[MUD-Dev] Re: darkness/visibility

J C Lawrence claw at under.engr.sgi.com
Tue Jun 30 16:40:42 New Zealand Standard Time 1998


On Tue, 23 Jun 1998 17:03:28 -0500 (CDT) 
Travis S Casey<efindel at io.com> wrote:

> On Mon, 22 Jun 1998, J C Lawrence wrote:
>> On Thu, 11 Jun 1998 08:21:01 -0600 Chris
>> Gray<cg at ami-cg.GraySage.Edmonton.AB.CA> wrote:

>>> Is there a real problem with the default situation? If the player
>>> types the command, then either the player knows the object is
>>> there, or is intending to try to find one. If there is one, and
>>> nothing prevents the action, then let it go ahead. Why get more
>>> complex, unless you really have something specific in mind to
>>> gain?

>> It actively encourages "guess the noun" games (as a variation on
>> "guess the verb".)  Players are now going to be wandering about the
>> place trying out all sorts of unlikely noun/verb combinations just
>> in case they find something.

> That's one problem.  However, there is another, more subtle reason
> to "get more complex" -- it helps to separate player knowledge from
> character knowledge.

Umm, true.  I usually think of them as synonymous (a side effect of my 
GoPish play style).  I'm rather uncomfortable with the very concept of 
the game attempting to retain an IC vs OOC knowledge disctinction
(mostly because it would bug the crap out of me as a player).

> The *player* may know that there's an object in that dark room, from
> previous play with another character.  However, the *character* may
> not know that, never having been in that room before.  In that case,
> you may not wish to allow the player to simply go ahead and have the
> character do something with that object.

This of course depends on whether your game/world distinguishes
between IC and OOC knowledge (no surprises there).  As happens some of
the Infocom titles did, and some didn't (IIRC HHGTTG varied wildly
within the game in this regard).  

<shrug>  

Play-style question -- we went over this this way back int he question
of the degree of seperation of the game character from the human
player.  Does the character have any existance outside of its human
player, or is it a pure thoughtless automaton as instructed by the
player's commands?  I favour the latter.

> Of course, now we can get even more complicated in terms of
> situations -- what if the character should know, based on
> information that's been given to him/her in-game, but not from
> personal experience?  For example, two adventurers talking in a
> tavern:

Bingo.  

> One problem that starts to come in is one of information storage --
> should the character remember the details of what's where in *every*
> room he/she has ever been in?  How do you store that much info if
> there are hundreds or thousands of rooms and many players?  A
> possible solution might be to limit the number of rooms the
> character "knows", possibly based on a trait of the character.
> Another method might be to change how the info is stored -- e.g.,
> for each room, record a "knowledge value" (the chance that the
> character remembers where things are in that room) and a "knowledge
> date" (the last time the character visited that room).  If you now
> timestamp objects when they are moved, you can easily check to see
> if something has been moved since the character was last there
> (although there's still a problem if it was moved and then moved
> back to the prior place).

Past a questionable arguable point simulation accuracy doesn't matter
any more.  

Heck, I occassionally lose the light switches on the wall at home --
and I see and use them every day.  Mostly this is due to my forgetting
that in the US flicking a light switch UP turns the light on, and in
England it turns the light off.  Ergo I've had to relearn (and still
occassionally forget) the habit of walking into a dark room and
sweeping my arm _down_ to turn the lights on to now (in the US) sweep
my arm _up_.  

Perhaps your above problem could equate to modelling my imperfect
light switch operating habits?

--
J C Lawrence                               Internet: claw at null.net
(Contractor)                               Internet: coder at ibm.net
---------(*)                     Internet: claw at under.engr.sgi.com
...Honourary Member of Clan McFud -- Teamer's Avenging Monolith...




More information about the MUD-Dev mailing list