How do you work when you create a game?

[Blue]

I'm in the early stages of creating a new adventure game, but this time I really want to make a real effort of it. With my first game (Fading Shades), I created a bunch of rooms and made up a story later. Now, I've started with the story, and I have a few characters and puzzles ready that I think will fit in nicely. My problem is, that I'm a lousy storyteller, and I'm probably making the story more complicated than it needs to be. At the same time I want everything you do to make sense.

So, I'm seeking inspiration from you guys: How do you work when you create a game?
Do you just make something up along the way, or do you have everything planned before you begin to create the rooms? I know you can't have everything planned in detail. There should be room for spontanius ideas for puzzles and rooms if they fit in.

This is how I'm trying to do it this time:
1) Story/Characters
2) Rooms and puzzles
3) Drawing the characters
4) Prototype graphics for rooms and inventory items
5) Coding the game
6) Redrawing the prototype graphics and inventory items

Where do you begin? And how do you work? And what inspires you to create the story/characters and puzzles?

[Igor Hardy]

It depends on if I'm creating it on my own or not.

If I work alone...

1) I just choose a game idea from my vast archives (I keep all the good ones written down) and start developing it in my head in terms of story and characters. While I'm not coming up with proper puzzles at this stage, I take note of potential "puzzle-like situations" in the story or reshape the story's structure to create more of them.

2) The story is far from being finished, but I start building/prototyping the game with ugly temp graphics. The most important goal at this point is coming up with good puzzles and other playable elements. The results (what works and what doesn't) allow me to make final decisions regarding interface, puzzles and graphics and some storyline related ones too.

3) I gradually keep introducing finished graphics (which ones and when depends on my whimsy). I do it mainly as a distraction from the other work and to finally get rid of the temp visuals.

4) polishing - creating all kinds of little neat details for the players pleasure

Overall, stage 2) is the one I come back to, again and again, and what I invest the most time in.

what inspires you to create the story/characters and puzzles?

Pretty much everything. The best bits are the result of long prototyping periods.

[Diath]

I have only finished one game but Ill tell you how I did my process.

I sat down and came up with a story(which was generally while I was bored out of my mind at work). I fleshed most of it out and whenever I reached a point that I thought could be a puzzle situation, I would put a question mark around those areas and would get back to them later.

Once I finished all of the basics of my story and most of the puzzle elements of my game, I started with what I thought would be my biggest downfall of me not finishing my game, and that was my background art. I knew that once I did most of my art that I would have enough motivation to continue the rest of the game making process in which I find most fun.

With the visuals came importing them into AGS and starting the scripting and dialoging, bringing the game some breath. This I thought was the most fun. Most of the script issues and challenges I went through many people before me had and thankfully were archived on these forums. During this phase I also figured out the final versions of puzzles and most of the art.

Once I managed to get the game playable from beginning to end I imported music, added anything else that I thought would make my game a little nicer looking (custom display text, custom icons, messages ect ect).

Next and nearly last phase I left my game alone. May sound weird but I figured getting my mind off of it may help me identify problems and spelling mistakes later on when I came back.


Well that was basically how I made my Game, I hope this helped and didnt seem like crazy ramblings. I also found it immensely helpful if you listen to something that gets you in the mindset of your game, which will obviously depend on what genre of game you are making.

[Dataflashsabot]

1. Have idea
2. Think about how cool the idea is
3. Draw a background
4. Open AGS
5. Remember that making a game takes work
6. Remember how lazy I am
7. Remember how bad I am at concentrating and just getting stuff done
8. Think 'maybe some other day'
9. Close AGS
10. Go do something else

[InCreator]

1) Having urge to make a game
2) Opening AGS or GM to take a quick look at unfinished projects
3) Being amazed how cool they are and how much work is put into them
4) Making a promise to finish them someday
5) Realizing that it's a lie
6) Feeling frustrated
7) Closing AGS or GM and retreating

[Dualnames]

Okay, first I have to say that if anyone is searching for a magic way, as far as I know only Vince has one, and that's because he's the Wizard of Oz. However this a great topic to share and copy ways of game process. Even of giving-up-certain-game process.

1 ) Watch something (film/movie/anime/porn) or find something on wiki, and excited about writing a good plot.
2 ) Releasing the plot isn't at good, and therefore giving up on it.
3 ) Randomly drawing stuff at photoshop for days without any purpose.
4 ) Suddenly I achieve a new drawing style out of nowhere
5 ) Retrieve that plot and put it on those backgrounds
6 ) Listen to some music in case I lose inspiration in the meantime
7 ) Plan scenes.
8 ) Posting about all that at the AGS forums

[LRH]

The basic timeline of one of my games being made:


1.) Make lots of coffee.
2.) Story is created.
3.) Timeline of events in the game is created.
4.) I start...'drawing' all required graphics. This part is generally the worst, as I'm no artist.
5.)I start putting the game together.
6.) (And this needs to stop, probably my greatest flaw.) I realize something I put into the story isn't really going to be put in to the game due to my inability to draw, or the idea being generally too complicated. I change many things to suit the new roles.
7.) Add music and sounds.
8.) Upload to my site.

That's about it, really.

[tzachs]

I've been refining my method via trial and error.
This is what I've got at the moment:

1. Get inspiration from MAGS to do a game.
2. Search your brains for an idea that you can do without having to draw walkcycles 
3. Create the story with the puzzles
4. Design (not draw) each room's composition, what are the main things in the room. Write it down.
5. Design all inventory items, and where are you going to get it from. Write it down.
6. Write down an entire walkthrough for the game. This is the place when things usually change for me.... here you can find logic flaws, puzzles that cannot be solved in this place in time, inventory items that can be used to solve a puzzle but were not intended to, etc.
7. Reiterate 3-6 until it all falls into place.
8. Only when you know exactly what you want to do, draw the backgrounds&characters (Or if you're working with an artist, send him your designs and work with replacements, same with composers).
9. Only when assets are all drawn, open up AGS and click on the "Make my game" button (3.1 and above).

[markbilly]

1) Have a story idea.
2) Think of some characters and locations.
3) Start drawing and scripting.
4) Add some puzzles as I go, expanding and changing the locations and story in order to do so.
5) Revise a load of stuff.
6) Slowly get everything done, which takes ages because I've expanded as I went.
7) The game is now playable from beginning to end.
8) Fill in all the gaps I left as I went and update sounds, etc.
9) Decide that was a dreadful way of making a game and make a note to myself to that effect.
10) Post about this completely undesirable method on a forum.

I 100% advise you do not use this method. It is pretty much how I do things, though.

[Wonkyth]

Haven't completed any games yet, but my flow of work for making not much goes:
1) Come up with idea for cool game mechanic/beginning/ending.
2) Do stuff.

That's about it. 

[Blue]

It's cool to see so many replies! Although I'd like a bit more detail on the process.
Perhaps, based on your latest game, you will explain how you got inspiration for the storyline and the characters in it?

[Jimbob]

I just read this about Scary Go Round and writing a comic is pretty much like an adventure isn't it? It's got plot, dialogue and lots of artwork? Only things its missing is puzzles and animation, but it's still a good read, and besides, Scary Go Round and Bad Machinery are awesome.

Planning a story

I'm liking the 'act' way of organising a story, it makes tying the plot up a lot easier and into manageable chunks. You can concentrate on your first act knowing that later on you know how it will be resolved, but what happens in the first can be quite self-contained. It can definitely be applied to adventures... and it's definitely how I'm going to work in the future.

Previously though (in the non-adventure games I've made as well), my games have mainly been built from a central theme/idea/prototype, then content blitzed through to produce something that has a start and an end state (I think this is VERY important), then just tweaking/remaking parts until I'm happy or I'm bored with it. That's very much how sync::routine was made as well, the mini flashback scene (if you saw it) was added in last .

And just to reiterate the end state point, if you are playing your game and you don't have a clear end state, it's very easy to get disheartened because you don't really know where a particular section needs to go. I guess this is another reason to go for the multiple act way of writing, because it provides these mini-ends to give you direction. Even if the mini end state is just 'He gets through to the next room', you know that all your efforts need to be aligned with that.

Anyhow, I've rambled enough already about that... (EDIT: and had some amazing typing errors... FIXED)

[Wonkyth]

It's cool to see so many replies! Although I'd like a bit more detail on the process.

Heh, I would give more detail, but quite honestly I have two or three ideas a day, and rarely get much farther than designing the parts of the game absolutely required for the mechanic.  :

[Wyz]

Well, my method is not the best, I always start by thinking I'll do it in a more efficient way but end up with this method anyway.

1) Sit back and start working on a concept.
2) Search for inspiration for the concept. (Google images mostly)
3) Write what I call a teaser, a short story that describes the beginning of the game from first person perspective. So it form the setting of the game.
4) Write the actual story leaving part open for puzzles.
5) Write script for cutscenes, dialogs and some other parts of the game that can be scripted. I use actual screenplay notation for this.
6) Design the puzzles, along with this I create a 'todo' for all sounds, cinematics, animations, backgrounds, sprites and music that I need. In this process I also make sketches of backgrounds and mark visually where things happen.
7) Drawing backgrounds. Adding walks, behinds and hotspots.
8} Drawing characters.
9) Animating characters.
10) Drawing props and inventory items.
11) Get everything so far in the game, also do some rudimentary scripting like GUI and menus. Also the interactions, witty comments etc. (I write these on the spot).
12) Designing sound effects. (I rip a lot from YouTube nowadays )
13) Making music. (Usually I already work on and off on some melody lines and chords as soon as I've written the teaser)
14) Script the rest of the game.
15) Test the game
16) Create intros and outros and cutscenes
17) Finish the game up (Menus, logos, credits etc.)

Some of the issues are:
- The first two steps really hold me back, it takes me days without really creating something.

- The teaser takes time but it is never ever end up in the game. Same for a lot of the scripts.

- Writing the story before there are any puzzles also makes designing puzzles harder, but I just suck at designing puzzles. They always feel like a compromise, laid on top of the story. It might be better to design the puzzles before the story, or at the same time.

- I spend days on writing and drawing backgrounds etc. before ever creating an ags project file, mainly because I hate spending time on mockups, but it has its downsides. Getting really stressed when the deadline is nearing, or not having a clue how the game will take shape and getting confused.

- Testing the game before adding the intors and outros will usually mean there are bugs in the beginning and ending of the game. I do it in this order because if I run out of time, I can leave them out.

[Blue]

I'm liking the 'act' way of organising a story, it makes tying the plot up a lot easier and into manageable chunks. You can concentrate on your first act knowing that later on you know how it will be resolved, but what happens in the first can be quite self-contained. It can definitely be applied to adventures... and it's definitely how I'm going to work in the future.

Interesting. I never thought of it that way. But you're right that there is a similarity between comics and adventure games. I will keep this in mind.

- Writing the story before there are any puzzles also makes designing puzzles harder, but I just suck at designing puzzles. They always feel like a compromise, laid on top of the story. It might be better to design the puzzles before the story, or at the same time.

Puzzles should be where the key elements of the story is, so that when you solve it you feel you've earned to progress in the story. But other than that puzzles can be placed everywhere along the line as long as it fits in with the story and has a purpose.

I already have one of the key puzzles ready for my game, but I don't know yet how big a role it is supposed to play for the game in whole or what happens when the puzzle is solved.
I have two characters ready for the game, but at this time I don't even know who the main char is.

[Wonkyth]

Making puzzles is like writing an essay, you just have to know what you're allowed to use, and what something should be used for.

[Jimbob]

Puzzles are definitely key elements, but the best games are where they come hand in hand with the story. Like I said, maybe the plot involves having to get access to a place, and the puzzle can very easy be arranged to that goal: bribing the doorman, breaking the lock, hacking the password, making a ladder to the open window etc...

I think that if you are making puzzles first, and bolting story and plot on later, it is not an adventure, it is a puzzle game. Like Professor Layton. If you want to make an adventure the two need to come hand in hand. Whilst I do enjoy solving puzzles of any kind (and I loved Layton) it always irks me if the puzzle in hand bears little or no normal relation to the world that you have been interacting with. If you are making a sliding tile puzzle, make it like the one in Machinarium, tied to making an electrical current go from one side to the other, don't put it in merely as a means of delaying the player to carry on the story, because that's just dull, frustrating and bad game design. Layton worked because it didn't pretend that it was anything but a 'solving riddles' game, and the bits of story it fed you for finishing them were satisfying.

Anyhow, I should probably write an article or something... I've barely scratched the surface here... but I should really get back to work...

[ThreeOhFour]

I change the way I work on games each time I start a new one.

It's fun and educational to try new things.

[cat]

C'mon Ben, you are the person in this thread with most finished games, give us some more information! 

(Btw, thanks for your AGS tips on your blog, I couldn't post there because the captcha didn't work)

[Dave Gilbert]

I'm with Ben!  Every time I finish a project I think I worked on it very stupidly and inefficiently, but I'll TOTALLY get it right next time!

Usually I start with a design - compile everything from start to finish into a design document.   It's much easier to break the production down into manageable tasks if you actually know what those tasks are going to be.  Of course, things are bound to change as you move forward, but it's a place to start.  It's much less overwhelming to work knowing your task for the day is "implement character switching function" instead of "MAKE THE GAME!"