Uneven cooling

hello there,im starting this thread to ask about my cooling temps in real temp.
i recently changed my mobo and cpu from a phenom mobo and 9850 black edition to a intel mobo and q6600 go and also brought a ocz vendetta to keep it cool and also got some arctic 5 to cool it.i have been using it a couple of days now and will add a pic to show you my real temp reading in idle mode

as you can see the first core is running hotter than the other three,could this be to do with the arctic 5 not being evenly on the cpu? i used the half pea method of placing a small ball in the middle of the cpu and letting the pressure of the vendetta spread the arctic 5 across the cpu when clamped on.

i know some will say i should have used other methods like the credit card or razor and using one of them to spread it across but this seems the best method to avoid getting air bubbles trapped inside.

what do you suggest i should do as i want to oc upto 3ghz but dont want to attempt it if one core isnt cool enough.

thanks in advance.

coolermaster cosmos 1000 silver/black
coolermaster 750 watt real power
crucial ballistix 1066mhz 8500 1gbx4
intel quad q6600 go stepping 2.4ghz
gigabyte 4870x2
vista prem 32bit
seagate barracuda 500gb 32mb
Gigabyte GA-EP45-DS4

Without looking I am not sure if that cpu is two separate dies or on die in either case if the thermal paste was the cause i think at least two of the cores would be overheating rather than just one.

I think you are fine as long as you keep the max temp in the low 50s. I have not had a single CPU that all the cores read the same.

There is always a bit of a difference in temps between cores, how are they under load.
Also try applying the TIM as shown in the pic. The full details can be found HERE.

this is what i got from playing crysis warhead maxed out

how much hotter would the cpu get after ocing to 3ghz?

well i downloaded prime95 and gave that a try on tortture test since alot of people use that

check out the results.first core shot up to 80celsius! after couple of minutes of use so i stopped the test cause i was worried that i would damage the cpu if i carried on,the second core seems not far behind.this cant be a good temp can it.

steveg123 said:

this is what i got from playing crysis warhead maxed out

Click to expand...

Run a program like Prime95 to load all 4 cores at once. Running a game will not give you even temperatures.

how much hotter would the cpu get after ocing to 3ghz?

Click to expand...

There are so many variables (voltage, cooling, case cooling, etc) that no one can accurately answer that question.

I wouldn't suggest running it faster if that is what you are getting now.

what bout the prime 95 results?

would reseating the cooler be the best bet?

Even running my Q9550 at stock 2.83 GHz with stock cooling my first core was 5-7 degrees C warmer than the others. Same thing once i got my water cooling going. Could be a flaw with the IHS on these quad core processors but I doubt it. Just one of those things i guess.

can people share your experiences with prime95?

80 celcius for the first core cant be good with the second core not far behind in the high 70s.

results are in the most recent pic of real temp i posted and that was after using prime95 for just a couple of minutes.

I would reseat the heatsink. 80c is way to hot for my comfort

my dual core will hit 58c @ 3.0Ghz on a HOT day with no A/C on

steveg123 said:

80 celcius for the first core cant be good with the second core not far behind in the high 70s.

Click to expand...

What kind of heat sink are you using? With my stock Intel heat sink I was reaching temperatures around there as well after a minute of Prime 95.

To me if you are running at stock with those temps something is wrong.
Firstly did you redo the application with the method recommended for HDT coolers.
Secondly what are your ambient temps. Lastly have you lapped the HDT cooler and/or CPU. Some people here will say never lap an OCZ Vendetta 2 as you will sand to far and ruin it.
Yes that can happen but you just need to be careful and lap slowly just to the point where it is as flat as you can get it. Finishing with about 1500 grit.
This is a screen shot I did the other night the temp in my study when I did the test was about 30°C and I ran Prime for about 15 mins. I know it was not run for long but gives a good indication. I did it for some people on OCAU to show what I was getting, in response to someone with a similar problem to yours.
I should add that I have a Sanyo Denki Fan that pushes about 110CFM on my OCZ at the moment. I am very happy with this and shows what can be done with a good application good cooler and TIM.

I imagine you need a reseat. Pull the heatsink off and check the contact patch of thermal material on the IHS. You want the entire IHS lightly covered during use but as thin as possible. You won't know if you can improve your temps until you try a remount so take a few minutes and give it a shot.

ok ive just remounted the vendetta,is it facing the right way?

If you have the air being pushed through the Cooler in the direction of the arrow you are OK.
Did you apply the TIM as shown for HDT coolers.
I would also look at some cable management, it is a little messy.

Cable management shmable shmanagement. (kidding )

You can tell from his fans' blades that your arrow indicates the current path of airflow.

firstly i find it a bit hard to to tell if the vendetta fan is exhausting the heat through the fins and out the case fan or if its merely blowing it backwards back into the case,i have put my hand close to it to feel but where the fan is weaker than the case ones its a bit hard to tell but im thinking its like the diagram on the picture below,blue arrows are for air being blowed that way and green arrows for exhausted air.

secondly i applied the arctic5 as you showed in the diagram.the second pic is the real temp reading after running for bout half hour mostly idle and browsing and using this forum.

lastly i will try and tidy up the cables more,is it dangerous to group the cables together heat-wise?

thanks for everyones input so far.

by the looks of your diagram your top front fan is sucking air in. Both your top fans should exhaust air for the best results. Idle temps don't really tell much of a story to wheather the cooler is mounted properly. I have a OCZ Vanquiser and the first time I installed it, one of the stupid push pin clips failed to hold. My idle temps were still low when idle but nearly fried my chip under stress. Now mounted properly my E4600 @ 3.3GHz runs at 35-38c idle and 55-60c under stress. Sounds like your Heatsink just isnt mounted properly. Now you have reseated it what are your full load temps? Any yes it's safe to group wires together.

can anyone tell me the right option to choose in torture test in prime95 that is right to try out my cpu and see how the recent temps fare?

Use second Option, where it says Maximum heat and power consumption. Its self explanatory.