OC and intelburntest

[Mnx4]

I've been running a stable OC for a while, e2180 @ 3.2, 10x, 320FSB, 1.4125V, ram multiplier 2.5. I was a noob and didn't test extensively to find the voltage sweet spot, tried prime at that voltage and it worked fine so I just ran with it.
Recently, I figured I'd try to find the best voltage to run at those speeds. I've seen intelburntest around and saw that a lot of people really like it cuz it's quick, easy, and gets better results. Couple of questions though.
1. Will the results always say how many tests passed? I ran in non-error reporting mode last night and the final note read to check to see if there was disparity between the expected result and the obtained result. There was a difference, but it seemed to run just fine. The ram isn't OC'd, so would a 1/4ram or 1/2 ram test be alright? This test ran fine at ~1.38v, but the computer restarted at ~1.36 during the middle of the test.
2. Each run took an average time of 1575s, and five runs lasted 2h10m. Is that normal? I did do the extreme, full ram test. I did a custom test after that using only 1 run and 1024MB ram, and it took about 148s and got the message it passed.

Temps under normal load don't exceed 45C. IBT would average about 48-50C.

 

[Zero82z]

1. Will the results always say how many tests passed? I ran in non-error reporting mode last night and the final note read to check to see if there was disparity between the expected result and the obtained result. There was a difference, but it seemed to run just fine. The ram isn't OC'd, so would a 1/4ram or 1/2 ram test be alright? This test ran fine at ~1.38v, but the computer restarted at ~1.36 during the middle of the test.

If you run in standard mode, it will not tell you how many tests passed, but you can see that information easily on your own. If even one value is off, your overclock is not completely stable. All of the residual (norm) values should be identical.

2. Each run took an average time of 1575s, and five runs lasted 2h10m. Is that normal? I did do the extreme, full ram test. I did a custom test after that using only 1 run and 1024MB ram, and it took about 148s and got the message it passed.

1575s is about 26 minutes, which is way more time than it should take. I find that sometimes setting it to use all available memory can cause issues, so I recommend that you set it to 2-3GB instead of the full amount. That should still be sufficient to test your FSB and memory stability.

 

[Mnx4]

Thanks for the reply. I upped the voltage to 1.39ish and it's went well with 1024 ram and 5 runs.
How many runs should it go through to be stable? I've seen anywhere from the regular 5 to the inane 120 runs.
Also, does adding more ram to the test make it more reliable? Or does it simply make for a higher chance for the test to screw up? Would it be alright to say memtest is better for ram overclocking stability, whereas this is primarily for the processor and thus is not reliable especially for a system with more than 2gb of ram?

 

[Zero82z]

How many runs should it go through to be stable? I've seen anywhere from the regular 5 to the inane 120 runs.
Also, does adding more ram to the test make it more reliable? Or does it simply make for a higher chance for the test to screw up? Would it be alright to say memtest is better for ram overclocking stability, whereas this is primarily for the processor and thus is not reliable especially for a system with more than 2gb of ram?

I usually do 5 to get a general idea, then I do 20 to test more thoroughly. Adding memory means that it will put more strain on your RAM so there's a better chance of it failing if it isn't stable. Memtest is more of an indicator of RAM integrity than stability, but you can use both to test.