Re: [CMRSRC Update] DG Events Documentation

From: Angus Mezick (angus@EDGIL.CCMAIL.COMPUSERVE.COM)
Date: 07/13/98


Oh good, was getting a little worried for a second there.  Guess I was reading
old mail about V1 dg-scripts.  Or grabbed an old patch.  Ours does something
like that.  We just keep all the events in a time sorted linked list and use the
time(0) +delay to figure out the time to call the event.  Then, each pulse, it
checks the time(0) vs the stored time and if the time(0) is greater the event
fires.  This goes on until an event is found that is scheduled for later than
time(0).
--Angus

______________________________ Reply Separator _________________________________
Subject: Re:  [CMRSRC Update] DG Events Documentation
Author:  INTERNET:CIRCLE@POST.QUEENSU.CA at CSERVE
Date:    7/13/98 3:38 PM

> Is this still the event system that decrements the counter on EVERY event
EVERY
> cycle?  If so, you might want to fix that first.  A nasty bit of overhead if
the
> event system is to be for more then DG-script delays.  If not, sorry, i am
> confused.  I am not sure if dg-events do the excessive decrementation or not
> because we had a system that was working before dg was installed and just
ported
> dg into it.

no, the dg events uses a global array of event buckets, and events are
hashed based on the global pulse that they are supposed to occur..this
cuts down the number of checks to 1/(number of buckets)..i
think..something like that anyway :)

siv


     +------------------------------------------------------------+
     | Ensure that you have read the CircleMUD Mailing List FAQ:  |
     | http://democracy.queensu.ca/~fletcher/Circle/list-faq.html |
     +------------------------------------------------------------+



This archive was generated by hypermail 2b30 : 12/15/00 PST