How much world do you create?

Seeing how I was in middle school when I began creating games, I'm not at all surprised that my tendencies have changed.  I am becoming more and more aware of a habit of mine though:

Creating a world before I create the game.

I mean, not just going into the map editor, making huge continents, and then filling in with cities, dungeons, and what-not.  I mean taking a look at the world I want my story to happen  in and ask a lot of why questions.

Why is there a kingdom there? 
Why does that cave exist?
Why is the city by the river not closer to the ocean?

And then I get really down into it and ask:

When did this kingdom become one?
Who are the character's parents, grandparents, etc.
What is the history of this region/organization/religion.

I've found myself creating a whole history for an entire world, individual regions, kingdoms, and cities. 

I think it helps in game creation, but some may think it's too much fluff that the player may never need to know (or care about).

What do you think?

I create games the exact same way, only I start off with the characters first then move on to question kingdoms and rivers and gods and goddesses. There was even one time when I questioned why the sun pops out in this game-world
"Surely that world is not as close to a sun as earth is, why does night time seem to be just as long? Should it not be longer? Shouldn't the characters freeze at this temperature?"

I like it though, thinking like that made me a series. I would create some massive reason why something existed or who this person was that people would mention in folklore; it didn't take long for that to turn into multiple stories that were connected... and one day I plan to actually make them.

and one day I plan to actually make them.

Yeah, about the finishing the game thing 

Well i do the complete oposite! lol!
i create the world first, and then i think what
character would be the most fun to see that
world,  for example:
this game i was making was about a light god
that created shadows to make people pray for him(kinda like ff x)
so i thought what kind of character would be the best
and my answer was the brother of the man choosen to became the paladin of light
so the player would end up seeing all the sacrifice the "younger brother" would do to became it
so when the climax agains the shadow lord ends and ure the only survivor and see the light god
ready to change ur brother into the next shadow lord it would make an even bigger impact
(this is kinda spoilerish but whatever) from each dungeon the brother would sacrifice something important
first his eyes(and since ure his brother u would became his eyes and the leader of the group)
then his voice, then his ears and finally his own body
so imagine the same history as the paladin and now as the main character(by the way he is a cleric of light a mix between a mage a healer and a warrior)

so i think it is better to make the world first then make ur characters as u will have a better way fixing him there
sorry for the long post ^^

I usually make stuff up as I go. All of the backstory for individual locations I make up when I make them, its har for me to plan ahead because I usually will end up changing my mind for something cooler...I have a short attention span

Lol!
i used to be like u but i took  a manga to the knee
well not literally...
i like to think about the story i want to tell
then the world it would be best
and finally the heroes
but some time i invert the order
make a hero i really want then what world it would be most fun
then finally i make ship up
it all depend on my mood/depression...

I always make a map first (hand drawn) and then begin making cities, or mountains, rivers, and what not.  I try to think of histories as I go on, and how things can relate to each other.

See if I just get going without planning out my world I eventually get to the point where I go: "Oops! I need to make another continent here for...."  Yeah.  Planning it all out just helps me get further along.

It used to happen alot to me(especially when i found a script or event i fallen in love with)
cant remember how many times i redid my demo with difenrent things and cbs and cms and tools
and crapalot
it went to a degree that i ended up not making because of it
damn you creative minds of (put the name of every one in the forum)

I start with story first, then plan the areas that MUST be there, these get priority in forming the world. And just because they are part of some character's past does not mean it occurs in game. I only focus on places that will actually be visited for the story. Next come places that are secondary to the story, but needed for gameplay (zones to farm materials, sidequests, minigames, ect). If it's not needed for story or gameplay, don't bother much. Focus on places with a purpose, and any extra bits can be addressed in dialog and item descriptions to add some flavor.

As always..KISS, the best advice in any gamemaking endeavor.

I've only made one game so far. It was fairly short and I made it up on the fly.

For the second one I've actually done an entire script and am building my world based on that. Similar to what you mentioned above I've ensured that every location is there for a reason and that it fits in with the overall story and "universe". Hell I even put toilets in the village houses and ensure there's a correct house to villager ration Not fair to assume EVERYONE lives in their shops right?

Seeing how I was in middle school when I began creating games, I'm not at all surprised that my tendencies have changed.  I am becoming more and more aware of a habit of mine though:

Creating a world before I create the game.

I mean, not just going into the map editor, making huge continents, and then filling in with cities, dungeons, and what-not.  I mean taking a look at the world I want my story to happen  in and ask a lot of why questions.

Why is there a kingdom there? 
Why does that cave exist?
Why is the city by the river not closer to the ocean?

And then I get really down into it and ask:

When did this kingdom become one?
Who are the character's parents, grandparents, etc.
What is the history of this region/organization/religion.

I've found myself creating a whole history for an entire world, individual regions, kingdoms, and cities. 

I think it helps in game creation, but some may think it's too much fluff that the player may never need to know (or care about).

What do you think?

I think you're taking the classic 'novelist' approach. That's exactly what most fiction writers do when they start work in a new setting - especially if they're planning on writing more than one book using that setting.

Do you have to get that detailed for games? That depends on how "deep" a story you want to tell. If you want novel type character arcs and plots that actually progress and make sense with conflicts and climaxes and all that, then it certainly doesn't hurt to have all that information handy. Even if the player doesn't see it directly, it can be felt through the things the player does see and hear.

You have inspired me my friend! O_O although not till after my mini-game....

I shape my World and Characters around the Story.

Being a humongous Elder Scrolls fan, I've always been drawn to worldbuilding. But to make something with the same scope as that series is so daunting for me to take on alone that I usually end up losing steam, and never finishing the game (if I'd even started work on it at all). This is a big part of why I don't make games anymore; it's impossible for me to make a humble attempt at creating one because i get so caught up in making the most grandiose, immersive title ever. I might come back to it someday. But I'd need a team of people, and I'd likely leave the hard-coding to someone else.

Usually, I try to avoid giving the player a huge, sprawling world to explore, and instead focus on visiting distinct locations. I kind of enjoy games that use similar setups because it gives me the sense that there is more out there, and it leaves me wondering if I'll ever get to visit them.

I like this topic.

I don't like how in most rpg games, it is usually very short to travel from town to town. Only certain dungeons are ridiculously long, and only because the developer cranked up the random encounters. 

So therefore in my game, the distance between places will be longer. I want people to feel as if they are exploring a world, going on an adventure, with things to do other than just defeat monsters. (but that will be a big part of it of course.)

I also don't like how in games, you either forgot your main objective (next part in the quest) or it's knocked over the head that you should go here. I like more, okay you know you're supposed to go north to rescue the princess (or whatever) but you COULD also go to the town west...

@OP
sounds like good creation to me, I normally build on the spot, what I need to make the story seem natural, every map I make is inspired by things like what style music I want in the map, what the atmosphere (weather, feel, etc.) is, things like that. So I can relate when it comes to thinking it through lol.

My only thought to add right now is considering making several games that happen in the same world. It would give history and perspective and would save the creator time when making a world for his game to happen in.

I think that all that history and lore definitely helps create a more fleshed out game. You just have to remember that not everything has to make sense, like why a cave is somewhere doesn't have to be thought about because it's not something a player will probably think about all that much and if you put something like spiders in it then they're going to be thinking "oh shit spiders fuck fuck fuck" and not "why are the spiders in this cave, it doesn't make any sense, this cave shouldn't even be here because the alignment of the planets and the moon just doesn't work and-" you get the idea, unless it's a cave near a volcano filled with frost spiders and there's lava everywhere but that's just lazy design unless you have an NPC or something introduce it or there's a bit of lore behind frost spiders that involves them taking cover in fire caves during they're spawning season or something. This is rather nonsensical but I like game design that has some thought behind it but too much thought can result in a world that's very large and sort of puts you off creating it because it seems like such a mission. And then you come up with another idea and well, I'm sure you can see where I'm going with that. . .

Yup.

I'm more bothered by games that do no explaining whatsoever. You can get overloaded for sure, but to have no backstory gets you in a "this town was created so that a warrior, Mage, thief, and obligatory fourth member could sleep here and buy better weapons" trap.

Not cool.

I'm a little rusty on current games, but star ocean 2 and 3 did a good job giving you history if you went looking for it. Otherwise you just played the game.

I'm in the process of making my first game (still), but lucky for me my game is based off a test map I made for a previously planned project, so most of the (small) world in my game is pretty random. I placed characters in the world without thinking why or who they are. I create a story and unique personality for each of them as I continue working on the game, even if the story doesn't make any sense whatsoever, but that's kind of the point of my game anyway. To be a funny, light-hearted dialogue-based game.
Buildings are also added as I go. The first building on my map was a prison, the second was the house my character lives in. From there on I added buildings to the town whenever I needed them. Like one of the quests you need to go to a farm. I didn't have a farm yet, so I created one.