How should I interpret Orthos errors?

[Fireye0]

So, I'm running pretty happy with my current overclock. It's not too extreme, temps are acceptable. E4300 running at 340x9, 1.375v, DDR667@816 4-4-4-12, 2.0v (I think, can't see it in cpu-z), on a Gigabyte P35-DS3L.

However, after approximately 9 hours, Orthos failed. How should I figure out what caused the failure, and what to change?

Test 7, 1200 Lucas-Lehmer iterations of M52331647 using 3072K FFT length.
FATAL ERROR: Final result was 89B6A890, expected: A1EAE85D.
Hardware failure detected, consult stress.txt file.
Torture Test ran 9 hours, 18 minutes 51 seconds - 1 errors, 0 warnings.
Execution halted.

 

[Noubourne]

An error in Orthos means CPU instability. You need to increase voltage slightly, or lower your overclock until you can pass all stability tests.

Or, you can call it 8 hr stable and leave it alone and pray that you never hit that weak spot during normal usage.

 

[The-Noid]

Orthos rounding errors are CPU related.

That however is a memory error.

 

[Fireye0]

Ahh, terrific. So, an increase to the mem voltage is in order, or loosening up the timings if that doesn't solve it?

 

[The-Noid]

Yes that will do it.

 

[Fireye0]

Terrific, thanks a ton for the advice. An extra .1v on the memory has allowed my system to run stable on Orthos for about 16 hours now.

 

[The-Noid]

No problem buddy have a great day.