I7 920, vcore in bios reads 1.225, in cpu-z 1.470

[acs24]

any reason why this would be.

trying to overclock to 4ghz on air and i'm at 190x20 right now with vcore set to 1.225. seemed to be running stable but i noticed my temps were pretty much maxed (70-80) while running prime95. quickly checked cpu-z and hwmonitor and both read cpu vcore 1.47?

 

[acs24]

fwiw at 180 x 20 with vcore set to 1.175 it reads 1.16 in cpu-z and hwmonitor.

why such a huge jump in voltage in the software? is this normal? i mean obviously i can't run it at these settings with those temps, what the hell.

 

[mcnuggetofdeath]

hyperthreading puts out crap loads of heat. Get a voltmeter and measure the voltage yourself. CPUz can be inaccurate.

 

[Shadow703793]

^Exactly HOW would you measure CPU voltage with a DMM (which contact points?)?

@OP: It's normal. It's vdroop. You wouldn't happen to be running a X58-UD3R would you?

 

[acs24]

it's weird, i upped to voltage one more setting and now it reads 1.27.

pc health status reads the same. not sure what the hell happened but it runs stable now. temps are at 47C.

^ nope, evga x58 tri-sli board.

 

[CompuTronix]

acs24,

Unfortunately, EVGA and Gigabyte boards have sloppy Vdroop characterisics, which makes overclock stability more difficult to achieve. Although I've been very impressed with these boards for their excellent features, sloppy Vdroop has always been the deal breaker. If you watch the Vcore signature in SpeedFan's "Charts" at idle, then run Prime95 Blend, you'll see just how bad it really is, which can exceed 64 millivolts. Each negative spike represents a potential BSOD crash point. When enabled, Load/Line Calibration can help to maintain an unchanging Vcore (little or no Vdroop) during fluctuating loads. The Asus boards will typically hold Vdroop to the bare minimum at only 8 millivolts, due to better voltage regulation.

Comp