Intel still had leading edge node supremacy at that point, and everyone go caught with their pants down with the end of Dennard scaling. AMD simply lucked out

Intel still had leading edge node supremacy at that point, and everyone go caught with their pants down with the end of Dennard scaling. AMD simply lucked out that they weren't designing a core to hit 10Ghz in the first place (because what would become GloFo didn't have the node chops to even pretend that they could hit that in the first place even ignoring Dennard scaling issues). AMD therefore didn't have to pull slower mobile core designs off of the back burner like Intel had to which took time to readjust. Intel's in a much worse position now. Their process woes aren't issues that the rest of the industry will hit in a year or so like the end of Dennard scaling was.