View Full Version : CPU Voltage unstable, Ideas?

10-19-07, 06:54 PM
I have a Abit QuadGT and a Q6600 and the CPU Vcore voltage is set to 1.275 and under load it is 1.22-1.25 When I OC it to 3GHzI have to set the Vcore to 1.350 for it to get 1.26-1.32

When idle the Voltage rises and so does the heat and under load it drops quite a bit. Tis is limiting my OC ability.

Is this a motherboard flaw?

10-19-07, 07:14 PM
It's called v-droop.


There are mods to reduce it, but you need to know how to solder.

10-19-07, 07:14 PM
VDroop, you will need a vmod to overcome this

10-19-07, 07:16 PM
I know of Vdrop but this is extreme from what I'm used to seeing :rolleyes:

10-19-07, 07:21 PM
"V-droop" on the 965 can be fixed with a "pencil mod" as well. Here is the screenshot of the mod:


It's the yellow resistor, 4 to 6 soft strockes with a B2 pencil over the full restistor should lower it nicely. The spot is between the CPU and system ram near the upper edge of the mainboard.

I did it myself (but didn't have such a strong voltage drop) and it was very simple and worked surprisingly well! :)

10-19-07, 08:46 PM
I wonder what causes the Vdroop in the first place.... well more interested in why ASUS would knowingly allow this...

10-19-07, 09:12 PM
Would a newer chip set have less Vdrop? How about a 12 phase Gigabyte motherboard?

10-19-07, 10:29 PM
I think vdroops are often done on purpose to reduce voltages but it's quite possible that they happen because of lower quality components and/or limited R&D (money saving techniques). I can imagine the costs for mainboards would be quite a bit higher if the best components were be used and with longer development time they would hit the retail market later as well. As long as the boards pass their QA tests, they sell them. I doubt they care a lot about OCers or know that they will fix problems like vdroop themselves.

Vdroop on the P5B with a quad core CPU can be quite a bit larger than with a dual core CPU so a new chip doesn't necessarly have to mean that the vdroop will be lower.

10-19-07, 10:42 PM
So if all I am doing is shorting out that one resistor... why is it even there in the first place? Surely it has its place.