>common mistake to conflate the efficiency of an algorithm in the abstract sense to how those benefits are manifested in the applied sense > While true, all th

>common mistake to conflate the efficiency of an algorithm in the abstract sense to how those benefits are manifested in the applied sense > While true, all things held equal a new algorithm will advance the field, providing "gains" as you said > but it's false to assume that algorithms are more important than understand computer architecture, full stop I wish there were two like buttons. I have nothing to add that you haven't said better then I could have.