From: user5857@newsgrouper.org.invalid   
      
   BGB  posted:   
   --------------snip---------------   
      
   > DIV uses Newton-Raphson   
   > The process of converging is a lot more fiddly than with Binary FP.   
   > Partly as the strategy for generating the initial guess is far less   
   > accurate.   
      
   Binary FDIV NR uses a 9-bit in, 11-bits out table which results in   
   an 8-bit accurate first iteration result.   
      
   Other than DFP not being normalized, once you find the HoD, you should   
   be able to use something like a 10-bit in 13-bit out table to get the   
   first 2 decimal digits correct, and N-R from there.   
      
   That 10-bits in could be the packed DFP representation (its denser and   
   has smaller tables). This way, table lookup overlaps unpacking.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)