From: muta...@gmail.com   
      
   On Sunday, July 18, 2021 at 9:30:47 AM UTC+10, Joe Monk wrote:   
      
   > > At the end of the day, the compiler needs to produce something.    
   > > If it isn't machine code that manipulates registers,   
      
   > The Burroughs machine architecture doesnt fit in the C hypothetical   
   > machine. It doesnt have registers. It has a stack architecture.   
      
   >> "Register machines routinely outperform stack machines".   
   >>   
   >> So there was an understandable pressure to move   
   >> from CISC to RISC, but not to abstract to a machine   
   >> programmed in Turtle Graphics or equivalent.   
      
   > https://en.wikipedia.org/wiki/Burroughs_large_systems#Stack_architecture   
      
   "Another point for speed as promoted by the RISC designers was that processor   
   speed is considerably faster if everything is on a single chip. It was a valid   
   point in the 1970s when more complex architectures such as the B5000 required   
   too many    
   transistors to fit on a single chip. However, this is not the case today and   
   every B5000 successor machine now fits on a single chip as well as the   
   performance support techniques such as caches and instruction pipelines."   
      
      
   Ok, thanks. So if the speed issue has been addressed, and   
   we can now perform work without the use of registers, is   
   that actually something that should ideally be done for the   
   next 1000 years or whatever of computing?   
      
   Based on what we know, which is always going to be   
   incomplete, is this the path forward?   
      
   Which apparently means ditching C90 in favor of Turtle Graphics.   
      
   I've been holding off on moving from C90 waiting to see   
   what Turtle Graphics actually is.   
      
   Thanks. Paul.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)