Game structure?

[soulcypher]

How to start game development so that in the middle of production you don't get stuck or you need to change majority of the game?

For example, if you script the room #2 and room #5, you figure it out that it's not cohesive with the next one or
it's not placed well with the story or some part of it.
Bascially, you would need to go over and redone the rooms.

So, how to structure the game before so you don't run into this problems.
In what order do you plan your game/puzzles when you flesh out the story?

[SilverSpook]

One option is to write the entire story beforehand.  Then just build the rooms that you need, like a director would scenes from a screenplay.

A lot of times (for me at least) I have an idea for the direction of the story, but things do change during development.  Also, things that sound good on paper don't work so well on screen or in AGS, a fairly limited game engine.  So just have a rough idea of where you want the story to go, and then start building the scenes.  I usually make up puzzles depending on when they make sense in the story. 

Simplest would be, there's a door in this room.  Maybe it's locked?  Player has to figure out how to open it.  But don't do that if it doesn't make sense, like if it's the person's own house.

You can always change the order of rooms, so like, let's say you wanted the player to start at their house (room 1) then go to the bank (room 2) and then go to the shopping mall (room 3) and you built them in that order.   Then you change your mind and want the player to go to the shopping mall before the bank.

In script you'd just do:

player.ChangeRoom(3);

at the end of room 1, and then

player.ChangeRoom(2);

at the end of room 3.  You can have the player move to any room at any time, they don't have to go to rooms in order.

[soulcypher]

Thank you.
Do you plan puzzles ahead or do you invent them on the fly?

[CaptainD]

Do you plan puzzles ahead or do you invent them on the fly?

Personally I use a combination of both.  Planning ahead is generally better though.

Flow charts are really helpful for keeping track of the game flow.  (Not that I'm much good at using them myself, but they are undoubtedly very useful!)

[Danvzare]

Thank you.
Do you plan puzzles ahead or do you invent them on the fly?

It's generally a good idea to plan puzzles ahead. If you spontaneously think of one, which you think would work well with the game, then by all means, implement it.
But you want to try and plan your puzzles. That way you can scale the puzzles, see which how many puzzles you want, how many steps should be in each puzzle, and want items and objects you need to draw and implement in your rooms.

Usually if you try making up puzzles on the fly, you end up with illogical shoe-horned crap. Not always though, so don't be afraid to make things up on the fly when you think it's appropriate.

And to avoid having to redo the entire game. First try implementing everything in placeholder graphics with very little dialog. That way you can see what works and what doesn't, and change as you see fit, without becoming attached to anything due to how much time you've spent on it.