From: minforth@gmx.net   
      
   Am 01.07.2025 um 21:56 schrieb Paul Rubin:   
   > minforth@gmx.net (minforth) writes:   
   >> Nobody seems to care about that time. Instead, the focus seems to be   
   >> primarily on code runtime, even though the difference is only   
   >> microseconds or less.   
   >   
   > I think in the Moore era, you got two speedups: 1) interpreted Forth was   
   > 10x faster than its main competitor, interpreted BASIC; and 2) if your   
   > Forth program was still too slow, you'd identify a few hot spots and   
   > rewrite those in assembler.   
   >   
   > Today instead of BASIC we have Python, and interpreted Forth is still a   
   > lot faster than Python.  That speed is sufficient for most things, like   
   > it always was, but even more so on modern hardware.   
      
   Today, you could go insane if you had to write assembler code   
   with SSE1/2/3/4/AVX/AES etc. extended CPU commands (or take GPU   
   programming...)   
      
   Even chip manufacturers provide C libraries with built-ins and   
   intrinsics to handle this complexity, and optimising C compilers   
   for selecting the best operations.   
      
   IMO assembler programming in Forth is mostly for retro enthusiasts   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)