Re: scheduling quandry

[Doug Donalson]

Paul,

   The origional purpose of the floating point scheduling was for the
implementation of event driven or asynchronous schedules.  When you have a
large number of events per time step and don't wan't to have ambiguity about
ordering of the events (ie. instead of ramdomizing) then they are useful.
They are not easy to work with however, and unless you need them for
simulation accuracy or there is no alternate solution, I wouldn't bother.

Cheers,

   D4

----- Original Message -----
From: Paul Johnson <address@hidden>
To: Swarm-support <address@hidden>
Sent: Tuesday, July 11, 2000 2:09 PM
Subject: scheduling quandry


> I got a scheduling mess, which you can see if you get the newest version
> of pjrepeater3.m.
>
(http://lark.cc.ukans.edu/~pauljohn/SwarmFaq/WorkingExampleCode/objc/pjrepea
ter3.m
> ) This is a multilevel swarm scheduling problem.
>
> Agents have their own schedules.  Those who do something a time X have
> their actions randomized because their schedules are activated in the
> model swarm.  And the model swarm has setSynchronizationType to
> randomize actions.  But at time X, I do not want actions from the model
> swarm itself "mixed in" with the actions that are "picked up" from the
> agent schedules.
>
> Due to the randomization at time X, I'm seeing the model swarm's action
> at X mixed in with the actions of the agents, as in:
>
>
> I'm model swarm, current time is 0
> I'm bugA, and the current time is 0
> I'm bugB, and the current time is 0
> I'm bugA, and the current time is 1
> I'm model swarm, current time is 1
>
> I keep wondering if floating point times would solve this problem, and
> Doug has described their usage somewhat, but I don't understand the
> meaning of it.
>
> The only fix for this problem I see now is to make sure that no agent is
> acting at a time when the model swarm is trying to record information.
> Actually, by total luck, the bigger project I'm working on is designed
> that way. But I'd still like to work this out more clearly.
>
> --
> 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.
>


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