apologies in advance just in case.

Right, i have Q6600 running at 3.01mhz and all 4 core temps are different, this i know is usual but...the difference is, i think too big to be normal.
Core 0 = 31
Core 1 = 33
core 2 = 37
Core 3 = 39
Core 3 is always 9-10 degrees hotter than 0 with the middle two somewhere between.
These are my idle temps, core 3 gets upto 71 under load while the others stay below 60. I have a Arctic Freezer Pro 7 fitted and have reduced the volts slightly to 1.35. Everything is stable, no hiccups, so should i take any action about the differences, or just "shut up and play"?
 
You could try to refit the AFpro7, see if that helps.

Q6600 (and all core2quads) are basicly just 2 core2due chips attatched together, so core 0/1 and core 2/3 have diffrent temps, as they are diffrent cpu's.

Mine is about 6C diffrence and has no real problems because of this.
 
ok basically the problem you are seeing here is that your CPU and heatsink are not flat against each other. This is often caused by the heatink not being mounted correctly, and also the fact the CPU and heatsink bases are often not machined perfectly flat.

I had the same thing with my Q6600 and Tuniq Tower. Cores 1&2 were about 4 degrees different from 3&4. The solution was to buy a lapping kit from a well known auction site and lap both of them to a flat polished surface. There is now only a 1 degree difference between all the cores, with it often reading the same temps across all 4 cores.

HTH
 
I would say if you were going to the effort of getting a kit that it would be worth doing both, especially as the CPU is much quicker to do than the heatsink.

I personally did both, but you would probably see a big benefit from just solely doing the heatsink as it's usually a base that are mass machined without care. When you've done a minute of sanding on the 'rough' paper you lift up the heatsink and you can see just how warped the surface is, despite it looking flat to the naked eye.

My tuniq tower needed a lot of work before I was happy with it, the CPU wasn't very flat either. It took an hour to do both and the temps were pretty much the same after that, but it didn't actually allow me to overclock any further :)
 
Now, my next worry is this is a brand new PC had it just 1 week, i'm a little concerned that if i attempt to take things apart, or lap the surfaces i could be throwing my 12 months warranty away?
 
Hi there

Don't bother if it's a PC you've just bought. Lapping is for those looking to tweak their PC's in every way.
If in doubt whip the heatsink off, clean off the heatsink pad and instead use a quality heatsink compound like OCUK sell. This alone should improve the disparity between core temps by a degree or two.

HTH
 
Thanks for the rapid reply's guy's!
I guess the answer is kinda what i had in mind, the temps are not overly hot so "shut up and play"
Just nice to have some more opinions to help put my mind at rest.
 
Dont forget that Windows will use one core as the idle core for keeping all processes in check, which is why all cores can be at 1% but one will always be around 4% as it will be doing stuff like Explorer.exe, svchost.exe's etc.
 
on the 3 q6600 cpus iv used now all showed a difference in temps like you have shown, either core 0 and 1 would run hotter or cooler than core 2 and 3 at idle but the gap closes in a little at load temps.
 
Both the Q6600s I have had showed a difference in the core temps. There was nearly 10 degrees on the first chip and even after lapping this only narrowed the gap by about 1-2 degrees. The second chip had less of a difference but it was still there.
 
I could do with a little more help on which test to run and at what settings, Intelburntest with 1/2 memory last night run the CPU 100% core 3 went straight to 98-99 degrees, just as i was thinking "oh ****" the PC re-started?
I've had a bad experience with this PC so far unfortunatly, twice i recieved it with problems straight out of the box (corrupted gfx on boot up) it's had 6! new gfx cards so far and a mobo upgrade from the Abit XP35 Pro to the Asus P5Q Deluxe, everythings been great since, i'm just a little concerned at these temps?
 
As others have said at idle you will have difference in core temps as each core is at different loads... see what your load temps are like... within 5c is nice but they can be as different as 10c.. unless you are going for some serious o/c i wouldnt worry all to much
 
I am happy enough with it at 3.01GHz, my temps just seem to be higher than most, and the differences can be huge, before it re-booted last night core 3 was 98-99 whilst 0 and 1 were "only" in the low seventies?
 
Spyke_UK said:
before it re-booted last night core 3 was 98-99 whilst 0 and 1 were "only" in the low seventies?
Click to expand...

What program are you using for temp monitoring and have you got any ASUS software installed? The sensors can easily go nuts on the P5Q (although it's usually the cpu temp and not any of the cores). Use the latest beta version of Everest if you can.
 
There's a 3c temp diff between cores 0/1 and 2/3 on my Q6600 under load, I've lapped the CPU and the HSF, no idea what the before temps were as I lapped straight out of the box :) Under P95 load my temps can go as high as 70 on cores 3/4.
 
Run Prime95 this morning, had 20-21 degrees diff between core 0 and core 3, core 1 was only 2-4 degrees diff from core 0, core 2 was only 2-4 degrees diff from core 3. Core 3 was the hottest again, hitting 90 degrees whilst core 0 didn't get above 69 degrees.
 
You must log in or register to reply here.
Back
Top Bottom