Author Topic: DNA lenght  (Read 2957 times)

Offline Drognan

  • Bot Builder
  • **
  • Posts: 56
    • View Profile
DNA lenght
« on: February 24, 2010, 02:49:46 PM »
Does anyone know what is the smallest lengh oft DNA that allows complex bot behavior?
I inserted 30 zerobots in sim, each 30 zeroes in DNA. I was thinking since most of zerobot DNA is junk DNA, and it's slowing computer down, maybe evolving bots with shorter DNA may speed up thinghs? Who knows, maybe bots with smaller starting DNA will have less junk, and more "useful" DNA?
I would like to hear your opinions...
Further info about sim expect in simulation emporium.

Offline Numsgil

  • Administrator
  • Bot God
  • *****
  • Posts: 7742
    • View Profile
DNA lenght
« Reply #1 on: February 24, 2010, 04:28:18 PM »
There were some zerobot viruses that worked by incrementing random memory locations.  So I'd say the smallest complex DNA strand is something like:

1000 rnd inc

or maybe

1000 rnd dec

Or something like that.

Offline Drognan

  • Bot Builder
  • **
  • Posts: 56
    • View Profile
DNA lenght
« Reply #2 on: February 25, 2010, 09:05:09 AM »
Quote from: Numsgil
There were some zerobot viruses that worked by incrementing random memory locations.  So I'd say the smallest complex DNA strand is something like:

1000 rnd inc

or maybe

1000 rnd dec

Or something like that.
ΒΈ
But is that real bot? I'm going for some interesting behavior like attacking target, runing away when attacked, feeding, reproducing and similar...

Offline Houshalter

  • Bot Destroyer
  • ***
  • Posts: 312
    • View Profile
DNA lenght
« Reply #3 on: February 25, 2010, 09:46:52 AM »
It depends on how you define "complex". This is like numsgils but it stores rather than a random inc/dec. Make a bot with 100 lines of this and see what happens and see what evolves. By changing the numbers and deleting the rnd it can evolve from randomness to an actual bot with "complex" behaviours.

Code: [Select]
200 rnd 100 sub 1000 rnd .store