Re: Allegro CL Floating Point

[tk@allegra.tempo.nj.att.com (Thomas Kirk)]

  >> I don't have a copy of Allegro, but I tried it in Lucid on a Sun 3/60
  >> and it ran in .18 seconds.
  
    But I'm guessing that it didn't actually run the trig functions, so this is
  just the time for the loop.

yes, the Lucid compiler computes away the exp and cos calls at compile time.
but it also recognizes that b isn't referenced, so it throws away the two 
expressions where b is set. kinda surprising that it doesn't go ahead and
throw away the loops as well.