3 ms·
> GSMArena bought two iPhone 15 Pro Max from different retailers to test the overheating issue and they were unable to replicate it. Did anyone manage to "repl
by sangnoir 3y ago
> GSMArena bought two iPhone 15 Pro Max from different retailers to test the overheating issue and they were unable to replicate it.
Did anyone manage to "replicate" the explosion of a Galaxy Note 7 under controlled usage? Assuming no one did, what conclusion wiuld we draw from s failure to replicate on n=2?
- nfriedly 3y agoFailure to replicate doesn't mean that it's not an issue. They even acknowledged that it has happened to other people. It's just another data point. IMO, it sounds like it's a real issue, but not one that affects the majority of users.
- nailer 3y agoThe way you handle inconsistent failures IRL is the same way you handle it in software: Run a hundred instances of the test. Measure the failure rate. Make the change. Re-measure. In bash it might be a for loop running the test a hundred times, in person it might getting a hundred phones.
- KolenCh 3y agoIf you have an overheated phone as a rare event then you will at worst wasted a few hours of your life, if you have an exploding phone as a rare event then you will have your remaining of your life wasted.