Messages

61

DNA - General / shootval and distance for -1 shots

« on: April 02, 2010, 06:06:19 AM »

I'm trying to figure out a formula for shootval so that the bullets are shot with the right distance.

But on the wiki sysvars i dont read how it behaves.
Here on the forums i read some words about   log2(distance)  ??
However log2 is not an operator in DarwinBots , what might come close to it is bit operation >>  
So far i failed to write a good formula and searched the forums and wiki, and the beastairy.

Has someone already solved it. ??
How to perfectly shoot at something with the right distance.

Also i think .mass impacts a shot distance ??,  but  'm not sure but bot size and shotlength dont seam to compare

I can figure out how to get the distance and angle with :
*.refxpos *.refypos dist .edist store
*.refxpos *.refypos angle .setaim store
-1 .shoot store


But now how to shoot exactly to it ? (or with a margin : *.edist 10 add .edist store)

62

DNA - General / food finding and eating genes

« on: March 30, 2010, 05:13:58 PM »

I'm trying to figure out good food eating and aiming genes

I've made below, besides that i wonder if the code is good (can i use multiple times .up ?)

What i would look is that it would move to 3/4  or maybe 1/2 of the distance between its target.
So no matter the speed or its own body mass (it requires some adjustments).
The shooting is allready quite good, and it less often does circle around algeas
And even while pretty big it speeds up a lot to for example to meetup with Efilans or others  (but will soon be outnumbred no repro so far)

The only problem is it can stay behind a enemy without ever coming closer if they move  (algea work fine)

here is the code so far.

Code: [Select]


'Polinator
'requires 2 types

def reproval 50    '10...100
def bigsmall 51    '10...100
def direction 52
def etarget 53
def eyetemp 54
def shootnow 55
def shootdist 56

'=================================================================
'GENE FOR FINDING FOOD
cond
*.eye5 0 =
start

'eye9 or eye1 are (each eye = 35 units, *4 = 130
'
' if we see something get fixed and turn into its direction.
' next the GENE FOR EATING  will run for it
'

*.eye9 *.eye1 sub sgn .eyetemp store
*.eyetemp 130 mult .aimdx store

0 .fixpos store
'  nothing insight +10 up
*.eyetemp abs -1 add abs 5 mult .up store
*.eyetemp abs -1 add *.mass mult .dn store
*.shootdist *.mass mult *.refvalscalar add .up store

stop


'====================================================================
'GENE FOR EATING   'using advanced shooting mechanism
cond
 *.eye5 0 >
' *.in1 *.reproval !%=
' *.in2 *.bigsmall !%=
'*.refeye *.myeye !=
start

*.refxpos *.refypos dist .shootdist store

' .shootnow   this mimics *.eye5 > 40  and gets a valeu of (0|1)
*.eye5 40 sub sgn 1 add sgn .shootnow store

*.refxpos *.refypos angle .etarget  store
1256 *.aim sub *.etarget add .etarget store
1256 *.etarget sub .etarget store
*.etarget .aimshoot store

*.etarget .aimdx store


*.shootnow .fixpos store
*.shootnow -1 mult .shoot store
'0 .fixpos store
*.shootnow 1 sub abs .shootdist mult 2 div 1 rnd add .up store

stop

end


63

Bot Tavern / developing self tuning bot based on evolution / dna

« on: March 29, 2010, 06:33:13 PM »

Wouldn't there be some problems with that if the bots use these values for conspec recognition? For example, bots that change their values to much won't have a chance because their own kind will eat them. You could also end up with copy cat bots. A single mutation in the part of that bot that tells it when to store those values, and it automatically changes every time it sees another bot. There could be other problems to. I think it would be best if you used racial memory and kept them secret.

I was more thinking of it as eco sim, so i'm less woried of conspec recognition.
Future siblings can change small amounts of the valeus, so in the end you endup with a sim of competing values, and find for those bots ideal values.
A kind of evolution of tuning, like how birds change wing sizes.

The reason a bit is that DarwinBots is a good enviroment for testing environmental cases
But since the language is quite advanced, some people say its too good for evolutionary processes since there is no outside pressure (rewarding) ==> see the wikipedia review.
I was wondering if it was possible using the language to write something that could do it in a limited way.

Thats also why i would like the plants not to grow to big, and why they having a seeding mechanism.


Hmm racial memory is also an option for this, might be better.
But still i'm wondering whats wrong with the aiming i use in these bots, there must be better aiming math, then what i'm using here, but it has been a while for me that i coded bots.


I will also check LionFish thanks Moonfisher

64

Darwinbots3 / DB3 Questions

« on: March 29, 2010, 10:29:16 AM »

vegies should get energy maybe based on light spots.
so whitin the range of the spots there is more light so you get green and les green areas.

nice would be an easy polunation from seed to plant, beeing for x cycles moveable with a speed y in direction z  (i make plants like that, now with code)

65

Bot Tavern / developing self tuning bot based on evolution / dna

« on: March 29, 2010, 10:05:27 AM »

I am not sure if this already has been done, i'm working on the following :

A simulation that competes for the best bot parameters.
For example a big impact of A and B in the following :

Cond
*.nrg > A
start
B .repro Store
stop


In my sim each plant has a random out1 for A and out2 for B
Robots startup and take the first valeus from the plant they see first (kinda random).
Also the robot will then put A for out1 and B for Out2 as a result their ofspring will also use those.
Since robots also determine A and B for checking if they are relatives, you get different species inside a single bot gene, with different behavior.
Both check with !=% (10% range) if other bot's are family on both A and B valeus and if not eat them.
Ciblings might change A and B values maybe later with 5%


.. i notice however that to much energy gets into the sim
so i need to make moves of bots cost more energy and also i think i should somehow limit plants not to become too big.
as with to much energy the sims then to get crowded by tiny bots.


What i wonder have more bots been made to try something like this ?.
For example to balance and find optimum eye distances or something else ?


included as zip file the work so far, note some parts have been marked out, as it seams it has some kind of move aiming problem  (Gene 7) of the 8 genes.
I often see it circle around a target and doing nothing, sometimes 9 a bit to often it doesnt attack, still trying to figure out why not, so there i placed some marks  to disable "not killing family"
The included vegies are part of this kind of sim, its default sim and the vegies dont need to be fixed.

66

Bugs and fixes / 2.44 Released

« on: March 29, 2010, 08:21:36 AM »

hi
It has been a while i since i have logged on here

 i wondered, the most recent versions seams 2.44.04  

which is somehow a much smaller compilation then any previous one ??  (whats the cause of that)

But also i wanted to know, have new DNA commands been introduced, is the wiki still up to date (sexrepro maybe, or viruses of multi genes? )

67

Evolution and Internet Sharing Sims / internet sharing problem

« on: January 21, 2009, 09:26:30 AM »

Welcome!

If you only get your own files back there might not be anyone else on.  It's been pretty quiet recently, so that wouldn't surprise me.  Internet Mode has also always been a bit buggy, so it's a use at your own risk sort of situation.

Maybe its an idea to keep a short list in there, like 20, or 50 or so.
Have people be able to upload, and send back the results of a battle field; if their species is stronger then it gets in the list and replaces the older.

hmm


that might actualy be a verry cool idea..

68

Biology / The first wet alife?

« on: January 21, 2009, 09:18:15 AM »

note that cristals also grow under the right environment, okay they dont replicate.
But how a bush fire, it does replicate to other trees.
I think with live, a definition is required, and that might get complex.


But I think it should respond to its environment, with a strategy to survive environment dangers.

For example the bushfire shouldnt run out of trees. So it needs a way to conserve itself.
A simple biological virus, folows simple rules often, but their numbers make them survive.
But then what about the wheater patern, its shifting and continously changing, and its a closed system on itself.
There should be a next rule I think that such systems are not called life, but be called environment.

An environment is a system that folows rules of its own, wich are generated by itself.
Its a system which is often chaotic in nature, and has no goal to survive, it simply is there, like space itself.


Then is darwinbots life ?
It is artificial life, in a sense that its enviroment is based on math, it folows simple rules so it can stay alive in that area.
The area is however verry limited, a bots definition doesnt work on a different planet with no computers.

So a next rule for more comonly known life might be :
That it can deal with its chemical environment, as if it contains a language, a system (biomolecular bioligy) to interact in its environment.

But maybe appart from known life it could also exist in, plasma, radiation, quantum fields, etc, things we now hardly understand.

69

Bugs and fixes / DarwinBot crash (ver 2.44)

« on: December 11, 2008, 06:46:29 PM »

The very very latest source is available from the FTP.  We don't have version L's source code, unfortunately, just h and 2.44.1

what is the difference with L and H
Where do those letters stand for ?

70

Bugs and fixes / DarwinBot crash (ver 2.44)

« on: December 11, 2008, 09:35:00 AM »

Darwin2.43.1L works good, I think the species forking stuff is the main issue with the newer versions.

Well the newer version 2.44.1 seams to do it better also
Dough it crashes at my pc on something else (when I want to start a simulation, but delete the last bot entry from the list), workaround is that I diasable the bots code execution.
Maybe dough I should post it in a new thread and not here.

71

Bot Challenges / simulation fish population evolution

« on: December 10, 2008, 05:14:00 PM »

Note there is also a graph for average age.

If it all works well, the fishing boats can be programmed to only eat small fish (like the science article) (based on refbody or refage ??)
Or in another sim to eat the big fish, I think its nice to see the system can be stable but also can be verry chaotic
That reflects is I asume the real fishing problem.

I'm running your sim now its nice

I often see the fishing boats swarming.. hmm maybe they multiply a bit to fast or so I'm not sure its hard to balance.
As otherwise the the algey loose from the fish... (we could let the algey eat fishing boats but nah thats aint realistic)

(edit) I now run it in a larger sim size 10 (runs faster) started with 30 algey (max 200) 15 fish and 5 boats..
 I seee now swinging population graph as you described, intresting as long there are algey the species come back wonder dough what it does over a longer period I keep it running for a while

72

Bot Challenges / simulation fish population evolution

« on: December 09, 2008, 06:57:02 PM »

 Great d-EVO !!  

Your the first showing Darwin bots can be used for species who evolve by finetuning an atribute.
This is the roadway of generating species  in bioligy I believe.
No longer it happens due to randomization of DB genetics.

It was argued at wikipedia not darwinbots pedia, this was something DB didnt do.
Your the first to show it can be done congratulations !  


I will look at the code later, your fish makes nice swiming paterns
Sometimes they go a bit to fast it seams, dough that maybe a setting of me which is wrong.

I saw sometimes the fishingboat nets clog with too many fish and then they stop functioning.
I will look at the fishboat code and put in the 360 degree auto aiming formula's I used in w6.txt
thats independent of a bots heading, and will always hit the 2 most nearby bots.

Its realy impressive you know, I'm impressed.
Also you have set a standard for these type of simulations, people can write fish based on it or other boats
Or people can change other mating atributes for example also take a look at body size etc.
It will go I think more to behaviour programing of species.

73

Darwinbots3 / Darwinbots3 Installer

« on: December 09, 2008, 02:37:16 PM »

ehm visual studio can create msi files to install something, doesnt require advanced installers, an msi is allready that..
More is not required, as you can alway de-install (trough ad en remove programs) any msi file.
And also if I remember well upgrades can go as an mst file
If the problem is that you want to put in extra files (like sample bots) just drag them in your project, like you would do with images.

heh and you can assign even msi files trough AD policies forcing the company you work for to use Darwinbots

74

Newbie / Eating Corpses

« on: December 09, 2008, 10:45:29 AM »

Skylmit,  I got some ideas to create a simualtion in which things would depend on eachother
              Its an idea written out as a fish population simulatin problem :

              http://www.darwinbots.com/Forum/index.php?showtopic=3059

 

   Basicly use inX and outX commands to identify species, so they al got an ID number
   Inlcuding the algey can have an ID set that way.
   And then you can have species eating specific species, or a range of species
   So only do a -1 shot if also   IN1  1000 <
   
   Also you one has to adjust eye routines to only search (turn to) for its specific food and otherwise ignore it (dont turn).

I had not much time to script it (my freetime is now filled with a language course), but d-evo is I think verrry close to show the idea, he allready makde some species as in the description of the problem (dough the problem was even a bit more complex then species control, also species adjustment was adressed in the problem).

75

Bot Challenges / simulation fish population evolution

« on: December 04, 2008, 07:43:46 PM »

some comments I had not had time to script yet
bussy work days

* Well I dont think pond mode would be required for this, (boats might also catch only big fish, or small fish its a rule we can set to ships)


* Its true I didnt mention "disable mutations", it was abvious to me.
  I personaly always disable it, altough nice for some true dna evolution acka darwin bots, starting from a zero veg, I dont like its polution.
  If you realy want that then use mrepro it has the same effect.

* its okay it has now been decided what out to use for species ID it think the next
  out1 could be used for for ships lets say anything between 8000 and 9000 can be a ship
  between 8000 and 9000 a ship
  between 4000 and 5000 a fish
  between 5000 and 6000 a shark

  Or simply if the number is higher then just stop bad luck let them eat you (more simple logic the same effect)
  Or if you loose me out of your sight I escaped and I continou   (combine eye5 and in1)
 

* I dont think males should eat females especialy fish (who only should eat veggies), both are required
  dough it does happen in nature like spiders etc, most fish dont do that

* there is allready refage
   But maybe its more handy to create an out2 wich turns on
   1 for to young to repro
   2 for male who is ready for mating     (at the same time we could avoid eating each here)
   3 for female who is ready for mating
   4 to old for repro ? or maybe some other state ? hungry ? (so a friendly species might feed it?) ...
   5 ....
   6 ..
 

I wonder how realistic we could make it
Maybe like salmon or so a species might dye after reproduction ? (keep it locked in place eventualy a shark will find it and eat it)
of example
   7 .. (I'm an algey but not only that I'm a place where fish mate)...  (we might do that later)



* there earlier math I explained (8+2) /2 + (sgn ( 2 - (8+2)/2)) , wasnt good I believe as rate for mating age would never be able to get lower.
   something else is needed, it has to be in the spirit of our fish.. it has a live it grows it reproduces at some age this simulation will gonna decide that age.
   If we can put it insight some smart singel gene or so this desicion process..
   Then that might be handy to also later test fine tuning for other valeus.

* therefore  lets reserve Out3 & out4  & out5 & out6  for any other things we might like to focus on later.
  (for example what would be the best eyes? or whatever...)