"Stability Testing" Practicality and Necessity Question

[vexion]

Hi guys,

I'm new to overclocking. I just built a new computer with an E8400 E0 and a Xigmatek HDT-S1283. I OCd it up to 4 GHz at 1.32V (Vdroop rendered it about 1.30V). It's keeping very cool at idle (35-40 Celsius) and not much hotter at load. I haven't done any real stability testing.

I did fire up Intel Burn Test and run just 5 iterations at the 50% (nice and easy) load setting. It bumped my core temperatures all the way up to 69. I realized my OC might not come across as fully stable if I run Prime95 for 24 hours or a harder Intel Burn Test.

So I might stability test later, but my question is, is there really a necessity for it? I'm not doing anything that will load my processor intensely on a frequent basis. I don't fold, I don't encode video. I mainly just game. For the last couple of days my computer hasn't had any crashes or hiccups, the OC seems "stable" from a general use point of view. Should I bother torturing it to make it fail?

 

[TooBooKoo]

I guess it really just depends on how "safe" you want to be. I usually just run Prime95 for about 4 or 5 hours to see where that gets me... and I like to use the blend mode. Although, I have heard stories of Prime95 giving errors after 18 or so hours in. Sometimes an unstable overclock can corrupt data on your hard drive as well... this is why I backup all my important data before giving the processor any more juice. Your temps and voltages look pretty good so hopefully you'll have a nice stable overclock there.

 

[Juggafat]

My computer won't pass orthos or prime95 but it will pass 30 runs of intelburntest and it has never crashed on me or BSOD so for me....its stable enough.

 

[Ocellaris]

Hi guys,

I'm new to overclocking. I just built a new computer with an E8400 E0 and a Xigmatek HDT-S1283. I OCd it up to 4 GHz at 1.32V (Vdroop rendered it about 1.30V). It's keeping very cool at idle (35-40 Celsius) and not much hotter at load. I haven't done any real stability testing.

I did fire up Intel Burn Test and run just 5 iterations at the 50% (nice and easy) load setting. It bumped my core temperatures all the way up to 69. I realized my OC might not come across as fully stable if I run Prime95 for 24 hours or a harder Intel Burn Test.

So I might stability test later, but my question is, is there really a necessity for it? I'm not doing anything that will load my processor intensely on a frequent basis. I don't fold, I don't encode video. I mainly just game. For the last couple of days my computer hasn't had any crashes or hiccups, the OC seems "stable" from a general use point of view. Should I bother torturing it to make it fail?

Really depends on how much you care about your data and PC reliability. The point of stress testing is to test your PC to ensure (within testing limits) that your OC is not going to fail. That way when you actually do something important on the PC, it does not choose that moment to blue screen or instantly reboot. Given that doing 10+ runs of Intelburntest on max does not even take that long (or you could just run that or Prime95 overnight), I do not see the reason for avoiding it. A little bit of extra effort now to ensure your 4GHz is fine could save you a lot of time down the road. And if you really can not find the time to run 10+ runs of Intelburntest on max and let Prime95 run overnight, you are really going to be screwed on time if you had to reinstall your games and apps.