Test

For the next test i've created a world with both horizontal and vertical movement.

For now I've only used clouds and stuff like that. I will add land, sea and other things later, but first I want to be able to get images from a external folder so I can add images external and i don't have to put them into the source itself. This makes it easier later on to make the graphics nicer and to put in more graphics. It also works better for making categories of worldtypes (land, sea, air, etc).

test

I've added seaworld to the last test.
When you approach a shore the computer knows that the following sprites will have to be sea sprites (and the other way around).

Test

For the next test i've created a wolrd which is build up random. If you go to the right you will always see a random piece of land, if you go back (or left) the land has a totally different configuration. So you will always have a unique travel.

Test 05

Test 04

Did a little spacing...

Test 4

Interesting?

some guys have started the lucidgaming group and claim to have found a way how you can play games in your dreams. They are currently beta testing tic-tac-toe...

http://www.dirtymechanism.com/lgg/aboutlgg.htm

Idea for worldbuild and user generated content

Ok, this could get a bit vague, but I will try to explain it as clear as possible:

Keep the grid system in mind. Untill now I have gone through a series of thought on how the world is going to be build:

1. First I thought that every gridsquare would be random until you approach/enter it. This meant that every grid was different. There was no cohesion.

2. Then I thought that every gridsquare should be more static when you gave it more attention, but still there was no cohesion. (see test 02).

3. Then I decided there should be intermediate forms between two (or more) random gridsquares. (see test 03)

But if you think from the perspective of 'gridsquare A' then the surounding random gridsquares (B, C and D) should only be random within a range of things that in somehow relate to 'gridsquare A'.

This could be called cognition, because the computer would only link things that relate to eachother. This is also what happens when you dream. In this way very strange things can be rendered. Example: in you dream your dad is also your teacher, because they are both a man.

So, for this you would need a giant database in wich everything has information of things that relate to itself. You could think of tagbased databases like flickr.com.
Some good friends of mine are currently working on a database called live.on.line. This system is a user generated symantic web of information.
In this system the user can travel through the information, generate new links and vote if they are correct or not. The system can also calculate which things are probably connected to eachother.
This 'thinking' can be seen as dreaming and the user response (the voting) as a dubble check to see if a connection is right.

This web of information can be traveld in billions of ways. It is multilinear, but the travel itself is always linear and from the perspective of the user...

-to be continued-

Test 03

I've made a little test based on the last test. In this test I tried to make a more consistent random world. The outer 4 squares are a random color, the rest is an average between these colors.

I you give attention to the 4 outer squares they will become more static.





Sadly enough I had to program each square appart. The next test will be totally generic I hope, so I can build a bigger world, or a world that expands when you move to the edge.

Rogue

Rogue: a key and lock game that creates random dungeons with random enemies. The player doesn't have any inluence on the game world. He/she can only explore.

Old games suck, but actually they rule.

Mojoworld

mmm, mojoworld's procedural generated landscapes are nice, but not very helpfull. One thing is certain: These types of procedural landscapes are too complicated for a realtime rendering... So I wonder what the current limits are.

Key and Lock Puzzles in procedural gameplay.

While researching through the references of the Case Study on Charbitat I found this paper. It is great. it describes how key and lock puzzles can be played in a procedural game world.

" Abstract:
This project is an attempt to model and simulate key and lock puzzles in the style of Shigeru Miyamoto games, specifically the Zelda and Metroid series. Much of the gameplay in these titles is oriented around encountering obstacles, and then locating an item that allows the player to circumvent the obstacle. The project is built on top of Charbitat, another project built on Unreal Tournament 2004, which produces procedurally generated space in response to the player's actions. " (source)