From: monnier@iro.umontreal.ca   
      
   Lawrence D’Oliveiro [2025-09-15 02:31:04] wrote:   
   > On Sun, 14 Sep 2025 10:52:10 +0300, Michael S wrote:   
   >> Personally, I never found the whole rounding modes business useful in my   
   >> practice.   
   > I remember one of Kahan’s writings suggesting it is useful for testing   
   > numeric stability: if your code gives results that differ only slightly in   
   > the four different rounding modes, then your calculations are *probably*   
   > stable; if the results vary a lot, then your calculations are *probably*   
   > unstable.   
      
   IIUC you can get the same result by adding a bit a noise to your inputs   
   and compare the output.  Maybe it's easier to change rounding modes than   
   to add noise to your inputs?   
      
      
           Stefan   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)