Author Topic: Amoeba  (Read 4721 times)

Offline bacillus

  • Bot Overlord
  • ****
  • Posts: 907
    • View Profile
Amoeba
« on: April 26, 2008, 11:18:27 PM »
Code: [Select]
cond
 *.robage 0 =
start
 2468 .out1 store
 2468 .tout1 store
stop

cond
start
 .tie inc
 *.nrg 3 div *.body sub dup .strbody store - .fdbody store
 .tout4 *.tout4 sgn mult dec
 2 .up store
stop

cond
 *.tin1 *.tout1 !=
 *.trefnrg 0 >
start
 -1 .tieloc store
 -1000 .tieval store
stop

cond
 *.eye5 0 !=
 *.out1 *.in1 !=
start
 *.refxpos .tout2 store
 *.refypos .tout3 store
 10 .out4 store
stop

cond
 *.tout4 0 =
start
 0 .tout2 store
 0 .tout3 store
stop

cond
 *.tin1 *.tout1 =
 *.tin2 0 !=
 *.tin3 0 !=
start
 *.tin4 .tout4 store
 *.tin2 .tout2 store
 *.tin3 .tout3 store
stop

cond
 *.tout2 0 !=
 *.tout3 0 !=
start
 *.tout2 *.tout3 angle .setaim store
stop

cond
 *.refnrg 0 >
 *.eye5 40 >
 *.in1 *.out1 !=
start
 -6 .shoot store
stop

cond
 *.nrg 350 >
start
 50 .repro store
stop
« Last Edit: May 02, 2014, 01:48:15 PM by Botsareus »
"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 rsucoop

  • Bot Destroyer
  • ***
  • Posts: 166
    • View Profile
Amoeba
« Reply #1 on: April 27, 2008, 12:49:37 AM »
Very nice bot, and your conspec genes are unrivaled, they last for hundreds of generations and point mutations, at no point did they become canibals.

Offline gymsum

  • Bot Destroyer
  • ***
  • Posts: 215
    • View Profile
Amoeba
« Reply #2 on: April 27, 2008, 03:17:25 PM »
Ok, I've tested your bot against the following with equal conditions for all btos:

Slim Evo 2 (ancient coms system)
My Algae plant
And Masterl

Slim Evo 2 out gunned it completely, and out maneuvered its tactics, mostly because Slim Evo has social genes for ancient sysvars lol.

Algae was easily harvested, but its virus caused your amoeba to crash in population.

Masterl still sucks as far as true combatant goes, but its more for evo sims, not what this was.

COnsditions:
F1 conditions
Thin fluid
Moon gravity
Total warp around pond mode, with 300 cycle day/night

Thwe concept is there, but take a look at the slim evo 2's tie communications, it has imbeded switches to keep the group focused on a target, and not eachother.

Offline bacillus

  • Bot Overlord
  • ****
  • Posts: 907
    • View Profile
Amoeba
« Reply #3 on: April 27, 2008, 05:56:59 PM »
This bot is only a rudimentary try at getting a multibot working; what I'mtrying to make is a bot that acts as a jellyfish, following a set path, and pulsating as it goes. The trick is getting the thing to pulsate without tie breakages.
Have you tried running it on map size 5, max veggies 500, pop threshold 250, veggies per repop event 100? You see what it should really look like.
Now I just need to reduce eyes to 3, add tentacles, a tie and a suit, and voila!
« Last Edit: April 27, 2008, 06:03:38 PM by bacillus »
"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 gymsum

  • Bot Destroyer
  • ***
  • Posts: 215
    • View Profile
Amoeba
« Reply #4 on: April 27, 2008, 06:10:48 PM »
you need to stifftie or you will break lose ties when moving. Seriously tho, look at the slim evo 2, I was able to get them to swarm in a circl by just changing a few values. It's very difficult to explain because I lack the patience to break down the entire genetic structure it uses, but it was built to be socialy aware, meaning it could adapt to have a universal conspec, and its movements are coordinated throughout the structure. Unfortunately, it likes to stick when tied to veggies, so its more of a spider then in hybernation. Its goal is to kill, collect veggies reproduce, and then send out scouts. I would say establish a set of goals for your bot and construct the genes that way. YOu have a nice start, but I would think about using one tout for movement purposes. I used stages to determine a unique behavior, and this was transmited throughout the entire multibot including vegies, otherwise trying to create unique behaviors are not alwyas used by the bot as it was best intended.

Offline bacillus

  • Bot Overlord
  • ****
  • Posts: 907
    • View Profile
Amoeba
« Reply #5 on: April 27, 2008, 06:55:25 PM »
Never heard of stifftie before, sounds good though. Thanks.
I'm working on an antbot at the momemt; a colony starts out as a young queen that evolves to a queen when enough energy has been aquired to start a colony. It uses drones to relay messages to the workers, which are harvest (feed until 5000 nrg is reached, then go back to queen and release nrg) and kill (when more than 200 ants, attack everything more fiercely until more ants are needed.)
Any suggestions?
« Last Edit: April 27, 2008, 06:59:53 PM by bacillus »
"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 gymsum

  • Bot Destroyer
  • ***
  • Posts: 215
    • View Profile
Amoeba
« Reply #6 on: April 27, 2008, 09:57:17 PM »
Quote from: bacillus
Never heard of stifftie before, sounds good though. Thanks.
I'm working on an antbot at the momemt; a colony starts out as a young queen that evolves to a queen when enough energy has been aquired to start a colony. It uses drones to relay messages to the workers, which are harvest (feed until 5000 nrg is reached, then go back to queen and release nrg) and kill (when more than 200 ants, attack everything more fiercely until more ants are needed.)
Any suggestions?

Start with some real basics; develop a hunter genome similar to Excalibur (as it is the best hunter around) but make sure you use some varibale(s) for a switch(es). Add some memory to remember a location you want to be the colony, say birthplace, or last known veg or, or last place they mated (something common to salmon). Develop a bot taht really tie feeds well, run it all with no mutations so you can really see how it works. Fine tune it with some tinkering and always compare genes before and after if its a complex set of genes for sasy communication. Make something unique for its conspec so it can last many generations; and establish what determines a queen or other member. Once your queen is produced, its up to you to develop each stage with a reset and switch. Slim 2 used only 3 or 4 stages, two of which involved collecting food and fixing to a spot; it is possible to develop said drones through reproductive-stagedependent genes taht produce other cells to make up some larger thing (in the case of certain ants, that would be a link to food to the colony, either transport or tie).

Offline Botsareus

  • Society makes it all backwards - there is a good reason for that
  • Moderator
  • Bot God
  • *****
  • Posts: 4483
    • View Profile
Re: Amoeba
« Reply #7 on: May 02, 2014, 01:49:41 PM »
Although I have seen a few like these,  moving to interesting bots anyway.