On Nov 29, 2007 9:11 PM, Jon Harrop
Mathematica uses a single arbitrary-precision integer to represent each generation of a 1D automaton. The rules to derive the next generation are compiled into arithmetic operations on the integer. The offloads all such work onto your big number library and, with GMP, will be as fast in Haskell as most other languages.
Does GHC already use the GMP library for Integer? It looks that way but I'm not positive. That'd be ironic, if the higher-level Integer representation is faster than a low-level bitwise one ... Still, I suspect accessing individual "bits" might kill me if I'm not able to move most of the calculation into a call to the library. Do you mind elaborating on how rules are compiled into 'arithmetic' operations or providing a link? Justin