[Icfp04-discuss] Re: haskell simulator performance

[Ben Lippmeier]

Dave,

>>My fabulous Haskell interperative simulator runs at about 400 steps / 
>>sec ! .. Put that down to massive abuse of lists and lack of destructive 
>>update.
>>    
>>
>
>My haskell simulator is about 16 times faster than that.  I figured this
>would be the case, and so wrote imperative haskell code :-)
>

What structure did you store the terrain in? After several false starts 
with one or other of the umteen different kinds of Haskell arrays, I 
ended up copping out and storing the whole lot in C land, using foreign 
ccalls to access it from the IO monad.

Ben.