[MUD-Dev] Statistics

Hans-Henrik Staerfeldt hhs at cbs.dtu.dk
Fri Nov 2 14:54:36 CET 2001


On Wed, 31 Oct 2001, Ben Chambers wrote:
 
> What statistics do you consider necessary to define a character?
> I am planning on using a 100% skill based system, with no levels,
> where the more you do something the better you get at it.  The
> problem is I was hopping that the skills would be defined by the
> host.  The set of skills that seem most logical to me are:
 
>   Strength - How capable of physical feats this character is
>   Constitution - How resistant to physical damage he is
>   Intelligence - How quickly he learns new skills Wisdom - How
>   well he retains learnt skills
 
> The problem is I can't figure out where spells and stuff would
> come in.  So I came up with this:
 
> If you reduce the system down to JUST the capability to learn and
> the retention of what is learned, Strength is no longer a
> statistic, but rather a skill.  The problem is too many things
> would be dependent upon this one skill.  The obvious solution (in
> my mind) is to define a basic skill for each item, and than any
> skill that you perform (attack) uses a modifier for how proficient
> you are with this weapon.  This however becomes to complex again.

I would think that reducing strength to a skill is not a bad idea,
but i would object to having your ability to 'learn strength' be
dependant of your intelligence and your ability to 'retain strength'
dependant on your wisdom.

I think what youre fishing for is the splitting of the bodys
physical descriptors into potential, aptitude and current level. For
mostly cognitive skills, i would think the wisdom would be a good
potential, intelligence would be the aptitude for this and each
skill would then have a 'current level'.

However for non-cognitive skills, such as those of the more physical
nature such as things requirering dexterity, strength and
constitution (to use some well-versed words) would require a similar
subdivision.

> What would happen if we defined TWO types of skills?  I could have
> skills that describe the proficiency with different things, and
> than skills which describe what you can do.  The key is that I
> can't do a jump kick with a magical attack, even if I am
> proficient in magic.
 
> Any insight as to how to make a good skills system, 100%
> customizable by the host, that represents ALL actions (other than
> social commands) that the player can do?

What you want is actually to have the principal components of the
skills you wish to implement. As some of these skills are most
aparrently fictional, this may be a problem (read: you'd have to
invent the principle components as well).

Now principle components is just a fancy word for clustering similar
things together that are influenced the same over a set of samples
(read; statistics show that if an athlete is good at 100m run, shes
often also good at the hurdles, but its less correlated with her
ability in the discos).

The problem here is to select the level of detail. The example of
the athlete would reflect a level of detail that, and that may be to
group leg-strength and arm-strength. You have to choose your level
of detail, and your choice may reflect how easy the game is to
play. Conder you do a 'status' and several parameters for each
single musclegroup, cognitive methodology, solcial aptitude towards
different psychic profiles etc etc. All too much work. I know that
someone on the list (i forget who, sorry) has designed an extremely
versatile skill system like the one you seek, and then try to hide
its complex representation for the player (thank god!) and represent
a simplified version.

Hans Henrik Stærfeldt   |    bombman at diku.dk    | work:  hhs at cbs.dtu.dk      |
Address:                |___  +45 40383492    __|__       +45 45252425     __|
DTU, Kemitorvet,        | Scientific programmer at Center for Biological     |
bygn 208, CBS.          |  Sequence Analysis, Technical University of Denmark|

_______________________________________________
MUD-Dev mailing list
MUD-Dev at kanga.nu
https://www.kanga.nu/lists/listinfo/mud-dev



More information about the mud-dev-archive mailing list