Messages

76

F1 bots / Rabidus (F1)(Vir)(Googlyeyesultra)-01.07.07

« on: July 13, 2007, 04:54:57 PM »

Settings should all be defaulted when you tick 'F1 conditions'.

I'll try the sim and see if I can find what is causing the difference.

77

F1 bots / Rabidus (F1)(Vir)(Googlyeyesultra)-01.07.07

« on: July 13, 2007, 09:04:25 AM »

Are you sure this is the same bot? First run I tried it lost to speedy and this time took 60+ matches to beat Ratita Australopitecus (30th place) before losing to speedy again, that would place it at 27th.

This is F1 conditions, run F1 league on start, DB v 2.43, using latest league download.

78

Off Topic / What happened to patch someone posted?

« on: July 11, 2007, 04:44:04 AM »

Somebody (excuse me for not remembering who) posted a patch for pond mode? Was when we had a little problem with forum and lost posts, just wondering what happened to it seeing as that post was lost when forum was reset.  

79

Bug reports / "Statistical Draw" after 3 out of 3 rounds with one victor

« on: July 11, 2007, 04:27:53 AM »

Here's a post where Nums explains how it's worked out; http://www.darwinbots.com/Forum/index.php?...dpost&p=1369961 plus he links it to a Wikipedia entry explaining more about the method.

Abyaly is absolutely right, 5 matches is the minimum you need atm to get a statistically reliable result (to 95% confidence).

As long as you start the league with 5 matches it will keep the minimum amount of matches at 5 even if a pair of bots fight it out for longer. It's matches that aren't league but are under F1 conditions that causes the increased minimum match amount.
(Am so used to seeing that I had completely forgotten about it! Will add it to list though).

If you turn off the graphical display when running leagues you hopefully won't notice the time it takes to run those extra two matches.

Maybe one day we'll be able to adjust the confidence level so we can run quicker tests but that'd be a future project for a programmer.  

80

Bugs and fixes / Known Problems with DB v 2.43

« on: July 10, 2007, 05:54:33 PM »

*.refpoison shows amount of poison in bot you see not the amount of poison commands it has.

Bots with DNA execution disabled have 0,0 for xpos,ypos, which results in refxpos, refypos being 0,0 when looking at them.

Lines that start '# (instead of just ' for commentary) cause an 'Invalid procedure call or argument' (quote Numsgill - # are for hashes. When you save the DNA of a bot a certain way, it'll save information about the bot (number of generations, etc.) as a hash. /quote)

Number of rounds in a contest is increased automatically when running under F1 conditions but without league if a pair of bots fight for longer than the rounds set. This increase applies to all following matches until reset manually.

- Mydn, mysx, refdn, and refsx all return 0 all the time, making them useless for determining if a bot is friendly. Also, refvelsx always reads 0 too, but refveldx works properly (so we don't really NEED refvelsx, since veldx is always equal to -velsx). (refveldn and refvelup both work though) Notably, velsx (getting my own sx velocity) works fine.

Of course, refdn and refsx may not actually be broken - they're just reading from broken mydx and mysx on the target bot.

Also (not really bug but); the def array (where you set def variable's) in bot programme is probably limited to 50 def's

Calling store with an address of 0 results in the value (not the address) remaining on the stack.
(I.e. - 99 0 store - leaves 99 as the top stack value)

The virus shot flies off in a random direction, and completely ignores aim commands.

There are a couple of 'instakill' methods, not really in the spirit of the game but it has been a long tradition to allow such things until they are fixed - I'll let you find out how to do them by reading the forum!

-bugs spotted by Trafalgar

81

Tips and Tricks / Hibernation

« on: July 10, 2007, 05:34:58 PM »

Was just a thought; the info on 1G bots you come up with is miles above the information (in the wiki) that is readily available atm.
The Wiki always seems to need updating though so if you don't get time for it then don't worry!  

82

Tips and Tricks / Hibernation

« on: July 09, 2007, 06:25:25 PM »

What are the chances of luring you into rewriting the conditionless bots part of the Wiki Abylaly? Apart from being (IMO anyway) the best 1g bot writer out there you also have a really simple and accessible way of explaining how it's done.
Not to mention your skill with viruses and venom...  

That post above showing how to convert a standard bot into a 1G bot really deserves to be added to the Wiki at least.  

83

Bugs and fixes / Known Problems with DB v 2.43

« on: July 08, 2007, 02:31:32 PM »

Robot data display shows slime when you make shell and vice versa.

-Thanks to Abyaly for correction and to whomever spotted bug first

84

Off Topic / Latest league update

« on: July 08, 2007, 10:43:09 AM »

Una 3.0 failed to get any further than last place in the shortbot league. (26th postion)

T. mosquito refuses to be entered in the SB league due to the fact that it causes the league to not accept it as a proper match!!? (Keeps beating Una 3.0 atm but once 5 rounds are won it replay’s the match...) Will have another go at sorting soon.

Originaloc, MB league, doesn’t do anything. It beat Vermis Inchy and Hexagonis though before losing to Botumor. (Looks like Botumor may be cheating a bit by feeding as a single cell but it gets to stay)

Abyaly’s Aura shot to the top of the SB league as expected, giving it a first in the F1, F2 and SB league! (This was before Etch, yet another bot by Abyaly took first in the F1 league).
One Man Bucket, also by Abyaly got all the way to 19th position in the SB league before losing to Dagblad Activia. (It reached postion 15 in the F2 league). That might not seem that impressive from somebody of Abyaly’s bot writing skill but you’ve got to remember that One Man Bucket has only ONE STORE instruction in the entire dna... Check out the expanded version under the same name in the bestiary if you want to see how it works a bit more clearly.
Spiral, an Interesting Behaviour bot, also by Abyaly, knocked Simplebot20 off the F2 league table and reached position 13 before losing to Swarm 2.0. In the F1 league it knocked Animal Minimaxis2 out and reached position 26 before losing to Proeliator Gryphus.

Speedy reached position 7 before losing to Furiae in the SB league, while in the F1 league it knocked Devincio Eversor out of the league and reached position 26 before also losing to Proeliator Gryphus.

Martian Tank2 knocked Umbra Draconis out of the F2 league and reached position 7 before losing to Excalibur. In the F1 league it knocked Vex Pefidiosos out and reached position 27 before losing to Speedy.
Martian Tank3 knocked J. Venomous Mobber out of the F2 league and ended up losing to Excalibur 1.12, it reached position 3. In the F1 league it also reached position 27 before losing to Speedy, it knocked out Virulus Spinnens.

Ratita australopetecus knocked Virulus Discretus out of the F1 league but only reached position 30 before losing to one of the other new entries, Spiral. (It is worth remembering that Tie bots are having problems feeding through ties at the moment and that probably didn’t help).

New downloads for these leagues are all now available. If I missed testing one of your bots then say so and I'll get it sorted.

85

Bot Tavern / Where have all the bots gone?

« on: July 07, 2007, 03:07:48 PM »

Bots that are to be entered into the leagues are supposed to stay queued up in "The Starting Gate" until they are run in the leagues. It is then the league manager's responsibility to move the DNA file into the respective league folder.

Please do not move DNA files that are intended to be league entries until they have been run in the leagues, otherwise we can't tell which ones are not entered yet.

If your bot has just disappeared from the Starting Gate and you are wondering where it went because the league listings haven't changed yet then don't worry. If it's been moved then it should be in the bestiary and a new league listing will be posted as soon as all the bots have been tested.   (It can take several days to run all the bots)!

86

DNA - General / Is there anyway for a bot to rewrite his DNA?

« on: July 06, 2007, 01:20:59 PM »

87

Darwinbots3 / New bot shape

« on: July 06, 2007, 01:12:36 PM »

3D would be very cool!

I think the second example of pivots looks pretty good, cells join on the outside of their surfaces. The first example, if in a 3D enviroment, makes me think the bot is linked on the outside of the cell still but our POV hides the link.

Your new shape for the bots looks like an extended circle, might be an idea to allow bots to modify their proportion along the centre axis as well and gain another means of propulsion. (change from circle to extendable cylinder at will while maintaining overall volume).  Unfortunately I'm not smart enough to do a pictoral example of that so I'll leave that idea to your imagination. Was thinking of muscle cells maybe...

88

Bot Tavern / Martian Tank

« on: July 04, 2007, 05:47:07 PM »

Martin Welgemoed is a Dutch name, are you from Holland?  

If the conditions are met then it adds -140 to the top of the stack; later on there is a

cond
start
.aimdx store
stop

gene which takes the top number off the stack and uses that to set the .aimdx value.

It's a cost cutting method that avoids using store too often. (First used by PY).

(I should have this tested by the end of the week btw Martin, just have to test some other bots first!)

89

Off Topic / MB league rules

« on: July 04, 2007, 05:39:28 PM »

Try it!

The MB league is quite old now, it hasn't really had many bots entered and with the current problems that v2.43 is causing tie bots it's about time we had a mass extinction event for the MB league.  

If your bot manages to get to the top of the MB league then that's just going to encourage other people to challenge you. If other people challenge you then that's likely to be using different methods. Once we have a few different methods of being a MB fighting it out in the league then we'll get a better idea of how the game is balanced with regards to multi cell bots and can make future changes with that in mind.


Botumor, the bot that stopped the bot I was running through the MB league earlier (Originaloc) didn't even seem to bother about that rule!

When there are lots of bots in the league and we can see what the balance is I'd suggest that that's the time to tighten the rules a bit more.

I think the league with the most rules at the moment is the F2 league, that has two rules... ((IIRC)No ties, no viruses).

90

Off Topic / MB league rules

« on: July 04, 2007, 02:31:07 PM »

Would be nice to have something like that (bang efficiency = 0). In the meantime, having run a bot in the MB league today and seeing how some of the other entries work I'm going to make that first rule even simpler...

MB league rule:
Feeding genes must have *.multi 0 != as condition

The reason for asking this is simply so I don't have to study the code or watch the bot on screen when it is entered into the league.
The reason for not wanting MB's to be able to feed as single celled units is that it gives them a massive initial advantage - even if they do turn into MB's later on in life.