That is why I mentioned I (used to) test for each (Small for the CPU and Blend for memory and CPU)..
As far as AIDA, I let it run at the default settings which tests a bit of everything, then check off just the CPU and test. I will also put up an hour of Realbench for giggles. For my uses, this has worked out quite well for me.
People put a lot of stock into being '100% stable' (whatever the heck that means), but unless you are crunching production math/science and truly need it, I don't see a point in all that testing anymore. For my PC uses, as mentioned above, my method works.
As far as AIDA, I let it run at the default settings which tests a bit of everything, then check off just the CPU and test. I will also put up an hour of Realbench for giggles. For my uses, this has worked out quite well for me.
People put a lot of stock into being '100% stable' (whatever the heck that means), but unless you are crunching production math/science and truly need it, I don't see a point in all that testing anymore. For my PC uses, as mentioned above, my method works.