Bots and Simulations > Internet Mode Commentary
IM is getting interesting
Testlund:
For the first time bots entering my sim don't get killed instantly. A whole bunch of species have been roaming my sim this time around. First Multiply4 was the most successful, but seem to have been replaced by VC and Flyfruit. I don't know if they evolved more successful or if you have redesigned them. Thought I should upload a screenshot of it.
Moonfisher:
I think it might be a compination of VC getting infected by the virus in FlyFruit and the infected bots mutating to improve survival of the infected bots...
I know the VC I've seen so far where infected by the virus but didn't seem to act like fruit flies... it was more of a fungus like behavior... but seemed to work pretty well....
Testlund:
And one of those bots infected my Autotroph with it's dna causing it to start multiplying quickly, increased CostX to over 2, then the population dropped down, lowering CostX to 0. I'll run it again tonight and see how it goes.
Moonfisher:
I tryed running an sim on IM with standard F1 costs and some regular alge in it.
Didn't take long for some Seasnakes to pop in from eric.
They look like this now :
[div class=\'codetop\']CODE[div class=\'codemain\' style=\'height:200px;white-space:pre;overflow:auto\']'''''''''''''''''''''''' Gene: 1 Begins at position 1 '''''''''''''''''''''''
cond
*111 431 !=
start
99 .sharenrg store
.dnalen store
300 .memloc store
336 973 store
*.body 5 >
*.nrg 80 >
dropbool
*.maxvel .dn store
50 .repro store
*.totalmyspecies 1000 <
*.nrg 200 >
and
60 mult not
*.body 1 >
and
*.body 0 sub 10 drop 10 91 floor 312 *.robage !=
0 *.numties 1 -3 >
.tienum inc
*.numties .up store
>
or
and
*.numties .deltie store
dupbool
*973 and
=
1 453 .tieloc store
1000 .tieval store
not
and
=
2 *973 dupbool
452 .tieloc store
-1 .tieval store
dropbool
*973 2 >
and
8 .tieval store
1 *.body =
0 *.numties =
and
dupbool
*.eye5 0 =
*.memval *.dnalen dec
=
or
and
350 .aimright store
*.eye5 0 *.dnalen !%=
and
*.refbody 20 >
*.totalmyspecies 600 >
or
and
*.refxpos *.refypos floor .setaim store
.tie inc
clearbool
dec
*.thisgene *.genes sgn add >=
*.genes 777 *777 1 sub -1 mult sqr dup *777 .genes dec
*.thisgene 1 sub *.thisgene 1 sub sgn mult *.thisgene 1 add *.thisgene 1 sub sgn 1 ceil dec
stop
'''''''''''''''''''''''' Gene: 1 Ends at position 179 '''''''''''''''''''''''
'''''''''''''''''''''''' Gene: 2 Begins at position 180 '''''''''''''''''''''''
cond
*104 431 2 !=
start
.eye5width store
336 .eye3width store
336 .tmemloc store
*.body 80 >
and
*.maxvel store
402 -133 1000 <
*.nrg 200 >
and
^ 60 *.body 10 mult sub .strbody inc
not
*.body 10 mult 10 sub 0 floor .fdbody store
*.numties 0 div *.robage 1 >
and
*.tiepres .tienum store
*.tmemval *.dnalen !=
*.numties 8 >
or
and
*.numties .deltie store
*.tmemval *.dnalen !=
452 .tieloc store
-1 .tieval store
*.numties 0 =
and
dupbool
*.eye5 0 =
*.memval dec
*.dnalen =
or
and
350 .aimright store
dropbool
*.memval *308 !=
and
*683
'''''''''''''''''''''''' Gene: 2 Ends at position 278 '''''''''''''''''''''''
'''''''''''''''''''''''' Gene: 3 Begins at position 279 '''''''''''''''''''''''
cond
.setaim store
.tie inc
clearbool
*.genes
'''''''''''''''''''''''' Gene: 3 Ends at position 285 '''''''''''''''''''''''
'''''''''''''''''''''''' Gene: 4 Begins at position 286 '''''''''''''''''''''''
cond
.genes dec
*.thisgene *.genes sgn add .mkvirus store
*.genes 777 *777 1 sub overbool
mult sqr dup store
mult *777 store
*.thisgene 1 sub sgn 1 *.thisgene *.shootval mult sqr *.thisgene add 7 1 sub clear add sgn -1 mult mult add .delgene store
else
'''''''''''''''''''''''' Gene: 4 Ends at position 331 '''''''''''''''''''''''
As far as I can tell it's been infected by the virus version of Fruit Flie twice and then mutations have caused it to adopt a very different behavior.
Inititaly fruit flies would consume and kill everything in sight, but this thing protects the alge instead and feeds on them without killing them... so basicaly it's breeding alga minimalis while forming a fungus like patern around them... I don't know if the alga minimalis in erics sim are infected with something, but in my sim it seems like the bots around the alge are enough to protect it against other predators.
Either way Fruit Flies was never buildt for mutations, so I think it's very cool that chance caused a bot to get infected twice, making mutations less harmfull and allowing it to evolve.
This thing is so badly mutated by now you can barely reconize it, the only way I can tell it's 2 fruitflies is from the remains of the virus key.
Actualy I think this may generaly be a good idea when mutating a hand authored bot, just copy paste evrything a few times, easy way to make it more stable during the early stages of it's evolution.
EricL:
--- Quote from: Moonfisher ---I don't know if the alga minimalis in erics sim are infected with something, but in my sim it seems like the bots around the alge are enough to protect it against other predators.
--- End quote ---
They aren't. I'm actually running a pre-version of 2.44.1 on my machine(s) that prevents veggies from getting infected with virus's when running in IM. Viruses are too powerful in my opinion. In IM at least, it strikes me as unfair that one can infect the all the veggies in a foreign sim. The incoming nrg in certain sims may be such that veggies are supposed to be long lived and the eco-system may be adapted to that. Infecting such long-lived veggies that essentially can't be killed is an enormous weapon, a sure way to insure everything in the sim gets infected no matter what.
Viruses are hard to see, hard to track, hard to watch speciate or compete as they change. I'd prefer to tip that balance back towards viruses as a heterotroph to heterotroph weapon used by bots without veggy intermediaries. Otherwise they essentially become the primary unit of evolution. I'm open to opinions here of course, but inclined to do this.
Additionally, I plan to re-visit the whole infection paradigm soon. I'd like to enable multi-gene viruses with a sequence-matching paradigm for infection targeting. This would would provide a much more fertile ground for arms-race evolution of infection and resistance strategies between bots and viruses.
Navigation
[0] Message Index
[#] Next page
Go to full version