What is the longest you have run a torture test for before receiving an error? E.g. if you had stopped the test one minute earlier you may have considered it prime stable.
They are. I would use a test that stresses all the controllers on the Cpu like real bench. Intel recommend Xtu which I often use with say heaven also running.
I've ran the likes of real bench overnight error free and had games like WoW lock up within minutes, I don't put much faith in any of the benching software now.
They are. I would use a test that stresses all the controllers on the Cpu like real bench. Intel recommend Xtu which I often use with say heaven also running.
I've ran the likes of real bench overnight error free and had games like WoW lock up within minutes, I don't put much faith in any of the benching software now.
If people have speedstep and all sleep states disabled then stress testers work. They are pretty hopeless otherwise though.
Just because a CPU is stable at it's full clock doesn't mean it will be stable at it's intermediate clocks which get used in moderate load situations. This is why stress testers pass a CPU (at full load), then it will quickly fall over in a game once a more 'dynamic' load is present and the clocks/vcore drop. For a gaming rig running games is the only real test of stability.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.