Mixture

Adam Wiggins nightfall at user1.inficad.com
Sat Mar 29 10:42:08 CET 1997


> [JCL]:
> >> I also like levelless numberless classless goal-oriented systems (and
> >> no, I'm not convinced that skill trees are the one true solution).
> On 27/03/97 at 10:27 PM, Adam Wiggins <nightfall at inficad.com> said: >
> >Well...ya gotta admit, they do work pretty darn well. :) 
> [JCL]
> True.  They just imply a world model I'm not comfortable with.

As in the idea of numerically rated skills themselves, or as in the
tree structure?  I agree the tree thing is a *little* bit simplified;
we considered making a 'skill web' where all the skills could be easily
interconnected in every which way, but decided that both for the ease
of coding and ease of player understanding a tree was better.

> Precisely.  Let the player do the comparisons.  If you feel you have
> to give him something to compare, give him an utterly meaningless
> value which he can use to feel good about.  Either way the value is
> useless in fact -- its just soporific.

Yeah, and this is fine, as long as they have _something_ to cling to.
Let's face it, a bunch of text scrolling by is abstract no matter what.
There's something concrete about a simple 'strength: excellent' that
players like, regardless of how relevant it actually is.  This is similar
to the skill ratings I was mentioning earlier - since they are based on your
class, several of your stats (stats which affect each skill vary by skill,
as well) and so on, it's only a really rough idea of where you stand at any
given moment, and one person's 'poor' might match up, numerically, with
another person's 'good'.  But even knowing all this, I get a strong feeling
of satisfaction when I see my rating in long blades go from 'good' to 'very
good'.  It's not so much knowing where you *are*, but knowing that you are
at least moving foward.  Since muds (and indeed, rpgs in general) are really
focused on character development, you have to have some way to be able to
see and gauge this development aside from 'Gee, I see to be parrying a lot
better nowadays than I was last week.'

> Fine.  Give meaningless values based on proportionate placement on the
> scale of what that player has seen recently.  Its fairly accurate, its
> meaningless, it can't be used to definitively predict outcomes, and it
> is delightfully subject to deception (players masquerading as
> higher/lower level than they are).

Yes, absolutely...it's not like you always know where you stand, things
like ego come into play.  For instance, our equivilent to the 'consider'
command takes into account their size quite a bit when checking their
strength.  So the little guy is always getting underestimated and vice
versa.

> >As much as I'd like to think that just
> >practical use is good enough, I kind of doubt it.  Our skill display
> >currently just tries to give a rough idea of your abilities at a
> >given level of the tree, like so:
> >
> >  > skills
> >  You are very skilled at combat, particularly offence and ranged
> >  attacks. Your stealth is quite good.
> >  Your knowledge of theology, particularly the goddess Celendria, is
> >  very good. You know a few things about thaumaturgy.
> ..
> 
> Right, and the problem is that those "values" are presented computed
> as absolutes.

What makes you say that?

>Admittedly they're better than most in that the max
> value is the max for that character-type, but it is still an absolute
> scale.  Once a player knows the maxes for each type, he knows the
> entire system.

Hmmm...the display doesn't have anything to do with the 'system.'  It's
just a list of skills you are familiar with, and roughly how good you are
with each one.  There's nothing about it being an absolute scale or
there being 'maxes'.  There's not really any such thing as an absolute max,
for starters, nor does the message actually reflect any 'fixed' value.
It's just an idea of where your character thinks she stands in a given area
of knowledge, much as I can give a rough estimate of how good of a programmer
I think I am in RL.  I may be *wrong*, of course, but I can still give
an estimate.  (And anyway, what is "wrong"?  It's all relative...)




More information about the mud-dev-archive mailing list