From: dxforth@gmail.com   
      
   On 19/07/2025 1:34 am, Anton Ertl wrote:   
   > dxf  writes:   
   >> So in mandating bit-identical results, not only in calculations but also   
   >> input/output   
   >   
   > I don't think that IEEE 754 specifies I/O, but I could be wrong.   
      
   They specify converting to/from external representation (aka ASCII).   
   If the hardware does it for me - fine - but as an fp implementer no   
   way am I going to jump hoops for IEEE.   
      
   >> IEEE 754 is all about giving the illusion of truth in   
   >> floating-point when, if anything, they should be warning users don't be   
   >> fooled.   
   >   
   > I don't think that IEEE 754 mentions truth.  It does, however, specify   
   > the inexact "exception" (actually a flag), which allows you to find   
   > out if the results of the computations are exact or if some rounding   
   > was involved.   
      
   AFAICS IEEE 754 offers nothing particularly useful for the end-user.   
   Either one's fp application works - or it doesn't.  IEEE hasn't changed   
   that.  IEEE's relevance is that it spurred Intel into making an FPU   
   which in turn made implementing fp easy.  Had Intel not integrated their   
   FPU into the CPU effectively reducing the cost to the end-user to zero,   
   IEEE would have remained a novelty.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)