Lately, I've been trying to put together the 'game' of Goblinson Crusoe. What I have now is a framework. A set of mechanics. A turn based structure that lets you cast spells, move, perform various actions, etc... But there is just no 'game' there, and I think it's getting to be that time. Time to figure out what it's all about.
One of the things that I've waffled about is whether or not to have an 'overmap mode'. In the beginning I did, then I eliminated it, then I brought it back, then I eliminated it. I'm thinking of bringing it back. I'm sort of struggling toward an idea that is only half-visible, and I still don't really have a strong sense of how the overarching gameplay is going to work, but here are my current thoughts.
The central basis of the whole thing, for me, is that the game follows the adventures of a young goblin who has been exiled to a remote archipelago in order to complete his training as a goblin wizard. He is given the fundamentals, but he must eke out his destiny on his own, at first in a simple fight for mere survival, and eventually in a quest to attain mastery of goblin wizardry. His quest will take him to various islands through the archipelago, as he explores (and plunders) the ruins of ancient goblin civilizations in order to learn their secrets, fights against pirates and human renegades, braves the depths of active volcanoes to wrest fire from the earth gods, and so forth. You know, the usual stuff a goblin wizard apprentice gets up to. Shenanigans.
My very first vision of it was ARPG, with each island being a complete game state. You explore the island, kill the baddies, loot the chests, go back to your raft, and sail away. With the switch to turn-based, however, I began to visualize it as more like, say, the classic JRPGs of my youth. You wander around on a world map, and are periodically cast into various randomized encounters as you roam. Unlike the JRPGs, each random encounter takes place in a location map with tactical maneuvering and whatnot: basically, the mechanics you can see at work in the current iteration. Expend movement points to maneuver, cast spells, whatever, and each unit takes their turn in sequence.
At some point, I eliminated the over world, and just had the goblin's raft staged as a device to allow traveling between island maps. However, this feels kinda... not right, I guess. The thing is, some of the island maps get big. Like... really big. Big enough to make it awkward and slow to try to explore the entire thing 10 movement points at a time. Some of the maps can be 256x256 or even 512x512 hexes, which might not sound like a lot but just try moving across it at 10 points per turn, especially with lots of combat critters being woken up as you go and lots of elevation forcing you along tortuous paths. It can literally take hours to explore some of the big maps, which might not necessarily be a bad thing but it is still kind of a long time to spend in one place.
My current thinking is to bring back the world map again. And I think that I would really like to implement it in the visual style of several iterations ago:
or
The thing is, I kind of miss that map style. It looks about right to me when I picture the world of GC. I'm wondering if I can make it work with that as the overmap, and the 3D look for the individual locations.
As far as how it all works, I'm still throwing ideas around. But my current thinking is that on the overmap the player is presented with a view of the entire island he is currently residing upon. Scattered around this island will be the various pre-determined randomized encounters that the island offers. Any given hex can be clicked upon to present an exploration dialog option, so for example if the player is running low on particular resources he/she can explore hexes to gather the materials he needs. Exploration takes you to the turn-based 3D view presented in the last several journal entries. Additionally, the randomized encounters (including an 'encounter' representing the goblin's landing location, where his raft resides and that can act as a home base) are turn-based locations just as the hex exploration, with additional 'stuff' pertaining to the encounter type. Some encounters might be a boss battle. Some might be a fortress assault. Some might be a treasure plunder. Some encounters can be attached, or concatenated with, other encounters. For example, a randomized 'base defense' encounter can be attached to the 'home base' location, allowing the player the option to play a 'base defense' mini-game if desired.
The idea is to break the 3D turn-based locations into much smaller pieces. This can accomplish two things: better performance when rendering, and tighter, more rapid gameplay when in the turn-based mode. I can have smaller 3D locations without sacrificing the perceived scale of the larger islands. The encounters can be classified and marked based on the encounter type, so that the player can decide which encounters he wants to experience. If he's looking for a nice boss-brawl, he can find one on the overmap, gear up, and go for it. If he just wants a shot at looting a random valuable treasure, he can look for a treasure quest. Random encounters can re-roll periodically, though some will be 'fixed' to the theme of the level. I think this sort of situation can more easily provide the player with the kind of pace he is looking for in any given play session. If he only has time for a brief adventure, he can choose one type of encounter, but if he is looking for a meaty dungeon delve he can choose another. I would probably not implement the kind of 'pull the player into an encounter at random against his will' featured in so many of the JRPGs of my youth. The more I think about that, the more I remember exactly how annoying it could be.
As far as game saving goes, my vision has always been to save only the current world seed and player stats/eq/inventory/progression flags. Not the complete location. It simplifies the game saving system, for one thing, and reduces the incidence of save scumming to overcome individual encounters. It's a choice that I might re-evaluate at a later time, but for now I'm fairly well decided.
As is usual, I'm going to put together a short demo of how it works and farm it out to my friends and family to try out. I've gotten plenty of good feedback from them before. And hey, maybe some day I'll even follow through on providing something for the general public to play. That's a hard thing for me to do, just as it's hard for me to expose my fiction writing to the world at large. Insecurities, you know.
Some turn-based tactical RPGs (Avadon: The Black Fortress comes to mind), give you infinite move points unless an enemy is within combat range (and alert to your presence). This would make is easier to explore your larger maps.
Additionally, you could give the player more movement points (say, 25 points), but make the grid cells around enemies cost more movement if you are within their radius (i.e. anything within 7 cells of them takes 2 movement points, but they still try to move toward and attack you if you are within 12 cells of them or up to, say, 25, if you've already attacked them or they already attacked you). This would change the dynamics of combat, but it could lead to some interesting mechanics.
That said, I do like world maps. Maps help give 'boundaries' to the world, and make things feel less overwhelmingly infinite. You can feel like you're making "progress" exploring.
I also like the hand-crafted feel of human-designed worlds. If you designed your archipelago, and manually placed the "locations" in the world, but when entering each location - such as "Great Goblin Magi Ruins" or "Honeycomb Waterfall Caves" - they are procedurally generated with the different assets associated with their area, you could possibly benefit from both forms of content creation (i.e. designer-crafted world, but procedurally-generated areas).
Whatever you do, I know it'll be enjoyable. And at the very least, beautiful to look at. :)