Author Topic: Viruses under control  (Read 2728 times)

Offline shvarz

  • Bot God
  • *****
  • Posts: 1341
    • View Profile
Viruses under control
« on: January 07, 2008, 04:54:11 PM »
You probably remember that my sims were plagued by viruses. There were two versions, one around 68 commands and the other around 21 commands. They were completely out of control, eventually wiping out my bots. Well, some time have passed and that does not happen anymore!  Viruses have not completely disappeared, there are plenty of them in genomes and I see recent infections in mutations, but not nearly as many as I used to.  I have no idea what happened or how.  The sim is attached in case someone wants to try to figure it out.
"Never underestimate the power of stupid things in big numbers" - Serious Sam

Offline Endy

  • Bot Overlord
  • ****
  • Posts: 852
    • View Profile
Viruses under control
« Reply #1 on: January 08, 2008, 06:21:25 AM »
Well, as long as the bots aren't completly wiped out, they can adapt to the viruses. I've had a few sims where the bots develop antiviral dna, from the virus itself.

Offline fulizer

  • Bot Destroyer
  • ***
  • Posts: 198
    • View Profile
    • http://www.freewebs.com/explodeimplode/
Viruses under control
« Reply #2 on: January 08, 2008, 08:50:53 AM »
evolving antivirus genes? that would be so cool to see watching a bot get infected then living even though it was doing nothing
"If this is coffee bring me tea, If this is tea, bring me coffee"

Offline shvarz

  • Bot God
  • *****
  • Posts: 1341
    • View Profile
Viruses under control
« Reply #3 on: January 08, 2008, 12:33:43 PM »
I know little of how bots can defend against viruses. What are the possible antiviral strategies?

Alternative explanation is that the virus became attenuated. It's fairly common in real life, I wonder if that's what happened here.
"Never underestimate the power of stupid things in big numbers" - Serious Sam

Offline shvarz

  • Bot God
  • *****
  • Posts: 1341
    • View Profile
Viruses under control
« Reply #4 on: January 08, 2008, 02:48:23 PM »
In that vein, here's a recent infection:

Quote
cond
 65 *999 !=
 start
 *.thisgene .mkvirus store
 *.myeye *.myshoot add *.myup add 31999 mult 50 4 337

The only thing that I can see is that the vtimer has been removed from the original virus.  Do you think this virus is going to be poor in some way compared to the original one?
"Never underestimate the power of stupid things in big numbers" - Serious Sam

Offline EricL

  • Administrator
  • Bot God
  • *****
  • Posts: 2266
    • View Profile
Viruses under control
« Reply #5 on: January 08, 2008, 02:57:14 PM »
I am currently using a farily strong antiviral strategy.  
As part of each gene N, I include the following DNA (where N is the specific gene number):

*.thisgene N >
*.flag 1 != and
 N .delgene store
1 .flag store

and in last gene

*.genes N >
*.flag 1 != and
N 1 add .delgene store

The .flag location gets set to 0 in the first gene.  This strategy does let the virus run one cycle (it must for the detection to work) but unless the virus tagets the flag location specifically or deletes another gene, this should serve to extracate the virus.




Quote from: shvarz
Do you think this virus is going to be poor in some way compared to the original one?
I makes me wonder whether this virus causes a virus shot to leak.  Does the total number of shots in the sim seem excessive?  Indulge me and try an experiment: Save the sim (so you can go back to it) then increase costs to something ridiculious so that everything dies.  Does the shot counter at the bottom still say there are shots left in the sim?
« Last Edit: January 08, 2008, 02:58:33 PM by EricL »
Many beers....

Offline shvarz

  • Bot God
  • *****
  • Posts: 1341
    • View Profile
Viruses under control
« Reply #6 on: January 08, 2008, 03:23:51 PM »
There is about 1500-2000 shots in the sim.  I did what you said and the number of shots went down together with bots, as soon as everything was dead, shots disappeared too.
"Never underestimate the power of stupid things in big numbers" - Serious Sam

Offline Endy

  • Bot Overlord
  • ****
  • Posts: 852
    • View Profile
Viruses under control
« Reply #7 on: January 09, 2008, 04:34:46 AM »
Vtimer is actually rather useless from a viruses point of view. It mainly just saves energy for the host. (yes, I admit this could help the virus preserve the host, but most of the time they don't seem to care about wasting the bot's energy)

I "think" there may be an internal race going on between the viruses within a bot, prompting a continual reproduction attempt on the part of the viruses. Haven't figured out a good way to actually test my theory though.
 
Kind of surprised the virus hasn't lost the rest of it's "junk" dna. You'd think that a streamlined virus would be more favored. Might be another virus out there though, or prehaps the virus found it better to let it's hosts live longer.

Quote
I know little of how bots can defend against viruses. What are the possible antiviral strategies?

Had one species evolve to store a huge negative number into vshoot(the viruses used .vshoot inc). For a long time these bots were able to take virus shots without making any more. Eventually though, the infected veggies allowed the virus to overpower the bot's defenses.

Might be cool to create a virus that attempts to delete other viruses in a bot.

Offline Botsareus

  • Society makes it all backwards - there is a good reason for that
  • Bot God
  • *****
  • Posts: 4483
    • View Profile
    • DJ Paul Kononov
Viruses under control
« Reply #8 on: January 10, 2008, 02:28:23 PM »
Quote
*.thisgene N >
*.flag 1 != and
N .delgene store
1 .flag store

and in last gene

*.genes N >
*.flag 1 != and
N 1 add .delgene store

Thats a bit problamatic in my system where each condition command starts a new gene and "As part of each gene N" simply does not work:
ex:

*.nrg 100 >
5 .up store
 *.thisgene N >
*.flag 1 != and
N .delgene store
1 .flag store

is actualy two genes...

see: http://www.darwinbots.com/Forum/index.php?...opic=2299&st=15 for details on what I mean...
« Last Edit: January 10, 2008, 02:41:15 PM by Botsareus »

Offline EricL

  • Administrator
  • Bot God
  • *****
  • Posts: 2266
    • View Profile
Viruses under control
« Reply #9 on: January 10, 2008, 03:48:39 PM »
Quote from: Botsareus
Thats a bit problamatic in my system where each condition command starts a new gene and "As part of each gene N" simply does not work:...

Well, since your "system" is fictional, there's not really much of a probelm now is there?
Many beers....

Offline Botsareus

  • Society makes it all backwards - there is a good reason for that
  • Bot God
  • *****
  • Posts: 4483
    • View Profile
    • DJ Paul Kononov
Viruses under control
« Reply #10 on: January 17, 2008, 01:36:28 PM »
Good point there Eric, but I figured it out: I'll use atg_start , atg_stop to attach code to genes; see: http://www.darwinbots.com/Forum/index.php?showtopic=2480
« Last Edit: January 17, 2008, 01:51:06 PM by Botsareus »

Offline fulizer

  • Bot Destroyer
  • ***
  • Posts: 198
    • View Profile
    • http://www.freewebs.com/explodeimplode/
Viruses under control
« Reply #11 on: January 24, 2008, 05:01:36 AM »
yes a virus that rather than killing the bot enslaves it then forces it to feed the shooter then hunt look for another member of the species that infected once it got some energy and feed that.
has anyone ever done that?
"If this is coffee bring me tea, If this is tea, bring me coffee"