swarm-support
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Test Program


From: Doug Donalson
Subject: Test Program
Date: Sat, 25 Mar 2000 20:37:06 -0800

First, I went to the Cygnus site and got the .dll and the "wait" stuff went
away.

Now to real issues.  I got this bug down to bare minimum.  No agents at all.
The program sets up an event schedule and then schedules an event
"addNewMussel" to itself (model swarm) at a time in the future.  The time is
drawn from an exponential distribution with mean 0.1.  First event time=
0+random draw.  When that event occurs model swarm creates a new time by
first making the same draw from an exponential and then adding it to the
present system time.  This should continue until "stop" is hit.  It starts
with 0 individuals and should increase linearly over time with a slope of
10.  Each time the event is executed it increments a counter which is shown
on the BLT graph.  It also prints a message on the console each time teh
event is executed.

What I see with the original Swarm.dll:  Linear increase with slope=10.
With new swarm.dll:  The event is executed twice and then it isn't visited
again.

My guess?  Somehow, the event created in the second call is inserted into
"the past" on the schedule thereby never getting accessed.  That is a shot
in the dark.

One other thing you shouldd know.  I have an initial event called "dummy".
It is used to force the system time on the unactivated event schedule to 0.
Otherwise my call to "getSystemTime" gets messed up when I set up initial
events on the unactivated schedule.

The only files that are "made" are observer and model.  I included all other
m and h in the directory from paranoia.  Last, make sure you know which
version of swarm.dll you are using, otherwise you could spend a bit of time
wondering why the bug fixed itself!  (As there are no removals from the
schedule, the origional 2.1 works for this test case.)

I won't be around most of tomorrow but if you have questions send me an
email and I'll respond when I get back.

Cheers,

   D3
*********************************************************************
* Doug Donalson                 Office: (805) 893-2962
* Ecology, Evolution,           Home:   (805) 961-4447
* and Marine Biology            email address@hidden
* UC Santa Barbara
* Santa Barbara Ca. 93106
*********************************************************************
*
*   The most exciting phrase to hear in science, the one that
*   heralds new discoveries, is not "EUREKA" (I have found it) but
*   "That's funny ...?"
*
*       Isaac Asimov
*
*********************************************************************


Attachment: forMarcus.ZIP
Description: Zip compressed data


reply via email to

[Prev in Thread] Current Thread [Next in Thread]