I've thought this way for a while. Once (if) we start feeling the diminishing returns in transistor scaling, there will be more incentive in the computer archit

I've thought this way for a while. Once (if) we start feeling the diminishing returns in transistor scaling, there will be more incentive in the computer architecture side to experiment with the relatively fixed transistor budget that we have.