Nforce SPP/MCP and HT nForce SPP <-> MCP Voltage Questions

[jackcrackerman]

Ok, so I've noticed that these voltages are being scaled with my FSB core voltage. My question is, can having the SPP/MCP voltages set too high cause instability with PCI-e cards. I've done the Stability tests on all of my hardware and they all pass... but for some reason when I play 1 older game it causes the system to crash and reboot however I can play Battlefield 2142 for hours on end and not have any issues what so ever. So far I've contacted the game devs forums to see if this is a common issue however no one else has the problem. Would you guys consider this a driver failure or something else? I've even dropped my overclock on my videocard to see if it was the cause however it didn't help.

So anyways, I've noticed my SPP sitting at a 1.4v and my MCP at 1.5v... What voltages would you recommend for these?

 

[morerevs]

Did you lock the PCI and PCI-Express bus? If these get OC/ed aswell they can cause all sorts of problems from artifacts to data corruption on Sata. Also, what older game are you playing that causes problems? Some games have trouble running on dual cores and have to be set to run on one core only.
As for voltages especially on northbridge (since these run hot sometimes) set them as low as system stability allows you to. This will take some experimenting with taxing benchmarks. I have mine set to 1.35 in bios (1.40 in voltage monitor) and southbridge @ 1.55 (1.60 v.m). I got to these values using 3dmark03 of all things as this would consistently crash during the second CPU test (trolls lair) even though prime95 was stable for several hours and 3dmark05/06,superpi and everest ran faultess.
Anyways.... it could still be drivers too

GL.

 

[jackcrackerman]

Well, I've ruled out my drivers being the problem as well as my video card being overclocked. I restored all of my system values to default and booted up and started playing the game, worked for a good 40 minutes flawlessly. BTW the game I was trying to play was From Sony's Station Pass (Cosmic Rift).

So, now I'm back to the drawing board on what may cause these issues.

What about the SPP <-> MCP voltage? should I leave that on auto?

(Also, I have my PCI and PCI-I locked at 100mhz (my 680i mobo does not change these regardless of FSB overclocking)

 

[jackcrackerman]

Ok, I've completely re-done my overclocking... my results have fixed the problem that I was previously having and increased my CPU to 3.4ghz

Hardware Voltage
--------------------------------------------------------
CPU V-core: 1.5v
Memory: 2.2v
FSB: 1.4v
PCI-E (SPP): 1.3v
nForce (MCP): 1.5v
HyperTransport (SPP<->MCP) 1.2v

CPU Temps

Idles at 34c
while running Prime95 Torture Test's max heat/Voltage, the core reached a max of 59c (of course, there will never be a point when the processor will ever reach this much usage, I ran dual instances allocated to each core to provide 100% across both cores)
Normal usage of the CPU (IE: Games and file compression) it never passes 45c.