Hi everyone,

I have a game idea for a world simulation game but I am not sure if I should go ahead and develop it. I am worried about how complicated it will be but I feel like I should forge ahead.

The player develop a city in a large game world. The city will be inhabited by people who will age and die. In order to expand their city they need to develop their infrastructure (food, power, water etc) to support their expansion. The people can go to school, get married, get jobs, start their own companies etc basically just lead ordinary lives. Players can also participate in world events like sports tournaments.

I will get someone else to do the artwork but I will be doing the programming.

What I don't know is how detailed to make this game. I don't want it to be way to detailed so that it will never be finished but I don't want it too simple.

Is it a good idea to go ahead. I really need some advice.

Thanks.

Just make sure you programme the game in such a way that it will be easy to add missing content later on, then make concentrate on making it work.

About the level of detail, i would say you were gonna give each citizen it's own stats + relations (to things, aka ownership, and to other people, aka friends family and partner)

Maybe some kind of favor-system if they help each other(give/trade things, creating friends) which doesn't need to be a requirement in any part of the game(aka gives room to add more content later)

Some kind of system of player-controlled/build buildings.

and some kind of system for events.

Just make sure you programme the game in such a way that it will be easy to add missing content later on, then make concentrate on making it work.

I think this is spot on. Just like any business plan, you need to start small. Use a mindmap to dot down everything you'd want to see in your game, and then put that mindmap aside. Start with some basic functionality and then build onto that later with more and more features.

I moved this to the Lounge because you were not asking a Game Design question. Whether or not you should develop your game depends on your reasons for and against doing it, what you would do with the game once developed, what you could do instead of developing it, and other factors. You should make a decision grid. http://sloperama.com/advice/m70.htm

I've been working on a similar game, and while it's been fun - I'm also struggling with the detail level. If you try and start out with some very basic lives scetched out, and some sensible parameters of life, you actually get something up and running pretty quickly (provided you have thought a minute or two about the size of game world and population to simulate in relation to your average PC memory!)

For one thing, I find that no matter how more complex you make the life simulations, they still end up either boring and mondane (unrealistically so), or simply totally unbelievable.

You need both time-relative statistics of lots of complete areas of a real world, and a proper AI (which is in skunk labs, but far from those of science fiction and not suited for our needs) to make a simulated life really realistic. But you could simulate parts of lives that are running in a "controlled simulation", where the user would choose individual persons or familys to open and control (like in Sims).

If leave you with some old pseudo-code from an earlier venture, before it got very complex and messy. It illustrates the game loop for a world simulation, where people live basic lives, get born, married and even go to war and die...

void world_sim() {
    for (int year = 1750; year <= 2100; year++) {
        for (int day = 0; day < 365; day++) {
            // people
            for (int p; p < people.count; p++) {
                if (!p.is_in_coma()) p.wake_up(); // fix: prevent people in coma from waking up every jan 1st
                if (day == 0 && p.age == 0) {      
                    p.get_born();
                }
                // here you can plug in your behavior
                p.have_breakfast();
                if (p.has_job() || p.age() > SCHOOL_AGE) {
                    p.goto_job();
                    if (p.is_lazy() && p.fucks_up()) 
                        if (random < 0.1) { 
                            if (p.is_child()) {
                                p.get_teacher().give_detention();
                            } else {
                                p.get_boss().warning_or_fire(p);
                            }
                        }
                    p.go_home();
                }
                // plug in behavior
                p.do_random_stuff();
                // plug in behavior
                p.have_dinner();
                if (p.is_married()) {
                    p.argue_with_partner();
                
                }
                if (p.get_partners().count() > 0) {
                    p.have_sex();
                    if (p.is_having_unprotected_sex()) {
                        p.make_baby_with(p.current_partner());
                    }
                }
                // plug in behavior
                p.sleep();
                if (rand() * p.age > 100) {
                    p.die();
                }
            
            // plug in behavior
            } // days
            
        // world events
        if (year % 100) major_catastrophic_event()
        if (year % 50) industrial_revolution();
        if (year % 25) major_war();
        if (year % 4) summer_games();
        // plug in behavior

        // for expansion
        // if (year && 20000) ice_age();
        // if (year && 500) new_religion();
    } // years
    
    end_of_world();
}

Sounds like a fun project.

Maybe start out programming a SimCity 1 clone, and than move on to a Dwarf Fortress style game, and eventually turn it into a Tropico 1 type of game.

Efficient path finding would be the biggest challenge in something like this, since it eats up a lot of the processor's time when you have many "people" in your game.

I think it also depends on your skill level if it is a good project to tackle now.

Guess you could write a little about where you stand.

Also what are your goals? Do you want the game to make money at some point?

If you do what you have written you will have a world ... but will it be a game?

I'd create a design document that gives an idea of what playing the game would feel like.

How does the player interact with the world exactly? I wouldn't start until I have more of a plan in place.