No one has mentioned Recettear? It did the "merchant" thing as well.
Yeah, Recettear is an obvious point of reference. I don't think I had played the game when I dreamt up the basic mechanics so far, but it probably is in there somehow, subconsciously. I don't intend to add the dungeons in to any real detail, always keeping them as abstracted little adventures based mainly on random events that lead the heroes either deeper in to the dungeons - to more peril - or back out again.
I would like to add in a means of directly communicating with the heroes though, making them go in to dungeons and fetch you stuff on 'contracts', perhaps based on whether you have a good relationship with the hero in question.
Just having a think about how the hero characters get a chance to interact with the dungeon in better ways based on their skills, got some ideas which I'll try and implement soon based on some 'special rooms' which all heroes will potentially get access to, but only certain heroes get to do anything with (e.g. because they have a given perk).
Secret treasure, summoning altars etc.
Key differences are that this game looks to be more gritty, and apparently will quantify RPG plots for their effects on economics. Or am I misunderstanding? When you say "killing dreams" and "bringing the wrong stuff out of the dungeon," are you going to have randomized RPG plots that can happen behind the curtain? Or will it be based off of, e.g., random possibilities attached to each adventurer's personality?
--Rexfelum
The main thrust of the 'killing dreams' comment is to try and get the world behaviour to change based on the behaviour and performance of the heroes. At the start of the game, it should be pretty normal out there, with some buried dungeons out in the woods that slowly grow in power.
As they grow in power, better and more extraordinary items get pulled from the depths. These either have a latent positive / negative effect on the weighted balance of good/evil order/chaos, and depending on that world score, depends on what goes on in the world. There is also a challenge in trying to make a clear link between the player and the performance of the heroes. It's a bit disconnected at the minute.
I have a basic good/order metric in - based on the total summed inventory of the worlds items, but it's clear that this is not going to be sufficient on its own to get any variance over time (it will just sit at effectively 50% due to the statistical weighting of the alignment of all averaged items).
So I need to get the power of the dungeons balanced right so they grow (at the minute they are getting 'cleansed' pretty efficiently, within two months or so, without killing enough heroes), and I need to try and get the increase in world evil power to increase the profitability of the stores, probably by linking the tax rate to 'order/chaos' (more chaos less tax) and also more monsters / higher dungeon levels = better loot from the dungeons.
Also visual cues like pixies / gremlins flitting about the screen in the forest.
Probably would have more desperate heroes as well, so you could mark up your goods to a greater level.
But this would come with a downside of having evil creatures springing from the dungeons, travelling across the town map, and causing havoc in town.
Ideally it would be a balance where ->
too much order = no more dungeons = no more heroes = no footfall (plus high taxes, quicker 'fail' condition from this angle)
too much chaos = too many monsters = collateral damage (but higher profits, more 'churn', lower taxes).
The order path would be stable and low risk, the chaos path would be unstable and high risk.
Good and evil would just really be flavour modifications, I think (unicorns generating in the dungeons rather than ogres, say)
I would like to try and achieve a system whereby you can't normally play the game for any longer than 6 months before a failure condition grabs you down. It would then poll your high score in terms of profit made (but hopefully it can go on for longer than this without any real issues).
In terms of the RPG plots, it is very much tending towards a whole host of random rolls tied in to the personalities. I think I could make this a lot more immersive by opening up the hero activity to be seen by the player (with a few hundred lines of flavour text associated with actions, hopefully this can be made to appear reasonable) and that will certainly be the first path I will be taking. It also fits in pretty well with where the codebase is at the minute, so all I would really need to do is whip up an individual hero interface sheet with a list of recent thoughts / activities and write a bunch of different possible sentences to describe the various things that can go on.