Re: Agent Pools

[Marcus G. Daniels]

>>>>> "RL" == Rob Leclerc <Rob> writes:

RL> Given that my predation model both creates and drops a lot of
RL> agents through the simulation, I am curious to know whether or not
RL> it would be cheaper if I created a large pool of agents at the
RL> begining of the simulation, and set a flag in the methods normally
RL> scheduled, such that if "isDead == true", that this would simply
RL> leave that method call.  This way I don't have to do dynamic
RL> scheduling OR object creation, but would the extra overhead be
RL> worth it? 

It's possible.  If it were true, it's also possible you could get the
performance back by reworking things.  Pretty much anything is
possible given the level of detail you've provided.  My advice is
don't just change structural aspects of your simulation (on vague
hypotheses) until you have acquired and rationalized profiling
information.


                  ==================================
   Swarm-Support is for discussion of the technical details of the day
   to day usage of Swarm.  For list administration needs (esp.
   [un]subscribing), please send a message to <address@hidden>
   with "help" in the body of the message.