My entry for Ludum Dare #24, with the theme of Evolution, is the game Conway's Revenge.
It is essentially Conway's Game of Life (http://en.wikipedia.org/wiki/Conway%27s_Game_of_Life) modified to work with multiple opposing teams of cellular automata. The rules of reproduction have been tweaked to make the game more interesting and 'game-like' by preventing deadlock scenarios and making sure it's not too hard to get your cells to grow and expand.
It has three game modes planned, two of which are implemented:
1. Campaign - A set of puzzle levels where you face off against progressively more difficult enemy formations (unimplemented). 2. Battle Royale - A bloody melee between four different teams of cellular automata. The grid is initialized with a random pattern, perfect for conflict! 3. Sandbox - A blank slate for you to explore and test your own creations with up to four different teams.
Watching the simulation update entranced me. Did seem more stable than Conway's. I'm not sure which color I was supposed to get to win or what input to the dynamic system would bring that about.
Neat variation on the original. I think you should have done more to make it a game rather than just a simulation though. Something where you only have a certain number of cells to place in order to try to 'defeat' enemy cellular arrangements.
Thanks for the comments. I agree it's more of a toy than a game at the moment, I ran out of time before I could implement the 'Campaign' mode that would actually make it into some kind of game.
Fun to play with. Would be cool to know the rules and how you managed to come up with them. My LD entry is somewhat based on cellular automata and it was really hard to come up with deterministic rules, so I eventually gave up and randomized.
A cool idea, especially with the sandbox, but the battle royale option doesn't work out so well. In the end, I'm just watching what essentially looks like a bunch of pixelated jelly wobbling around until one color takes over the others. That could be remedied by maybe toning down the starting density, and putting families of different colors in their own little areas to start - you know, a dab of blue here, a pile of green there. That way there's a better chance for something less chaotic and more entertaining (to watch at least) to emerge.
As for now, I'm gonna say, while this isn't terrible by any stretch of the imagination, it could have been done a bit better.
I came up with the rules mostly through a longish cycle of testing and thinking.
I started out with the basic rules from Conway's Game of Life, then iterated from there, seeing what worked (met my specific goals) and what didn't.
Here's a summary of the rules. A cell's neighbours are the 8 cells that surround it (i.e. diagonals count too). This is different from the original Game of Life.
The rules are roughly: - Cells who have at most 0 or 1 neighbours of the same colour die. - Cells who have between 2 and 5 inclusive total neighbours (no matter the colour) live. The strongest player gets the cell. - Cells who have more than 5 total neighbours and don't have at least 5 neighbours of the same colour live. This rule specifically targets border conflicts to make them more interesting and resolve faster. - Any other cells die.
Ties are resolved by giving the cell to the player with the most cells on the board. I initially didn't like making global board state a part of the decision-making, but it eliminates deadlock scenarios (which were all too common) quite nicely. It's also intuitive - if green has 1/4 of the board and red has 3/4, it doesn't make sense for green to be able to resist indefinitely.
Nice job. I used the game of life for my inspiration too. I like that you have a sandbox mode where you can experiment with the game rules outside of battle mode.
Wow this is so weirdly addictive! At first I thought it was a bit plain, but I ended up playing this for a few minutes! It was fascinating looking for particular patterns that could last!
Some spins on the game of life. Presentation is clear and the additional mode(s) is quite interesting. Is it a game though, probably not, but still well done.
Interesting to play with for a few minutes, but it's missing something to make it fun. Perhaps another CA ruleset might be more interesting, but as you noted, multiple colors makes it a little harder to balance. Giving each color a different rule might be fun. Not sure how you turn this into a game? Slow it down and let the player introduce patterns? Anyway, I think it's a great starting point for CA experimentation.
Comments29
cadin: your description is actually exactly what I had planned for campaign mode! :O Proof: http://nickknowlson.com/blog/2012/08/26/ludum-dare-24-finished/
As for now, I'm gonna say, while this isn't terrible by any stretch of the imagination, it could have been done a bit better.
I started out with the basic rules from Conway's Game of Life, then iterated from there, seeing what worked (met my specific goals) and what didn't.
Here's a summary of the rules. A cell's neighbours are the 8 cells that surround it (i.e. diagonals count too). This is different from the original Game of Life.
The rules are roughly:
- Cells who have at most 0 or 1 neighbours of the same colour die.
- Cells who have between 2 and 5 inclusive total neighbours (no matter the colour) live. The strongest player gets the cell.
- Cells who have more than 5 total neighbours and don't have at least 5 neighbours of the same colour live. This rule specifically targets border conflicts to make them more interesting and resolve faster.
- Any other cells die.
Ties are resolved by giving the cell to the player with the most cells on the board. I initially didn't like making global board state a part of the decision-making, but it eliminates deadlock scenarios (which were all too common) quite nicely. It's also intuitive - if green has 1/4 of the board and red has 3/4, it doesn't make sense for green to be able to resist indefinitely.
I liked it
Neat idea. The only bad point is the lack of real gameplay (at this time!)
Good job!
Is it a game though, probably not, but still well done.