Code center > Suggestions
New Mutation Paradigm
Numsgil:
Definately pointed out the two major problems Carlo.
I'll leave type to type mutations unworked out for now. Most problems seem to stem from that. However, number -> *number and back again should probably still be possible.
Endy:
Would it be possible to just have a readback command? Something like:
ref 50 could readback the value of 50
Not sure how hard it'd be...
Numsgil:
I was thinking something like that.
Have a * command, so that
*50 is the same thing as 50 *.
shvarz:
Good idea with the * command.
I was thinking about your discussion of junk DNA...
First of all, I want to comment on Carlo's notion that we already have junk DNA. This is not exactly true, becasue true junk DNA is not expressed at all, while any "junk code" that evolved bots acquire actually does affect bot's behaviour - at the very least it places some numbers on the stack, which may affect real genes later on. So, when I am talking about junk DNA, I mean DNA that is not processed by program at all.
Originally I thought, just like Nums did, that having junk DNA would be very beneficial for the same reason it is beneficial in real organisms: duplicate gene >> turn one copy off >> mutate for a while>> turn it back on>> got a new protein.
But then I thought about it some more... In DBs we are forced to use relatively high mutation rates, becasue we don't have time to wait for millions of generations for something good to appear and our population size is several hunderds at most. At these mutation rates, any inactive gene will go into complete mess very quickly and is unlikely to become anything even remotely useful for a bot.
So now I don't have a very strong opinion on this. My inclination would be to allow junk DNA and see where it goes. If it becomes a huge burden on processor resources and not too useful, then turn it off. But it may turn out that all said above was wrong and having junk DNA would actually be useful.
Endy:
I was kind of thinking that the * command would be used for changing the location referenced, currently the bots are pretty much stuck with whatever *locations they have at birth. Should be good from both a combat and mutation standpoint.
Would make some great conspecific recognition genes possible, a bot could litterally cycle through all the ref/myvars making worries about misrecognition a thing of the past. :)
For mutations it would help to neutralize some of the higher/unused numbers by returning values of zero for most of them.
Endy B)
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version