smbergie200

Distinguished
Mar 23, 2008
23
0
18,510
Hello - I have a question regarding the Q9450 temps. Here is my motherboard and heatsink:

asus rampage formula
zalman 9700 nt with arctic silver 5 (applied with thin line like manufacturer says)

I run realtemp 2.60 and get these temps:

Idle: 50 50 44 44

Load: 59 59 55 55 with Prime 95

Now I have everything set to auto and have used the onboard cpu level up to a slight overclock to 3.2ghz
I also get these same readings without an overclock.

My question is are these temps normal?

I just want to game with all settings on auto and not have to mess with voltages and stuff. Everything runs great and stable and I can run Prime 95 with no errors for 12 hours.


 

smbergie200

Distinguished
Mar 23, 2008
23
0
18,510
The only benchmark that made the CPU too hot was OCCT 2.0.0a. I had the cutoff set to 70 and it cut the program off after a few hours. Shouldn't I be able to pass OCCT with flying colors?
 

kad

Distinguished
Feb 29, 2008
524
0
18,980

What is your ambient (room) temp. ??
Temps look OK but difference 4 degrees between cores temps might indicate improper HS seating or uneven TIM application or HS convex base that needs lapping
With same price you were able to get TRUE heatsink which is better
However check also you CPU fan if running at full speed 100%
Try also reducing Core voltage step by step (If you prefer to keep on Auto it's up to you) you'll spend lot of time and effort to adjust voltage
What is your core voltage now??
One thing more
Be patient for 200 working hours cause AS5 needs 200 hours curing , then your temps might go down
Another thing best way of applyong TIM is X mark, check this review
http://benchmarkreviews.com/index.php?option=com_content&task=view&id=170&Itemid=1&limit=1&limitstart=3

Square_TIM_Application_Cross.jpg

 

iluvgillgill

Splendid
Jan 1, 2007
3,732
0
22,790
well if the base is even on both the heatsink and CPU then it doesnt matter which way you apply the thermal paste.doesnt matter you do a line in the middle or X mark like above.

the link you gave is biased because the amount of thermal paste being put on the heatsink is different.if the X line style above with much thinner lines of thermal paste the result will be the same as all the other style.plus the 2 line in the middle give the same result as the X line style above.

and if both heatsink and CPU is 100% flat then no thermal paste would be the best as thermal paste will just add gaps in between the 2 perfect flat surface.but in real world 100% flat on both is impossible so small amount of thermal paste is still needed on a lapped heatsink and CPU.
 

smbergie200

Distinguished
Mar 23, 2008
23
0
18,510
core voltage is 1.184

Should I forget about the as5 and use zm-stg1 zalman thermal grease that came with the heatsink?

I am just new to building and I just want to make sure my temps are all correct.

If everyone thinks these temps are ok I will leave as is. I hated mounting the heatsink and would prefer not to do it again, but if necessary I will.
 

Zuwie

Distinguished
Jun 4, 2008
4
0
18,510
I have the Q9450 with the Abit IP35 Pro board and stock heatsink OC'd to 3.0(8x375, stock voltage) and my temps are no where near that. I've tried it up to 3.2 , but it wasn't completely stable and Prime95 failed after about 30 minutes and I don't want to increase the voltage. Also, why use realtemp over coretemp? I've seen several people say use realtemp. With realtemp I Idle anywhere between 27-32C and load 37-41C. On coretemp the temps are ~10C higher.
 

shadowthor

Distinguished
Jun 2, 2008
1,128
0
19,280
Realtemp has the correct TJ Max where as coretemp doesn't. Plus it has been tested by majority of everyone at xtremesystems.org .Realtemp has detected 45nm processor correctly plus it can check if the temp sensors on the cpu is stuck or not. If it is stuck, then it might not register temps correctly.