Children Are Immune to Parent's Shots

[Endy]

There seems to be a limit in DB's code preventing the child from being affected by a parent's shots for 10 cy. Got annoyed after my bots lost their firing restriction coding and went looking for answers.

Endy B)

[Numsgil]

Yes, that is a recent change, and is mostly there to prevent a parent from injuring a child if it reproduces while firing a continuous stream of shots.

[PurpleYouko]

I didn't know about that one Num.

Is that a good idea?  :(

If the robot is firing a continuous stream of shots and it kills its offspring then surely that is one of the consequences of bad genes.
It should be selected against by evolution rather than artificially minimized by the program.

[Sprotiel]

Yes, that is a recent change, and is mostly there to prevent a parent from injuring a child if it reproduces while firing a continuous stream of shots.

IMHO, this is wrong. A bot that shoots its child deserves to get extinct.

[Endy]

I personally am not that much offended by the code, since the bots don't normally acheive much distance after birth, the continual shooter would still wind up hurting it's offspring.

Mainly it was just interesting that the code preventing both child and parent from firing at refage/robage 0 = was bypased. Presumably this allowed child to essentially feed from it's parent, and the parent to hunt other bot's offspring.

Your call either way.

I'm not sure how much the bots need tons of speed. For the most part they select for slowness since higher speed makes hunting more difficult and involves more nrg. I had one sim that the evolved alot of Motionless bots then a fast Canni group wiped them out, only to evolve into a slower moving Canni later.

As long as all the vegs are eaten(or moved about) the bots just have to wait until they repop near them. All the nrg can then be devoted to reproduction.

Endy B)

[Numsgil]

The parent can't hurt child code was put in specifically because:

1.  Bots are immune to their own shots
2.  When a bot reproduces the child just kind of 'appears' in front of the bot, quite magically.  Meaning it will pop into existance on top of a stream of shots.

Mammal children are protected from their mom's immune system while they are gestating.  So I was going along the same lines.  10 cycles was arbitrarily chosen, but the idea is simple enough: the baby bot should still have whatever protein markers the parent has that prevents the parent from being injured by its own shots.

This code doesn't prevent parents from trying to shoot their kids, or vice versa.

I didn't think it was that big a deal, or I would have certianly dicussed it with you guys.  If you think it's artificial, it's easy enough to remove.
The mutations code is, to be frank, a mess.  Combined with the simplicity of the environment (yes, a destroy-the-birth-tie will be more successful, but only slightly.  In DB environment, that edge is negligable, so it rarely develops or even survives the selection process).

[Greven]

Mammal children are protected from their mom's immune system while they are gestating. So I was going along the same lines. 10 cycles was arbitrarily chosen, but the idea is simple enough: the baby bot should still have whatever protein markers the parent has that prevents the parent from being injured by its own shots.

But still this, protection of the child, lies in the genes, it is not an universal rule of the universe!!! This should be developed by the evolution to select against parents killing its offspring.

I think we need to discuss the DNA/Genome structure... Because I think most of the bad mutations we get is not caused by lack of a complex envoriment, but bad structure of DNA/the genome.

[Numsgil]

I've modified the source code so that the age is now set to 0 or 1.  That is, shots that exist at the time the bot is born won't hurt it (unless they belong to another bot than its parent) but shots fired by the parent after its birth will.