Re: Smashes (was: Another combat factor)
Kevin Davis Connery (firstname.lastname@example.org)
Tue, 20 Sep 1994 19:00:54 -0700 (PDT)
> It seems like this throws things really far in favor of an arbitrary
> level of outclassing. Someone else on the list suggested weighting
> attacks to the attack factor (speed=af), which has a similar effect,
> but one which takes over more gradually, without the hard threshold.
Continuous functions are generally far preferable to those with discrete
breakpoints; fewer strange boundary conditions. A log-type function
for example, rather than 'no more than n times', induces less probability
that some weird edge condition will crop up.