I've said before that I didn't have a real plan or strong vision for Delve when I began coding. I figured I'd have some fun coding up a roguelike game and see how far I could get. The local directory (and github repo) I build in is "yarl2" (yet another roguelike)...
That said, I did set out with the vague idea I wanted to have an overworld with a village with NPCs the player could interact with. And I had in mind a game play loop where the player explores a bit, returns to town, explores deeper. Kind of like Darkest Dungeon maybe, or how I played older games like Bard's Tale. In CRPGs, I enjoy having a to-do list or a small goal as part of a larger mission. "This session I want to find the stairs to the next floor", "Let me get enough XP to level up my wizard", "Try to find the place to use this crystal key I found.", "The blacksmith wants you to find her dog who disappeared into the dungeon"
Unlike real life, video game chores are just very satisfying to complete.
In traditional roguelikes you tend to just plunge into the dungeon. Many of them have no town or overworld and there's usually minimal backtracking. Even in Angband the usual advice is to explore as deep as possible and only return to town when you need absolutely need to. And recent versions have eliminated hauling expensive items out of the dungeon to sell, reducing motivation for returning to town.
So since I began development without a strong vision of what the game would be, I mostly did things the way the traditional roguelikes I loved did things. I built Delve with a pretty standard roguelike dungeon, but in the back of my mind I also wanted that explored/retreat game loop and looked for ways to incentivize returning to the town. This was why I added the half-assed stress mechanic (stolen from Darkest Dungeon) and eventually split the main dungeon into a few smaller ones.
I'm not sure I really like the stress mechanic. It isn't fleshed out and I don't know that it fits well with the sillier tone I began to add to the game's text.
Side quests all the way down
I do wish I'd spent more time thinking about the kind of game I wanted to make before I began coding. But to be fair, this was began on a whim. I had no idea how much effort I was going to put into this.
Apparently quite a bit, it turns out...
Needless to say, with more forethought I'd likely have come up with a different game.
So I've been mulling this all over. I have this vision of a game that's a little like...Stardew Valley maybe? A village full of NPCs to interact with. Some of the villagers give you quests within the dungeon along with the over-arching goal of exploring deeper to eventually deal with the evil demon monster or whatever that lurks at the very bottom. You'd unlock and expand the town with new shops and such. Repeated chats and such with villagers might unlock more sidequests and story.
I wouldn't attempt proc-gen dialogue and villagers because hand-written ones will always be more compelling. Ideally, I'd have a roster of villagers and at the start of the game select a few of them for that run. And Stardrew Valley demonstrates that if the characters are well-written, people will be happy seeing the same ones over and over.
I think this game would not have persistent levels. Each time you entered the dungeon, the levels would be generated anew. You'd explore for a while and when your HP gets low, return to town and rest at the inn.
What to do?
So that's kind of my dilemma. Do I steer Delve in this direction? Do I finish Delve in a more classical roguelike form and then make a whole other game? What if I want to move on and work on a non-game project after this? What if I push Delve in this other direction and then decide I don't like it? (I mean, I have git but it would feel real bad to waste days or weeks of effort)
I won't really know what is the right decision for a few weeks of development time and it's a little scary :O I'm not sure if this post is me trying to talk myself into it, or out of it...
