From: tkoenig@netcologne.de   
      
   Anton Ertl  schrieb:   
   > So adding two dense matrices tends to be memory bandwidth bound, but   
   > stride-based prefetchers help to avoid getting any extra latency   
   > beyond that coming from the bandwidth limits (if any).   
   >   
   > Likewise, John McCalpin's Stream benchmark uses dense vectors IIRC,   
   > but is memory bandwidth limited.   
      
   CFD codes are usually memory bandwidth limited; they use sparse   
   matrices.   
      
   --   
   This USENET posting was made without artificial intelligence,   
   artificial impertinence, artificial arrogance, artificial stupidity,   
   artificial flavorings or artificial colorants.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)