Molecular Dynamics with mixed Quantum/Classical Forcefield

   Hi all,
   I will be doing simulations using the mixed or combined quantum/classical
 forcefield in CHARMM.  I have several questions, and would appreciate your
 help:
 1) Does anyone know of improvements in combined-forcefield codes as
 regards efficiency?  I am asking because I may designate a non-insigificant
 number of atoms as quantum atoms, and so the CPU time will go up drastically.
 2) Does anyone know how the CPU time/md step changes as a function of the
 number of atoms designated as quantum atoms?
 3) Is there any information on the relative amounts of time spent on
 calculating the interactions between quantum atoms, vs. the time spent
 on interactions between quantum and classical atoms?
   I will summarize the results and post to the list.  Thanks.
   Howard
 --
   Dr. Howard Alper
   Moldyn Inc.
   955 Massachusetts Avenue
   Fifth Floor
   Cambridge, MA 02139-3180
   617-354-3124 x19
 - helping molecules find happiness for over a 10th of a century.