Problem overclocking opteron 165

[Thasp]

Hello. Even with CPU temps in low 40s and smobo temp in low 30s, I have iissues where it turns of. I tried at first, 1.425v, 9x300 with 2:3 mem divider. Then 1.45v Then 1.475v.

 

The issue is that, the 1.45v primed for six hours then turned off, while the 1.475v did a mere four hours of encoding then turned off.

 

I am trying long encodes now at stock speeds just to make sure the setup works at stock.

 

What should I try? I seriously never had the computer just turn off before, nor have I had an issue where higher voltage results in slightly less stability when temps are in the proper range.

[Thasp]

I began an encode at stock speeds at 4 AM, 8:30 AM it turns off.

 

So now I know it's not an overclock problem.. this sucks.

 

I reverted back to the august bios from the march one. I also saw that I had a graphics card ram heatsink on the PWMIC.. it was glopped with far too much thermal paste, so I reapplyed with a thin layer, similar to how I do the CPU.

 

If this doesn't work, I'll try my Venice again. If the Venice works, either the DAGF can't support an opteron, or this is a faulty chip.

 

Any suggestions or assistance would be highly appreciated.

[soloz2]

have you tried your OC at lower frequencies? you make it sound like you just booted at 300x9 right away. not all 165's are capable of those speeds

[rapty]

For a start what Stepping do you have?

 

Sounds to me its a cooling problem the heat has to be controlled alot more with Dual Core's so the Cores can work in sync with eachother Air Cooling isn't the best as the Temperature Fluctuates alot more letting the Cores become imablanced then leading to isntability problems, thats what it sounds like here try Primeing at a Lower Voltage with LDT at 1.4

 

It wont matter if the Temperatures are in range on the majority of Dual Chips its about Temperature Control Single are so much easier to OC even though the Fab of DC's are now excellent and will be getting better and better as the year progresses to December

[Thasp]

As my second post stated, the issue took place at stock volts/speeds too.

 

I reverted to the august BIOS and fixed the PWMIC heatsink up. It's been encoding for a few hours now overclocked, and at a cool temp, so I hope it doesn't shutdown again.

[red930]

I think that might be a PSU issue. When you put alot of load on your system, if the PSU can't handle it, it will usually shut off. Try a different PSU if you have one around for further testing. I hope this helps...

[l3lackh0le]

turning off sounds like it could be the psu crapping out esp if you're running stock. heat is obviously not your problem. when i OC to 3ghz i'm hittin mid 50's on air and no bsod or shutting down anything like that

[l3lackh0le]

I think that might be a PSU issue. When you put alot of load on your system, if the PSU can't handle it, it will usually shut off. Try a different PSU if you have one around for further testing. I hope this helps...

 

if you don't have another psu to test w/ you can try unplugging some parts like ur optical drive and stuff

[Thasp]

I don't know if it was fixing the PWMIC heatsink I put on, or reverting to the old BIOS, but one of the above worked.

 

I've been encoding for almost a day now without issue, and a cool PWMIC. 2.7 GHz, 1.45v.

[sauron18]

I'm having like the same sort of problem i'm gonna change my PSU soon, putting a evercoolg vc-re and something to cool down the pwm ic, my psu is a thermaltake 420watts 20pin with the adapter to 24pin, when i have my opty in 2.5 and when i increase the HTT to like 2.6 when i enter in a game there is some sort of interference in the sound.

[Thasp]

I hope you can fix it as easy as I could.

[Thasp]

The shutdown occured again while running prime, don't know why since all the fans were on high so nothing could have possibly been overheating. However, this time, it was able to encode tons of stuff for at least a day before the shutdown.

 

I tried the newest BIOS again, and went back to stock speeds, CPU center spread spectrum on. I put the auto-shutdown temperature as high as it can go(85c) incase that was malfunctioning and causing the issue.

 

If anyone has a suggestion, I'd be glad to hear it.