Can you help me simplify this RPG/Strategy idea?

[supermikhail]

The core of my idea is that you, the player, can choose a character out of an existing population which is organically evolving according to sort of strategy rules. I started game dev as a hobby relatively recently, and tried approaching this idea from a couple angles, but I think I’ve underestimated the scope both times. This last time I decided to represent the world as a hex grid. I have 4 tile models according to “elevation”: mountain, hill, plain and water, then semi-randomly distribute resource markers, such as trees and fish. From these I derive the amounts of resources that are present in tiles. Then I place a random population in one tile. This is where it starts falling apart.

I’ve been looking up the demographics during the Middle Ages for appropriate parameters, and a good territory for a village seems to be 1 square kilometer. I like these parameters for a tile, but it confuses the notion of a turn. I originally thought one turn would correspond to one day, and for tasks like farming and building it makes sense, but what if you move between tiles? It’d be ridiculous if it took a whole day. And in general you’d probably want finer control than spending a whole day doing one task.

Next is a problem in representation. What happens to tree models when people cut forest down? One model was to symbolically represent a bunch of tree resources, so when do you remove one, and which one? What happens if people decide to build a village in a tile? Is its representation placed randomly and wipes out an arbitrary number of trees?

And finally, I started with a basic number of resources, a la RTS, such as wood, stone, food, but for characters and especially the player to have things to do there needs to be more stuff. For example, I have thought of having a task called “housekeeping”, one of the results of which would be improving the quality of food. I implemented that as simply doubling a certain amount of the food resource per day. This leads to people being able to indefinitely produce food out of thin air, unless I add the distinction between raw and cooked food. Along with these other issues it seems like the idea is getting more and more messy and unwieldy. And there’s still no gameplay, only generation.

Just an hour ago I thought that maybe I should get rid of a graphical representation of the world entirely. But somehow it has to be generated and have a structure, and I don’t know what that would be. And if this becomes essentially a text adventure, is Unity the right engine?

[Antony-Blackett]

It sounds like you haven’t even scratched the surface of your ideas and you don’t have a solid idea of what it is you’re building. Before you start programming you should nail down exactly what it is you’re trying to make otherwise you’ll end up hacking features on top systems they were never designed to handle them and you’ll get halfway before deciding to abandon the project or start again…

I would suggest playing out your ideas as a board game first. A lot of RPG and RTS games can be simplified down to boardgames with a few icon pieces and grid or hex board to represent the world. It’s often a good test to see if your idea is over scoped, if you can’t play it as a board game then you may find the scope is too large to complete as a video game by yourself.

I find it useful to think of a core idea or two that the game hangs on. If you think of an idea for a feature then you can ask yourself a question. “Does this new idea/feature help bring out the core ideas of the game?” “Does this new idea/feature hinder any of the core ideas of the game?”. Using simple tools like this can really help drive your decision making.

And remember that less is more:
“Perfection is Achieved Not When There Is Nothing More to Add, But When There Is Nothing Left to Take Away” - Antoine de Saint-Exupery

[ZakCollins]

@supermikhail

I’ve been looking up the demographics during the Middle Ages for appropriate parameters, and a good territory for a village seems to be 1 square kilometer. I like these parameters for a tile, but it confuses the notion of a turn. I originally thought one turn would correspond to one day, and for tasks like farming and building it makes sense, but what if you move between tiles? It’d be ridiculous if it took a whole day. And in general you’d probably want finer control than spending a whole day doing one task.

If your goal is to make a fun game, and not just a historically accurate simulation, then realism doesn’t matter much at all. Think about the Civilization series. At the start of the game in Civ, it can take hundreds of in-game years to build a granary or a wall, or to move a couple tiles. That makes absolutely zero sense whatsoever, and yet very few players notice or care. The balance of the game and being fun are infinitely more important. Imagine if tv shows and movies spent half the movie with the character sleeping and using the bathroom. It is realistic, but not entertaining at all.

Make your turns last however long you want, make towns take up as many tiles as you want, and don’t tell the player the real world scale of the tiles if you are worried about them doing the math and getting upset at how they can only travel x miles in y number of hours (but virtually no players will care). This sort of realism that the player will only notice if they do a bunch of research and pull out their calculator doesn’t really matter.

@supermikhail

Next is a problem in representation. What happens to tree models when people cut forest down? One model was to symbolically represent a bunch of tree resources, so when do you remove one, and which one? What happens if people decide to build a village in a tile? Is its representation placed randomly and wipes out an arbitrary number of trees?

I’m not sure exactly what your tiles currently look like so it is hard to give advice, but I don’t think this matters at the game’s current stage. Make the game fun and then figure stuff like this out. For simplicity, I would just say if there are any trees in a tile, then just have a forest in the tile. Don’t worry about the number of trees in the forest correctly representing the number of trees actually left in the tile. If there is a village in the tile, then just represent the village and don’t worry about trees. Early in a game’s development you should really just focus on the core features and making those fun.

[supermikhail]

Thanks for your replies.

I started implementing this idea while just trying to see whether I have enough brains to even set up a hex grid… and I sort of don’t since I had to copy the first script off the Internet. I guess the point is that I can only lay out an idea as far as the technologies and algorithms I currently know, which leads to problems like the one I’m trying to solve here.

I don’t play board games, so for that I’d have to essentially learn another skill or something, but the rest has definitely helped me start figuring things out.

The only part I still have no clue about is the resources. At the start of my current “game document” I wrote about money as the only resource that people could accumulate, but I parted with this concept after a couple days. Maybe rock, wood, etc. seemed more intuitive, but it feels like it’s going to spiral out of control, and I’m going to end up with different kinds of fruit and vegetables. I guess it’s alright if that’s just assets that could be added after the core is established.

[theANMATOR2b]

@Antony-Blackett

I would suggest playing out your ideas as a board game first.

@supermikhail

I don’t play board games, so for that I’d have to essentially learn another skill or something, but the rest has definitely helped me start figuring things out.

What this means is you should “paper” prototype before starting the game in engine. Work out the game completely on paper or board game or using tooth picks and dice, and rubber bands. A game like this can be played entirely on paper, and/or a board, so it is HUGELY beneficial to work out all these problems you are theorizing over - on paper first.
ONLY After that - can you confidently start generating ideas on how to visually represent resources, population of units per grid, movement time and distance, types of resources and how they are gathered.

@supermikhail

I’ve been looking up the demographics during the Middle Ages for appropriate parameters, and a good territory for a village seems to be 1 square kilometer. I like these parameters for a tile, but it confuses the notion of a turn. I originally thought one turn would correspond to one day, and for tasks like farming and building it makes sense, but what if you move between tiles? It’d be ridiculous if it took a whole day. And in general you’d probably want finer control than spending a whole day doing one task.

I reference the game Nobunaga’s Ambition regularly because imo I think it was done well (first games I do not know about the current versions)
In Nobunaga’s Ambition each turn represented a season, and this seemed to work well. Things that moved around the map like troops, dignitaries and vendors to sell rice would take approximately one season to get from here to there. And this is approximately realistic since traveling back in those days was not a trivial matter.

@supermikhail

Next is a problem in representation. What happens to tree models when people cut forest down? One model was to symbolically represent a bunch of tree resources, so when do you remove one, and which one?

How bout removing a branch from the tree that represents a set amount of wood resources. When all the branches are gone - no more wood resources.

[supermikhail]

I prefer ZakCollins’ idea for trees (and other features), partially because I’d thought of it, too. A few very approximate pre-made tile models, including forest, would spare me from probable jankiness of different model types interacting procedurally (like the aforementioned trees and settlements)… I guess he meant the prototype stage, but I’m ready to embrace the completely symbolic nature of this world representation.

[ZakCollins]

I think going symbolic is perfectly good for the final version of your game as well.

[Kiwasi]

@supermikhail

I don’t play board games, so for that I’d have to essentially learn another skill or something, but the rest has definitely helped me start figuring things out.

You really need to rectify this if you want to be a game designer. The art of game design is expressed in its purist form in table top games. It’s there that you strip away all the fancy graphics and hidden calculations and get to the heart of the game systems.

Play a bunch of board games in your genre. And paper prototype out your mechanics before moving to the computer. It will save you a lot of rework.

On to specifics:

@supermikhail

I originally thought one turn would correspond to one day, and for tasks like farming and building it makes sense, but what if you move between tiles? It’d be ridiculous if it took a whole day.

So let a move action cover more then one tile. Perhaps a character can move three tiles on a day, and five if a road exists. Or whatever balances out appropriately.

@supermikhail

Next is a problem in representation. What happens to tree models when people cut forest down? One model was to symbolically represent a bunch of tree resources, so when do you remove one, and which one? What happens if people decide to build a village in a tile? Is its representation placed randomly and wipes out an arbitrary number of trees?

Who cares? This is a graphics problem. Solve it later on when the core gameplay is sorted. Which brings up an interesting question. What is your core gameplay? Both Skyrim, Warcraft and Civilisation all involve moving between villages.

@supermikhail

And in general you’d probably want finer control than spending a whole day doing one task.

Do you? The finer control you have, the smaller the overall scope of the game. Ultimately it’s a choice. Do you want to simulate one day in a medieval village? Or a hundred years of warfare? You can pick either, but you can’t do both.

@supermikhail

And finally, I started with a basic number of resources, a la RTS, such as wood, stone, food, but for characters and especially the player to have things to do there needs to be more stuff.

Resources should perform a specific purpose in your game. But ultimately this comes down to my earlier question. Are you designing a grand, strategic game? Or are you making a detailed medieval life simulator?

So which is it, Skyrim or Civilisation?

[Ryiah]

@Kiwasi

You really need to rectify this if you want to be a game designer. The art of game design is expressed in its purist form in table top games. It’s there that you strip away all the fancy graphics and hidden calculations and get to the heart of the game systems.

Additionally if your game works great in a table top fashion you might want to consider trying out conventions like UNPUB which are focused around playtesting them.

http://unpub.net/
http://www.bgdf.com/forum/game-creation/playtesting/game-conventions-playtesting#comment-77413

[theANMATOR2b]

@Kiwasi

Play a bunch of board games in your genre. And paper prototype out your mechanics before moving to the computer. It will save you a lot of rework.

Suggestions to consider paper prototyping seem to have been purposefully dismissed. This might be a lesson that has to be learned the hard way.

@Kiwasi

So let a move action cover more then one tile. Perhaps a character can move three tiles on a day, and five if a road exists. Or whatever balances out appropriately.

Which could be resolved in a matter of minutes with paper prototyping.

@Kiwasi

Solve it later on when the core gameplay is sorted. Which brings up an interesting question. What is your core gameplay?

Seems to be unknown. Would be already known with paper prototyping.

@Kiwasi

The finer control you have, the smaller the overall scope of the game. Ultimately it’s a choice. Do you want to simulate one day in a medieval village? Or a hundred years of warfare? You can pick either, but you can’t do both.

Knowledge gained through prototyping.

Did I mention paper prototyping?

[supermikhail]

Re: board games. I guess I’d ask for advice on how someone like me would introduce them into their life. Problem is, I don’t have any friends, and I only have one table which is almost completely occupied by my computer. It seems like it at least would be a hassle… Wait, paper prototyping involves just writing and drawing on paper? Yeah… it’s one of those activities that is rather annoying to engage in with my arrangement. Although I should say that I quite enjoyed it when I had the space.

Oh, I also came here to ask what steps you think I’d take next to continue in the vein of a simple prototype. I’ve got the grid, a random population in a random hex, you select a person out of that population (their only distinguishing trait is their name so far), and perform several different tasks in a hex or move to a different hex. Of course, there’s barely a graphics component to it all, most information is conveyed through an on-screen log. I feel like the next logical step is NPC AI, and on the one hand I already started on that in the previous, overly-complicated iteration, and it seemed fairly straightforward, on the other hand, maybe that’s a rabbit hole.

Sorry that I haven’t replied to everyone directly. I feel really uncomfortable with long posts, and often just seem to be unable to express my thoughts concisely enough.

[theANMATOR2b]

@supermikhail

and often just seem to be unable to express my thoughts concisely enough.

Don’t worry about it. I lurked on this forum for almost 2 years before making my first post. This community is pretty forgiving and nice. And is inviting to nearly all, even non-coding wanna be developer, animators like me.

Consider putting together some stuff that your working on/towards - in a video/images and create a wip thread.

[supermikhail]

In the spirit of that suggestion I’ve been redoing the UI for the past 2 days… I feel like I still should add NPC AI before any WIP threads, since right now you basically “beat” the game in a couple dozen clicks. And I don’t have a name. And I still can’t tell if I have good gameplay or not. I feel like it can be tense, but I just haven’t added enough options. And for some reason I have trouble coming up with the options, just as with the name… I hope this isn’t off-topic, by the way.

[theANMATOR2b]

You can always setup a build and post it in feedback friday thread. A lot of useful information can be gained from other devs playing what you already have.

[supermikhail]

I went through a sort of crunch thinking I’d make it to the mentioned feedback friday, but in the end I’m not sure it’s the right place for me, because the game feels terrible to play, at least to me, and it has to do with my AI algorithm (and the lack of animations on some crucial things). I’m not sure where the right place to ask for advice about that is. In case here is one of them, in this screenshot:

on the left are all the tasks that are possible to do in the current tile. For the AI I take a random task, simulate the state of the world upon its completion, then repeat to build a sequence of tasks. I take several sequences and evaluate the end state of each with respect to needs. The problems are, 1) right now most of the tasks are travel, 2) I redo the algorithm each turn, so NPCs “discard” their previous “plan” every turn. So they end up running around like headless chickens, and not contributing to the state of the current tile, and in fact just showing up to eat all the food I’ve managed to produce.

There’s a couple simple solutions: 1) just planning for one turn, but that would mean that NPCs would never manage to build a house because it requires the intermediate step of cutting wood, which doesn’t add to needs. 2) Saving task sequences, but that would result in spam, because there’s no communication and NPCs don’t know if everybody has queued up to build a house next turn, which would be possible to do, but what’s the point of everyone owning a home if we’re going to starve to death.

I’ve thought of another option, but the post is long enough, and maybe someone out there has some advice.

[samnarain]

One of best things to do when it comes to world design, and most importantly, how to make world “economics” work, I’ve as general rule of thumb four games I look at that solve many design “problems”.

1. The Settlers II
   This game is all about creating a successful economy and expanding borders. It also, unlike modern games have a pretty cool mechanic that influences how you as a player would design your infrastructure. Definitely something you want to check out just to design your own virtual worlds.

2. Dungeon Keeper 1
   One of the best games ever made, in my opinion. Again, all about the “world” economics that determine the success rate of victory. Especially how the AI behaves and using time mechanics (“pay day”) to overrule certain behavior.

3. Warcraft 2
   Anything is made out of basic resources. But, in this case it is not so much about the game, but its level editor. Take a look how most maps are designed. And how much space buildings are taking in. Even though far from realistic, it is pretty clear “estate” usage.

4. Age of Empires
   The game is similar to the others, only it is about “evolving” to a next era. Those ages mean that the player is able to expand their base with new units and buildings, which makes town centers evolve into defensive bases.

So how does this apply to your game design? Well, everything in your original posting is solved in these classic games quite elegantly. All I can do is recommend you do some research for your game, by playing some

[supermikhail]

I happen to be an avid Age of Empires (II HD) player, but I feel that its AI player logic is quite a bit out of my reach. Anyway, I’ve played or at least looked at those 4 games, and drew some conclusions.

First, it seems that the strategy part is executed in a top-down fashion, instead of bottom-up as I’ve been doing. What I mean is, in the majority of those games the player decides on the placement of structures and/or production of units, or it’s some overarching AI stratagem that moves units like chess pieces. I’m not sure what to do with that, I think that’s up there with stuff beyond my reach.

Second, units have distinct roles which encompass a few or just one task. At least along these lines my brain quickly started working. I’m not sure yet if it’s in the right direction, but I decided to implement jobs for my people. Although I only started paper-prototyping yesterday, and today I’m still not sure I’m ready for implementation. But I thought I’d post to let you know I didn’t ignore your advice.

[supermikhail]

I’m stuck again.

Whatever I do, jobs are getting spammed, food and resource exhaustibility is messing with jobs, play experience has devolved into pressing “Next” over and over…

First, jobs I have, whose names should be somewhat self-explanatory: Nomad, HunterGatherer, Worker (this last one is responsible for cutting wood and building houses, which is sort of a problem in itself). Advancement of a hunter-gatherer to worker is dependent on food surplus. Depending on the parameters of the Hunt task, workers either get spammed (then eat all the food and flip back to hunter-gatherers without achieving anything useful), or don’t appear at all.

I’ve also watched some board-games on Youtube. The trouble is I’m trying to combine a strategy with an RPG and whenever I think I’ve come up with a good abstraction for one layer I realize that it won’t work for the other one.

[Kiwasi]

@supermikhail

The trouble is I’m trying to combine a strategy with an RPG and whenever I think I’ve come up with a good abstraction for one layer I realize that it won’t work for the other one.

Ever wonder why no one has ever successfully pulled that off before? Some game designs just don’t work very well.

Don’t try and change the world on your first design.

[supermikhail]

Yeah, I’m thinking I should just make a straight strategy and see from there.