My story with the EVGA 680i motherboard and a QX6700 is a strange one. Perhaps some people here can give me some insight.

System specs: eVGA 680i rev A2 (Quad core friendly)
Bios P27 (but everything I will write about occurs with P26 and P25 as well)
CPU: QX6700 A stepping
Ram 4G OCZ PC6400 Platinum XTC Rev2
SLi 8800GTX
SIlverstone Olympia 1000W power supply
DTec Fuzion WB, DDC2 with Petras Top, MCR320-QP rad with 3 Yate loon ML
All of the usual speed limiting parameters in the bios are off (speed step etc, GPU ex off also)

I am currently running at 3.5 GHz, FSB 1400 10x multi.
CPU Voltage is set at 1.5V in bios, the bios monitor shows 1.46V, and in windows, speedfan shows 1.46v idle and a drop to 1.42 under load.
Idle temperatures are at 39,39,33,35. Under load with orthos on two cores and prime 95 large fft's on another and prime 95 small ftt's on another temperatures are at 63 maximum over a 14 hour period. The system is very stable.
Ram is running unlinked at 800 mHz 4-4-4-15 2T at 2.1V (stock), but can also run at 850 prime stable.

All voltages are at auto. FSB is 1.4, SPP is at 1.5, MCP is at 1.5, and HT at 1.2.

Ok all indications from this is that I have a fine overclock, with no need to push any part of the system to high voltages.

However, something does not seem right to me.

My FSB is limited to 360 no matter what I do. I can sometimes boot into windows, but then the system will lock up even at a low cpu speed. (I've tried 1600,1601, 1625, 1650, 1700, 1750, 1800...all behave the same) I have increased all of the voltages to high levels (as documented on evga's site) and this does not change at all even with a low multiplier on the cpu. Linked vs unlinked memory makes no difference.

I can adjust the multiplier and FSB(Lower) to get my cpu up to 4 GHz but it is not stable at anything over 3.5GHz. At 3.8 - 3.9 GHz it can boot into windows but will usually lock up and require a shutdown. This is the same at stock voltages or with boosted voltages including the CPU up to 1.6V. Even at 1.6V I cannot get 3.6G stable.

If I try to increase the FSB to get 1600 and using a 9x multiplier, it will usually crash during windows bootup but if I do log in it will freeze. Even with lower multiliers. There seems to be no correlation at all with voltage selections in the bios.

One interesting point, is that the MCP temperature never seems to change in Speedfan no matter what the voltage. But this could just mean it is not actually measring the MCP temperature.

I have exactly the same issue with trying to overclock the ram. It is stable from 800-850MHz 4-4-4-15 2.1V but not at any other speeds such as 900 5-5-5-17 (2.2V) irrespective of the settings of the other voltages. Even with using optimum timings (as determined by the bios).

I am now wondering it there is the possibility that the voltages are actually not changing even though I set them in the bios, and the bios monitor shows that they have changed (although strangely, only the CPU voltage shows any kind of vdroop and the other voltages at all exactly the same as they are set).

It would seem that perhaps the voltage adjustment circuitry is doing nothing. Does this make sense? Is it possible that the CPU voltage can be changed but nothing else?

Or, is this just a characteristic of my particular CPU?

I would appreciate it if the experts here can shead some light on this.
Thanks!