Q6600 OC on XFX 680i... Stable OC no longer stable.

[jerreece]

So I had my system OC'd to 3.3Ghz and things ran fine for 2 evenings of Battle Field 2142 without a glitch. This evening while playing BF2142 my system has frozen 8-10 times. I've tried moving FSB up and down a few notches, everything from 3.2Ghz to 3.32Ghz. Still the same thing after 15-20minutes.

Interestingly enough, I had my vcore at like 1.475 before. I just moved down from 3.3Ghz to 3.27Ghz and put all the voltages on Auto. The system ran fine with the BIOS settings vcore at 1.4 even. Eventually it froze again.

I had set my FSB voltage to 1.5v previously, but <Auto> is setting it at 1.4v. Again system runs just as long before freezing. So apparently I must have had all my voltages set higher than I really needed. This stupid 680i LT is annoying me. 😉

Anyhow, granted I have a 680i which isn't the best for OC'ing... but anyone have ideas as to why I can't get it completely stable? Prime 95 used to run for over an hour on my initial 3.3Ghz setup without failing.

I could just go back to stock settings, but who wants to do that?

Current BIOS settings:

FSB @ 363.5x9 = 3271MHz
vcore 1.4v (auto) CPU-Z = 1.416v @ idle
FSB 1.4v (auto)
RAM 1.85v (auto)
nForce SPP 1.4v (auto)

Memory 5-5-5-18 (stock is 4-4-4-12) @ 800Mhz (PC6400)
No video card OC.

I just thought of something. Could I possibly be running into a hot MCP or something on the mobo? I've heard they run hot on the 680i anyhow. I've got plenty of fans blowing at my CPU to keep it cool, and Core Temp and the CPU-ID Hardware monitor don't show me mobo related temps. If it got hot after 15-20 mins would that cause my freezes?

 

[ausch30]

Heat could be the issue, I had a P5N32-E SLI and I bought a fan and mounted it to the NB because it got so hot. 3.3 on a 680i with a quad is very good and I would be shocked if you kept it stable at those speeds.

I had a E6300 overclocked to 2.8 on my P5N for about 6 months and one day it just wouldn't boot. I ended up going back to the stock speed because no matter what I tried I couldn't overclock.

I would clean everything out with some compressed air and get some cooling on your NB, that might help.

1hr isn't stable, you should run Orthos or Prime95 for at least 10-12hrs (I run it for 24hrs) and if it's still running strong then your stable.

 

[jerreece]

I'm trying to figure out how to monitor the temp on the NB itself. I'm reading that the MCP on the 680i's runs hot. Playing with Sandra Lite right now, as it has a "burn in" ability that is supposed to monitor temps. But all it shows is CPU, Board, and several Auxiliary temps. Only 3 temps show on the graph at all.

Trying to specifically find the MCP or NB cause I'm wondering if that's the deal. Prime 95 runs over an hour, yet a simple game like BF2142 (which doesn't push me at 100%) freezes the system after 15-20 mins.

 

[ausch30]

I checked the temp by touching the heatsink and when it burned my finger I knew it was too hot. I don't know of a way to check the temp of the MCP except for a thermal probe or a laser thermometer.

 

[roadrunner197069]

If you want to OC stable buy a better mobo and rest with prime minimum 8 hours. 1 hour barely runs any tests.

 

[jerreece]

Something is wrong here. I reset everything on the Mobo back to Defaults. Running everything at STOCK 2.4Ghz on my Q6600.

I opened up Everest Ultimate Edition Trial and told it to run it's Stability Test. (I tried this at 3Ghz and it froze)

Running it at stock, Core Temp .97 shows all my CPU cores in the 44c-45c range with all cores at 100%.

After a couple minutes Everest shows "CPU" at 101c! Suddenly a fan inside the PC kicks on pretty loud, yet Core Temp shows the cores are in the 40's. So is Everest reading "CPU" wrong? Or applying "CPU" to something else?

EDIT: At Idle it is showing "CPU" at 94c, "MCP" 43c, "Motherboard" 28c. Everest correctly shows the individual cores in the 30's to 40's as it sits idle right now. As I'm running "Default" settings I have C1E turned on, so the CPU cycles down to 1.6Ghz at idle, and 2.4Ghz under load.

I'm using the latest Everest Ultimate Trial from download.com. It can't be reading "CPU" correctly. 94c is way to freaking hot, and obviously Core Temp wouldn't show me 31c as my highest core right now if CPU was really 94c.

What could Everest be reading as 94c????

 

[ZOldDude]

Fix the OS to -halt on all errors- so you can see -why- the system is crashing then fix the issue rather than grasping at straws over what the problem is.

 

[jerreece]

LOL I opened up CPUID Hardware Monitor, and it is showing the following:

SYSTIN 81c
CPUTIN 94c

SYSFAN 5400 RPM
CPUFANIN0 3900 RPM
CPUFANIN1 2600 RPM

Core 0 31c
Core 1 27c
Core 2 26c
Core 3 30c

 

[jerreece]

Not sure what the deal is. I booted up my PC this morning, and am sitting here at idle.

SYSTIN -1c
CPUTIN 30c

I don't understand why lastnight at idle these same things were 81c & 94c respectively after running Everest, and then closing Everest and sitting at idle for 20 minutes.

If I run Prime 95 right now I get:

SYSTIN 25c
CPUTIN 46c

Shutting down Prime 95 results in those temps immediately going back down. I just began running Everest's stability test again. And all temps appear okay, until 2 mins into test when CPU suddenly hits 103c & the fans in the PC audibly charge up.

Everest must be testing something Prime 95 isn't testing, that ultimately is stressing something on my motherboard. What I don't understand is why CPUID Hardware Monitor shows CPUTIN & Everest shows CPU @ 100+ while both (and Core Temp) show each core in the 30's to 50's when idle or full load.

After Everest is stopped CPU hangs out around 98c... where before using Everest it was 30c. Where's the glitch?