Sign in with
Sign up | Sign in
Your question

Massive vdroop.

Last response: in Overclocking
Share
June 22, 2012 12:47:48 AM

Hello. I'm new to this site but not new to overclocking. I have what I believe to be advanced knowledge, but by no means am I an expert.

First, a little history. Up until last week I've been running a Q6600 on a Gigabyte ep45-ud3p with water cooling happily at 3.51 ghz for a couple years. After doing some research I recently upgraded to a G1.sniper 3 board, gskill sniper ram (2x4 gb 2133 mhz), 2500k, and zotac GTX 670.

I have my 2500k stable at 4.9 ghz but I feel there is more left to get from it. My problem is vdroop. It's massive, and I don't quite think it is normal. I have my voltage set to 1.52 (max safe as my research has found) and while idle vcore actual is 1.512, which is fine. Under load it drops to 1.452v. Any lower than 1.52 commanded and it will BSOD with a 101 code, which is low core voltage. If I bump the vcore to 1.525 command it will idle around 1.521-1.523 which I do not want. Is there any way to deal with the vdroop? Is running higher than 1.52v safe when it is idle? It still drops to just under 1.0v when the cpu clocks all the way down.

I have the H100 cooler, so no problem with temps. With my current settings, they are around 66-70c full load in p95 after 4-5 hours. I have arctic silver 5 paste on it.

More about : massive vdroop

a c 110 K Overclocking
June 22, 2012 2:32:44 AM

First off, 1.52V may be Intel's max recommended voltage, but I've seen reports of less than that even, killing chips within a matter of weeks (even DAYS, in some cases). 1.45V is as high as I would recmmend going and even that causes issues sooner rather than later. I have a personal limit of 1.40V, myself. I'll never go higher, except for a possible suicide Validation run.

Second, to stop Vdroop, enable LLC.
June 22, 2012 2:50:08 AM

Thanks for the advice. I had to set my llc to extreme for it to do anything at all. It was just on auto before. I'm only dropping about .012-.014v now. I put the volts down to 1.46 for now, seems stable enough until I have time to properly stress test it tomorrow.
!