Man, I'm running off of 5 hours sleep because I stayed up late last night trying to get some lower temps... this is starting to turn from being a hobby into an addiction
So first thing's first. I attempted the different HSF direction and placement, where I was attempting to pull air up into my PSU outake, and it actually increased temps (guess the HSF was pulling air faster than the PSU could). It's quite a shame because I really preferred that placement, as it made room for much better airflow.
I didn't have time to do any lapping yet, and thanks for the head's up on lapping the V2, I'm wondering if I'm going to bother doing that at all (and if so, I'll probably be so cautious that the difference it'll make will be negligible. Better safe than sorry).
I did add a second outake fan which I hadn't taken the time to install yet. Unfortunately, due to space limitations, I have to have the generic/weaker fan (80mm) above the 92 mm Scythe Kama fan. I might see if I can find one of those fan shrouds you mentioned so I could install my Scythe S-Flex E fan as my outake fan instead. Or, even better, I am seriously thinking of buying a new case this weekend with more room, and larger/more fan intake/output holes.
I reapplied my AS5, trying to fill the valleys between the heatpipes of the V2. I also have some Arctic Cooling MX2 on hand, and didn't think it was any better than AS5 from the looks of it, but I'm beginning to read reviews which suggest it is. Guess it's a good thing I am going to be lapping my CPU tonight, since that will give me a chance to switch compounds.
I also shifted my RAM to the dual channel slots that were further away from the CPU (again, since they do the dual channels on alternating slots, we're only talking about 1 CM difference), but I reinstalled the fans on them. I'll try removing them first thing when I get home to see how much difference they make... should be interesting.
Sadly, after all my efforts, I really did not see any difference in temps, but that's without allowing the AS5 to cure, so I'm hopeful that there may have been a slight gain (which would be more apparent in time), and will be even more with my efforts tonight.
Some additional comments to your response
-I do have the RAM set to 2T
-As for the VDROOP, I used a multimeter to measure the changes I made. With it set to 200Kohms, the initial readout was 96.1. My current readout is 26. I did get it down to 14 on my first attempt, but I felt that might be too low, so I erased some and got it to my current setting. I might try bringing it lower, but I figured the most important thing to achieve that there's no fluctuation in V between idle and load, and I've achieved that, so I'm mostly content with it. However, I know some people get their VDROOP down to 0.008, so I might try to achieve that.
As for my overnight testing, instead of testing my 3.33GHz settings with the slight V increase (I did save the profile, though, so I can test it anytime), I decided to try something a little different. I went for a 3.5 GHz test (8 x 425) with 1.36 VCore. Woke up today to find the test errored after the 1 hour mark (but no lock up). I bumped it down to 3.45 GHz before leaving for work today... would be amazing to come home and find it stable!
Which does somewhat bring me back to the original inquiry I had with this thread... testing error vs locking up, what are your thoughts on this? Is one a better sign than the other? My interpretation is when Windows locks up, it could be any single factor, whereas when I get a testing error, it's just that the VCore is too low. Does that sound accurate at all?
Anyways, thanks again for taking the time to help! You've been providing me great feedback and advice. I have a feeling I'll have a stable setting before the weekend. I'll give you some rep, but if this board is like others, my post count is probably too low for it to mean anything
Edit: Wow, didn't realize I was writing a novel there.Edited by RShannonCA - 6/18/08 at 8:02am