Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[The Moonlit Knight]

The politics and intrigue start to make it not a (roguelike) game anymore.

Nope. Also, double nope.

[Granite26]

Just my opinion...

[Granite26]

Time:

I'm running into this one.  I want to have infinitely additive stats, including speed.  Modifiers aside, the faster your speed, the faster your turn should come up.  Since the stats are infinite, I can't use the traditional 'Add your speed to your initiative counter, when it reaches X you get a turn' method.

I'll also want to have round effects, such as poison that lasts a set amount of time before wearing off.  That time should be absolute, not relative.  I don't want to base it on the afflicter or the afflictees move rate.  (At least I don't think... The possibility that poison only hits once every 10 character moves late in the game is kind of obnoxious, as is a stat boost that lasts too long).

Anyway, I anticipate the SPEED characteristic ranging from 0 to about 250-400, and again with a logarithmic scale.  worst case scenario your opponent should go 10 times before you do.  (That's pretty ridiculous even then, since combos and extra attacks could push this up into 50-60 attacks)

I'm not sure how I feel about deterministic either (roll init at the start and keep, or roll every turn).

Potentially I could find the highest speed on the map, add 100 pts and go from there. (So something with a 100 Speed as the fastest would go every other 'turn', while his 50 pt opponent would go every 4.)  That gets ugly fast, unless I scale it (take the sqrt of the stats?)

Any other thoughts?

[Servant Corps]

The politics and intrigue start to make it not a (roguelike) game anymore.

Nope. Also, double nope.

A roguelike just need randomness. If you can randomly generate new "enemies" to talk to, well...

Come to think of it, should Sibboth be updated for the modern age?

[The Moonlit Knight]

Just my opinion...

Hah, sorry, didn't mean to be snippy or anything, I've always just kind of bristled at the idea that roguelikes are always repetitive hack-and-slash affairs and nothing else. One of my first really good experiences with RLs was actually Dwarf Fortress, so I'm not used to the definition being so narrow. And hell, if you broaden it just a little, the "political intrigue" idea could be of the same mindset as hack-and-slash; rat people out to gain higher security clearance, become more trustworthy to the Friend Computer, gain further access to sensitive systems and information (in order to hunt down the Communists tampering with all of it, of course), that sort of thing.

Any other thoughts?

I remember someone on r.g.r.d. had an idea for "infinite" skills/stats, though they really had a peak of 100; the trick was that as stats approached 100, the increments they increased by got progressively smaller, going from multiple whole numbers to increments of 1 to mere fractions, always approaching 100 but never reaching it. Sort of a halfway-of-halfway thing; halfway from 0 to 100 is 50; halfway again is 75 (not 100, as half is no longer 50); halfway again is 87.5, 93.75, 96.875, so on and so on. Probably not what you have in mind but it can't hurt to pitch the idea I guess.

As for roll init at start and keep vs. constant re-rolling, I think they're pretty different from a gameplay perspective, obviously. Constant re-rolling leaves the potential for sudden changes in how the fight is going; if a monster unexpectedly gets 3 turns to your 1 (or vice versa) it can tip the tables. I also think it gives more of an advantage to characters/monsters with high speed; in a roll init at start system, a bad roll could set back a quick monster for the duration of the fight no matter how speedy it supposedly is. With constant re-rolling, the monster would be able to use its speed to its full extent and advantage. If I understood your meaning correctly, that is.

[Aqizzar]

Okay, so, I'm getting into this whole "teach yourself to code and make a roguelike" business.  I've batted it around in my head, and come to the conclusion that yes, I can do it.  I understand the functional logic of programming, I just need to retrain my obsolete JAVA fluency into a working grasp of C++.  That tutorial posted in the OP, while incomplete, did at least teach me how to make everything actually display on screen, which was my biggest stumbling block.  Everything else is just read-ins and math.

I've got a couple big ideas on how to make my game, and put together a function by function plan of action.  Get one basic feature (like walking, or walls) working at a time, and move down the necessary list.  If anybody has any thoughts/suggestions/comments/advice/words of discouragement, speak up, I'm new.

First, I want to make a roguelike that involves advanced movement - jumping and climbing specifically, along with terrain interaction.  The set-up behind my idea is that the Player is a mad scientist, using genetic engineering and cybernetics to make monstrous servants which you send out to do your bidding, using various abilities you build into them for the job.  Sort of a combination of Liberal Crime Squad and Alphaman.

Second, and more importantly, came to me while thinking about that word "climbing".  I want to make a 3D roguelike, and the plan I've got seems simple in concept.  It revolves around a few basic pieces - the map itself would just be a 3D array of integers.  The level would be drawn by reading those integers against a big catalog of tile-type information, and any interaction with a tile is done by calling the location's tag, or changing that tag when needed.  Essentially, the only object would be the map itself.  Line of sight and distances would calculated by basic Pythagorean math.  Actor AI would be a little complicated, but really just big piles of simple functions.  The levels will be finite, probably 50x23x9 tiles or so.

Looking through the big roguelike history sites, I don't think anyone has made a simple yet 3D gameworld outside of DF Adventure mode.  I'm probably wrong.

[Granite26]

Just my opinion...

Hah, sorry, didn't mean to be snippy or anything, I've always just kind of bristled at the idea that roguelikes are always repetitive hack-and-slash affairs and nothing else. One of my first really good experiences with RLs was actually Dwarf Fortress.

I get the exact same way about music genres.  I'd certainly never say 'it's bad because it doesn't fit it's genre conventions.  Hell, I'm not even planning on making a roguelike.  Those two throwaways WERE (the entirety of) my thoughts on the genre

Any other thoughts?

I remember someone on r.g.r.d. had an idea for "infinite" skills/stats, though they really had a peak of 100; the trick was that as stats approached 100, the increments they increased by got progressively smaller, going from multiple whole numbers to increments of 1 to mere fractions, always approaching 100 but never reaching it. Sort of a halfway-of-halfway thing; halfway from 0 to 100 is 50; halfway again is 75 (not 100, as half is no longer 50); halfway again is 87.5, 93.75, 96.875, so on and so on. Probably not what you have in mind but it can't hurt to pitch the idea I guess.

That's what's going on any place stats get compared to each other.  The problem is the need for a constant timescale .

As for roll init at start and keep vs. constant re-rolling, I think they're pretty different from a gameplay perspective, obviously. Constant re-rolling leaves the potential for sudden changes in how the fight is going; if a monster unexpectedly gets 3 turns to your 1 (or vice versa) it can tip the tables. I also think it gives more of an advantage to characters/monsters with high speed; in a roll init at start system, a bad roll could set back a quick monster for the duration of the fight no matter how speedy it supposedly is. With constant re-rolling, the monster would be able to use its speed to its full extent and advantage. If I understood your meaning correctly, that is.

You read what I was saying correctly, although if I made it deterministic, I probably wouldn't use a roll at all.  (Always use plain init bonus only)

[Willfor]

Little is likely to come of this. Especially since it is my first try doing it in something other than that fancy tutorial (which kind of got me to where I am), and tries at text based things.

Spoiler (click to show/hide)

I am aware I am doing things wrong. It's still sort of in progress. (and I will progress to do things wrong in a more efficient fashion in the future. >_>;; )

[Fualkner]

Well, I tried to write a program that would output thousands of repetitive names, all numbered and such... Simple program is below:

Spoiler: It's in Python! (click to show/hide)

thing = open("item_list.txt", "a")
amount = int(raw_input("Start where?"))
amount2 = int(raw_input("To where?"))
for a in range(amount, amount2):
    s = str(a)
    thing.write("i" + s + ",")

And here's my output:

Spoiler (click to show/hide)

ど椬ⰱ㉩椬ⰳ㑩椬ⰵ㙩椬ⰷ㡩椬ⰹㅩⰰㅩⰱㅩⰲㅩⰳㅩⰴㅩⰵㅩⰶㅩⰷㅩⰸㅩⰹ㉩ⰰ㉩ⰱ㉩ⰲ㉩ⰳ㉩ⰴ㉩ⰵ㉩ⰶ㉩ⰷ㉩ⰸ㉩ⰹ㍩ⰰ㍩ⰱ㍩ⰲ㍩ⰳ㍩ⰴ㍩ⰵ㍩ⰶ㍩ⰷ㍩ⰸ㍩ⰹ㑩ⰰ㑩ⰱ㑩ⰲ㑩ⰳ㑩ⰴ㑩ⰵ㑩ⰶ㑩ⰷ㑩ⰸ㑩ⰹ㕩ⰰ㕩ⰱ㕩ⰲ㕩ⰳ㕩ⰴ㕩ⰵ㕩ⰶ㕩ⰷ㕩ⰸ㕩ⰹ㙩ⰰ㙩ⰱ㙩ⰲ㙩ⰳ㙩ⰴ㙩ⰵ㙩ⰶ㙩ⰷ㙩ⰸ㙩ⰹ㝩ⰰ㝩ⰱ㝩ⰲ㝩ⰳ㝩ⰴ㝩ⰵ㝩ⰶ㝩ⰷ㝩ⰸ㝩ⰹ㡩ⰰ㡩ⰱ㡩ⰲ㡩ⰳ㡩ⰴ㡩ⰵ㡩ⰶ㡩ⰷ㡩ⰸ㡩ⰹ㥩ⰰ㥩ⰱ㥩ⰲ㥩ⰳ㥩ⰴ㥩ⰵ㥩ⰶ㥩ⰷ㥩ⰸ㥩ⰹㅩ〰椬〱ⰱㅩ㈰椬〱ⰳㅩ㐰椬〱ⰵㅩ㘰椬〱ⰷㅩ㠰椬〱ⰹㅩ〱椬ㄱⰱㅩ㈱椬ㄱⰳㅩ㐱椬ㄱⰵㅩ㘱椬ㄱⰷㅩ㠱椬ㄱⰹㅩ〲椬㈱ⰱㅩ㈲椬㈱ⰳㅩ㐲椬㈱ⰵㅩ㘲椬㈱ⰷㅩ㠲椬㈱ⰹㅩ〳椬㌱ⰱㅩ㈳椬㌱ⰳㅩ㐳椬㌱ⰵㅩ㘳椬㌱ⰷㅩ㠳椬㌱ⰹㅩ〴椬㐱ⰱㅩ㈴椬㐱ⰳㅩ㐴椬㐱ⰵㅩ㘴椬㐱ⰷㅩ㠴椬㐱ⰹㅩ〵椬㔱ⰱㅩ㈵椬㔱ⰳㅩ㐵椬㔱ⰵㅩ㘵椬㔱ⰷㅩ㠵椬㔱ⰹㅩ〶椬㘱ⰱㅩ㈶椬㘱ⰳㅩ㐶椬㘱ⰵㅩ㘶椬㘱ⰷㅩ㠶椬㘱ⰹㅩ〷椬㜱ⰱㅩ㈷椬㜱ⰳㅩ㐷椬㜱ⰵㅩ㘷椬㜱ⰷㅩ㠷椬㜱ⰹㅩ〸椬㠱ⰱㅩ㈸椬㠱ⰳㅩ㐸椬㠱ⰵㅩ㘸椬㠱ⰷㅩ㠸椬㠱ⰹㅩ〹椬㤱ⰱㅩ㈹椬㤱ⰳㅩ㐹椬㤱ⰵㅩ㘹椬㤱ⰷㅩ㠹椬㤱ⰹ㉩〰椬〲ⰱ㉩㈰椬〲ⰳ㉩㐰椬〲ⰵ㉩㘰椬〲ⰷ㉩㠰椬〲ⰹ㉩〱椬ㄲⰱ㉩㈱椬ㄲⰳ㉩㐱椬ㄲⰵ㉩㘱椬ㄲⰷ㉩㠱椬ㄲⰹ㉩〲椬㈲ⰱ㉩㈲椬㈲ⰳ㉩㐲椬㈲ⰵ㉩㘲椬㈲ⰷ㉩㠲椬㈲ⰹ㉩〳椬㌲ⰱ㉩㈳椬㌲ⰳ㉩㐳椬㌲ⰵ㉩㘳椬㌲ⰷ㉩㠳椬㌲ⰹ㉩〴椬㐲ⰱ㉩㈴椬㐲ⰳ㉩㐴椬㐲ⰵ㉩㘴椬㐲ⰷ㉩㠴椬㐲ⰹ㉩〵椬㔲ⰱ㉩㈵椬㔲ⰳ㉩㐵椬㔲ⰵ㉩㘵椬㔲ⰷ㉩㠵椬㔲ⰹ㉩〶椬㘲ⰱ㉩㈶椬㘲ⰳ㉩㐶椬㘲ⰵ㉩㘶椬㘲ⰷ㉩㠶椬㘲ⰹ㉩〷椬㜲ⰱ㉩㈷椬㜲ⰳ㉩㐷椬㜲ⰵ㉩㘷椬㜲ⰷ㉩㠷椬㜲ⰹ㉩〸椬㠲ⰱ㉩㈸椬㠲ⰳ㉩㐸椬㠲ⰵ㉩㘸椬㠲ⰷ㉩㠸椬㠲ⰹ㉩〹椬㤲ⰱ㉩㈹椬㤲ⰳ㉩㐹椬㤲ⰵ㉩㘹椬㤲ⰷ㉩㠹椬㤲ⰹ㍩〰椬〳ⰱ㍩㈰椬〳ⰳ㍩㐰椬〳ⰵ㍩㘰椬〳ⰷ㍩㠰椬〳ⰹ㍩〱椬ㄳⰱ㍩㈱椬ㄳⰳ㍩㐱椬ㄳⰵ㍩㘱椬ㄳⰷ㍩㠱椬ㄳⰹ㍩〲椬㈳ⰱ㍩㈲椬㈳ⰳ㍩㐲椬㈳ⰵ㍩㘲椬㈳ⰷ㍩㠲椬㈳ⰹ㍩〳椬㌳ⰱ㍩㈳椬㌳ⰳ㍩㐳椬㌳ⰵ㍩㘳椬㌳ⰷ㍩㠳椬㌳ⰹ㍩〴椬㐳ⰱ㍩㈴椬㐳ⰳ㍩㐴椬㐳ⰵ㍩㘴椬㐳ⰷ㍩㠴椬㐳ⰹ㍩〵椬㔳ⰱ㍩㈵椬㔳ⰳ㍩㐵椬㔳ⰵ㍩㘵椬㔳ⰷ㍩㠵椬㔳ⰹ㍩〶椬㘳ⰱ㍩㈶椬㘳ⰳ㍩㐶椬㘳ⰵ㍩㘶椬㘳ⰷ㍩㠶椬㘳ⰹ㍩〷椬㜳ⰱ㍩㈷椬㜳ⰳ㍩㐷椬㜳ⰵ㍩㘷椬㜳ⰷ㍩㠷椬㜳ⰹ㍩〸椬㠳ⰱ㍩㈸椬㠳ⰳ㍩㐸椬㠳ⰵ㍩㘸椬㠳ⰷ㍩㠸椬㠳ⰹ㍩〹椬㤳ⰱ㍩㈹椬㤳ⰳ㍩㐹椬㤳ⰵ㍩㘹椬㤳ⰷ㍩㠹椬㤳ⰹ㑩〰椬〴ⰱ㑩㈰椬〴ⰳ㑩㐰椬〴ⰵ㑩㘰椬〴ⰷ㑩㠰椬〴ⰹ㑩〱椬ㄴⰱ㑩㈱椬ㄴⰳ㑩㐱椬ㄴⰵ㑩㘱椬ㄴⰷ㑩㠱椬ㄴⰹ㑩〲椬㈴ⰱ㑩㈲椬㈴ⰳ㑩㐲椬㈴ⰵ㑩㘲椬㈴ⰷ㑩㠲椬㈴ⰹ㑩〳椬㌴ⰱ㑩㈳椬㌴ⰳ㑩㐳椬㌴ⰵ㑩㘳椬㌴ⰷ㑩㠳椬㌴ⰹ㑩〴椬㐴ⰱ㑩㈴椬㐴ⰳ㑩㐴椬㐴ⰵ㑩㘴椬㐴ⰷ㑩㠴椬㐴ⰹ㑩〵椬㔴ⰱ㑩㈵椬㔴ⰳ㑩㐵椬㔴ⰵ㑩㘵椬㔴ⰷ㑩㠵椬㔴ⰹ㑩〶椬㘴ⰱ㑩㈶椬㘴ⰳ㑩㐶椬㘴ⰵ㑩㘶椬㘴ⰷ㑩㠶椬㘴ⰹ㑩〷椬㜴ⰱ㑩㈷椬㜴ⰳ㑩㐷椬㜴ⰵ㑩㘷椬㜴ⰷ㑩㠷椬㜴ⰹ㑩〸椬㠴ⰱ㑩㈸椬㠴ⰳ㑩㐸椬㠴ⰵ㑩㘸椬㠴ⰷ㑩㠸椬㠴ⰹ㑩〹椬㤴ⰱ㑩㈹椬㤴ⰳ㑩㐹椬㤴ⰵ㑩㘹椬㤴ⰷ㑩㠹椬㤴ⰹ㕩〰椬〵ⰱ㕩㈰椬〵ⰳ㕩㐰椬〵ⰵ㕩㘰椬〵ⰷ㕩㠰椬〵ⰹ㕩〱椬ㄵⰱ㕩㈱椬ㄵⰳ㕩㐱椬ㄵⰵ㕩㘱椬ㄵⰷ㕩㠱椬ㄵⰹ㕩〲椬㈵ⰱ㕩㈲椬㈵ⰳ㕩㐲椬㈵ⰵ㕩㘲椬㈵ⰷ㕩㠲椬㈵ⰹ㕩〳椬㌵ⰱ㕩㈳椬㌵ⰳ㕩㐳椬㌵ⰵ㕩㘳椬㌵ⰷ㕩㠳椬㌵ⰹ㕩〴椬㐵ⰱ㕩㈴椬㐵ⰳ㕩㐴椬㐵ⰵ㕩㘴椬㐵ⰷ㕩㠴椬㐵ⰹ㕩〵椬㔵ⰱ㕩㈵椬㔵ⰳ㕩㐵椬㔵ⰵ㕩㘵椬㔵ⰷ㕩㠵椬㔵ⰹ㕩〶椬㘵ⰱ㕩㈶椬㘵ⰳ㕩㐶椬㘵ⰵ㕩㘶椬㘵ⰷ㕩㠶椬㘵ⰹ㕩〷椬㜵ⰱ㕩㈷椬㜵ⰳ㕩㐷椬㜵ⰵ㕩㘷椬㜵ⰷ㕩㠷椬㜵ⰹ㕩〸椬㠵ⰱ㕩㈸椬㠵ⰳ㕩㐸椬㠵ⰵ㕩㘸椬㠵ⰷ㕩㠸椬㠵ⰹ㕩〹椬㤵ⰱ㕩㈹椬㤵ⰳ㕩㐹椬㤵ⰵ㕩㘹椬㤵ⰷ㕩㠹椬㤵ⰹ㙩〰椬〶ⰱ㙩㈰椬〶ⰳ㙩㐰椬〶ⰵ㙩㘰椬〶ⰷ㙩㠰椬〶ⰹ㙩〱椬ㄶⰱ㙩㈱椬ㄶⰳ㙩㐱椬ㄶⰵ㙩㘱椬ㄶⰷ㙩㠱椬ㄶⰹ㙩〲椬㈶ⰱ㙩㈲椬㈶ⰳ㙩㐲椬㈶ⰵ㙩㘲椬㈶ⰷ㙩㠲椬㈶ⰹ㙩〳椬㌶ⰱ㙩㈳椬㌶ⰳ㙩㐳椬㌶ⰵ㙩㘳椬㌶ⰷ㙩㠳椬㌶ⰹ㙩〴椬㐶ⰱ㙩㈴椬㐶ⰳ㙩㐴椬㐶ⰵ㙩㘴椬㐶ⰷ㙩㠴椬㐶ⰹ㙩〵椬㔶ⰱ㙩㈵椬㔶ⰳ㙩㐵椬㔶ⰵ㙩㘵椬㔶ⰷ㙩㠵椬㔶ⰹ㙩〶椬㘶ⰱ㙩㈶椬㘶ⰳ㙩㐶椬㘶ⰵ㙩㘶椬㘶ⰷ㙩㠶椬㘶ⰹ㙩〷椬㜶ⰱ㙩㈷椬㜶ⰳ㙩㐷椬㜶ⰵ㙩㘷椬㜶ⰷ㙩㠷椬㜶ⰹ㙩〸椬㠶ⰱ㙩㈸椬㠶ⰳ㙩㐸椬㠶ⰵ㙩㘸椬㠶ⰷ㙩㠸椬㠶ⰹ㙩〹椬㤶ⰱ㙩㈹椬㤶ⰳ㙩㐹椬㤶ⰵ㙩㘹椬㤶ⰷ㙩㠹椬㤶ⰹ㝩〰椬〷ⰱ㝩㈰椬〷ⰳ㝩㐰椬〷ⰵ㝩㘰椬〷ⰷ㝩㠰椬〷ⰹ㝩〱椬ㄷⰱ㝩㈱椬ㄷⰳ㝩㐱椬ㄷⰵ㝩㘱椬ㄷⰷ㝩㠱椬ㄷⰹ㝩〲椬㈷ⰱ㝩㈲椬㈷ⰳ㝩㐲椬㈷ⰵ㝩㘲椬㈷ⰷ㝩㠲椬㈷ⰹ㝩〳椬㌷ⰱ㝩㈳椬㌷ⰳ㝩㐳椬㌷ⰵ㝩㘳椬㌷ⰷ㝩㠳椬㌷ⰹ㝩〴椬㐷ⰱ㝩㈴椬㐷ⰳ㝩㐴椬㐷ⰵ㝩㘴椬㐷ⰷ㝩㠴椬㐷ⰹ㝩〵椬㔷ⰱ㝩㈵椬㔷ⰳ㝩㐵椬㔷ⰵ㝩㘵椬㔷ⰷ㝩㠵椬㔷ⰹ㝩〶椬㘷ⰱ㝩㈶椬㘷ⰳ㝩㐶椬㘷ⰵ㝩㘶椬㘷ⰷ㝩㠶椬㘷ⰹ㝩〷椬㜷ⰱ㝩㈷椬㜷ⰳ㝩㐷椬㜷ⰵ㝩㘷椬㜷ⰷ㝩㠷椬㜷ⰹ㝩〸椬㠷ⰱ㝩㈸椬㠷ⰳ㝩㐸椬㠷ⰵ㝩㘸椬㠷ⰷ㝩㠸椬㠷ⰹ㝩〹椬㤷ⰱ㝩㈹椬㤷ⰳ㝩㐹椬㤷ⰵ㝩㘹椬㤷ⰷ㝩㠹椬㤷ⰹ㡩〰椬〸ⰱ㡩㈰椬〸ⰳ㡩㐰椬〸ⰵ㡩㘰椬〸ⰷ㡩㠰椬〸ⰹ㡩〱椬ㄸⰱ㡩㈱椬ㄸⰳ㡩㐱椬ㄸⰵ㡩㘱椬ㄸⰷ㡩㠱椬ㄸⰹ㡩〲椬㈸ⰱ㡩㈲椬㈸ⰳ㡩㐲椬㈸ⰵ㡩㘲椬㈸ⰷ㡩㠲椬㈸ⰹ㡩〳椬㌸ⰱ㡩㈳椬㌸ⰳ㡩㐳椬㌸ⰵ㡩㘳椬㌸ⰷ㡩㠳椬㌸ⰹ㡩〴椬

... So, can anyone help me out or rewrite my program? I unfortunately can't code in Mandarin.

[Willfor]

I'm guessing item_list.txt isn't filled with the characters your program just output? I'd help, but my python skillz are at roughly... zero.

[Mondark]

My own python skillz being rather less than 0.001, there's a much better than average chance I'm wrong here, but it looks like the problem lies with the 's = str(a)' bit.  I'd guess that the 'a' variable is a number, and the str function is converting it to some unicode string you don't want.  Try having it print out the amount variables before starting the loop, and maybe trying to set them to something corresponding to english in unicode (or whatever encoding).

At least, that's my guess.

[Sowelu]

What's your expected output for this little snippet?  Is it:

i100,
i101,
i102,
i103,

?  If that's the case, yeah, I see what's going on.  Just a misunderstanding of str(a).  You think that str(123) will produce "123", but instead it produces the single character that sits at position 123 on the character table.  Look for a function like itoa(), which is the C version of what you want.  Or maybe just a print-type of function that will print a number, as a string, INTO another variable.

[Fualkner]

Ah, that does make sense... thanks guys, I'll try it out.

[deej]

I've got three things for you to add to the top posts.

My DF inspired 7DRL, Excitable Digger, whos discussion and downloads are found here:
http://www.bay12games.com/forum/index.php?topic=32304.0

A console input/output emulator library for Java which can be found here:
http://code.google.com/p/libjcsi/

And a Java engine for creating RLs and CRPGs, which uses the above library here:
http://code.google.com/p/serf-engine/

Both of those last two were created by Slashie of http://slashie.net/news.php fame.

The libjsci library has been expanded and improved from where it started by Slashie and Myself, so if anyone runs into issues with it, I can probably help.

[Dasleah]

Man, it's been ages since this thread was active. Let's change that.

The roguelike I was working on stalled (gasp! I hear you say. Who'da thunk it) mainly because I was too busy being lazy and fooling around with Google Apps. Recently though, I've been looking at Boo and by extension, .NET. It's really quite interesting, and fortunately, a .NET binding of libtcod exists (and thanks! I know there's a person who contributed to it somewhere on this forum, so kudos to your work).

Boo is similar enough to Python for it to be really comfortable for me, which makes it very attractive. It's just taking a while to get used to some things (compiling?! What the hell!) and of course instead of being able to use Python's standard library, it's off to dig through the .NET references to use something. But apart from that, it's going well.

I think the main reason why my previous project died was that it was just coding for coding's sake - I had an idea for what kind of game it might have been, but I didn't have a plan that I could follow, a justification for coding beyond "well this thing is common to all roguelikes so I might as well code it". So I'm going to change that and before I do anything, I'm going to write out a design document for it, detailing not only what I'm coding to code, but why.

Thanks for the links deej, I'll add those straight away.