XPost: comp.os.linux.misc, alt.politics, alt.economy   
   XPost: talk.politics.misc   
   From: tnp@invalid.invalid   
      
   On 04/11/2024 18:20, John Ames wrote:   
   > On Mon, 4 Nov 2024 13:01:01 +0000   
   > The Natural Philosopher  wrote:   
   >   
   >> However fabrication limits are getting stuck at 10nm and below,  and   
   >> clock speeds are stuck at a few GHz which means that the   
   >> power-performance ratio is pretty much the same for Intel and ARM   
   >> architectures. Only by having fewer transistors and implicitly doing   
   >> less, can the power be reduced.   
   >>   
   >> I.e. Moore's law has basically stopped representing reality. And ARM   
   >> is no longer fantastic power performance compared with Intel. Its one   
   >> advantage is it doesn't have to support a legacy architecture. And so   
   >> its probably cheaper and less buggy.   
   >   
   > I've long held that necessity will ultimately force a serious rethink of   
   > programming practices w.r.t. resource-efficiency once Moore's Law runs   
   > afoul of pesky real-world physics principles, i.e. "eighteen inches is a   
   > nanosecond" vs. "you can't cram an arbitrary amount of stuff into a   
   > finite space without creating a black hole." Gonna be real interesting   
   > when we finally hit the wall.   
   >   
      
   It may be that computing as we understand it is simply a mature   
   technology, and there isn't much more to actually do.   
      
      
   --   
   There’s a mighty big difference between good, sound reasons and reasons   
   that sound good.   
      
   Burton Hillis (William Vaughn, American columnist)   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)