Thanks for that Sven. I came across a folder on the Swarm ftp site a few
months back called multi2d.tar.gz and promptly forgot about it. This
contains MultiDiscrete2d.hm and MultiObject2dDisplay.hm. I'm not sure who
wrote these but the comments indicate that it they were written to allow
multiple agents in one cell with Discrete2d so I'm testing them out with
my
current model. I'll also have a look at your code - the more the merrier
(right?). I found multi2d.tar.gz in:
ftp://www.swarm.org/pub/swarm/apps/objc/contrib/
for anyone else interested.
C
----- Original Message -----
From: "Sven N. Thommesen" <address@hidden>
To: "Swarm Support" <address@hidden>
Sent: Sunday, August 28, 2005 3:32 PM
Subject: Re: [Swarm-Support] Multiple Agents in One Cell
>
> Crile,
>
> I remember having similar problems myself a while back (in 1997!) and
> I created some extensions to Grid2d for the purpose. These may be of
> use to you. They are found in the 'contrib' directory for Swarm
> contributions, but the file is so small I just appended it here. Hope
> you find it useful. (You should be able to extend the procedures that
> calculate neighborhoods to figure your slopes ...) Of course, the
> code may need to be tweaked a little to work with the latest release of
> Swarm.
>
> Cheers,
> Sven Thommesen
>
> At 11:23 AM 8/24/2005, you wrote:
>>Hi there SwarmList,
>> I'm trying to work out multiple agents in a cell using lists, as
>> suggested by some previous posts (if that doesn't work, I'll try
>> using MultiGrid2D). The primary agents are volumes of water
>> flowing over a riverbed (the cell surface, a Discrete2d which
>> contains the cells), so I need each cell to hold an elevation value
>> as well as a list of agents. I have borrowed some methods from
>> MultiGrid2d (my thanks to Paul Johnson and Opinion2.0). The
>> elevation values come from an ascii file originating as a GIS
>> raster grid and are read into each cell after the agent list has
>> been created but before any agents are added (the collection is
>> created when the cell is created, well, during -createEnd a la
>> Opinion, to be exact). During runtime, a method gets called from
>> the agent to the cell surface that determines which cell is the
>> best to move to by calculating local slopes from that particular
>> cell. The problem is that where before I could easily read in the
>> elevations, calculate the slopes and move the agents around (until
>> two ended up in one cell, of course), I now get NaN returned when
>> the cell surface queries a cells for its elevation values (and a
>> subsequent crash). A -getCellAtX: Y: method is used to access the
>> cell's elevation value - this worked fine prior to implementing the
>> list. It seems like creating the list has either made the
>> elevation value inaccessible, or perhaps the list has overwritten them.
>> So, have I been sloppy in creating my lists? My gut feeling is
>> that my cells should be able to hold both elevations and lists and
>> I simply haven't implemented this well. Could the order in which
>> the list is added and elevation values read in be causing a
>> problem? Any thoughts?
>>Thanks -
>>Crile
>>
>>Dr Crile Doscher
>>Natural Resources Engineering
>>Lincoln University
>>Canterbury
>>New Zealand
>>
>>"What if the hokey pokey is what it's all about?"