XPost: comp.ai.genetic   
   From: jritacco@adelphia.net   
      
   I have recently completed development of a new GP language/engine which I   
   firmly believe to be the fastest in the world.  I have benchmarked against   
   Discipulus (TM) - the reigning speed champion -- in as much of an "apples   
   for apples" comparison as possible, and the new engine (currently code-   
   named "MGIP") is showing a performance advantage of between 2.7 times and   
   4.3 times the processing speed.  (Comparisons were done both on a gen-by-   
   gen basis and a resulting-fitness basis.)   
      
   In addition, MGIP achieves approximately 3 times the genetic-code density   
   of Discipulus, which allows for about triple the population size for a   
   given amount of RAM.   
      
   I have tested this engine with a number of linear regressions, the   
   Fibonacci sequence, factorial, Boolean-6 multiplexer, Boolean-11 mux,   
   Gaussian data, and sequence induction, with what I feel are astoundingly   
   good results.   
      
   Because I am trying to patent some elements of this creation, I am not in a   
   position to disclose the novel aspects of the engine.  If possible, though,   
   I would like to get together with someone who is actively using GP for   
   commercial/industrial applications for an alpha-test of my engine under the   
   auspices of a non-disclosure agreement.   
      
   If you are an interested party, please contact me at:   
      
   jritacco@adelphia.net   
      
   Thanks!   
      
   -- John R.   
      
   [ comp.ai is moderated ... your article may take a while to appear. ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)