I wonder if x86 really is inherently inefficient, or if it could get there given enough investment (i.e. not just put a 2nd tier team on it for a few months but

I wonder if x86 really is inherently inefficient, or if it could get there given enough investment (i.e. not just put a 2nd tier team on it for a few months but really push it with major resources). It feels like they just did not want to bother, not that it was impossible.