Re: [Repast-developer] getTickCount/getCurrentTime - why typed asdoubles?

[North, M. <no...@an...>]

Les:

This is a good question.  Users can schedule events with real numbered
values.  This is useful for modeling processes with known or
approximately known physical timing (e.g., modeling the arrival of
packets in a network).

Mike

> -----Original Message-----
> From: rep...@li... [mailto:repast-
> dev...@li...] On Behalf Of Les Gasser
> Sent: Monday, February 12, 2007 1:46 PM
> To: rep...@li...
> Cc: ga...@ui...
> Subject: [Repast-developer] getTickCount/getCurrentTime - why typed
> asdoubles?
>=20
>=20
> I'm implementing some extensions to Repast.
> Regarding uchicago.src.sim.engine:
>=20
> The TickCounter interface specifies the method "getTickCount" which
> returns a double.  The SimModel class specifies the method
> "getCurrentTime" which also returns a double.
>=20
> Question:
>=20
>   Both these methods refer to simulation ticks or times. Why
>   are their return types specified as "double", which is a floating
>   point type, and not int or long? When would a RePast-computed tick
or
>   time count ever be fractional, i.e., anything but an integer?
>=20
>   (If I've got this right, the method of updating sim ticks is through
>   reading "nextTime" fields in BasicAction.java; BasicActions are
>   placed in a priority queue called ActionQueue (nb: Java 1.5 now has
>   java.util.PriorityQueue). Sim time is advanced to the minimum time
>   (next) entry in this queue - so it can leap forward by more than one
>   tick depending on the times for which the BasicActions were
>   (re)inserted in the ActionQueue. Nonetheless, all the time
specifiers
>   I could find in the RePast code are integer values, typically 0 or
1,
>   and it's hard to envision when a float would be needed for a
>   simtick.)
>=20
> Suggestion:
>=20
>   While the inheritance and implementation graphs for these interfaces
>   are clearly different, both methods seem to be intending to return
>   the current simulation tick or time. Why the difference in names
>   then?  It seems like it might be a good idea to unify the concept of
>   getting access to the current simulation time/tick, unless there
>   really is some underlying conceptual difference that I'm not seeing.
>=20
> Thanks,
>=20
> - Les
>=20
> (not a list subscriber, so please respond directly as well as to the
> list - thanks.)
>=20
>=20
>=20
>
------------------------------------------------------------------------
-
> Using Tomcat but need to do more? Need to support web services,
security?
> Get stuff done quickly with pre-integrated technology to make your job
> easier.
> Download IBM WebSphere Application Server v.1.0.1 based on Apache
Geronimo
>
http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D120709&bid=3D263057&dat=3D=
121642
> _______________________________________________
> Repast-developer mailing list
> Rep...@li...
> https://lists.sourceforge.net/lists/listinfo/repast-developer