> Even on 32nm, instruction decoding was a tiny % of die space! Exactly that. Looking what the actual problems with the processors that tried to be better than

> Even on 32nm, instruction decoding was a tiny % of die space! Exactly that. Looking what the actual problems with the processors that tried to be better than x86-line were, the good or bad instruction "decoding" was certainly not what made them worse than x86-line, at the end.