[MUD-Dev] DGN: Absolutes, percentages and ranges

Brian Hook hook_l at bookofhook.com
Wed May 18 04:25:02 New Zealand Standard Time 2005


On Fri, 13 May 2005 15:05:21 +0000, cruise wrote:

> What happens if instead, there is a percentage chance of
> winning/losing, instead of the absolute?

This is a fairly common argument: deterministic
vs. non-deterministic mechanics?  Chess, RPS, checkers, tic tac toe,
etc. are deterministic by nature, whereas most RPGs are
non-deterministic.  The relative weighting of randomness can play a
huge roll in the flavor of a particular game.

> anything, but at say 90%, how much does that occasional rare
> surprise win add, if anything?

It can add a lot, and it can also be very frustrating.  If you're
about to die but have a 5% chance of doing something to save
yourself, it's exhilirating.  By the same token, if you woefully
outclass your enemy and he gets in a lucky hit that takes you down,
you'll be very pissed.

> In D&D, a +1 sword is basically a 5% bonus, but no one would pass
> one up. Does a 5% resistance to fire really affect gameplay in any
> significant fashion, or is it too small to really matter?

Yes, it makes a difference, so long as you have enough points to get
over a threshold of success vs. loss.  For an example of playing
very small odds differences for big gains, look at card counting in
blackjack.

> Personally I favour bigger effects, for the simple expedient of
> making player choices matter.

Sure, but large differences in parameters can be easily exploited.

> granuality, how much difference is there really between 63/100 and
> 6/10?

For prototyping, not much.  For measuring advancement, it can be a
big difference.  In fact, for an MMOG that is advancement oriented,
using 0-1000 can make a psychological difference since players can
see observable change (STR =3D 877 vs. 876) constantly.

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


More information about the MUD-Dev mailing list