From: steve@seesignature.invalid   
      
   On 11/23/2023 5:29 AM, Woozy Song wrote:   
   > Where X and Y are both real, and W is double precision. I presume they   
   > wanted extra precision for the arithmetic, but then it gets quantised   
   > back to 32-bit float.   
   > By the way, this code came from a large oil company with a team of   
   > programmers.   
      
   The multiplication by 1.0D0 of each term has no effect whatsoever. If   
   those were removed, the multiplication of Y*W would convert Y to double,   
   due to "mixed-mode arithmetic" rules, before doing the operation. Then   
   when X is added, it too would be converted to double before the   
   addition. Then the whole result would be rounded back to single   
   precision for the assignment.   
      
   If I were writing this, I'd use the standard conversion intrinsic   
   functions to make it clear what is happening, as I hate seeing implicit   
   conversion. For example:   
      
   X = REAL(DBLE(X)+(DBLE(Y)*W))   
      
   --   
   Steve Lionel   
   ISO/IEC JTC1/SC22/WG5 (Fortran) Convenor   
   Retired Intel Fortran developer/support   
   Email: firstname at firstnamelastname dot com   
   Twitter: @DoctorFortran   
   LinkedIn: https://www.linkedin.com/in/stevelionel   
   Blog: https://stevelionel.com/drfortran   
   WG5: https://wg5-fortran.org   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)