From: bop@gmb.dk   
      
   Seungbeom Kim wrote:   
   > On 2012-02-10 14:14, Francis Glassborow wrote:   
   >> How an actual system represents values is irrelevant, the C++   
   >> Standard requires that an implementation conforms to the abstract   
   >> machine. In C++ this abstract machine is required to represent   
   >> integer types in a pure binary form (there are three options,   
   >> two's complement (most common) ones complement (I think this is   
   >> very rare these days), sign and magnitude (again rare). There are   
   >> no other options. The bitwise operators will operate on the 'bits'   
   >> even if the underlying system uses some other representation.   
   >   
   > Then, is the result of a bitwise operation implementation-defined,   
   > depending on the representation of negative numbers?   
   > For example, how is the result of (3 & -3) defined -- can it be   
   > any of 0, 1, or 3?   
   >   
   > -3 = 11..1101 (2C)   
   > -3 = 11..1100 (1C)   
   > -3 = 10..0011 (SM)   
      
   Yes.   
      
   It's perhaps clearer if we just change the sign bit. That will   
   obviously change the result in a way that depends on the   
   representation.   
      
      
   Bo Persson   
      
      
   --   
         [ See http://www.gotw.ca/resources/clcm.htm for info about ]   
         [ comp.lang.c++.moderated.    First time posters: Do this! ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)