• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

QX9650 problems

R

Raap

Gawd
Joined
Jul 12, 2008
Messages
717
I've problems OC'ing my CPU. Well, not really problems as such, but I need an incredibly high core voltage to get it stable. I'm wondering if I just got unlucky with my CPU, or if I'm doing something wrong, since this is the first computer I've tried to OC.

My system:
QX9650
850W CoolerMaster PSU
Asus Maximus Extreme
4*2GB Corsair DDR3 1333( 9, 9, 9, 24 )
9800GX2


At the moment my CPU is at 3.8ghz( 10* 380 ), memory at 1525mhz( 8, 8, 8, 20 ) and MB FSB at 333mhz. This OC is stable, both the RAM and the CPU( getting about 38degrees idle and 60-65 degrees celsius when running Prime95, according to RealTemp with TjMax95C ).

Now to the problematic part. To achive this OC I need the following Bios voltages:
vCore@1.450( 1.408 in Windows )
Memory@1.92
NB&FSB@1.50'ish

Isn't that *very high* for the vCore? My OC seems to be fine, but the reason I brought it up was because I noticed other people could do a 4ghz OC on this CPU with only 1.30-1.38 Bios vCore. For me, on the other hand, the computer will fail when loading windows if I bring the vCore down to less than 1.42 or so. Any ideas? Am I ruining my CPU running the vCore this high? And I don't suppose there's any chance getting it to 4ghz, considering that would probably take at least 1.475v?

Funny thing is, when I tried a lower OC some time ago, 3.6'ish ghz I believe, I didn't even need to mess around with the voltage at all, I could just let it be at Auto and it would be perfectly stable.
 
That is a really high voltage, and I recommend you lower it, since you could damage your CPU with over 1.4V. From what I can tell, the most likely culprit is that your CPU simply isn't that great of an overclocker. However, considering you have a QX9650, it is very odd that you're having so much difficulty overclocking it. I suggest you take the voltage down to ~1.35-1.36V and play around with the other settings in your BIOS, like GTL settings, VTT, and PLL voltage, as well as lowering your RAM ratio and making sure that you aren't being limited by that.
 
I have my voltage set at 1.52v (CPUZ) to get the 4.0. It isn't so much the voltage but how hot it is running. I have a dual 120 radiator that keeps my temps in high 30's idle and low 50's maxed out. It has been running stable at this voltage for 4 months now. I have read that 1.6 is as high as you want to go but you need good cooling. I know a lot of people get 4.2 or more with less volts but mine won't go beyond 3.9 without higher volts.
 
fentzinator said:
I have my voltage set at 1.52v (CPUZ) to get the 4.0. It isn't so much the voltage but how hot it is running. I have a dual 120 radiator that keeps my temps in high 30's idle and low 50's maxed out. It has been running stable at this voltage for 4 months now. I have read that 1.6 is as high as you want to go but you need good cooling. I know a lot of people get 4.2 or more with less volts but mine won't go beyond 3.9 without higher volts.
Click to expand...

1.4V is generally the accepted limit for 45nm chips. Anything over 1.45V will permanently damage the CPU.
 
I updated my bios and messed around some with the other settings( following mostly the Maximus Extreme guide here http://www.anandtech.com/mb/showdoc.aspx?i=3172&p=10 ), and it seems to have improved things a bit. At 1.40vCore it now runs stable at 3.8ghz, and even manages to get into Windows with 4ghz. I still can't seem to make 3.8ghz stable at anything less, though, so I guess the chip isn't the greatest. I'm also wondering if my NB running hot could be affecting my OC; even at Post it's at 50-60 degrees, but supposedly these boards do run hot without water cooling.

Crazy differences though, if some people can do 4ghz at 1.30 vCore or less, and my chip can barely do 3.8ghz at 1.40 vCore.
 
I know I'm pushing the limit at 1.52 volts but the 1.4 volts limit for 45nm is for the VTT voltage as I understood it.
 
Raap said:
Crazy differences though, if some people can do 4ghz at 1.30 vCore or less, and my chip can barely do 3.8ghz at 1.40 vCore.
Click to expand...
That's down to the stepping really. I don't know of any C1 or C0 quads that can do 4GHz at 1.3V, and most can't do 4GHz altogether. 3.8GHz is the upper limit for most C1s, although you do seem to need quite a bit of voltage, but as long as it's stable, you're fine.
fentzinator said:
I know I'm pushing the limit at 1.52 volts but the 1.4 volts limit for 45nm is for the VTT voltage as I understood it.
Click to expand...

You understood absolutely wrong. That limit is for the Vcore, not VTT. Intel's own datasheet specifies that 45nm Core 2 CPUs are not guaranteed to function altogether at voltages higher than 1.45V.
 
You must log in or register to reply here.
Back
Top