Has anyone else noticed this similar issue?  Take a look at my screenshot.
 

 
Notice how the run time reported in LinX is quite different than the run time reported by RealTemp (15:50 vs 14:59).  I opened both programs within seconds of eachother.  The longer I stress tested, the greater the variance became.
 
I have noticed this for quite a while now and never paid much attention to it.  It exhibits itself on the screen as the programs appearing to lock up.  Both program counters stop for a brief period of time every now and then.  But the LinX clock jumps forward to the right amount of time once it starts counting again, while the RealTemp clock just keeps counting from where it left off (hence the time discrepancy).
 
This didn't concern me before because I figured it was probably pretty normal and was due to LinX having a large amount of system priority and causing other services to lag behind.  But now it is concerning me a bit more because now I am starting to notice that my video card driver crashes every now and then while stress testing the CPU.  The screen will go black and the monitor will blink on and off a few times, and then come back on saying that the video card driver has recovered from an error.  But I don't have video card driver errors in normal 3D use or even 2D use.
 
I am wondering if I need to raise chipset voltages to stabilize the video card during high CPU load or if the video card symptom is also a normal symptom of the video card driver timing out due to not receiving data in time due to LinX having a high priority.
 
What are your thoughts?
 
EDIT:  See post #10 for results.
http://forums.evga.com/fb.ashx?m=647208

Thanks.  I will give it a try.

YerBuddy

I had the same thing the other night.  I was not running LinX or RealTemp at the time.  I had not noticed the run time differences before.  Interesting find.  Please let us know if the vtt increase works...or whatever you do.  I'm waiting on a replacement MB right now or I'd take a look myself.

I raised VTT from +50mv to +125mv.  It hasn't had a video card driver crash yet during CPU stress testing.    But who knows. 

emersj

Doesn't the IOH vcore need to be 1.2v to stabilize the graphics cards?

I don't know.  Does it?

Evga Precision and MSI Afterburner caused this issue on my system even at stock. Try closing these programs.

I had the same 758 x58 as you and with 3 260's needed to run an ioh of 1.3V - 1.4V depending on what benchmark I was running. Temps of the nb / sb on that board are really good so you don't have to worry too much.  Vreg on the other hand is a different story on water that thing could hit 80c easy...
 
Aslo if you think you aren't getting as high of an overclock with your CPU you probably aren't. I've gone through three 758 boards. With my 920 D0 the highest I could OC on DICE was 4.6GHz and this was just for CPU-z lol no benching. No matter where I put or took away volts I couldn't get past blck of 222. So I that I had a bad batch CPU. Now that I have a 4-way classy with the same exact CPU I'm able to OC and bench SuperPi @ 4.7GHz on water. Just thought I'd give you a heads for encase you have OC issues.
On the other hand the 758 was great for me in vantage with Ocing the GPU's, my 260 could run 800 core in vantage. With my 4-way I haven't gotten past 756 core yet, still new to the board though....
Since your bot I wanted to share

 

Don't get me wrong the 758 for everyday uses and gaming is great but since I love benching for boints this board wasn't cutting it for me.
I got my 4-way classy for $300 which is something I couldn't pass up. Although if I had the chance I'd go with the REX III from Asus because the fact that my 260 won't OC as high but I've yet to post my bios temp for help so it could be user error