 |
Author |
Topic: Pollen, alife (Read 1449 times) |
|
Mike Davis
|
Pollen, alife
« on: Jan 9th, 2003, 9:36am » |
|
Pollen
http://www.lightcycle.org/workspace/pollen/index.php
Five cellular experiments. Some take a little while to display their behavior. Click to reset.
The examples all use essentially the same code, which can serve as a platform for simple alife experiments. Source for each variation is linked.
Each lit pixel is a cell, which takes actions depending on it's surroundings. Cells run one at a time, in random order. The coordinate space wraps around the screen.
Some (1, 2, 3) have one type of cell, others (4, 5) use two types of cell. Some (1, 3, 5) use the wraparound coordinate space, while others (2, 4) draw a line for a floor resulting in a cylindrical space. There is a lot of room for variation and extension.
The code is commented heavily, but feel free to ask me anything.
|
| « Last Edit: Jan 9th, 2003, 11:22am by Mike Davis » |
|
|
|
|
REAS
|
Re: Pollen, alife
« Reply #1 on: Jan 10th, 2003, 12:13am » |
|
Good work Mr. Davis. This is exciting. I'm looking forward to making some explorations with your architecture.
|
|
|
|
|
mrb
|
Re: Pollen, alife
« Reply #2 on: Jan 11th, 2003, 11:15pm » |
|
I agree -- excellent work. It is really nice to add in a bit of interactivity -- I chose to simply draw lines with the mouse to channel the cells -- a lot like Zach Lieberman's excellent "Pouring Buckets" applet on http://www.theremediproject.com/ -- check it out if you haven't seen it. The interesting aspect here, of course, is that the pixels are programmed with behaviors more complex than in the instance of Zach's piece, if I am not mistaken. Because of the large amount of things happening at once in the original code, line drawing can be kind of crude with just the simple code that i added in, which just draws lines from the pmouseX,pmouseY to mouseX,mouseY. Maybe implementing vector drawing would be more efficient? The interactivity allows for more programmed behaviors -- cells can be taught to "avoid" the lines you draw if you tell them to change directions if there is the stroke color of the line within 10 pixels in any direction of their movement. This leads to some visually interesting behaviors.
mike
|
|
|
|
|
Mike Davis
|
Re: Pollen, alife
« Reply #3 on: Jan 14th, 2003, 10:22am » |
|
Thanks guys, I appreciate the feedback. I look forward to seeing what you come up with.
Here's my first attempt at interactivity: http://www.lightcycle.org/workspace/pollen12/
The position of the mouse changes the parameters to some of the rules, leading to a reactive system. I think I'll try the interactive drawing you suggest mrb; it'd be more intuitive and fun for users to affect the system by adding new boundries and obsticles, especially with cells that climb.
Currently I'm looking at hive building by letting cells 'pick up' an adjacent dirt pixel and deposit it when encountering a certain situation. These rule sets are partially coming from the Santa Fe Institute's Swarm Intellegence. So far, they just block off the dirt source from unladen cells and construction stops...
I intend to try something with evolving rule sets and cooperation, but Winter break just ran out.
Thanks again. Feel free to call me Mike; I'm still an undergrad. 
|
| « Last Edit: Jan 14th, 2003, 10:28am by Mike Davis » |
|
|
|
|
Glen Murphy
|
Re: Pollen, alife
« Reply #4 on: Jan 15th, 2003, 4:46am » |
|
This is ace.
I can imagine it working with something like the mine-control system. Or as a laser light show.
*thinks* .. cor. If I was out at a club and saw the lasers doing that, I would positively split in two.
|
| « Last Edit: Jan 15th, 2003, 4:46am by Glen Murphy » |
|
|
|
|
|
