I really wonder however, how you manage to have THIS many hardware failures?!
Easy:
1) He's got a lot of hardware, and plenty of hardware is shipped faulty. So it's numbers
2) He's stressing the hardware, mining really puts hardware through it's paces, and quite a bit doesn't make it.
1) I don't consider 15 mainboards "a lot" - and there were more than 1 DOAs amongst them, reportedly. I don't even blame tawsix for anything, it's just very strage, that the hardware failure rates seem through the roof to me and far above anything I would expect.
2) Then he does something wrong - as he won't disclose exact specs, one can only speculate (no pun intended...). Might be for example a too cheap PSU model that is running too hot or something else like a simple BIOS switch or whatever.
Yes and no. One needs more info for saying whether this was funny. For example if they got all delivered in one run from one dsitibutor it may have been a bad batch and / or a transport issue.
I regularly buy enterprise drives. not so long ago I returned a complete purchase order within 3 weeks as one after the next drive went dead.
Statistical distribution is not normal here, often you have glitches on storage, transport or production.