Topics

16

Bugs and fixes / .vshoot not affecting virus range

« on: November 19, 2006, 01:37:02 PM »

.vshoot doesn't seem to affect the range of virus shots. All viruses, whether they're ...

cond
start
 *.thisgene .mkvirus store
 1 .vshoot store
stop

... or ...

cond
start
 *.thisgene .mkvirus store
 100 .vshoot store
stop

... have exactly the same range. Even if a bot is storing values upwards of 10,000 into .vshoot, the virus doesn't go any further.

17

Bugs and fixes / Bots stop reproducing

« on: November 19, 2006, 10:33:00 AM »

EDIT: See the bot below

18

Suggestions / Tie feature request

« on: November 19, 2006, 10:05:20 AM »

First off, am I right in thinking that .tienum can be changed inside cycles like this:

cond
start
1 .tienum store
314 .setang store
.tienum inc
628 .setang store
.tienum inc
942 .setang store
.tienum inc
1256 .setang store
stop

So tie 1 gets set to 314, tie 2 gets set to 628 etc.

*********

Secondly, I have a feature request for 2.4 ...

.sendgene

.sendgene can be set to a specified gene, like .mkvirus, which is copied through the tie to which .tienum is set. The bot at the recieving end gets the selected gene inserted into it's genome like a virus

cond
*.numties 0 >
start
*.tiepres .tienum store
*.thisgene .sendgene store
stop

^ That would be a 'tie virus' that spreads as bots tie together

This would make horizontal gene transfer possible to code (me and Endy were trying to do this with viruses a while back), and would make possible all manner of interesting new kinds of virus, and other phenomena.

It's simple, useful, and limitlessly open-ended

19

Multi-Bots / Duo (MB)(Elite)-16.09.06

« on: November 19, 2006, 04:53:11 AM »

Code: [Select]

' Duo
'
' 2-Bot MB
' Moves using ties

cond
*.numties 0 =
start
50 .repro store
stop

cond
*.robage 0 =
start
.tie inc
stop

cond
*.numties 1 =
start
1 .tienum store
628 .fixang store
50 .sharenrg store
stop

cond
*.eye5 0 >
*.eye5 40 <
*.refeye *.myeye !=
start
*.refxpos *.refypos angle .setaim *.robage sgn mult store
-8 .shootval store
-6 .shoot store
stop

cond
*.eye5 39 >
*.refeye *.myeye !=
start
*.refxpos *.refypos angle .setaim *.robage sgn mult store
8 .shootval store
-6 .shoot store
stop

cond
*.trefage *.robage >
*.robage 10 mod 5 >
start
*.tiepres .deltie *.body 500 sub sgn 0 floor mult store
1000 .fixlen store
1 .fixpos store
stop

cond
*.trefage *.robage <
*.trefage 10 mod 5 >
start
0 .fixpos store
*.eye9 *.eye1 sub 2 mult .aimdx store
stop

cond
*.trefage *.robage >
*.robage 10 mod 5 <
start
1 .fixlen store
0 .fixpos store
stop

cond
*.trefage *.robage <
*.trefage 10 mod 5 <
start
1 .fixpos store
stop

cond
*.waste 10 >
start
*.waste .shootval store
-4 .shoot store
stop

cond
*.nrg 2000 >
start
100 .strbody store
stop

cond
*.nrg 1000 <
start
*.body 100 ceil .fdbody store
stop

20

Suggestions / Command-ize everything

« on: October 24, 2006, 06:30:32 AM »

I'm sure this had been discussed briefly before

How about steadily transfering the 'do something' sysvars into commands. This gives them greater flexibility and more compactness

For example:

10 .up store -> 10 up

-1 .shoot store -> -1 shoot

This way, you can also pack more functions into one command too, for example:

-1 .shoot store
314 .aimshoot store

becomes

314 -1 aimshoot

And

300 *.eye5 mkvenom

Makes 300 units of venom that affects an enemy bot's eye5

And maybe we could also have some readback commands, like some eye and ref ones. For example:

7 eye

Returns the value of *.eye7
And

6 refeye

Returns the *.refeye of the bot in eye6

21

Darwinbots Program Source Code / Modifying the code

« on: September 05, 2006, 04:05:01 PM »

I'm thinking of doing some tinkering with the program of my own  

I took the source code of Endy's v2.37.7 and modified the 'shots.bas' using a standard text editor, to comment out the code granting conspecs' immunity to other conspecs' poison. So how do I now wrap that up into a neat executable program?

22

Suggestions / Transposition

« on: August 01, 2006, 05:25:16 PM »

Simple. Take a run of BP and deposit them somewhere else in the DNA at random

And how about a 'point transposition' that works like a point mutation

23

Suggestions / Sexual Reproduction Focus Group

« on: July 31, 2006, 02:24:41 PM »

Hiya everyone, just returned from holiday  

Let's get together and decide what we want to do about sexual reproduction in DB

Here's my idea to start things off:

Firstly, meiosis: reproduction that splits the bot into two bots, splitting the DNA of the bot at a certain point and giving each portion to one of the bots. This would be achieved via the sysvar .meio, with the number stored indicating where in the DNA the split occurs

Secondly, give bots the ability to shuffle their genes, moving genes around their genome. This way they can 'line up' their genes into two halves for meiosis.
Bots should also have the ability to copy their own genes. The importance of this will become clear later.

Finally, recombination: two bots can, via a tie, merge into one bot and merge their DNA together, with the bot that initialized the recombination's DNA simply being added to the end of the other bot's DNA
.reco will fire a tie forward, which, if it hits anything, will initiate recombination

*********

Here's how sexual reproduction would work:

First, line up your genes into two halves (if necessary)
Next, split your DNA using meiosis
Then, have each 'half-bot' to find a 'half-bot' of the 'opposite sex'
Then, have the halves recombine into one bot

It works better still for MBs, which can afford to have specialist sex cells

*********

Here's the best part - implications of this system:

1) Different species cannot breed easily

If one bot accidentally recombines with a bot of a different species you're left with a bizare hybrid, which will probably be missing large chunks of vital DNA, and be run by two conflicting DNA 'half-bot' programs

2) Viruses emergy as a natural consequence

A virus would be a single gene (or several) that copied itself, sent the copy to the end of the genome, and split it off with meiosis, only to have the new SG bot recombine with the first bot it saw.

3) Incresed evolutionary benifits of sex, including true speciation

*********

What does everybody think of that?

24

Biology / Laugh at the ID-ers

« on: June 30, 2006, 02:48:17 PM »

http://www.socialaffairsunit.org.uk/blog/archives/000427.php

I had a guy throw this at me while debating about evolution

The author of the article has absolutely no idea about how evolution works, and spouts nearly every fallacy about evolution ever concieved. He drivels on about cats giving birth to kangeroos and racoons and how eyes form from nowhere, and cites this as proof that evolution is flawed.

Reliable source ... er, no

Point and laugh people  

So what fallacies and stupidity have you seen when debating with ID-ers/creationists (I know PY does this quite a lot)

25

Suggestions / Discussion: Ties

« on: June 23, 2006, 11:43:40 AM »

Relevant old topics:
New Tie Structures
Tieports

Nums' new tie paradigm for the C++ version

We need a new system of controling ties for the Visual Basic version. The curent system is very unwieldy, can only control one tie at once, and isn't very powerful or simple to program/evolve.
For example, to transfer energy through a tie under the current system you must:
[tie phase] .tienum store
-1 .tieloc store
[nrg transfer value] .tieval store
Which is far too much for such a simple action IMO, and nearly impossible to evolve

What's the community consensus on this?

Discuss  

26

Bugs and fixes / Ties in 2.4

« on: June 15, 2006, 11:23:02 AM »

Here's what's working:
.tie
.deltie
.tienum
.fixlen
.sharenrg and other sharing commands
.tieloc
.tieval
.fixang

Here's what's not working:
.tielen1-4
.tielen
.tieang1-4
.tieang

But the sysvars .tieang1-4 are really useful for organising MBs

Also, the vision grid doesn't show ties underneath it, which is kinda annoying

There also seems to be no limit to the number of ties a bot can have, which might pose a problem for fixing the tie ports

EDIT: In fact, the limit is 9 ties

27

Bot Tavern / Reciprocating bots

« on: June 14, 2006, 03:08:57 PM »

Prompted by an idea Endy had in another thread

Any ideas?

Here's an idea: Have a bot spin until it sees something, and then check the angle that the aother bot is pointed at. If the other bot is pointed in its direction, it 'locks on' until the bot is pointing elsewhere or out of range

If attacked (*.shang and *.shflav) it could spin to face the attacker

If the bot it is looking at has no eyes then it could assume the bot is a veg and attack anyway

Thoughts? Ideas?

It would be great for resisting cannibotism, and would be a much more realistic conspec recognition for evosims. The reason we're getting cannibotism is that bots aren't retaliating if a conspec attacks them, giving cannibots a huge advantage.

28

Tips and Tricks / Vision and Turning

« on: June 06, 2006, 02:13:08 PM »

Here's a handy chart:

Code: [Select]

eye1    -140
eye2    -105
eye3    -70
eye4    -35

eye5    0

eye6    35
eye7    70
eye8    105
eye9    140

The number on the right is the clockwise turn you need to make, in DB turning units, so that you turn towards the eye listed on the left

So, if you want to turn towards your eye9 then just store 140 in .aimdx

The vision range of a bot is a 90 degree arc with the eyes spaced 10 degrees apart

In DB units:
360 degrees = 1256
90 degrees = 314
10 degrees = 35

29

Suggestions / Waste

« on: June 03, 2006, 09:08:28 AM »

I was digging through some old posts, and found this:

Do you think I should make waste a mutagen? So building up waste causes more mutations? That would be cool, and I'd be all for it. Sort of random

Excellent suggestion - go Num  

No that we've got point mutations in 2.4 then you could have waste simply accelerate the point mutations and eventually trash the DNA. Maybe this would be a better solution than altzimers?

So once the waste reaches the waste threshold the point mutations are multiplied until the bot's DNA gets mutilated beyond repair until the bot dies.

What do you think?

30

Veggies / Alga Cohesum (V)(Elite)-01.06.06

« on: June 01, 2006, 11:21:52 AM »

' Alga Cohesum
'
' Forms large colonies
' and sends out spores
' to form new colonies

cond
 *.robage 0 =
start
 .tie inc
 0 .fixpos store
 100 .tielen1 store
 100 .tielen2 store
 100 .tielen3 store
 100 .tielen4 store
stop

cond
 *.robage 2 >
 *.velscalar 0 =
 *.fixed 0 =
start
 1 .fixpos store
stop

cond
 *.robage 5 >
 *.fixed 1 =
start
 .repro inc
 1256 rnd .setaim store
stop

end