On 22 January 2011 06:13, James Paige <[email protected]> wrote: > On Sat, Jan 22, 2011 at 01:04:31AM +1300, Ralph Versteegen wrote: >> I wanted to add some more detailed options to the equipment+hero >> elemental resist merging formula. But like I said earlier, we can't >> hope to cover every reasonable possibility -- eventually we'll add an >> option to use a script in lieu of that -- so I don't want to >> overcomplicate things. So I want to see what others think first: >> >> -Add a new formula "Average of non-100% values". Takes the average of >> everything which is not exactly 100%, use 99.9% if you want to average >> in 100%. >> >> -New formula "Most significant value": use the damage from one piece >> of equipment, whichever is the "greatest" in some sense. But I'm not >> sure how to pick it -- whichever is further away from 100%? Isn't 20% >> (1/5) damage a more significant difference change than 200%? And what >> about absorb? > > What about the Final Fantasy 6 method? > > * First check failure conditions > * Then check absorbs (if there are any, stop) > * Then check resists (if there are any, stop) > * Then check weaknesses > > Roughly paraphrased from > http://www.gamefaqs.com/snes/554041-final-fantasy-iii/faqs/13573
Ah yes, it make a lot of sense for absorption (and immunity) to override anything else. But another sensible way to decide between resistances and weaknesses would be to choose the resist if the reciprocal of the resist (< 1.0) damage is greater than the weakness (> 1.0) damage. Or, there's no reason we can't add two formulas and have it both ways. >> -Originally I wanted to add two parameters to all or most formulas to >> add caps or decreasing returns on stacked damage. But it would be >> cleaner to add a selectable second, independent formula for bounding >> the result of the first formula, with upper bound and a lower bound >> options. The 4 choices I thought up are: >> *None >> *Bound, ie. max(min(result, upper), lower) >> *Approach in Limit, which uses a sigmoidal function (I would piece >> together one with a linear section in the middle 50% or 67% or >> something like that, so that you don't get annoying results like 100% >> damage actually becoming 97% >> *Decreasing Returns, which above (below) the upper (lower) bound has a >> steadily flattening curve such the squareroot function, but still goes >> to infinity. > > That is a pretty cool idea! I like it. > >> -Also, what about duplicating the equipment merging stuff for merging >> together the percent damage due to all the different elements, which >> are currently just multiplied together? Again, I suggest it due to >> concerns about excessive stacking. > > I like that idea too. It would keep elemental mixing from being so > strong. > > --- > James > _______________________________________________ > Ohrrpgce mailing list > [email protected] > http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org > _______________________________________________ Ohrrpgce mailing list [email protected] http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
