If we're looking at a post-Moore's Law interregnum in process improvements before we find some new substrate then that might be a good opportunity to explore things like clockless designs when efforts can take longer to pay out.
And the truth is that we are. Here is a bunch of data on Intel's processes that I've pulled together from various public domain sources. The yield curves are a very rough estimate, so take them with a grain of salt.
Comments
If we're looking at a post-Moore's Law interregnum in process improvements before we find some new substrate then that might be a good opportunity to explore things like clockless designs when efforts can take longer to pay out.
And the truth is that we are. Here is a bunch of data on Intel's processes that I've pulled together from various public domain sources. The yield curves are a very rough estimate, so take them with a grain of salt.
https://www.nedbingham.com/intel_max_transistor.png
https://www.nedbingham.com/intel_switching_frequency.png
https://www.nedbingham.com/intel_tdp.png
https://www.nedbingham.com/intel_gate_delay.png