Math is hard

Last night and today were spent on a bit of a diversion, but an important one.

In working on the Twisted Divide AI, I kept thinking to myself that it could be implemented much more cleanly with a certain piece of AI functionality, which I'll call 'X'. The desire for X is far from just an isolated random wish, however; all of the more complicated AI I've designed (crafting, the Tabernacle, and the Trials of Karmova) tries to "hack together" a form of X with our current tools (which are by no means meager, by the way - Nodeka has some crazy strong AI).

It works, but the problem is that things get really complicated. For example, the live crafting mob AI alone (just the handful of mobs with the portion of the crafting system that is live) is over 200,000 lines long. When things go wrong, sometimes it can be REALLY tough to figure out where the issue is. (For example, the crash issues with the seasonal quests last Christmas, or that pair of morality Trial bugs that still pops up from time to time.)

As my plans for the Divide expand, it has become apparent to me that it would be a huge benefit if I worked on implementing X first. Having X would greatly simplify my work on the Divide and just about every other piece of future AI I've been planning.

Thus, my current focus is on adding that AI functionality. I've already implemented the base, so right now I'm in testing and bugsmashing mode.

The flipside to this is that, if I can get X working, it would enable me to easily finish up a pvp-based Trials extension that I've had on the design table for a while now - much sooner than I could possibly have the Divide ready. Thus, I may work on whipping that out for you in the near future. If I can get X working soon, expect an entry by this weekend chronologically outlining the details of what major area/event based additions to Nodeka will be coming next.