Hello, I am currently attempting to stabilize my slightly overclocked computer. Through some Prime95 testing, I've come to the conclusion that I need to be running at higher than normal voltages in order to be stable.
It is to my understanding that:
1) Intel states that the maximum safe voltage for VDIMM is 1.65v.
2) Intel states that the maximum safe voltage for VTT is 1.35v.
2) VDIMM and VTT should be within .5v of each other.
 
Things worth noting:
1) I am attempting to become stable with 12GBs of ram running at 2000MHz.
2) I am using a 6 core i7 3.2GHz processor (overclocked to 3.6GHz).
 
I had been trying to become stable at 1.65v VDIMM and 1.35v VTT because those are the maximum safe voltages according to Intel. The problem was that Prime95 was continuously returning errors no matter what I was tweaking. I gave up and decided to simply set everything to [AUTO]. I still edited the base clock and ram multiplier to achieve 2000MHz for my ram.
 
After doing this, I ran Prime95 once again. I did not run Prime95 for an extended period of time, but none the less it yielded no errors for the time it was running. I am using Prime95 v26.6 so it had 12 threads running for 6 cores with Hyper Threading.
 
The problem is when I went back into the BIOS and noticed the voltages. [AUTO] had set the VDIMM to 1.8v and the VTT to 1.4v, which are obviously quite high compared to Intel's stated maximums. The VCore, on the other hand, was quite low. [AUTO] had set it to less than 1.2v; I don't remember exactly what it was, but I want to say 1.175v. I am assuming this is due to a low base clock, which I should probably explain. My base clock was set to 143 (ram multiplier set to 2:14) because I wanted to make use of my processor's 25th clock multiplier (Yay for special features!).
 
Regardless, the main point of my question regards the VDIMM and VTT. Would running the VDIMM at 1.8v and the VTT at 1.4v be too high (damage my computer)? I ask this because those voltages are obviously higher than Intel's stated maximums, yet I am achieving stability through Prime95 with those higher voltages. Once I get some feedback about this issue, I will start doing some extensive testing at those voltages with Prime95. I didn't want to attempt it now because I was afraid of damaging my computer.
 
Thank you for any and all help with this issue, it is much appreciated! Thank you!

Sorry, I moved this thread from a different section of the forum. I have edited the original post accordingly.
 
I am using 6 sticks of 2GB ram.
 
I guess the basis of my question was whether Intel's stated maximums of 1.65v VDIMM and 1.35v VTT were absolute and should never be elapsed, or are they able to be set at higher voltages with no ill effects to your computer?

Don't use prime95 use linx instead.  Also as WD stated you are going to need more than 1.35 VTT to stabilize 6 sticks of RAM also may want to try altering the timings with a command rate of 2T.  Most of the XMP profiles call for 2T command rate on 2000mhz RAM.  also you may want to try 1.66 or 1.67 DimmV.  as long as you keep VTT and DimmV within .5 of each you're safe.  Intel states 1.65 DimmV and 1.35 VTT because they also state that the CPU can only run 1066/1333mhz RAM speed.  so attempting to go over that you will need to increase VTT and may even have to increase QPI PLL to compensate for the increased QPI speed(also known as NB).  So you might want to throw the idea of only using 1.35 VTT out the window.   if you want to use less VTT i suggest using only 3 sticks.  Even then its the luck of the draw as to how good your IMC is and if it can even handle 2000mhz.  So best advice manually set your timings for RAM manually change your VTT to either +200 or you might even need to go as high as +300 but this comes down to testing at each voltage.