From: marcov@stack.nl   
      
   On 2005-07-30, Skybuck Flying  wrote:   
   >   
   > In reality the data bit and meta bit are grouped together as a single entity   
   > which can be read into the cpu since otherwise the cpu would not know where   
   > to start reading the data or meta bits. Now it simplies start with the first   
   > data + meta bit pair.   
   >   
   > Because a cpu might need to know the length of the bit field up front the   
   > cpu/algorithm works simply as follows:   
   >   
   > The cpu starts reading data and meta bits until it reaches a meta bit of 1.   
      
   (ignoring the obvious "what is the (B)Pascal angle?" for a moment)   
      
   "scanning" is to my best knowledge not easily paralellisable. Hence your   
   proposal would already lead to severe performance degradation.   
      
   Some other reasons:   
   - memory use would double.   
   - memory to cpu bandwidth would halve.   
   - what if I have more metabits than register width? IOW what if I try to load   
    a 64-bit value on a 16-bit cpu?   
   - If I per accident give only zeroes as meta, what would happen?   
      
   P.s. I'm no electrical engineer. If I can shoot so many obvious holes in your   
   reasoning,   
   you don't even want to know what an EE could do :_)   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)