From: ldo@nz.invalid   
      
   On Sat, 4 Oct 2025 11:15:05 +0100, The Natural Philosopher wrote:   
      
   > I think there is an issue at the core of all this as to whether, given   
   > similar fabrication densities, a CISC computer is inherently more power   
   > hungry than a RISC one.   
      
   Yes, it is. That’s why you see x86 at an inherent disadvantage in low-   
   power situations. That’s in the real world, not based on theoretical   
   considerations.   
      
   > I suspect that in the end they come out the same: the major power   
   > consumption is FET state transitions per second, and if CISC has more   
   > FETs, but they aren't being used all the time, I see no inherent reason   
   > why more power should be drawn ...   
      
   But those extra transistors are being used a lot of the time. Intel has   
   previously talked about some kind of “RISC-like core”, so extra   
   transistors are needed to translate from the programmer-visible CISC   
   instruction set into these “core” instructions. And they are needed all   
   the time, as new instructions come in.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)