Jump to content

Probability fixes

Recommended Posts

Another long-term project of mine complete: probability fixes. Pretty much every item or spell with an x% chance to do something is wrong1. Monster Summoning spells that are supposed to summon two monsters 40% of the time only do so 39% of the time. Celestial Fury's 5% chance to do extra electrical damage is actually 6%. It's a small, pervasive bug that's almost everywhere. While fixing it, I also found a couple of other bugs:

  • The Heavy Crossbow of Searing was only adding its two fire damage on 49% of hits instead of all of them.
  • The demon wraith will level drain one or two levels (49/51 split), but the combat feedback would always report one level drained.
  • The drow 'whip' weapons had overlapping probabilities for their special effects (e.g. they could cause two bad things on hits instead of just one)
  • The Cacofiend wild surge is accompanied by a lighting effect that only plays half the time
  • The Wand of Apprenti was not actually balanced between the three elements.

Otherwise, the rest were pretty straightforward. About 100 items and spells have been fixed, including several items used as touch attacks by enemies (e.g. the attack of vampiric mists, celestials, shambling mounds, etc.).


1 So, the technical bits. Every effect has a pair of probability fields that set the upper and lower bounds (inclusive) for occurrence on a d100 roll. Using the arcane spells Conjure Air/Earth/Fire Elemental as an example, the spell is supposed to summon a 12-HD elemental 60% of the time (set in the effect as 0-60), a 16-HD elemental 35% of the time (61-95), and a 24-HD elemental 5% of the time (96-100). The problem is that the effects were set everywhere with the assumption that the d100 results are 1-100, but they're actually 0-99. (This is not new information as the IESDP has had it listed forever, it's just never been fixed.) So you only receive the 24-HD elemental on rolls of 96, 97, 98, and 99, or 4% of the time. Since 0 is a valid roll, it means the 12-HD elemental is actually 61% likely, not 60%. Generally everything that was anchored at the bottom of the probability spread--which is almost everything--had an extra 1% chance to occur, e.g. Ravager +6 caused decapitation on 11% of hits (0-10) instead of the intended 10%. Not everything was wrong, but the exceptions were few and far between.


Link to comment

Cam are fixes from core 12 beta incorporated into EE? And if not do you plan to apply the missing fixes to EE's at some point?


The overwhelming majority of Fixpack 11 stuff is either already fixed in EE or slated for the next patch cycle. The handful that aren't will be passed along, yes--it just takes a while since I have to make repo cases and other stuff since the QA is a lot more formal.

Link to comment


This topic is now archived and is closed to further replies.

  • Create New...