OK, so I'm still curious about the underlying efficiency. Is there a better test of single core efficiency at the same TDP? Does Apple Silicon provide any noticeable improvement over other x86 or Arm chips in that scenario, or is it really just a straightforward function of process and node size?