From: no.email@nospam.invalid   
      
   dxf  writes:   
   > Perhaps I misunderstood.  So we agree Forth locals are unlikely to   
   > ever match C locals for performance?   
      
   This I don't know.  If the issue is parameter passing in registers,   
   maybe a fancy enough Forth compiler could do that.   
      
   > I don't know whether it's possible to make forth code using locals as   
   > efficient as forth code using stack operations.  What I do question is   
   > the necessity for it and the wisdom of it.   
      
   I think in case of an interpreter, locals might be more efficient, since   
   as the thread title says, they treat the stack as an array.  The   
   hardware is built to do that, so why not use it?  With an optimizing   
   compiler, I think they should usually be equivalent in principle.   
      
   > Certainly Forth Inc's early successes didn't rely on the existence of   
   > a standard.   
      
   In those days there was only one significant implementation ;).   
      
   > https://www.ultratechnology.com/antiansi.htm   
      
   I remember that from a while back and will look at again.  The context   
   though was a Forth chip with stack hardware, being compared against a   
   software interpreter.   
      
   I miss Jeff but must also remember that he was sometimes prone to   
   hyperbole.   
      
   Do you still use blocks instead of files nowadays?   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)