Some real quick thoughts:

• All Entities (player characters and monsters):
  stats - strength, agility, dexterity, intelligence
  classes - fighter, mage, thief, cleric
  level based or skill based?
• Monsters - standard undead types, giant rats (I know, so cliche), goblins,
• Items - what weapons, magical loot, armor, etc should be included? Track player progress by value of loot obtained?
• Levels - traps, chests(?), how randomly generated
• Online "high score" site? Allowing trading online (possibly 2.0 version )?
• Other stuff?

_________________
Bored? Head on over to my blog and see what I'm up to.

Microsoft XNA MVP

- For stats I'd love to see luck added.
- Which would DnD be classified as level or skill based?
- Well since it's going to be all dungeon crawling all the loot would have to be usuable. Unless you meet a character in the dungeons that would be willing to trade... Also, having the weapons randomly created would be neat too. Kind of Diablo-esque.
- Levels can/should be completely generated based on the user level. Never playing the same level twice would be awesome.
- The trading idea would be pretty solid, I have no idea how we'd go about it though.

_________________
In brightest day, in blackest night. No evil shall escape my sight. Let those who worship evil's might, beware my power... Green Lantern's light!
Twitter!

Agreed. Could be used to increase what kind of loot is found and in combat.


Depends on which edition of DnD you're talking about. I believe 3.x is a hybrid. Consider NWN vs. Dungeon Siege.


Unless a trading system is implemented the loot generator would have to take into account the character so the majority of the loot is useful. Not really a problem though.


As long as the level generator creates interesting levels yeah. Diablo did a good job of this, but they also had a huge amount of resources to choose from.


I was thinking either a web service for in-game stuff or an auction type site.

_________________
Bored? Head on over to my blog and see what I'm up to.

Microsoft XNA MVP

I was always a fan of the way that KoTOR did it's leveling. (I think it's based on 3.5 rules.) Where your character does gain levels, at every level you gain points to dump into abilities (lock picking, trap disarms, etc) and every 4 levels you get to pick a new feat or skill.

Using a website for trading would be awesome, and as an added bonus, it doesn't have to be done right away. (Coded with a plan to implement, but like you said, it can be done in 2.0, if we even get there.) But... If we're going to create completely random loot you're right, we either need the trading system handled right off the bat, or we need NPC's that will buy / trade items.

Considering we're looking at an absolutely flat 2D dungeon crawler creating resources shouldn't be that difficult. (I don't think.)

_________________
In brightest day, in blackest night. No evil shall escape my sight. Let those who worship evil's might, beware my power... Green Lantern's light!
Twitter!

While I don't care much how it's done, I've never really like that ability to increase skills that were never used. You could become an expert lock picker without ever having picked a lock simply by dumping points into the skill. Games I GMd back in the day I never let players increase a skill unless they'd attempted to use it either by just trying or seeking out a trainer to learn the basics. That may just be me though.

The code I have for the RPG for my book does use a system similar to 3.5 or Rolemaster though, just because it's the most popular.

I wonder if it would be worth looking into the System Reference Document open ruleset from WotC.

_________________
Bored? Head on over to my blog and see what I'm up to.

Microsoft XNA MVP

I'd be game for that. Kind of the same way Oblivion did it then.

_________________
In brightest day, in blackest night. No evil shall escape my sight. Let those who worship evil's might, beware my power... Green Lantern's light!
Twitter!

Up for what? I had 3 separate things in the last post?

_________________
Bored? Head on over to my blog and see what I'm up to.

Microsoft XNA MVP

Err. The skill leveling.

_________________
In brightest day, in blackest night. No evil shall escape my sight. Let those who worship evil's might, beware my power... Green Lantern's light!
Twitter!

Morrowind was much better with lockpicking than Oblivion. Oblivion was just a minigame and real-world skill overrode the game skill. In Morrowind, increasing your skill level meant you could get into harder locks in less time and with lower resource cost. (wearing/breaking picks)

My vote...

I think randomly generated maps would work well. The biggest issue I have with dungeon crawlers is once you mapped the dungeon and know where all the super secret hiding places are, there is just no reason to continue to play.

If a store system is going to be implemented then item wear should be implemented to create a money sink for the players.

Character advancement...
I personally prefer the skill route. Implement a set of trainers that the player can go to to raise there skill in a given area.
IE: A sword trainer will increase a players sword ability by 1 point for a small fee of course.
The only issue here is when dynamically creating dungeons, knowing what mobs to include in that dungeon. It wouldnt be much fun if every time you walk into a dungeon you are confronted with a great wyrm and your sword ability is only 2.
If a small world map was created with the town in the center (to house the shops and trainers) then you could create different dungeon entrances for different skill levels. IE: Easy, Moderate, Hard, Suicide.
The only difference in the dungeons would be the types of mobs and treasure created.

This might be a bit over reaching...
As far as the trading site goes, implementing something along the lines of the auction house in WOW would work well.

Going the hybrid (level/skill) route would allow matching up of the character and mobs.

Thinking about the skill training and everything I quickly got into some scope creep and had to throttle back.

I'm starting a wiki page for this here.

_________________
Bored? Head on over to my blog and see what I'm up to.

Microsoft XNA MVP

Since I don't know the C-sharp language, and don't know anything about these mud dungeons, I'm feeling like I'm going to be a bit left out on this one... Which is okay.



There are some things I can do to help with the project if you want me to. Feel free to decline if you'd prefer c-sharp programmers. I don't mind.

But for example, I could develop an algorithm for randomly generating the dungeon maps. But you'll have to convert it to c-sharp yourselves. If you give me specifications like sizes and complexities, how linear, how maze-like, how much backtracking is necessary, how many dead ends, whether it is rooms and passageways, or just rooms, or just passageways, how big rooms are, whether rooms are square, rectangles or have an irregular boundary, how straight or twisty the passageways are, etc.

From what I know of you I think you'd be able to pick up C# in no time at all. That's up to you though.

As for levels, if you check the wiki page you'll see the link I posted to crappy programmer art for rooms and hallways. I'm thinking rooms can be any shape, but perhaps limited to 10 squares or so. Hallways maybe a max of 10 squares and can turn any number of times. Minimal dead ends.

_________________
Bored? Head on over to my blog and see what I'm up to.

Microsoft XNA MVP

Ok, that's some of the variables. I'll see what I can do with those. I'll guess the rest

Do you have an example map, just to give me a feel for the layout/density?

Also, I see from your graphics that you want tile-edge-type walls rather than full-tile-type? Full-tile-type is easier

I'll see what I can throw together.


I'm trying to visualize what you're talking about but I'm having trouble. What would full tile look like as far as passages linking to rooms?

_________________
Bored? Head on over to my blog and see what I'm up to.

Microsoft XNA MVP

Full-tile-type walls:




Tile-edge-type walls:

I would think the tile-edge walls would be easier to work with as you're dealing with just 1 tile to determine collision, not 2. Maybe it's just me though.

_________________
Bored? Head on over to my blog and see what I'm up to.

Microsoft XNA MVP

It depends how its stored. It's easy to test +/- (x,y) to see if a tile is wall or not wall.

With edge walls, it's harder for automated map making, because every tile is potentially walkable, so we have the problem of threading the walls around tiles. It's harder because the geometry of walls isn't simple: there are two types of walls: those which go horizontally and those which go vertically.

You don't get that with full-tile-type walls.

The benefit to edge-type is that you can have two rooms next to each other and a door between them is direct without the slightest bit of passage. Also you can have U turns with 1 side step rather than 2.

The benefit of full-type is easy construction. Walls are always thick solid blocks, not wafer thin drywalls.

In history, I have a feeling that full-tile-type games tend to have been more popular... as well as being easier to make.

For a dungeon, full-tile-type would make it feel more solid.

But it's your game design. I'll go with what you say. But I'd feel more confident with full-tile-type walls

If it's easier using full-tile-type walls that's fine. It doesn't matter much to me. If I'm thinking correctly that changes the number of room tiles to just the 1.

edit: as far as the number of rooms in a level, I'm wondering if we can just do some kind of formula based on the character level. Any ideas anyone?

_________________
Bored? Head on over to my blog and see what I'm up to.

Microsoft XNA MVP

How many levels does a character have? Lets say 100 character levels.

How many rooms is right for level 1? -- 2 rooms?
How many rooms is right for level 10? -- 5 rooms?
How many rooms is right for level 100? -- 20 rooms?

eg, rooms = 2 * sqrt[level]