Post by Margaret Dax in Lesser Journeys: prototyping a procedurally generated subgame

Viewing post in Lesser Journeys: prototyping a procedurally generated subgame

From what I played this feels pretty interesting. I made a table of interactions which I'm including and built bc it demonstrates what I feel is something kinda missing. All the actions with zero at the intersection feel a little.... empty? Balancing things that interact in a destructive or constructive way is really interesting, but since they do that on one dimension I can understand not wanting to make all of them interact in that way. Have you considered other dimensions for them to interact on? For instance you could have ones with zeros impact the gold click reward up and down, so you could for instance click water next to bones, and now bones cost down to -2 gold to click, or forest next to flower increasing the gold click of flower up to 4. Dunno maybe that's not an interesting vector. The other I was thinking about is spreading, so maybe ones that don't interact with neighbors replace them with a 1. So click bones next to fire and it replaces the fires with bones, but if there were flowers they grow. Sorry if any of this is unwelcome, just thinking outloud mostly :) Look forward to seeing where this goes!

droqen5 years ago (1 edit)

Ah, no, not unwelcome at all -- sorry about the slow response, but it's rly good to see your take on it!

The most important thing right now is that the system can be intuited roughly and interacted with in a sort of pleasant way, so keeping the number of abstractly different interactions to a minimum is pretty important to me. That is, I want someone to be able to enter a zone and figure out a couple important rules and not have to worry about all the rest!

These vectors sound really interesting to me but I'm concerned about visibility, about too much complexity. Without giving away too much right off the bat, this isn't a system that's meant to remain interesting across an entire game, but to provide a locally interesting puzzle that can be solved without a ton of trouble. So I'll be making more areas with different rules, some more complex than this.... and probably some simpler!

I'm going to think more about what kind of vectors will be interesting to work with, because ultimately I do want to have wild & varied rulesets, which means I'm gonna need a lot of wild & varied rules.

for organization's sake:

alter HP of nearby tiles
alter GOLD REWARD/COST of nearby tiles (vis: show a 'stack of gold' or something indicating the reward level? could be interesting to maximize the gold, if it increases exponentially or something)
change tile-type of nearby tiles*
revive nearby empty tiles, filling them w/ a tile again
give player gold / take player gold
other player-targeting effects. stun them for a bit, take away hp, return hp?
any of the above but with broader reach, or temporarily - e.g. boost all tiles' gold rewards by x 2, but only for 1 minute. stacking effect.

* I did try this one already, actually, and it was very very disorienting... but, I did have it happening pretty randomly. 'Spreading' is a good metaphor. HP was my first attempt at 'changing tile-type' in an intuitive way. Rather than gold -> bones -> water, I was doing big tree -> mid tree -> li'l tree. But I never got to the point of treating each HP-level as a completely different tile.

Margaret Dax5 years ago

Treating different HP as different tiles would be really fun to try out! So there's different interactions at different stages. Maybe some of what I experienced as dull could be solved through level gen though. Yeah I was thinking like one more vector of interaction max, because you're completely right it'd get very confusing.

I think tile interactions changing based on XP could add a lot of depth in a pretty simple way so I'd be interested in trying that out. Could see going from person to warrior to skeleton or something haha. Probably having a mix of ones that are a quantity of a single thing vs a quantity of stages of things could be cool. Applying real temporality to something player-stepped sounds pretty interesting though maybe slightly disorienting. That dynamic could be cool of having to beat the clock in real life vs making minimal sub-optimal decisions from rushing to maximize that temporary benefit.

Definitely weird to think about this idea of a subgame, straight up a ruleset without a clear labeling of what the player is doing as good or bad, winning or losing. Thinking about this in context of minigames like the FO4 password hacking, or skyrim lock picking. You can probably get better at this, but it doesn't have a clear win/lose condition. I think if you had it as you have to reach the door, or reach the 3rd door, or something, would make it a game. If you had limited resets it's a game of how many doors do you make it through. If it's 1 room it's how much gold can I get / space can I clear (SpaceChem graphs and all). But since there isn't an end and there's this notion of suboptimal decisions (higher # resets = bad) and optimal decisions (maximizing destructive interactions towards gold / exits) but no clear labeling of an overall performance that makes it disorienting. I wonder how you can take relative-limits based goals away from things that currently exist.

Mario lives as a ++ counter instead of a -- counter
Halo deathmatch -death +damage score

I've gotta get going but I'll be chewing on this for sure :)

itch.io

Viewing post in Lesser Journeys: prototyping a procedurally generated subgame