Q9550 & EP45 EXTREME STABILITY PROBLEM

Griff7520

Griff7520

Griff7520

Associate
Joined
2 Jul 2007
Posts
2,219
Location
Cheshire East
Hi all

I need a bit of help

I have managed to get my Q9550 to 4.0 ghz on vista 64 bit os and it seems to be running fine in normal use (I am on it now as i type)
The problem is when i try to check stability with OCCT prog it crashes after about 2 min of the test. Randomly it says core 1 error or core 3 error. I think i am really close to getting this stable and am wondering if it is a voltage problem.

Here are my settings in bios at the moment.

CPU VCORE .................1.3265
CPU TERMINATION ......1.32
CPU PLL .....................1.60
CPU REF .................... 0.84
MCH CORE ..................1.40
MCH REF..... ...............0.836
MCH/DRAM .... ............1.010
ICH I/O ......................1.500
ICH CORE .... ..............1.100
DRAM VOLTAGE ...........2.10
DRAM TERMINATION ......auto
CHANNEL REF ..............auto

i am running at 8x500 fsb with gskill pc8000 memory at 1/1 divider at 1000 mhz this gives 4.0 ghz. Temps are low 30's at idle and 50 to 52 at full load (using water)
vdroop goes down to about 1.285 on cpuz

any ideas if i need to tweak any of these settings to get it more stable or are any of these settings to high and dangerous? :)

thx in advance for any help
 
It looks alright so far. You could probably up the vcore slightly and see if that helps. You're still well within safe bounds. Anything up to 1.4 is fine. I have my PLL set to 1.5 although I don't know too much about it other than this board, if left on auto, will set that way too high.
 
OK.....changed to 8.5 x 371, upped cpu voltage to 1.375, set pll to 1.50 and ran OCCT for 15 min ok...I know thats not long but usually if its going to fail...it does it in first 5 min.

so far lookin better.
now going to try less cpu volts and upping the cpu term a bit....see if i can get the temps down a bit more...they were at 55 under load during the test.
 
update

Ran 4 hrs last night on COD5 without a problem. But as soon as you try to run a test like OCCT it fails after 3 or 4 min...for normal use it seems to be ok...but under a lot of stress it isn't. is this ok or is there more tweaking the voltages needed to get it more stable

thx
 
You must log in or register to reply here.
Back
Top Bottom