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

Q9450 + P5E3 - big differences in the temp of cores

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

Necrus

New Member
Joined
Mar 25, 2014
Location
Poland
Hello,

I have the following set :
Q9450
P5E3 Pro
2x2GB DDR3 1333 CL9 Goodram
Plextor SSD 128GB + Seagate 250GB M5Pro SATA
PSU Corsair HX520
Palit Sonic 1GB GTX560ti
Win7 Pro

Due to the work and the fact of having a small child I do not play games much, but when I had some time for myself my stuff mentioned above allowed me to play interesting games for me in more or less high settings. But reading the comments about X48 chipset I was tempted to have some fun with OC . I would like to add that this is my first OC .

After reading a million online guides about overclocking " for Dummies " I started releasing my Q9450 @ 3200MHz. The CPU @ 3,2GHz , RAM @ 1333MHz 1.5V. This is not complicated to achieve such a result - only changing the FSB to 400 - the rest on Auto and it worked. But being green I tried to do it on box cooling , so the CPU temperature reached to 80 C degrees in the stress - and this interrupted me OCCT tests .
I decided to buy some better cooling. I decided to budgetary SiletiumPC Fera2 cooling as I have read good opinions about it.
And here , on the one hand, the problem is solved , because my temperature in stress after OC oscillate around 70 degrees, and without OC arround 50 degrees. In Idle I have a 35 - 37 degs. But there is a problem with tempreature of cores. After installation of the new cooling I have noticed that the cores heat unevenly . # 0 and # 1 have correct tempearture at the start - the 35-37 degrees and then go up, while # 2 and # 3 are starting with 51 degrees. Then, the temperature increases in stress analogously to the # 0 and # 1 , but after going back to Idle - the first two cores cools to the starting temperature , while the latter two are at 51 degrees and do not descend below . Earlier on the box cooling everything was evenly on 4 cores .
I would add that I read the temperature using three different programs - HwInfo64 , RealTemp and said OCCT and everywhere the same. I generate load using OCCT or Prime.

Before I decided to write here I have read the various forums and found three possible causes:
Suggestion 1 a lot of people have written that the core # 0 is generally warmer than the others, because often " takes more work " than others - this situation does not concern me - for me # 0 is just one of the cooler ones

Suggestion 2 I applied unevenly paste or badly installed cooling. I took that to heart - I'm fresh , so I decided that this probably is the cause - I removed the cooling - carefully cleaned paste (by the way - I use this attached to cooler) - I put it all over again trying to evenly distribute . Tightened the back heatsink ensuring that both sides were tight and unfortunately nothing has changed - again # 2 and # 3 started with 51 degrees.

Suggestion 3 I've seen reviews that temperature sensors on the CPU can be problem as they can work inproperly. This option is most worrying for me . Unfortunately I am unable to verify it without your help .

And here 's my request - after reading what I have wrote above (BTW sorry for my english - it is not my mother language) are you able to suggest what is wrong or what I'm doing wrong? I would add that I do not care for extreme OC - I'd like to set the CPU to 3.2 or 3.3 MHz as I can reach it without tunning the memory. I do not know how these Goodram sticks are capable for OC , so I prefer to let them around the nominal settings . For more "extreme" memory now I do not have the budget right now.
Currently, the CPU goes stable @ 3200 and no overheating (unless around 70st in the stress is too much then please let me know, but 51 degrees on two cores at Idle worries me a lot.

Is there any possibility of verifying whether it is actually a problem with the sensors ? And if so, can I somehow fix it? And if not sensors then what is it? Is there a possibility that I could somehow physically damage these sensors while installing new cooling ? Because as I mentioned earlier while doing the first steps on the box cooling all cores had more or less the same temp.

I will be grateful for hints.
 
OP
N

Necrus

New Member
Joined
Mar 25, 2014
Location
Poland
Hi,

thanks for the answer.

I used the "test sensors" in real temp, but the only thing what it does is to load the CPU to 100%, then to 80% and so on until idle. The temp. were going down and #2 and #3 were left on 51 degrees and the test said nothing :(

What worries me most is that before starting the fun with the OC everything was OK - all cores have more or less the same temp - in stress and in idle mode. DO you think that the sensor could be broken while doing OC? Or maybe there is some method to reset the sensors? I was trying to find some info on the Intel page but unsuccesfully.
 
OP
N

Necrus

New Member
Joined
Mar 25, 2014
Location
Poland
OK I have read the article, and there is nothing I can do to reset these snsors - I will have to live with them as they are :( They have just stuck and it happens with CPUs 45nm.
 
OP
N

Necrus

New Member
Joined
Mar 25, 2014
Location
Poland
And as I mentioned I am green in OC - the temps arround 70C in stress @3,2GHz are allright and safe for my CPU?
I have also read that it is better to set the Vcore manually than leave it to AUto - what would you suggest?
 

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
Yes, they are.

Whatever the minimum voltage is needed to be stable. Every CPU is different, so you will have to test and find out. Find what it says on LOAD (Prime 95) in CPUz and set that manually in the bios and work your way down.

It will help you to read the overclocking guides we have for that CPU... shake a lot of that green off!!
http://www.overclockers.com/forums/showthread.php?t=515316