Author Topic: Global mutation rates  (Read 2680 times)

Offline Schigum

  • Bot Neophyte
  • *
  • Posts: 15
    • View Profile
Global mutation rates
« on: April 02, 2007, 09:50:21 AM »
Hi,

I recogniced, that the number of mutations within my zero-bot sim allmost perfectly matched with the estimated number, calculated with the mutation-rate. I used the standart rates (1/5,000,000 for point-mutation) and had an average of 190 mutations at 4,000,000 circles (DNA length is 231). Estimated number is about 185, if I'm right.
The global mutation rate is at 32X, so I wondered, if maybe those 1/5,000,000 are meant to be at 32X and I ran a simulation with a global mutation rate of 1/32 with the same result (average of 45 mutations after 1,000,000 cycles).
At last I ran a simulation with an oscillating mutation rate and watched the mutation-graph, wich should have built steps. But it returned a straight line so there is no change in mutation rate.
I searched the forum without any result, cause I thought I might have missed that mutation rates are disabled.
So it seems to be a bug, but I can't imagine noone else found this before.

I use Version 2.43 of course

m.f.G.
Schigum

Offline MacadamiaNuts

  • Bot Destroyer
  • ***
  • Posts: 273
    • View Profile
Global mutation rates
« Reply #1 on: April 02, 2007, 11:38:46 PM »
I've had the same sensation for veeery long but just ignored it as I always used fixed mutation rates.
Sometimes you win, and sometimes you lose...

Offline Numsgil

  • Administrator
  • Bot God
  • *****
  • Posts: 7742
    • View Profile
Global mutation rates
« Reply #2 on: April 02, 2007, 11:42:01 PM »
This is probably a real bug.  Mutations code hasn't been played with much since the 2.4X code series was started (which involved a complete rewrite of the code base for mutations and DNA).