I don’t know of any games, but I think that a simple cellular automata based generator would produce cave-like veins. For example, in the picture below, the white could be empty or soft stone and the black could be hard stone or undiggable stone.
I think that a perlin noise generator would work, too, as they often are used for natural-looking patterns.
In either case, this phase would just generate the veins in the rock. The rest of the features of the mountain (lava, water, lairs with existing tribes, etc) would just be dug out from whatever this cave-vein-generation phase creates using the existing algorithms you have.
Reason behind this is that I don’t think it’s interesting to dig a dungeon from a blank canvas every single game, it’s too easy. There was a good post on the Steam forums a few months back by some other guy that echoed a similar sentiment.
As for seeing stuff in advance, the system that exists now, seeing two or three squares into the blackness, seems to work well enough.
I’ve thought about this before, clearly something that works for MOBAs could make an interesting combat mechanic in KeeperRL. I wonder if it’s not possible to do this by differentiating melee and projectile damage? These things are already present in the game, and would only require some changes in formulas (I guess projectiles would bypass armor and strength, and would use a separate stat and class of items to calculate defense). What do you think?
Well, my suggestion would still be to make damage type and weapon range being two orthogonal categories so that damage type A and damage type B could both be dealt at any distance. Maybe it’s just because I’m used to playing games that do it like that.
That’s a possibility. Maybe researching basic sorcery, for example, will also automatically give all of your creatures access to a complimentary package of basic spells that don’t need to be found. The way I see it, the game is going to have a large number of spells that will all be divided into “tiers” (e.g., basic, advanced, expert, master, to use the existing terminology). Any spell of any tier could be potentially found on the map, but by researching “basic sorcery,” it would be like giving you a one-time of bonus of 4-6 complimentary basic spells. Perhaps these are the same 4-6 basic spells each game or perhaps they are different each game.
Only when you find the master sorcery book you get the more advanced spells.
To be clear, I was suggesting that “master sorcery” (or “master abilities,” as I think it should be called) should still be researched at the cost of mana, but the individual spell or skill is what you would find in all of these books that are littered around the world map. So, even if you found a really powerful spell early in the game, you wouldn’t be able to use it immediately because you haven’t “unlocked it” in the the tech tree.
But maybe a cleaner way to do what I’m describing would be to require that “good” spellbooks can only be placed in higher-tier libraries (so we could have wooden shelves, iron shelves, steel shelves), which would be unlocked with ironworking/steelworking. So if you found a powerful spellbook, but you only had wooden shelves, you’d have to wait until you unlocked steel bookcases. So, then you wouldn’t need the “sorcery” research branch any more. Well, it’s just another idea. If you think it’s horrible, I won’t take offense :p
If you get a minion with some ability, is it able to pass it to other minions or do you still need to get a book?
Generally, one book can teach infinitely many minions (but maybe some books have limited use). When one of the player’s raiding parties comes home, it would drop any books that it has collected at the entrance to the dungeon just as it does with all of its other loot. The only difference is that books get placed on library tiles instead of going into general equipment storage.
Would minions do any learning automatically or all of it needs to be micro-managed?
When you want to assign an ability to a minion, you would do so from the minion inventory window. Let’s say an orc has a free ability spot available; you click on the free ability spot and tell it to learn “strength boost.” It goes over to the library tile that has the strength boost book inside of it and it starts training to learn strength boost.
I don’t see much point to automating the learning process. I suppose that you *could* make an addition to the immigration UI panel that tells X creature to automatically learn Y ability when recruited, but idk if that would be of much use.
I don’t like the fact that you need to do additional research on top of finding abilities. Maybe just add minion level or stat requirements per ability?
This would work, too, and it would be less tedious than waiting for a minion to explicitly train a new ability like I suggested. I just think that minions should have a limited number of ability slots (maybe they get more as they level up).
So then, here’s another proposal for learning spells. This one doesn’t use magic weapons or items at all. It also involves what the game calls “skills” (e.g., knife throwing), so I’ll collectively refer to spells and skills as “abilities.” Here’s the idea:
– The total number of abilities in the game is brought up into the hundreds.
– Some creatures have some natural abilities along with their natural traits…
– … but most abilities must be “found” on the world map as books that must be retrieved and stored in a library tile. In order to accumulate this knowledge, the player must constantly grow his library.
– Once in the library, an ability can be studied by a minion. Just like crafting, learning takes time and the player has full control over what he wants to teach to whom, but minions can only learn so many skills and they take their knowledge to the grave, so study wisely!
– Every ability that is found has a tier (e.g., basic, advanced, expert, master). The “sorcery” research items are still present in the research tree, but they no longer grant a package of spells and they are renamed to “abilities.” In order for a minion to study an ability of the “basic” tier, “basic abilities” much be researched. (This is simply to prevent the player from being able to use an overpowered ability early on in the game; Brogue does this by making heavy weapons require lots of strength, and FTL does this by making powerful weapons require lots of energy to use). It’s sort of like how you can build a forge, but you still have to research “two-handed” weapons in order to make a battle axe.
– Certain abilities are restricted to certain types of creatures. (Yes, this means that there will be thousands of potential creature-ability pairs that need to be defined. But I or other people could help to define these pairs if need be)
– Naturally, the player is only going to see a small subset of the available abilities in each game. This is what helps to make each replay a bit different.
– As a subtle point, floors affect learning rate, and once an ability is placed inside a library tile, it’s magically bound to that bookshelf forever!
– This system could also be extended to crafting stuff. Eg: the recipe needs to be discovered on the world map, and then learned by a specific unit.
– Beasts probably wouldn’t be able to learn new things by studying because they can’t read. In this case, you could recruit a proxy unit (e.g. satyr, centaur) who could learn abilities and teach it to the beasts. Or do something else clever like that. idk
All of that covers how abilities are learned. But it still doesn’t cover how powerful an ability is going to be, or how much its cooldown is. If an ability is scalable, then it will do so as a function primarily of the attribute values of the creature. As I said earlier, I would suggest borrowing from MOBAs and creating an abstract attribute called “intelligence” or “ability power,” which would be used by most abilities to determine the scaling power of the ability. However, some simple abilities might be independent of this ability power stat (e.g., a stat boost ability which simply multiplies an attribute by some constant factor for a short period of time).
