Author Topic: Tie Movement  (Read 5317 times)

Offline intron

  • Bot Neophyte
  • *
  • Posts: 7
    • View Profile
Tie Movement
« on: March 14, 2010, 10:01:46 AM »
Did you ever heard of "Dawin Pond"?
It's a evosim simulating small bacteria like things with a variable number of subdivided tails. after some time, real gliding behaviour evolves.
i think it doesn't seems too complicated to implement this in DB. By setting the volutary movement cost to a high level, cooperation between bots would be crucial for survival.

Try out "Darwin Pond" - this kind of estetic behaviour in DB would be great!

[a href=\'index.php?showtopic=0\']http://www.ventrella.com/Darwin/darwin.html[/a]


Offline Houshalter

  • Bot Destroyer
  • ***
  • Posts: 312
    • View Profile
Tie Movement
« Reply #1 on: March 14, 2010, 10:51:36 AM »
Sorry, it gave my this:
Quote
The error returned was:
Sorry, the link that brought you to this page seems to be out of date or broken.
« Last Edit: March 14, 2010, 10:52:08 AM by Houshalter »

Offline intron

  • Bot Neophyte
  • *
  • Posts: 7
    • View Profile
Tie Movement
« Reply #2 on: March 14, 2010, 12:10:54 PM »
hmm.. sorry.
don't know, why it won't work.
just copy-paste it.
it's worth it!

Offline Houshalter

  • Bot Destroyer
  • ***
  • Posts: 312
    • View Profile
Tie Movement
« Reply #3 on: March 14, 2010, 12:46:39 PM »
Quote from: intron
Did you ever heard of "Dawin Pond"?
It's a evosim simulating small bacteria like things with a variable number of subdivided tails. after some time, real gliding behaviour evolves.
i think it doesn't seems too complicated to implement this in DB. By setting the volutary movement cost to a high level, cooperation between bots would be crucial for survival.

Try out "Darwin Pond" - this kind of estetic behaviour in DB would be great!

[a href=\'index.php?showtopic=0\']http://www.ventrella.com/Darwin/darwin.html[/a]
Testing, testing, this link SHOULD work

Offline Ammeh

  • Queen of the Internets
  • Bot Destroyer
  • ***
  • Posts: 169
    • View Profile
Tie Movement
« Reply #4 on: March 14, 2010, 01:38:55 PM »
I think this could be adapted somewhat. Get rid of .up etc altogether, and make movement dependant on *doing something*.

It will open the door for so many more movement patterns and new behaviours, and become more true-to-life.

For example, firing a shot would send you in the opposite direction (newton's laws), moving a tie would drag through the "water", propelling you forward. Perhaps there could even be a new variable for growing little hairs to manipulate.

This shouldn't be too difficult code, although it would probably require some modifications of DB's physics.

Offline ashton15

  • Bot Builder
  • **
  • Posts: 99
    • View Profile
Tie Movement
« Reply #5 on: March 14, 2010, 01:59:19 PM »
Quote from: Sammeh
I think this could be adapted somewhat. Get rid of .up etc altogether, and make movement dependant on *doing something*.

It will open the door for so many more movement patterns and new behaviours, and become more true-to-life.

For example, firing a shot would send you in the opposite direction (newton's laws), moving a tie would drag through the "water", propelling you forward. Perhaps there could even be a new variable for growing little hairs to manipulate.

This shouldn't be too difficult code, although it would probably require some modifications of DB's physics.

Adding a third dimension and more advanced physics and then you might have something like Karl Sims or 3DVCE, with a large simulated envrioment player coding and all the new things that db can offer the combination of the two would proberbly be a hundred times better, more modification still and it could be used as an eingeneering/academic program like some other evolution sims which would give darwinbots the pubility it needs... if ever it gets that far.  

Offline Houshalter

  • Bot Destroyer
  • ***
  • Posts: 312
    • View Profile
Tie Movement
« Reply #6 on: March 14, 2010, 06:12:36 PM »
Im confused  
What exactly is this topic about?

Offline ashton15

  • Bot Builder
  • **
  • Posts: 99
    • View Profile
Tie Movement
« Reply #7 on: March 14, 2010, 06:54:16 PM »
Quote from: Houshalter
Im confused  
What exactly is this topic about?

I belive it's about implementing physics-based fluid dynamics into DarwinBots...   Though I could easily be wrong
« Last Edit: March 14, 2010, 06:54:51 PM by ashton15 »

Offline bacillus

  • Bot Overlord
  • ****
  • Posts: 907
    • View Profile
Tie Movement
« Reply #8 on: March 15, 2010, 12:57:23 AM »
In DB3, there won't be ties as such-each bot will be made of two joined circles, each acting as a pivot for other cells to cling on, so a 'tie' will be sort of sticking two circles together. With the proper physics, this can easily be manipulated into such motion, due to conservation of momentum and energy within the system. Another thing that will be solved by specialization of cells by fat, muscle etc...
"They laughed at Columbus, they laughed at Fulton, they laughed at the Wright brothers. But they also laughed at Bozo the Clown."
- Carl Sagan

Offline intron

  • Bot Neophyte
  • *
  • Posts: 7
    • View Profile
Tie Movement
« Reply #9 on: March 15, 2010, 01:25:24 PM »
is there somebody who got darwin pond working?
i think i fucked the link up, so: Darwin Pond Download

yea, a better physics engine would be nice and i like the idea about connecting bots with their surfaces. different sorts of "antigens" would be interesting: common ones for feeding and virus adhersion, specified for multibots (and differentiated organs).

muscle simulation would be possible if bots where able to adjust their radius somehow.

and neural networks could easily be simulated with suppressing and amplifying "potential spreading" along MB.
ceck out The Origin of the Brain
gets interresting in the last quater. it's a counter creationist video lol.

Offline bacillus

  • Bot Overlord
  • ****
  • Posts: 907
    • View Profile
Tie Movement
« Reply #10 on: March 15, 2010, 02:33:03 PM »
Basically, each of the cell functions will be enhanced by certain elements that can be built within the cell. This makes DB3 heavily biased towards multibots - howeever, there's been a lot of discussion about how this will be done. Check out the DB3 discussion subforum.
"They laughed at Columbus, they laughed at Fulton, they laughed at the Wright brothers. But they also laughed at Bozo the Clown."
- Carl Sagan

Offline Numsgil

  • Administrator
  • Bot God
  • *****
  • Posts: 7742
    • View Profile
Tie Movement
« Reply #11 on: March 17, 2010, 04:21:34 PM »
Yes, this is something I'll be working on for DB3 and multibots.  I asked Ventrella once how he set up the physics for Darwinpond, but it sounds like it's mostly a hack.  But I've done most of the research necessary to implement a physically accurate, if simplistic, fluid model for swimming for DB3.  I haven't done any proof-of-concept tests with it yet, though.