From: user5857@newsgrouper.org.invalid   
      
   Stefan Monnier  posted:   
      
   > > So, would it not be easier and faster to simply make a densely-packed   
   > > 128-bit Fixed-Point decimal function unit ?!? A rounding mode, and a   
   > > number of digits below the decimal point, in a control register ???!!!   
   >   
   > I'm not sure I understand your proposal correctly,   
      
   It is not so much of a question, as it is my mind rambling through the   
   possibilities. Loosely based on IBM 360--except 2× or 4× as long and   
   stored in densely packed decimal instead of 4-bit digits. No decision   
   on whether data is stored in registers or processed via memory.   
      
   >                                                    but the number of   
   > digits below the decimal point should not be a global setting because   
   > several computations can commonly happen at the same time with different   
   > number of digits below the decimal point.   
      
   OK, forgot that COBOL has each number defined with its own decimal   
   location.   
      
   Should each calculation have its own "rounding mode" or "what to do   
   with bits that fall off below the defined Lower-end" ??   
      
   It just seems to me that once the "container" is big enough to deal   
   with world GDP (of 2100) in the least valuable currency in the world,   
   that making the decimal point "float" adds little value.   
   >   
   >   
   >         Stefan   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)