ID:154169
 
Yes it is I, James of Many Projects (Completer of None). Anyway, anybody out there playe WH Quest or HeroQuest? Fun little board games. My concept is this: Players create a character to go down into a randomly generated dungeon in search of glory and loot. Unfortunately randomly generated monsters and traps also await. I picture it as being somewhat turn based IE you can move or attack or drink a potion or get something, etc. for your turn and then have to wait for the next turns (turns are maybe 5 secs apiece). What I am having problems with is how to generate the rooms. Dan's maze lib is good, but I am looking more for a generator that selects from a list of room 'components' and the furnishes it. IE the list may include a hall, a square room, a circle room, a rectangle room, etc. Kinda like the room tiles in WH quest. I am not sure of a good way to do this, especially sense it means the dungeon could sprawl in any direction, thus making maxx and maxy determinations kind of tricky.

Any thoughts?
Sounds fun, but turned based would suck. And we should be able to group up with our friends like in Delve! That game rocks! Also, a turned based version has been done with these words:

LEXY QUEST

Don't make it turn based, it would be slow and I wouldn't play it.
I've been wrestling with this myself, though I've been more attempting to divvy up a preset rectangular region equally using randomized rooms. It's not an entirely dissimilar problem, though, especially in that both problems suck.
Also, maybe sd_mapmaker will help you out!
You might want to check out my darke_MapMaker library if you have a BYONDscape subscription. The makerooms() proc in particular is similar to what you are looking for. MapMaker doesn't work particularly well with irregularly shaped rooms at the moment, but if you set the room_x and room_y up and make the node_x and node_y each 1, you will see round and rectangular rooms within the maze.

You could fiddle with the library a bit to make it draw particular room shapes in large nodes instead of defaulting to rectangular nodes. If you're patient enough to wait on me, I can work on it myself. Hard telling when I'll get to it though.
In response to Sariat
Well, I've never tried LEXYQUEST, but ideally the turns would be short, so that usually you don't even notice them say 3-5 seconds. If you can move 4 "tiles" per turn and 1 tile per second, you would only wait 1 second before your movement was refreshed. Combat might be a bit slow as would picking up an item, but as anyone who has played on DIKU muds knows tics/turns usually aren't a big deal if they are on a suitable interval.

-James
In response to Jmurph
My beef with basic diku timing is that the clock is simultaneously too rigid in some places and not rigid enough in others; if you want something that runs faster than strict turns, I think I'd prefer a smoother, more organic timing system where all actions flowed on an independant time delay system, rather than having some actions evaluated in realtime and others evaluated on rigid universal intervals.
In response to Shadowdarke
Oooh, I had forgotten about this. Yes, the 3x3 preset in the demo produces a very nice result! I think I shall fiddle with this....


-James
In response to Leftley
Oh, good point. That's why I used the "turn" idea. You could do a certain amount of "stuff" (IE move, attack, etc.), but once the world's turn timer clicked over, everybody's "stuff" would be renewed so they could take another action. Make sense?

-James
In response to Jmurph
Having played quite a bit with simultaneous turn evaluation, I think that while it's a great idea and can make for astronomically better play than either rigid rotation turns or full realtime when done properly, you do have to be very careful. From my experiences both with my own games and other games that run on similar principles, it works best by far when you want a real feel of randomness and uncertainty, especially with the timeframe you're talking about--there's not an awful lot (by which I mean any whatsoever) in the way of think time allowed by that timing scheme. That can make for an exciting game, but one of the best aspects of the sort of boardgame you're talking about is often the way different player characters can interact and cooperate. You also might want to think about how faithful you want to be to the original boardgame idea; in my opinion, if you're going to move things that quickly, you might as well just make it realtime with substantial delays attached to actions. Really I seem to remember those types of boardgames as being oriented towards around 3-4 players, so the sort of 5-second timing you're talking about would still go very tolerably quickly even with strict turn rotation.
In response to Leftley
Another fine point, Leftley. I do want to maintain the 'feel' of the original games. Hmm, maybe a strict turn based system would be better. IE Player1, Player2,..., Monsters. Maybe use Dan's turn lib and slap a limit of some sort so if someone is taking too long it can bump to the next player.Ironically, I initially considered my original scheme partly because of how well it worked in Delve! I must say Delve! is certainly one of the best games on BYOND IMHO, and one of the more fun ones I have *ever* played. But I am a really big fan of Talisman, too.

-James
In response to Jmurph
Er... I think you're thinking of Treasure Seeker (now Treasure Seeker Classic, pending the development of Treasure Seeker Not-So-Classic), not Delve! (which was fully realtime).

The real irony here is that I'm cautioning against this sort of system specifically from my experience with Treasure Seeker. It was a novel system, but it really took a lot from the game and was, by all accounts, a rather clumsy expedient to keep a "turn-based" game from becoming too slow. Nyctophobia and Gughunter's (I think) Conflict use the idea much better due to their nature; in essence, they are more complex, multiplayer versions of Rock, Paper, Scissors. There are other situations where I think the idea could work; a game on a smaller scale than Treasure Seeker, where moves encompass a few steps or a single sword swing, could use it to good effect--especially if it were a cooperative game, as HeroQuest and similar games tend to be. The decision there is where you want the cooperative emphasis to be: on effectively coordinating your movements in a hectic, almost frenzied combat systems, or on strategically using your strengths and weaknesses to complement each other (doing both, of course, would be great but massively difficult).
In response to Leftley
Heh, you're right, of course, with the names. TREASURE SEEKER was a blast, though. What problems were you running into with they system you used?

-James
In response to Jmurph
Well, I suppose you COULD make an argument for preferring the style of play it produced, but it certainly wasn't anything where I was aiming for--which is a rather bad job all around, seeing as I was aiming for any of several different results! I had two primary complaints: Navigating the dungeon was too sequential, and the room-by-room level movement combined with the system made for REALLY bad chases. Then assuming you did manage to chase someone down, you were probably in for disappointment, because the way combat rounds had to be conducted meant that anyone you did catch up to would have ample opportunity to escape. (There were other problems with the combat system that compounded this, but it's still altogether not nearly so elegant as the PvP combat in, say, Talisman, and that's not too elegant in the first place).