Author Topic: natural diversity VS simulated  (Read 9401 times)

Offline Carlo

  • Bot Destroyer
  • ***
  • Posts: 122
    • View Profile
natural diversity VS simulated
« Reply #15 on: March 12, 2007, 01:57:06 PM »
Quote from: viplex
But in the sim, why do all species but one extinct? Time doesnt help. On the contrary..

It's an old problem. My opinion is that the enviornment DB offers is way too small and homogeneous to allow the existence of more than one species at a time. Simply, there isn't any reason for different species to share the same environment in a simulation; a few thousand cycles are always enough for evolution to tell the best one and make the others go extinct.
Real world is different because of specialization to different environments or different niches in the same environment. The complexity of real chemistry and physics allows numberless different niches to coexist in the same physical space. Spatial segregation, due to distances or natural barriers, is another source of diversity.

We discussed for a long time the opportunity to add complexity to the environment by superimposing to the field a so-called "environment grid". But there were different opinions about which properties of the field had had to be specified by the grid and nothing was done in the end.

Offline Numsgil

  • Administrator
  • Bot God
  • *****
  • Posts: 7742
    • View Profile
natural diversity VS simulated
« Reply #16 on: March 12, 2007, 03:54:41 PM »
I find that features like the env grid are hard to program because you never feel that the discussion is "done".  You keep reinventing it, and eventually it just gets put on the backburner.  Probably a downside of design by comittee

Offline rsucoop

  • Bot Destroyer
  • ***
  • Posts: 166
    • View Profile
natural diversity VS simulated
« Reply #17 on: January 19, 2008, 10:52:19 PM »
The biggest contributor to the precission lost in the simulation is the fact that sexual reproduction is impossible. It appears that the best way to create a stronger dna strand is not to copy the same one over and over with mistakes and hope it works, but to splice the genes and mix up the commands/values resulting in a mutant of the two parents. This splicing is not 50% exactly even in reality, and is 100% random everytime. So things like more complex multi-organisms from two seperate organisms would be possibly in nature, but not necessairly in the simulation. Until sexual reproduction is properly used, the simulation will never truely match the evolutionary process of the Earth.

Offline Peter

  • Bot God
  • *****
  • Posts: 1177
    • View Profile
natural diversity VS simulated
« Reply #18 on: January 20, 2008, 03:40:20 AM »
On earth species have evolved long enough without any sexual reproduction. Is not necesarily needed for evolution.
It would be nice to have sexual reproduction but it isn't there.

And what do you mean?

Quote
This splicing is not 50% exactly even in reality, and is 100% random everytime.
The splicing is 100% exact in reality, or do you mean things like crossover that change every side somewhat.
Oh my god, who the hell cares.

Offline rsucoop

  • Bot Destroyer
  • ***
  • Posts: 166
    • View Profile
natural diversity VS simulated
« Reply #19 on: January 20, 2008, 10:03:31 AM »
I refer to the selection of genes. Everytime this happens in life, it is 100% random and independent of the result of the previous gene that was copied. When I say 50% exact, I mean a perfect cross between parents resulting in a perfect 50% child. That's not always thre case, a black family with the correct recesive genes could potentially produce a red headed light skin child; not necessairly a mutation, but a very new type of person from the parents dominant/recesive genes.
« Last Edit: January 20, 2008, 10:05:03 AM by rsucoop »

Offline Peter

  • Bot God
  • *****
  • Posts: 1177
    • View Profile
natural diversity VS simulated
« Reply #20 on: January 20, 2008, 11:53:07 AM »
Quote from: rsucoop
I refer to the selection of genes. Everytime this happens in life, it is 100% random and independent of the result of the previous gene that was copied. When I say 50% exact, I mean a perfect cross between parents resulting in a perfect 50% child. That's not always thre case, a black family with the correct recesive genes could potentially produce a red headed light skin child; not necessairly a mutation, but a very new type of person from the parents dominant/recesive genes.
Ok, I thought you where speaking of chromosomes.
The selection of the chromosomes is in many cases not 100% random, and maybe even dependable of the earlier gene, with the splitting of chromosomes and the crossover of genes an really 50% exact cross between the parents seems hard too. Making a exact selection of the genes of 50% each and randomness for every gene possible, pretty plausible, everything is dependable of eachother. Genes aren't for nothing together in desame chromosome.

And the last sentance, that's not a case of genotype but of fenotype. It is true two cows with a hetero-zygote black gene could have a brown cow, chance is 25% roughly.

The whole dna-story in a whole is pretty complicated. But simply said there is nothing that is 100% random. If there are certain genes you took from your father you can predict that some others are too if they are in desame chromosome. I heard some genes even try to cheat to get in the dna.

The good point of sexual reproduction is indeed the mix of genes that didn't knew eachother before and they really work working together, you're right there.
What really is better then the 'normal' reproduction that depents for a big part on trial and error, altrough bacteria seem to exange eachothers genes too.

It has been stated before, but the reason DB hasn't got a complicated ecosystem and nature does. Is becouse nature is just a 'little' more complicated then DB.
I heard they tried to simulate a very small virus in a water drip, some time ago wih a supercomputer. That took some weeks to simulate it for a few seconds. The computer simulated literaly everything from every molecule. And nature can do it in real-time, and also then look how long it took to come this far.
« Last Edit: January 20, 2008, 11:54:22 AM by Peter »
Oh my god, who the hell cares.