• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

CPU Stability Test

Prime95 does the job fine for me. Very handy with the Speedfan UI for a visual representation of which cores have failed and been cut from the test.
 
Prime95 doesn't use modern instruction sets though so it doesn't prove 100% stability.
 
I wish there was a definitive answer to this question.

Ultimately one comment will tell you to "use x" while another will say "x isn't enough you should use y". Then you hear that "y is a power virus" and "even though the cpu passed y the system was unstable in games. You should use z".

I can't give you an answer but don't worry too much over it. Pick your poison and let it run, then if all is stable start your software of choice and if all is still good then set and forget.
 
Depends on the platform.

A mixture of XTU, AIDA and Realbench are sufficient for a gaming rig. There isn't a definite answer as different architectures fall over in different ways. What you don't want to be doing, however, is hammering AVX Prime on high core count CPUs (namely on Intel's side).
 
Cheers everyone for the feedback, gave 'Intel Burn Test' a go since i had never used it before. Seemed to the do the trick and a very simple program with an equally simple layout.
 
The best test is just using it normally. If you get a crash, bsod or whatever then tweak more. Don't spend half your life benching and running CPU tests.
 
As a quick test bf1 or BFV quickly show instability in overclocks, if it is stable for an hour or more you are heading in the right direction. I have had my machine pass hours of realbench and memory tests but fail fairly quickly on bf1 / bf5.

As above it depends what your intended usage is.
 
DarkHorizon472 said:
As a quick test bf1 or BFV quickly show instability in overclocks, if it is stable for an hour or more you are heading in the right direction. I have had my machine pass hours of realbench and memory tests but fail fairly quickly on bf1 / bf5.

As above it depends what your intended usage is.
Click to expand...

Exactly! Stress test isn't real life.

People who do 24 hour stress test have to much time on there hands
 
what do you do with your system ? if its game . just test with games.intel burn test like many other stupid tests just put unneeded stress that you will never replicate on your system. bench the same stress you will use your system for.
 
koooowweeee said:
The best test is just using it normally. If you get a crash, bsod or whatever then tweak more. Don't spend half your life benching and running CPU tests.
Click to expand...

Trying to find your highest clock without using quick stress tests, from which to know what to back off from, would be a right pain.
Normal use is then the final decider, but if I find through that I'm not stable, then one notch down will usually do it.
I use OCCT.
 
Last edited:
Using 1.330v (Skylake) at the moment but have noticed with adaptive mode i spike to 1.360v which i know is still safe but was wondering if a manual setting of 1.330 for 24/7 is resonable for long time use?
 
Safetytrousers said:
Trying to find your highest clock without using quick stress tests, from which to know what to back off from, would be a right pain.
Normal use is then the final decider, but if I find through that I'm not stable, then one notch down will usually do it.
I use OCCT.
Click to expand...

Yeah I use a quick test. But that's it, a quick test.
 
You must log in or register to reply here.
Back
Top Bottom