Optimization (was Re: Scheduling

[Paul E Johnson]

> Rob Leclerc wrote:
> Part of my implementation is that I have
> each agent schedule their own activities as opposed to having the
> model itself activating those 
In the ExampleCode archive (in SwarmFaq directory of my site), I have a
series of programs "pjrepeaterX" which show various scheduling concepts
and I too get a big slowdown when each agent is a swarm and maintains
its own schedule and they are merged theough the hierarchy.

It runs much faster if you keep a master schedule in the model level,
and allow agents to hand actions on that schedule.

On a related note, one of my students has written a really slow model,
because each agent surveys its environment in a 7 square moore
neighborhood before each step.  We wrote this in the ordinary loop
fashion, like in heatbugs, where the agent sums the neighborhood. but
now I'm wondering there must be a faster way.  Anybody have a working
example of a code base that recalculates sections of a grid by "slice"
somehow, not by individual square?
-- 
Paul E. Johnson                       email: address@hidden
Dept. of Political Science            http://lark.cc.ukans.edu/~pauljohn
University of Kansas                  Office: (785) 864-9086
Lawrence, Kansas 66045                FAX: (785) 864-5700

                  ==================================
   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.