The goal of Blind Mapmaker is to procedurally generate dungeons throughout the world, and modify how vanilla Minecraft progression works.
I will update this post periodically as I make progress.
Phase I - Floorplan Generation
The goals of Phase I are quite simple: generate a floorplan and translate it into Minecraft.
Blind Mapmaker uses a set of algorithms based on ideas taken from various individuals whose writings on PGC I've been studying over the years. I was uncertain I would be able to take the raw ideas and turn them into code, hence Phase I's mildly underwhelming goals. However, this phase is crucial, as it lays the foundation for everything that will come after it; it has to work perfectly if anything else is to be done.
The following video gives an introduction to the mod and an overview of the results of my efforts for Phase I. If you want to skip the background stuff, skip ahead to roughly the 5 minute mark for the demonstration.
Phase II - Doors, Zoning, and Lock-and-Key
Phase II is a bit more complex.
Having completed Phase I, the goal is to drill doors between rooms and divide the entire floorplan into zones, accompanied by some implementation of "lock-and-key" and - behind the scenes - a means to track the puzzle tree. This will allow Blind Mapmaker to create dungeon flow and control tension through the experience of traversing one of its dungeons.
Phase II is now complete.
Phase III - Dungeon Entrances, Dimensions, and Progression
Phase III moves away from the dungeon itself and focuses more on the Minecraft-specific aspects of the mod. Dungeon entrances need to be generated throughout the world in a sensible manner, and will be tied to villages: the goal is for each village to have between two and six dungeons associated with it, which will be tied together through key items acquired at each dungeon's end. These key items will be required to unlock another dungeon in the Nether, at the end of which will be a chest containing either an Eye of Ender or highly-enchanted gear.
A new villager type will be introduced, whose sole purpose is to allow players to trade for key items that they may have lost that are associated with that village.
Strongholds will also be modified, removing the Ender portal frame and replacing it with an entrance to a dungeon, at the end of which will be an Ender portal frame whose blocks are always empty. Thus the player will have to complete at least twelve Nether dungeons to be able to reach the End.
An interesting idea. Some of it remind me a little of a mod I once had an idea for that would have been called "crazy quests" or something similar -- but that had smaller ambitions, more of a madlib-like inspired by an old fist edition AD&D book about rolling up idea for adventures and campaigns. I shelved that idea based on wanting to work on other projects and uncertainty about things like getting enough map info about distant chunks without generating them, concerns about things I considered ideal such as putting quest item in other people dungeons. I think Notch consdered quests (he told the Todd Howard he did a video discussion), but of course that never happened. However, none of this really had the expansive plan of changing the course of the game, just giving more to do and a reason to go out and tackle dungeons (for no reason beyond boredom anyway, even leaving the loot after my storage rooms were full). For longer gameplay I usually turned to Minefantasy for slower progression -- though that tends lead to less exploration and focus and building up bases and processing facilities for grinding away on tools.
An example of overworld dungeon entrance generation, which will probably be part of Phase III. The code currently generates these structures randomly, as a test of embedding them into terrain; I need to do a lot more work to get them to generate in a sensible manner.
Phase II is almost entirely complete at this point. I still have some design decisions to make, but zoning works more-or-less perfectly (in smaller dungeons, fewer zones are generated than I would like, but that's just a matter of math), door placement is good (including stair generation for doors that lead up or down a floor), and room intensity calculations have panned out pretty well (which will be used in determining presence of mob spawners and keys for doors).
I'll probably be posting the Phase II overview video sometime soon.
I really like this concept. Will you be adding key consumption upon use on the doors, and a portal at the end of the dungeon to get you back out/pressure plate linked to a command block to teleport you to a random location at the surface above the dungeon, or something of the like? Keep up the amazing mods, GnomeWorks!