Re: Time steps

[Marcus G. Daniels]

>>>>> "JG" == Jacek Gomoluch <address@hidden> writes:

JG> Alternatively, I could "discretise" my model in a way that one
JG> milliecond corresponds to one timestep (with large gaps in the
JG> simulation time where nothing is happening). Will the performance
JG> of the tool be degraded then ?

If it is possible to scale timesteps into a 32 bit integer, then I'd suggest
using this approach.  Swarm collapses time, so sparseness won't be a problem
for memory size or performance.

A more elaborate `two tiered' approach can be seen below.  This will
remove the range limitation.

  http://www.santafe.edu/projects/swarm/archive/list-archive.0002/0152.html


                  ==================================
   Swarm-Modelling is for discussion of Simulation and Modelling techniques
   esp. using Swarm.  For list administration needs (esp. [un]subscribing),
   please send a message to <address@hidden> with "help" in the
   body of the message.
                  ==================================