Oh yeah, not trying to say this level of failure is normal. Puget getting 5% failures, or thereabouts, is a typical historical failure rate and they are getting it by being more conservative than others. Ending else is running more aggressive defaults.

I was just trying to provide a few examples of real first hand failures. And most OCing doesn't break anything, but every once in a while you set some voltage and one part never works again, hard not to conclude it was the OC when the failure perfectly coincides. I suppose it could be coincidental, but that stretches credulity.