Orthos/SP2004 error or my CPU??

[kent]

System:

Opteron 165 CCBBE 0617
MSI K8N Neo4-FI motherboard with latest BIOS
2x1GB Ultra DDR400
1 IDE HDD
2 IDE Opticals
X600SE PCI-E (don't ask...)
Antec TruePower Trio650 (Seasonic built)



Anyways, I've been Prime stable at stock voltage for 24 hours at the following speeds: 1.8GHz (duh, gotta make sure she works!), 2.2GHz, 2.5GHz, 2.6GHz

I bumped it up to 2.7GHz and I was curious of my temp when I only loaded up one core so I opened Orthos and in the Task Manager set the affinity to CPU 0 only and started Orthos.

It failed within a second.

So I let Orthos use both cores, it's ran for 5 minutes stressing both cores A-OK.

So I stopped it and reset the affinity to Core 0 only, again it failed within a second.

Reset affinity to Core 0+1, it's been running 5 minutes again.



Is this an Orthos error or what?

Using CPU #0
Beginning a continuous self-test to check your computer.
Press Stop to end this test.
Test 1, 800000 Lucas-Lehmer iterations of M172031 using 8K FFT length.
Test 2, 800000 Lucas-Lehmer iterations of M163839 using 8K FFT length.
Torture Test ran 5 minutes 28 seconds - 0 errors, 0 warnings.

Using CPU #1
Beginning a continuous self-test to check your computer.
Press Stop to end this test.
Test 1, 800000 Lucas-Lehmer iterations of M172031 using 8K FFT length.
Test 2, 800000 Lucas-Lehmer iterations of M163839 using 8K FFT length.
Torture Test ran 5 minutes 28 seconds - 0 errors, 0 warnings.
Execution halted.

Execution halted.

Using CPU #0
Beginning a continuous self-test to check your computer.
Press Stop to end this test.
Test 1, 800000 Lucas-Lehmer iterations of M172031 using 8K FFT length.
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.
Torture Test ran 0 minutes 3 seconds - 1 errors, 0 warnings.
Execution halted.

 

[kent]

Well, it didn't error out at stock clocks.

Increased voltage 3.3% (BIOS calculated 1.394 but it was a little lower actually), it failed after a minute and some change.

Increased voltage 6.6% (BIOS calc'd 1.44 but again, it's lower, 1.392-1.408 is typical, some higher spikes recorded but they're brief and no higher than 1.44 like the BIOS called for)

So far I'm at 3 min and some change.


Strange how on stock voltage it would stress both cores just fine for 5 min+ but when I instructed it to stress only Core0 it errored out immediately


Any idea, this doesn't make sense? Is this a sign of system instability with my overclock? How can stressing the cooler running core cause an error so quickly?

Do I really need to increase voltage?

 

[kent]

Bump damn. I know SOMEONE knows

It's not like Orthos is vaporware

 

[Arcygenical]

This is why I dislike Orthos...

2 Instances of Prime 95 really is the best way to test IMO...

 

[NordicRX8]

One instance of Orthos tests both cores at the same time... why would you set affinity to only one core?

Aren't you basically, manually failing one of the cores (as far as Orthos sees it)?


I'm not a programmer, so I don't know... which is why I'm answering your questions with more questions.

 

[PliotronX]

Indeed, ORTHOS is meant for SMP. Use the original SP2004 frontend if you're interested in testing one core of a multi-core setup.