2010-10-02T00:03:08  *** Zeiris <Zeiris!~Zeiris@d206-116-158-135.bchsia.telus.net> has joined #aichallenge
2010-10-02T00:05:16  *** delt0r___ <delt0r___!~delt0r@188-22-161-104.adsl.highway.telekom.at> has joined #aichallenge
2010-10-02T00:07:32  *** delt0r_ <delt0r_!~delt0r@188-22-18-134.adsl.highway.telekom.at> has quit IRC (Ping timeout: 276 seconds)
2010-10-02T00:17:55  *** amstan_ <amstan_!~alex@75-119-241-68.dsl.teksavvy.com> has joined #aichallenge
2010-10-02T00:17:55  *** ChanServ sets mode: +o amstan_
2010-10-02T00:19:05  *** amstan <amstan!~amstan@76-10-164-254.dsl.teksavvy.com> has quit IRC (Remote host closed the connection)
2010-10-02T00:19:11  *** amstan_ is now known as amstan
2010-10-02T00:23:40  *** krokokrusa <krokokrusa!~zvold@89.31.118.163> has quit IRC (Ping timeout: 252 seconds)
2010-10-02T00:24:13  *** krokokrusa <krokokrusa!~zvold@89.31.118.163> has joined #aichallenge
2010-10-02T00:44:25  <bhasker> http://72.44.46.68/canvas?game_id=266834
2010-10-02T00:47:59  <Sylph> :|
2010-10-02T00:48:00  <Sylph> I want to play
2010-10-02T00:48:06  <Sylph> i need to add my defensive capability first though
2010-10-02T00:48:16  <Sylph> just finished the attacking function
2010-10-02T00:48:35  <Sylph> need to work on defense, then maybe front line defense stuff, then it'll be usable
2010-10-02T00:52:10  *** Cyndre <Cyndre!~Cyndre@DOCSIS-181-75.cable-lynx.net> has quit IRC (Quit: Leaving)
2010-10-02T00:58:39  <contestbot> Top 10 players: Hazard(3805), cfaftw(3644), rsergio(3638), EBraun(3633), felixcoto(3623), dmj111(3622), sequoh(3609), _iouri_(3601), davidjliu(3555), sighbot(3538)
2010-10-02T01:20:44  *** Queue29 <Queue29!~Queue29@cpe-70-112-194-192.austin.res.rr.com> has quit IRC (Remote host closed the connection)
2010-10-02T01:58:41  <contestbot> Top 10 players: Hazard(3820), EBraun(3671), cfaftw(3663), rsergio(3654), felixcoto(3637), dmj111(3633), sequoh(3619), _iouri_(3605), davidjliu(3563), sighbot(3549)
2010-10-02T02:12:24  *** HakanD <HakanD!~hakand@85.105.10.191> has joined #aichallenge
2010-10-02T02:12:43  <HakanD> morning
2010-10-02T02:12:50  <antimatroid2> morning
2010-10-02T02:13:00  <antimatroid2> i just beat dmj for a second time finally
2010-10-02T02:15:55  <Sylph> Grats :)
2010-10-02T02:16:04  <Sylph> I can't wait to actually put in my bot >.<
2010-10-02T02:16:12  <Sylph> working on the defense now :/
2010-10-02T02:16:25  <Sylph> Trying to think of the perfect/best way to do it.
2010-10-02T02:23:26  <antimatroid2> man, if you make one bad move against dmj he just pounces
2010-10-02T02:23:31  *** yasith <yasith!~yasith@112.135.84.49> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T02:23:40  *** yasith_ <yasith_!~yasith@112.135.84.49> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T02:27:14  <Sylph> lol
2010-10-02T02:27:15  <Sylph> yea
2010-10-02T02:27:49  <Sylph> my bot only attack atm, need to get it to defend then it'd be perfect
2010-10-02T02:28:14  <Sylph> still trying to figure out the best method that take into acccount every possibility :|
2010-10-02T02:29:11  <HakanD> and I broke the 30 wpm barrier (:
2010-10-02T02:29:54  <Sylph> lol
2010-10-02T02:29:56  <Sylph> grats
2010-10-02T02:30:42  <Sylph> you are making me switch back to colemak :)
2010-10-02T02:32:15  *** Cyndre <Cyndre!~Cyndre@DOCSIS-181-75.cable-lynx.net> has joined #aichallenge
2010-10-02T02:32:24  <HakanD> hehe\
2010-10-02T02:34:13  <Sylph> I can't deal with less than 60-80 wpm atm though :|
2010-10-02T02:35:43  <HakanD> well, 30 wpm is enough for me to switch completely
2010-10-02T02:36:08  <HakanD> i just need  1-2 weeks i think
2010-10-02T02:36:20  *** yasith_ <yasith_!~yasith@112.135.91.108> has joined #aichallenge
2010-10-02T02:36:35  <HakanD> to get my full speed back
2010-10-02T02:36:56  <Sylph> I reserve the right to whine on the difficulty of making a perfect bot :p
2010-10-02T02:36:58  <Sylph> mm
2010-10-02T02:36:59  *** yasith <yasith!~yasith@112.135.91.108> has joined #aichallenge
2010-10-02T02:37:13  <Sylph> It's near perfect on attacking now, just need defense :|
2010-10-02T02:39:30  *** rul <rul!~rul@190.188.115.15> has quit IRC (Ping timeout: 265 seconds)
2010-10-02T02:40:03  <HakanD> and than your defense strategy probably break your attack strategy (:
2010-10-02T02:40:07  <Sylph> nah
2010-10-02T02:40:29  <Sylph> a few simpler methods would work with no problem on attack strategy
2010-10-02T02:40:34  <Sylph> on my attack strategy
2010-10-02T02:40:48  *** rul <rul!~rul@190.188.115.15> has joined #aichallenge
2010-10-02T02:41:14  <Sylph> Problem is thinking of the perfect one
2010-10-02T02:41:19  <Sylph> like imagine you got 20 ships at planet 1
2010-10-02T02:41:26  <HakanD> there is a balance between perfect attack and perfect defense, can't have both (:
2010-10-02T02:41:38  <Sylph> growth rate is 5 per turn, making it 70 ships in 10 turns.
2010-10-02T02:41:54  <Sylph> enemy coming in with 70 ships
2010-10-02T02:42:04  <Sylph> most defense codes probably would just sit and do nothing
2010-10-02T02:42:14  <HakanD> yea
2010-10-02T02:42:17  <Sylph> I want to be able to move the ships around nearby planets to defend othe planets if needed
2010-10-02T02:42:27  <Sylph> before it have to be located in that planet.
2010-10-02T02:42:39  <Sylph> A lil complex to think about, hrm
2010-10-02T02:43:26  <Sylph> nm, that might be easy with my future projection as it is
2010-10-02T02:43:41  <Sylph> there's the next step too, possible ships that enemy  will send in to reinforce
2010-10-02T02:44:09  <Sylph> and yeah on the balance, I was talking about perfect attack within the constrain of perfect defense :p
2010-10-02T02:49:14  <Sylph> http://72.44.46.68/canvas?game_id=267909  -> you'll notice that I always take the planets I attack
2010-10-02T02:49:20  <Sylph> no matter what
2010-10-02T02:49:29  <Sylph> just need to add defense algorithm atm, still trying to think of the best one
2010-10-02T02:49:31  <Sylph> method
2010-10-02T02:49:36  *** sinan <sinan!~sinan@87.238.132.0> has joined #aichallenge
2010-10-02T02:49:52  <Sylph> well, it doesn't work when enemy take my planets before I send the units out :/
2010-10-02T02:49:56  <Sylph> that part isn't written yet.
2010-10-02T02:58:41  <contestbot> Top 10 players: Hazard(3794), rsergio(3678), EBraun(3669), cfaftw(3643), felixcoto(3629), dmj111(3610), sequoh(3602), _iouri_(3572), davidjliu(3551), sighbot(3540)
2010-10-02T03:00:00  *** JCS^ <JCS^!jcs@201-92-156-159.dsl.telesp.net.br> has quit IRC (Quit: \x00)
2010-10-02T03:04:32  <Sylph> meh i'll just go simple for this defense part >.<
2010-10-02T03:04:50  *** keyne <keyne!~AndChat@203.236.22.120> has quit IRC (Quit: Bye)
2010-10-02T03:05:09  *** keynes <keynes!~AndChat@203.236.22.120> has joined #aichallenge
2010-10-02T03:18:45  *** Naith <Naith!~nickbarth@S0106001b38137d1b.ed.shawcable.net> has joined #aichallenge
2010-10-02T03:32:17  <HakanD> KISS is a good philosophy (:
2010-10-02T03:32:29  <Sylph> especially when it start to give me headache :p
2010-10-02T03:32:37  <HakanD> hehe
2010-10-02T03:32:39  <Sylph> I think I got an idea of how complex I want to make it.
2010-10-02T03:41:33  *** hellman <hellman!~hellman@nat232-248-205-109.tvoe.tv> has joined #aichallenge
2010-10-02T03:52:37  *** Appleman1234 <Appleman1234!~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au> has quit IRC (Remote host closed the connection)
2010-10-02T03:58:41  <contestbot> Top 10 players: Hazard(3811), rsergio(3714), cfaftw(3665), EBraun(3664), felixcoto(3642), dmj111(3628), sequoh(3620), _iouri_(3585), davidjliu(3559), sighbot(3557)
2010-10-02T04:13:16  *** sinan <sinan!~sinan@87.238.132.0> has quit IRC (Ping timeout: 265 seconds)
2010-10-02T04:22:55  *** Sylph <Sylph!~x@adsl-76-234-137-205.dsl.sfldmi.sbcglobal.net> has quit IRC (Ping timeout: 265 seconds)
2010-10-02T04:26:48  *** Sylph <Sylph!~x@adsl-76-234-137-205.dsl.sfldmi.sbcglobal.net> has joined #aichallenge
2010-10-02T04:30:13  <Sylph> half done with the defense algorithm :)
2010-10-02T04:30:24  <HakanD> nice
2010-10-02T04:30:29  <Sylph> already winning fights on the tcp without defense though
2010-10-02T04:30:55  <Sylph> can't wait til I get done with the defense :)
2010-10-02T04:30:59  <Sylph> it's simple and stupid though :/
2010-10-02T04:31:02  <Sylph> May work on it again later
2010-10-02T04:31:14  <HakanD> and im stabled around 27 wpm (:
2010-10-02T04:31:17  <Sylph> Grats
2010-10-02T04:31:20  <Sylph> are you on windows or linux
2010-10-02T04:31:27  <Sylph> linux have a super nice app that got me to 50+ wpm on dvorak
2010-10-02T04:31:39  <Sylph> I almost switched completely
2010-10-02T04:31:43  <HakanD> linux now
2010-10-02T04:31:45  <Sylph> Then realized..... it was useless
2010-10-02T04:31:58  <HakanD> which app?
2010-10-02T04:31:59  <Sylph> ctrl+c/ctrl+v/ctrl+t/etc are not at convenient places
2010-10-02T04:32:04  <Sylph> I forgot, hrm
2010-10-02T04:32:09  <Sylph> it was a command line i found on the apt-get :|
2010-10-02T04:32:28  <Sylph> should be on freebsd too
2010-10-02T04:32:32  <HakanD> dvorak7min?
2010-10-02T04:32:34  <Sylph> search synaptic or whatever you use
2010-10-02T04:32:34  <Sylph> ah yeah
2010-10-02T04:32:38  <Sylph> i think that's the one i liked
2010-10-02T04:32:45  <Sylph> i remember the 7 letter
2010-10-02T04:32:50  <HakanD> Yep its cool
2010-10-02T04:32:52  <Sylph> was about the only thing i remember
2010-10-02T04:32:56  <HakanD> (:
2010-10-02T04:33:41  <HakanD> are you using colemak now?
2010-10-02T04:33:45  <Sylph> nope lol
2010-10-02T04:33:48  <Sylph> i don't like <30 wpm :|
2010-10-02T04:33:52  <Sylph> especially before I'm done coding this
2010-10-02T04:34:01  <Sylph> i am now
2010-10-02T04:34:18  <Sylph> but i wish i got dvorak7min for windows
2010-10-02T04:34:27  <Sylph> so slow :/
2010-10-02T04:35:00  <HakanD> have tons of work to do still
2010-10-02T04:35:24  <HakanD> I need to learn emacs also
2010-10-02T04:35:30  <Sylph> ah
2010-10-02T04:35:46  <Sylph> I am hoping my algorithm can get me top 10 place :/
2010-10-02T04:36:03  <Sylph> But still like 3-5 functions to do before what I am hoping. >.<
2010-10-02T04:36:19  <Sylph> adding defense on this should get me to pretty good place though.
2010-10-02T04:36:52  <HakanD> hope so
2010-10-02T04:37:06  *** choas <choas!~lars@p578F6CD8.dip.t-dialin.net> has joined #aichallenge
2010-10-02T04:37:21  <HakanD> good competitors is always good (:
2010-10-02T04:37:29  <Sylph> yeah :)
2010-10-02T04:38:09  <Sylph> I'm still liking to see the wins with just attacking lol
2010-10-02T04:38:12  <Sylph> http://72.44.46.68/getplayer?player=Sylph-CSharp-AttackOnly
2010-10-02T04:38:41  <Sylph> i wonder how many of the bots i won against (other than ragebot obviously) is not just something people mess around with
2010-10-02T04:40:59  *** contestbot_ <contestbot_!~supybot@75-119-241-68.dsl.teksavvy.com> has joined #aichallenge
2010-10-02T04:40:59  *** contestbot <contestbot!~supybot@75-119-241-68.dsl.teksavvy.com> has quit IRC (Disconnected by services)
2010-10-02T04:42:45  *** Cyndre <Cyndre!~Cyndre@DOCSIS-181-75.cable-lynx.net> has quit IRC (Ping timeout: 265 seconds)
2010-10-02T04:42:45  *** contestbot_ is now known as contestbot
2010-10-02T04:42:45  *** ChanServ sets mode: +o contestbot
2010-10-02T04:42:48  <HakanD> btw, why you are not supplying the front planets?
2010-10-02T04:42:57  <Sylph> that's after defense
2010-10-02T04:43:08  <Sylph> i only finished the attack algorithm so far
2010-10-02T04:43:16  <HakanD> k
2010-10-02T04:43:16  <Sylph> working on defense, then front line
2010-10-02T04:44:34  <Sylph> wanted to test out my future prediction algorithms, and attacking/defending is pretty much the first to go with that  :)
2010-10-02T04:46:42  *** iFire <iFire!~kittens@unaffiliated/ifire> has quit IRC (Read error: Connection reset by peer)
2010-10-02T04:47:06  <Sylph> it'll be plugable easily though :)
2010-10-02T04:47:13  <Sylph> anyway, nite nite
2010-10-02T04:47:19  <Sylph> sleep time now
2010-10-02T04:48:28  <HakanD> night
2010-10-02T04:48:39  <Sylph> changed mind, don't want to sleep yet :|
2010-10-02T04:48:39  <Sylph> hrm
2010-10-02T04:48:51  *** jesionaj <jesionaj!~dongs@174-18-5-185.tcso.qwest.net> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T04:52:05  *** sinan <sinan!~sinan@82.212.70.90> has joined #aichallenge
2010-10-02T04:54:55  <HakanD> lunch time for me (:
2010-10-02T04:55:09  <Sylph> cya :)
2010-10-02T04:55:13  <Sylph> have fun
2010-10-02T04:57:18  *** keynes <keynes!~AndChat@203.236.22.120> has quit IRC (Quit: Bye)
2010-10-02T04:58:41  <contestbot> Top 10 players: Hazard(3822), rsergio(3704), cfaftw(3671), EBraun(3671), felixcoto(3655), dmj111(3638), sequoh(3629), _iouri_(3594), davidjliu(3573), sighbot(3564)
2010-10-02T05:01:31  *** yasith <yasith!~yasith@112.135.91.108> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T05:02:02  *** yasith_ <yasith_!~yasith@112.135.91.108> has quit IRC (Ping timeout: 264 seconds)
2010-10-02T05:08:58  *** pgpaskar_ <pgpaskar_!~phor@nat/uwaterloo/x-wauebiznthmjbesq> has quit IRC (Ping timeout: 245 seconds)
2010-10-02T05:09:48  *** pgpaskar_ <pgpaskar_!~phor@nat/uwaterloo/x-cmcefygwbdttjobl> has joined #aichallenge
2010-10-02T05:13:52  *** Blkt <Blkt!~user@93-33-140-101.ip44.fastwebnet.it> has joined #aichallenge
2010-10-02T05:15:14  <Sylph> done with basic defense, next is front line stuff :|
2010-10-02T05:26:24  *** BtbN <BtbN!~BtbN@p54BB7467.dip.t-dialin.net> has joined #aichallenge
2010-10-02T05:34:50  *** iFire <iFire!~kittens@unaffiliated/ifire> has joined #aichallenge
2010-10-02T05:37:26  <Sylph> oh you're on it, bhasker  :p
2010-10-02T05:37:33  <Sylph> Wish my bot is closer to done ..<
2010-10-02T05:37:39  <Sylph> barely half way atm, if even :/
2010-10-02T05:58:42  <contestbot> Top 10 players: Hazard(3817), rsergio(3671), EBraun(3668), cfaftw(3666), felixcoto(3649), dmj111(3634), sequoh(3624), _iouri_(3591), davidjliu(3562), sighbot(3558)
2010-10-02T06:04:51  *** Javache <Javache!~javache@unaffiliated/javache> has left #aichallenge
2010-10-02T06:06:31  *** wh1teside <wh1teside!~admin@adsl-70-131-113-154.dsl.emhril.sbcglobal.net> has quit IRC (Read error: Connection reset by peer)
2010-10-02T06:06:53  *** sinan <sinan!~sinan@82.212.70.90> has quit IRC (Ping timeout: 252 seconds)
2010-10-02T06:07:12  *** RainCT <RainCT!~RainCT@ubuntu/member/pdpc.student.rainct> has joined #aichallenge
2010-10-02T06:08:03  *** wh1teside <wh1teside!~admin@adsl-70-131-113-125.dsl.emhril.sbcglobal.net> has joined #aichallenge
2010-10-02T06:18:42  *** Zeiris <Zeiris!~Zeiris@d206-116-158-135.bchsia.telus.net> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T06:33:26  *** sinan <sinan!~sinan@94.249.36.78> has joined #aichallenge
2010-10-02T06:56:20  *** HakanD <HakanD!~hakand@85.105.10.191> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T06:56:56  *** DanielVF <DanielVF!~DanielVF@67-197-185-240.dyn.comporium.net> has joined #aichallenge
2010-10-02T06:58:34  *** smellyhippy <smellyhippy!~asmodean@about/csharp/regular/smellyhippy> has joined #aichallenge
2010-10-02T06:58:43  <contestbot> Top 10 players: Hazard(3827), rsergio(3679), cfaftw(3675), EBraun(3664), felixcoto(3657), dmj111(3642), sequoh(3632), _iouri_(3601), davidjliu(3570), sighbot(3567)
2010-10-02T07:08:38  <antimatroid2> sigh: you got a new bot for the weekend yet?
2010-10-02T07:09:10  <sigh> nah, my ideas keep sucking, so taking a break
2010-10-02T07:09:41  <antimatroid2> yeah fair enough
2010-10-02T07:11:57  <antimatroid2> zaphus's new bot beat me
2010-10-02T07:18:00  *** megalomania <megalomania!~Didi@78-105-207-174.zone3.bethere.co.uk> has joined #aichallenge
2010-10-02T07:22:58  <antimatroid2> awww, bitmetv is down :(
2010-10-02T07:27:41  *** Guest95353 <Guest95353!kay@free.blinkenshell.org> has quit IRC (Remote host closed the connection)
2010-10-02T07:30:02  *** megalomania <megalomania!~Didi@78-105-207-174.zone3.bethere.co.uk> has left #aichallenge
2010-10-02T07:42:36  *** sinan <sinan!~sinan@94.249.36.78> has quit IRC (Ping timeout: 252 seconds)
2010-10-02T07:55:06  *** Itkovian <Itkovian!~Itkovian@2002:4e16:e2a0:0:21f:5bff:fec8:6e2c> has joined #aichallenge
2010-10-02T07:57:10  *** Migi32 <Migi32!~migi@d54C128C5.access.telenet.be> has joined #aichallenge
2010-10-02T07:58:43  <contestbot> Top 10 players: Hazard(3823), rsergio(3692), cfaftw(3667), EBraun(3659), felixcoto(3651), dmj111(3638), sequoh(3630), _iouri_(3594), ruilov(3571), davidjliu(3567)
2010-10-02T08:00:18  *** dmj111 <dmj111!~user@pool-108-1-247-64.altnpa.east.verizon.net> has joined #aichallenge
2010-10-02T08:03:43  *** Itkovian <Itkovian!~Itkovian@2002:4e16:e2a0:0:21f:5bff:fec8:6e2c> has quit IRC (Quit: Itkovian)
2010-10-02T08:05:16  *** delt0r_ <delt0r_!~delt0r@188-22-161-210.adsl.highway.telekom.at> has joined #aichallenge
2010-10-02T08:06:59  *** delt0r___ <delt0r___!~delt0r@188-22-161-104.adsl.highway.telekom.at> has quit IRC (Ping timeout: 272 seconds)
2010-10-02T08:07:49  *** FlagCapper <FlagCapper!45a5adac@gateway/web/freenode/ip.69.165.173.172> has joined #aichallenge
2010-10-02T08:14:42  *** Naith <Naith!~nickbarth@S0106001b38137d1b.ed.shawcable.net> has quit IRC (Quit: Naith)
2010-10-02T08:29:33  *** ichti <ichti!~ichti@x1-6-00-24-b2-9f-a2-1a.k429.webspeed.dk> has joined #aichallenge
2010-10-02T08:38:16  <Migi32> is there anyone who has made a decent minimax bot?
2010-10-02T08:40:55  <Migi32> or is there anyone who has made a decent bot that would play (near) perfect given infinite computing time?
2010-10-02T08:42:26  *** jimmerton <jimmerton!~jamespick@142.166.63.87> has joined #aichallenge
2010-10-02T08:48:26  *** sinan <sinan!~sinan@82.212.70.90> has joined #aichallenge
2010-10-02T08:50:35  <sigh> Migi32: define "perfect" in the context of this game
2010-10-02T08:52:04  <Migi32> sigh, well, I mean that as it's given more computing time, it would play better and better, and the limit of how good it plays would surpass any currently existing bot
2010-10-02T08:52:26  <Migi32> like minimax
2010-10-02T08:53:10  <sigh> it's not immediately obvious that a minimax bot would satisfy that criteria, athough it is most likely true
2010-10-02T08:53:30  <sigh> an minimax bot would be making pessimal assumptions
2010-10-02T08:53:36  <Migi32> such as?
2010-10-02T08:53:52  <sigh> your opponent knows what moves you make
2010-10-02T08:54:18  <Migi32> ah yes, true
2010-10-02T08:55:20  <sigh> I wasn't in the tron contest, but minimax worked well enough there... so it's not completely bad :P
2010-10-02T08:56:20  <sigh> in a game with this much flexibility, making assumptions like that might paralyse your bot
2010-10-02T08:56:57  <sigh> so back to what I said, it's not immediately obvious :)
2010-10-02T08:56:57  <Migi32> good point
2010-10-02T08:58:43  <contestbot> Top 10 players: Hazard(3814), ruilov(3738), rsergio(3708), EBraun(3673), cfaftw(3669), felixcoto(3652), dmj111(3636), sequoh(3623), _iouri_(3601), davidjliu(3571)
2010-10-02T08:59:56  <Migi32> a1k0n one said "turns out, knowing the move your opponent made before you make yours gives you a huuuuuge advantage"
2010-10-02T09:01:12  <sigh> yup
2010-10-02T09:01:22  <Migi32> then he said his alpha-beta search didn't work out "basically for the 'minimax incorrect' reasons from the tron contest"
2010-10-02T09:01:51  <Migi32> so did minimax work out for Tron or didn't it?
2010-10-02T09:02:18  <antimatroid2> speak of the devil
2010-10-02T09:02:28  <antimatroid2> i am in the process of trying to set up a minimax bot and see how it goes
2010-10-02T09:02:43  <antimatroid2> Migi32: minimax was very much the winning strategy for tron in the end
2010-10-02T09:02:47  <bhasker> http://72.44.46.68/canvas?game_id=272178
2010-10-02T09:02:48  <bhasker> :D
2010-10-02T09:02:53  <antimatroid2> although a1k0n is convinced one could have beaten it
2010-10-02T09:03:18  <antimatroid2> yeah, not sure what my front line planet thinks he's doing sending ships off like that :P
2010-10-02T09:03:36  <sigh> Migi32: this is a1k0n's write up of the tron contest: http://a1k0n.net/blah/archives/2010/03/index.html#e2010-03-04T14_00_21.txt
2010-10-02T09:03:48  <bhasker> heh my front planet does that to
2010-10-02T09:03:48  <sigh> it seems that a lot of people were using minimax
2010-10-02T09:04:11  <bhasker> the problem with minimax i feel is your prediction of whats best for the enemy is not well defined
2010-10-02T09:04:16  <sigh> He does say this though: "I had long ago come to the realization that a better evaluation heuristic will always beat deeper minimax searches, because a deep minimax search using a flawed evaluation heuristic is self-deluded about what its opponent is actually going to do, and will occasionally favor moves that lose to moves that win, simply because it can't tell the difference. Anything you can do to make your ev
2010-10-02T09:04:19  <sigh> aluation function smarter will result in improved play in the long run."
2010-10-02T09:04:34  <sigh> so based on that, his final bot didn't use minimax?
2010-10-02T09:04:41  <antimatroid2> we used minimax, starting with us moving first, and them picking optimally agains tus
2010-10-02T09:04:41  <bhasker> exactly pretty much what i feel about minimax
2010-10-02T09:04:43  <sigh> dunno, I'll have to ask him when he is around :P
2010-10-02T09:04:53  <sigh> a1k0n: you here?
2010-10-02T09:04:54  <bhasker> yes but how do you know in this case that your pick is optimal
2010-10-02T09:04:54  <Migi32> no I'd say it did use minimax but with a better heuristic
2010-10-02T09:05:06  <bhasker> what if you picked a suboptimal move for your oppoonent
2010-10-02T09:05:13  <bhasker> your whole evaluation falls apart
2010-10-02T09:05:14  <antimatroid2> sigh: it did, he just thinks one could have exploited the fact that a bot was using minimax
2010-10-02T09:05:34  <sigh> ah, ok...
2010-10-02T09:05:37  <antimatroid2> at least that's my understanding of what he thinks, i can't fully speak for him :P
2010-10-02T09:05:41  <sigh> :)
2010-10-02T09:06:12  *** Umsturz <Umsturz!~Umsturz@nat-wh.rz.uni-karlsruhe.de> has joined #aichallenge
2010-10-02T09:06:37  <antimatroid2> bhasker: you considered all possible moves in tron to a certain depth
2010-10-02T09:06:49  <Migi32> but antimatroid2 how are you implementing this minimax? Isn't the "branching factor" waaay too high?
2010-10-02T09:06:51  <bhasker> yea but in this thats not really possible
2010-10-02T09:06:59  <bhasker> too many possible moves
2010-10-02T09:07:06  <antimatroid2> yes, i have to "select" a small subset of the possible strategy space
2010-10-02T09:07:16  <antimatroid2> that's why i'm just trying it out to see how it goes
2010-10-02T09:07:19  <bhasker> exactly so that selection strategy may not be correct ^
2010-10-02T09:07:25  <antimatroid2> ^^
2010-10-02T09:07:26  <bhasker> if that is not correct, it falls apart
2010-10-02T09:07:40  <bhasker> though i should check http://72.44.46.68/canvas?game_id=272651
2010-10-02T09:07:48  <bhasker> why my bot decided that going for that far off planet was a bright idea
2010-10-02T09:07:57  <bhasker> probably because of its 8 size
2010-10-02T09:08:01  <bhasker> 8 ships ^
2010-10-02T09:08:27  <antimatroid2> i have a few ideas to get the tree generation going in the right direction
2010-10-02T09:08:33  <antimatroid2> so i will just see how it goes
2010-10-02T09:08:44  <Migi32> well you can select your own strategies but you simply can't select your opponent's moves unless you know for sure all non-selected moves are worse than your selected moves
2010-10-02T09:08:53  <antimatroid2> i'm still not sure what the optimal moves per turn/depth of tree ratio will be
2010-10-02T09:09:29  <sigh> bhasker: you seem to have a off-by-one error, your bot send 20 ships to a 20-ship neutral
2010-10-02T09:09:30  *** Haggis <Haggis!81d73b91@gateway/web/freenode/ip.129.215.59.145> has joined #aichallenge
2010-10-02T09:09:38  <bhasker> no
2010-10-02T09:09:42  <antimatroid2> the idea is to pick moves i might make, then have them make their moves based off them "already knowing what I did" and then eliminating moves they will easily exploit
2010-10-02T09:09:46  <bhasker> i launched from two planets
2010-10-02T09:09:47  <bhasker> i think
2010-10-02T09:09:59  <bhasker> so there musbe a 1 size fleet somewhere
2010-10-02T09:10:00  <antimatroid2> and then repeat as long as i can
2010-10-02T09:10:14  <antimatroid2> it's awefully messy, still not entirely sure how it'll come together
2010-10-02T09:10:15  <sigh> bhasker: yeah, it's lagging behind a lot though
2010-10-02T09:10:22  <sigh> not well coordinated
2010-10-02T09:10:36  <antimatroid2> sigh: i do that as well, i tried making them coordinate arrival, but it just makes the bot send more
2010-10-02T09:10:41  <bhasker> yea i don't coordinate attacks
2010-10-02T09:10:41  <Migi32> well antimatroid2 say you want to look 20 moves ahead, more than 2 optimal moves / turn is probably not going to work
2010-10-02T09:10:45  <sigh> send more?
2010-10-02T09:10:52  <antimatroid2> 20 moves ahead isn't going to happen
2010-10-02T09:11:01  <bhasker> i basically sort the list of ships available by proximity to planet
2010-10-02T09:11:04  <sigh> haha 20 moves
2010-10-02T09:11:07  <bhasker> and send from closest planets
2010-10-02T09:11:12  <bhasker> when i try to pile up ships
2010-10-02T09:11:13  <Migi32> aren't some maps like 20 turns in diameter?
2010-10-02T09:11:15  <antimatroid2> you'll be lucky to get 5
2010-10-02T09:11:17  <bhasker> it actually does worse
2010-10-02T09:11:33  <bhasker> sending partial fleets actually works better imo
2010-10-02T09:11:47  <antimatroid2> it's surprising how fast game trees explode in size
2010-10-02T09:11:55  <sigh> hmm... my bot does better when I coordinate
2010-10-02T09:11:58  <antimatroid2> even tic tac toe game trees are quite large
2010-10-02T09:12:08  <sigh> the other bots at the top seem to do it as well
2010-10-02T09:12:10  <bhasker> the problem with coordinating
2010-10-02T09:12:19  <bhasker> i know adum's doesn't coordinate
2010-10-02T09:12:26  <antimatroid2> there's over 2 million unique subgame perfect nash equilbibria in the trivial 2 player 2x2 tic tac toe game
2010-10-02T09:12:28  <sigh> yeah not all
2010-10-02T09:12:38  <antimatroid2> might be 1 million, i could work it out if you insist :P
2010-10-02T09:12:51  <bhasker> as i was saying problem i found with coordinating is you lose time
2010-10-02T09:13:01  <bhasker> and that is a cost that needs to be applied
2010-10-02T09:13:07  <sigh> how do you lose time?
2010-10-02T09:13:15  <Migi32> antimatroid2, seriously? wow
2010-10-02T09:13:25  <bhasker> well a) you have to funnel ships to a planet from where you can launch the attack
2010-10-02T09:13:45  <bhasker> so all your ships have to travel to the tip of your attack
2010-10-02T09:13:52  <bhasker> vs going straight for the attack
2010-10-02T09:13:55  *** rul <rul!~rul@190.188.115.15> has quit IRC (Ping timeout: 265 seconds)
2010-10-02T09:14:03  <bhasker> so overall launching a non coordinated strike is faster
2010-10-02T09:14:05  <sigh> bhasker: no, what I mean is that you send ships so they arrive at the same time
2010-10-02T09:14:18  <sigh> the way you describe is slower, yes
2010-10-02T09:14:40  <bhasker> so you launch from further out planets before you launch from nearer ones
2010-10-02T09:14:41  <bhasker> ?
2010-10-02T09:14:45  <sigh> yup
2010-10-02T09:14:52  <bhasker> i tried doing that but i can't seem to get the math right ^
2010-10-02T09:14:56  <sigh> ah ok
2010-10-02T09:14:58  <bhasker> and my bot might switch priorities midway
2010-10-02T09:15:18  <sigh> :)
2010-10-02T09:15:23  <bhasker> so either i have to add some attack logic that remembers that it started attacking something and remembers to complete it
2010-10-02T09:15:32  <bhasker> basically if i leave ships where they are
2010-10-02T09:15:38  <bhasker> they get redirected
2010-10-02T09:15:39  <bhasker> ^
2010-10-02T09:15:40  *** rul <rul!~rul@190.188.115.15> has joined #aichallenge
2010-10-02T09:15:43  <sigh> my bot doesn't have state, it just recalculates every move
2010-10-02T09:15:55  <sigh> so it naturally redirects if required
2010-10-02T09:16:05  <Migi32> bhasker, if you start an attack it's a "sunk cost", so it might not always be the optimal move to complete the attack fully
2010-10-02T09:16:11  <bhasker> yea
2010-10-02T09:16:13  <sigh> ^^ this
2010-10-02T09:16:14  <bhasker> hmm
2010-10-02T09:16:28  * sigh was pointing at Migi32
2010-10-02T09:16:28  <bhasker> should try coordinating it
2010-10-02T09:16:40  <bhasker> my math is not working out for that ^
2010-10-02T09:17:07  <antimatroid2> yeah, it's 1327104 pure strategy subgame perfect nash equilbria
2010-10-02T09:17:11  <antimatroid2> ridiculous
2010-10-02T09:17:15  <sigh> it's also a good test of your prediction code, very obvious if things are off :D
2010-10-02T09:17:59  <antimatroid2> i agree with the notion of it being a sunk cost, but whenever i tell it to wait for other ones it does worse
2010-10-02T09:18:25  <antimatroid2> i do have it redirecting fleets to planets closer to enemies such that the journey would finish in the same time as the furtherst "source" planet though
2010-10-02T09:18:26  *** rul <rul!~rul@190.188.115.15> has quit IRC (Read error: Connection reset by peer)
2010-10-02T09:18:28  <bhasker> i don't know i am not yet convinced that holding ships off is a better idea
2010-10-02T09:18:29  <antimatroid2> that seems to work well
2010-10-02T09:18:38  <sigh> when attacking the enemy, the benefit is not as clear cut
2010-10-02T09:18:47  <sigh> as you may want to reduce units as fast as possible
2010-10-02T09:18:48  <antimatroid2> bhasker: it allows you to change the move if the optimal move changes
2010-10-02T09:18:49  <bhasker> yea cause he is reinforcing too
2010-10-02T09:18:52  <antimatroid2> technically it is better
2010-10-02T09:18:54  <bhasker> hmm
2010-10-02T09:19:06  <antimatroid2> that's a good point
2010-10-02T09:19:25  <dmj111> bhasker: on a given turn, if I attack from multiple planets, I "launch" from them all.   Only the ones furthest out really launch, but internally, my code creates a fleet for every planet involved, so I don't overload any planet with future launches.
2010-10-02T09:19:27  <sigh> against a neutral, I can't see any disadvantage *unless* you are about to get taken out by enemy
2010-10-02T09:19:28  <antimatroid2> i shall see if that's why it does worse later
2010-10-02T09:19:28  <bhasker> oh well let me see if i can try getting the coordination logic correct
2010-10-02T09:19:46  <dmj111> like sigh, my code does not track state between turns, so each turn it can change plans.
2010-10-02T09:19:54  <bhasker> hmm k
2010-10-02T09:20:06  <sigh> stateless code is just easier to manage :)
2010-10-02T09:20:08  <bhasker> makes sense
2010-10-02T09:20:10  <antimatroid2> i just subtract those ship amounts without adding a fleet :P
2010-10-02T09:20:19  <antimatroid2> and store fleets between turns
2010-10-02T09:20:20  <bhasker> i should try that out
2010-10-02T09:20:33  <dmj111> sigh: i agree, the benefit is not clear cut.   I prefer launching later, in case the enemy does not decide to reinforce, I don't commit forces.  But, it might keep my bot from doing some of the dumb things it does.
2010-10-02T09:21:08  <bhasker> do any of you do fleet merging
2010-10-02T09:21:24  <sigh> dmj111: yeah, I've just found that waiting improves my performance... I haven't tried mixed strategies yet
2010-10-02T09:21:24  <dmj111> antimatroid2: I create a fleet, because I recalculate the future state of the game based on the fleets to decide the next target to attack on the turn.  This was a hack to keep me from re-attacking a planet again on the same turn.
2010-10-02T09:21:24  <bhasker> one thing i was thinking was if i have two neutrals to capture same turn
2010-10-02T09:21:28  <sigh> merging?
2010-10-02T09:21:29  <bhasker> and in the same direction
2010-10-02T09:21:33  <bhasker> merge the fleets to one neutral
2010-10-02T09:21:58  <bhasker> not sure if that works out better
2010-10-02T09:22:03  <sigh> I'm not sure what you mean, how is that different from what we were talking about?
2010-10-02T09:22:16  <bhasker> what i mean is this
2010-10-02T09:22:20  <bhasker> lets say you launch from A
2010-10-02T09:22:21  <dmj111> bhasker: maybe, but the shortest path between two points cannot include a pit stop.  It will take longer, but maybe it will add a surprise factor.
2010-10-02T09:22:33  <bhasker> well its harder to counter dmj
2010-10-02T09:22:40  <bhasker> since i send 2x the ships to one planet
2010-10-02T09:22:41  <Migi32> antimatroid2, I don't know much about Nash equilibria, but if there's only 24 ways that this 2x2 tic tac toe can play out, how can you have more equilibria?
2010-10-02T09:22:48  <bhasker> so enemy can't counter
2010-10-02T09:22:48  <dmj111> good point
2010-10-02T09:23:17  <bhasker> basically if you have two planets that are within say a few degrees of each other in the same direction
2010-10-02T09:23:17  <antimatroid2> Migi32: a strategy is defined as the move each player might make at every possible node in the tree they might reach
2010-10-02T09:23:27  <bhasker> merge the fleets headed to the two of them and send them to just one of them
2010-10-02T09:23:27  <sigh> bhasker: ah, I understand
2010-10-02T09:23:35  <dmj111> i guess i was thinking of two neutrals you could clearly take, not ones where there is possible contention from the enemy... because my bot is a chicken and never tries that anyway :)
2010-10-02T09:23:38  <bhasker> though it would probably work only early game
2010-10-02T09:24:00  <bhasker> late game you rarely take two neutrals at same time
2010-10-02T09:24:07  <bhasker> so not sure how much benefit there is in that
2010-10-02T09:24:26  <bhasker> your bot is funny to watch its like a predator
2010-10-02T09:24:35  <bhasker> it waits then jumps at you in one shot
2010-10-02T09:24:38  <bhasker> and wipes you out
2010-10-02T09:24:48  <sigh> :)
2010-10-02T09:24:51  <antimatroid2> you wouldn't ever calculate all of those for an ai bot, i just wrote such a function for a library i'm putting together, so was testing it out
2010-10-02T09:25:01  <bhasker> http://72.44.46.68/canvas?game_id=272838
2010-10-02T09:25:05  <antimatroid2> no surprises the function didn't work on 3x3 tic tac toe
2010-10-02T09:25:08  <bhasker> i don't understand why your bot sent 93 ships
2010-10-02T09:25:11  <bhasker> to a 1 growth planet?
2010-10-02T09:25:17  <antimatroid2> i couldn't even find a nash equilibria for 4x4
2010-10-02T09:25:27  <bhasker> when the planet has 8 ships on it?
2010-10-02T09:25:29  <antimatroid2> or 3x3x3
2010-10-02T09:25:38  <dmj111> bhasker: dmj111 == chicken hawk
2010-10-02T09:25:58  <sigh> bhasker: I assume because it's closer to the enemy
2010-10-02T09:26:12  <bhasker> i think dmj restricts the space to search initially
2010-10-02T09:26:13  <sigh> and dmj111 decided it needs some extra protection :)
2010-10-02T09:26:14  <bhasker> it didnt find any planets
2010-10-02T09:26:23  <bhasker> so it sent everthing up front
2010-10-02T09:26:32  <antimatroid2> dmj111 waits patiently till he can just waltz over and fuck you up
2010-10-02T09:26:49  <antimatroid2> i still haven't been able to properly counter it whilst not getting ruined by other more agressive bots
2010-10-02T09:26:52  <bhasker> these vortex.* variants are annoying
2010-10-02T09:27:07  <antimatroid2> i get him most of the time atm :D
2010-10-02T09:27:16  <dmj111> antimatroid2: thanks...  I feel proud of that statement... but a little weird for it.
2010-10-02T09:27:24  <sigh> haha
2010-10-02T09:27:51  <antimatroid2> most of your wins are under 100 moves, and it's usually a result of the other player expanding too fast and allowing your bot to just go in for the kill :P
2010-10-02T09:27:51  <dmj111> sigh: I think you are right.  I have a couple of different priorities in the code, and the interact strangely at times.  Your explanation makes the most sense.
2010-10-02T09:28:10  <dmj111> oops ... might have been bhasker, not sigh.
2010-10-02T09:28:31  <dmj111> bahhh.... it was sigh.  I am not allowed to type any more.
2010-10-02T09:28:36  <sigh> dmj111: I have a bot that behaves similarly :)
2010-10-02T09:28:52  <sigh> in that respect, not overall
2010-10-02T09:29:02  <sigh> (otherwise I would upload it :P)
2010-10-02T09:29:18  <dmj111> :)  That behaviour has given me some losses too, though.
2010-10-02T09:29:27  <antimatroid2> my current bot better make top 10 on official site aha
2010-10-02T09:29:42  <antimatroid2> it hasn't lost on there yet, but still above 200
2010-10-02T09:29:44  <bhasker> dmj http://72.44.46.68/canvas?game_id=266834
2010-10-02T09:29:54  <bhasker> my bot counters your surprise ^
2010-10-02T09:30:02  <bhasker> sometimes ^
2010-10-02T09:30:17  <dmj111> nice :)
2010-10-02T09:30:39  <bhasker> wow the top 3 bots are at 3800 elo now ^
2010-10-02T09:31:26  <bhasker> the goal to hit top 10-20 seems further and further away /cry
2010-10-02T09:31:33  <sigh> bottom bots are down to 378, almost 3500 point range
2010-10-02T09:32:10  <antimatroid2> bhasker: the bots keep getting better and better every day, it's very annoying
2010-10-02T09:32:22  <sigh> yup, I've been kicked out of the top 10
2010-10-02T09:32:31  <bhasker> heh
2010-10-02T09:32:34  <bhasker> i see new names up there
2010-10-02T09:32:36  <Migi32> aren't you able to set standard deviation in the ELO algorithm?
2010-10-02T09:32:49  <Migi32> isn't it a parameter?
2010-10-02T09:32:53  <antimatroid2> woah, new leader
2010-10-02T09:32:58  <sigh> @rankings
2010-10-02T09:33:00  <contestbot> sigh: Top 10 players: ruilov(3821), Hazard(3808), rsergio(3755), EBraun(3680), cfaftw(3663), felixcoto(3653), dmj111(3638), sequoh(3620), _iouri_(3609), davidjliu(3569)
2010-10-02T09:33:07  <sigh> haha, yeah
2010-10-02T09:33:09  <bhasker> look at the difference
2010-10-02T09:33:15  <antimatroid2> he just beat hazard
2010-10-02T09:33:27  <bhasker> the 4th bot is almost 200 elo points lower
2010-10-02T09:33:49  *** keynes <keynes!~AndChat@211.234.145.8> has joined #aichallenge
2010-10-02T09:34:17  <antimatroid2> ruilov hasn't played very many games
2010-10-02T09:34:36  <Migi32> yea I noticed that too
2010-10-02T09:34:38  <dmj111> the true skill idea is probably a good one.
2010-10-02T09:34:40  <Migi32> he submitted today
2010-10-02T09:34:46  <bhasker> oh he beat hazard
2010-10-02T09:34:47  <bhasker> thats why
2010-10-02T09:34:54  <antimatroid2> he beat quite a few top bots
2010-10-02T09:34:57  <bhasker> yea
2010-10-02T09:35:00  <antimatroid2> but how'd it get paired against them so fast?
2010-10-02T09:35:11  <bhasker> luck?
2010-10-02T09:35:21  <bhasker> since top bots get paired against lower bots
2010-10-02T09:35:23  <bhasker> every now and then
2010-10-02T09:35:35  <bhasker> like my bot plays people at 150-200 rank
2010-10-02T09:35:40  <bhasker> even though its at 78
2010-10-02T09:35:47  <Migi32> I think in ELO, your first 10-ish games make your ELO go up/down faster than later games
2010-10-02T09:35:48  <antimatroid2> my bots won 8/8 games and it's ranked 230
2010-10-02T09:35:52  <Migi32> and he won all first 10
2010-10-02T09:36:01  <Migi32> oh no wait he didn't
2010-10-02T09:36:29  <antimatroid2> although i have absolutely spammed submissions last few days, oops
2010-10-02T09:36:36  <bhasker> lol http://72.44.46.68/canvas?game_id=273000
2010-10-02T09:36:55  <bhasker> you take the 1 growth in middle and pile up ships there
2010-10-02T09:37:14  <sigh> hmmm... in his game against me he won by controlling the center :( http://www.ai-contest.com/visualizer.php?game_id=5222313
2010-10-02T09:37:25  <sigh> it was close for a while
2010-10-02T09:37:45  <bhasker> antimatroid that was for you sorry
2010-10-02T09:38:29  *** Itkovian <Itkovian!~Itkovian@2002:4e16:e2a0:0:21f:5bff:fec8:6e2c> has joined #aichallenge
2010-10-02T09:38:42  <antimatroid2> yep, hence why i decided to move on and try minimax
2010-10-02T09:38:52  <antimatroid2> my bot still loses, it's not acceptable :(
2010-10-02T09:39:38  <bhasker> i remember why i didnt coordinate attacks
2010-10-02T09:39:43  <bhasker> i launch from the closest planet first
2010-10-02T09:40:26  <antimatroid2> the other thing about sending parts later is you keep that as private information
2010-10-02T09:40:29  <bhasker> maybe i need to queue up the orders, then only execute a few ones
2010-10-02T09:40:45  <antimatroid2> if that hasn't been mentioned
2010-10-02T09:41:18  <bhasker> let me quickly implement it and see
2010-10-02T09:41:21  <antimatroid2> same with redirection part of the way when possible
2010-10-02T09:41:40  <sigh> antimatroid2: yah, that is a big part of it
2010-10-02T09:41:44  <antimatroid2> so you reveal that information at the latest time possible while still executing the intended move
2010-10-02T09:41:57  <sigh> if your enemy moves some units to protect, you can switch targets
2010-10-02T09:42:10  <sigh> where as if you are commited, then you can't do anything
2010-10-02T09:42:17  <antimatroid2> sigh: your enemy wont even necesseraly know what you're intending :P
2010-10-02T09:42:38  <antimatroid2> they'd do a damn good job to know somehow
2010-10-02T09:43:35  <sigh> yeah, but you can make assumptions based on the fact that you are already under attack :)
2010-10-02T09:43:54  *** Itkovian <Itkovian!~Itkovian@2002:4e16:e2a0:0:21f:5bff:fec8:6e2c> has quit IRC (Quit: Itkovian)
2010-10-02T09:44:17  <antimatroid2> yeah, but they wont necessarily know you're about to take a neutral near teh centre till you make your last move closer to the centre
2010-10-02T09:44:44  <antimatroid2> is anyone currently picking to move to those planets cause the other player sent some ships there from a planet further away? i doubt it
2010-10-02T09:44:57  <sigh> heh
2010-10-02T09:46:11  <antimatroid2> DRAW is so annoying, but it'd be worse if it was LOSS aha
2010-10-02T09:47:20  <antimatroid2> i wonder if it'd be possible to write a bot that when winning could then try to cause a draw at the end
2010-10-02T09:47:26  <antimatroid2> that's be cool
2010-10-02T09:47:30  <antimatroid2> that'd*
2010-10-02T09:48:11  <antimatroid2> wouldn't be hard to write, how well it'd go though i dunno
2010-10-02T09:48:41  <sigh> draws are hard to manufacture
2010-10-02T09:48:59  <antimatroid2> you'd have it make moves based off the end turn
2010-10-02T09:49:11  <antimatroid2> and hope the other player doesn't mess up the intention for right at the end
2010-10-02T09:49:11  <sigh> it becomes easier if there are neutral planets to suicide into
2010-10-02T09:49:16  <antimatroid2> would need to be in front though i think
2010-10-02T09:49:22  <antimatroid2> yeah
2010-10-02T09:50:04  <sigh> I would be very impressed if you made a bot that could create draws
2010-10-02T09:51:51  <antimatroid2> bhasker: well i don't always lose to you :)
2010-10-02T09:52:01  <bhasker> i know:p
2010-10-02T09:52:09  <bhasker> thats why i feel happy when i beat one of the top bots
2010-10-02T09:52:12  <bhasker> mostly i get crushed ^
2010-10-02T09:52:22  <antimatroid2> i don't think i'd say i have one of the top bots
2010-10-02T09:52:41  <bhasker> well on tcp at least:)
2010-10-02T09:52:59  <antimatroid2> yeah, hardly anyone's running their bots on there atm though
2010-10-02T09:53:55  <bhasker> there is dmj /you /vortex.*/golfin /voidptr/locutus and bartwe
2010-10-02T09:54:02  <bhasker> i would say thats a decent competition
2010-10-02T09:54:09  <bhasker> and then there is DRAW
2010-10-02T09:54:18  <bhasker> and all the whiteside variants
2010-10-02T09:54:23  <sigh> hmmm... I'll put up one of my bots, I don't know how it will do
2010-10-02T09:54:29  <sigh> it's too aggressive
2010-10-02T09:54:53  <antimatroid2> i can't find the right balance between agressive and patient
2010-10-02T09:55:09  <antimatroid2> it's there somewhere
2010-10-02T09:55:20  <sigh> it depends on what the opponent is doing
2010-10-02T09:55:33  <antimatroid2> yeah, but it gets very complicated
2010-10-02T09:55:37  <sigh> currently my bot ignores the opponent for the most part
2010-10-02T09:55:42  <antimatroid2> bugs pop up everywhere
2010-10-02T09:56:13  <antimatroid2> i am not paying attention to the moves they make, but that's going to change
2010-10-02T09:56:27  <antimatroid2> dev_tcp is playing again now too
2010-10-02T09:56:31  <antimatroid2> and i beat the moungrel
2010-10-02T09:56:36  <antimatroid2> he was owning me before
2010-10-02T09:57:53  <antimatroid2> http://72.44.46.68/canvas?game_id=272869
2010-10-02T09:57:54  <antimatroid2> wiw
2010-10-02T09:57:56  <antimatroid2> wow*
2010-10-02T09:58:01  <FlagCapper> Wow I'm on a lucky streak
2010-10-02T09:58:08  <bhasker> hah nice 7 turns
2010-10-02T09:58:19  <sigh> antimatroid2: ah 7 moves, like I did to bartwe :D
2010-10-02T09:58:21  <FlagCapper> Nice
2010-10-02T09:58:45  <contestbot> Top 10 players: ruilov(3823), Hazard(3807), rsergio(3732), EBraun(3690), felixcoto(3660), cfaftw(3658), dmj111(3634), sequoh(3619), _iouri_(3608), sighbot(3567)
2010-10-02T09:59:05  <sigh> hey, I kicked davidjliu out of the top 10
2010-10-02T09:59:15  <antimatroid2> party time? :P
2010-10-02T09:59:21  <sigh> :P
2010-10-02T09:59:38  <antimatroid2> i've almost breached top 100
2010-10-02T09:59:48  <sigh> hurry up :)
2010-10-02T10:00:08  <antimatroid2> it's probably lose soon and take even longer
2010-10-02T10:00:30  <antimatroid2> wow, my english is all over the show tonight
2010-10-02T10:01:29  <bhasker> your bot should be in the top 40-50 my bot is 76
2010-10-02T10:01:36  <bhasker> and been holding there for last couple of days now
2010-10-02T10:01:42  <bhasker> and your bot does much better than mine
2010-10-02T10:01:48  *** yasith <yasith!~yasith@112.135.78.217> has joined #aichallenge
2010-10-02T10:03:23  <antimatroid2> oh, i know it'll get that high, just hoping for top 10 :P
2010-10-02T10:03:38  <sigh> this was an interesting battle: http://72.44.46.68/canvas?game_id=273356
2010-10-02T10:03:45  <sigh> a lot of back and forth sparing
2010-10-02T10:04:45  <antimatroid2> i'm still prioritising defense over anything else
2010-10-02T10:05:08  <antimatroid2> oh here we go sigh
2010-10-02T10:05:27  <antimatroid2> huzzah
2010-10-02T10:05:29  <sigh> :P
2010-10-02T10:06:25  *** jmpespxoreax <jmpespxoreax!~notauser@CPE002369b3cbac-CM001868e7f01e.cpe.net.cable.rogers.com> has left #aichallenge
2010-10-02T10:06:34  <sigh> haha, my defence sucks
2010-10-02T10:07:03  <antimatroid2> this game is longer
2010-10-02T10:07:06  <antimatroid2> that usually means i lose
2010-10-02T10:08:04  <sigh> nah, I lost
2010-10-02T10:08:15  <antimatroid2> lol still hasn't told me :P
2010-10-02T10:08:21  <antimatroid2> just came through
2010-10-02T10:08:28  <sigh> I'm watching my logs :P
2010-10-02T10:09:36  <sigh> lol, that was a boring game :P
2010-10-02T10:09:42  <antimatroid2> it was
2010-10-02T10:10:10  <antimatroid2> lol, i went down in ranking on the main site from a win
2010-10-02T10:10:56  <Migi32> is that even possible?
2010-10-02T10:11:04  <antimatroid2> yes, play a lower ranked person
2010-10-02T10:11:10  <antimatroid2> or other people you played drop in ranking
2010-10-02T10:11:17  <sigh> you can do up from a loss too
2010-10-02T10:11:23  <sigh> *go
2010-10-02T10:11:24  <antimatroid2> yep
2010-10-02T10:11:39  <antimatroid2> draw on a stupid map is most painful
2010-10-02T10:11:44  <sigh> you lost to dmj111! you must be really strong!
2010-10-02T10:11:52  <Mathnerd314> I thought that was going to get fixed using lower-bound instead of upper-bound...
2010-10-02T10:12:11  <antimatroid2> i have beaten him at least twice today
2010-10-02T10:12:22  <Mathnerd314> s/upper-bound/most likely score/
2010-10-02T10:12:26  <antimatroid2> well, yesterday*
2010-10-02T10:12:26  <Migi32> oh you mean you went down in ranking, not in ELO
2010-10-02T10:12:42  <antimatroid2> i don't track the elo score sorry
2010-10-02T10:13:23  <antimatroid2> wow, 3 loses in a row
2010-10-02T10:14:11  *** medeman <medeman!4fc84033@gateway/web/freenode/ip.79.200.64.51> has joined #aichallenge
2010-10-02T10:14:17  <medeman> hi
2010-10-02T10:14:27  <sigh> hi
2010-10-02T10:14:34  <medeman> uhmm
2010-10-02T10:15:30  <medeman> here (http://ai-contest.com/simple_strategy_guide.php) it says
2010-10-02T10:15:34  <medeman> down there
2010-10-02T10:15:51  <medeman> the thing with fleets in the air if you attacked then 1 if not 3
2010-10-02T10:16:00  <medeman> when i do it like that and compile
2010-10-02T10:16:26  <medeman> it gives me an symbol error
2010-10-02T10:16:29  <medeman> i have java version
2010-10-02T10:16:38  <medeman> and im using mac os x 10.6.4
2010-10-02T10:16:42  <medeman> why is that error?
2010-10-02T10:16:44  <sigh> does your PlanetWars class have a Production method?
2010-10-02T10:16:57  <medeman> i dunno im new
2010-10-02T10:16:57  <sigh> i think that was missing in the start packs
2010-10-02T10:17:06  <medeman> oh
2010-10-02T10:17:08  <medeman> how to add it=
2010-10-02T10:17:13  <medeman> ?
2010-10-02T10:17:28  <sigh> hang on, I'll check the java starter pack to be sure
2010-10-02T10:18:06  <medeman> okay thank you
2010-10-02T10:18:23  <antimatroid2> 6 wins to follow makes me feel a bit better
2010-10-02T10:18:31  <antimatroid2> but now i have to verse dmj
2010-10-02T10:18:36  <antimatroid2> that wont go well
2010-10-02T10:19:01  <sigh> medeman: look at the PlanetWars.java file in the examplebots folder, that has the production method
2010-10-02T10:19:35  <antimatroid2> no surprises there
2010-10-02T10:20:20  <medeman> ...
2010-10-02T10:20:54  <medeman> okay that means
2010-10-02T10:20:57  <medeman> ...
2010-10-02T10:21:10  <medeman> i change this:
2010-10-02T10:21:10  <medeman> if (pw.Production(1) >= pw.Production(2)) {
2010-10-02T10:21:14  <medeman> to this:
2010-10-02T10:21:26  <medeman> if (Production(1) >= Production(2)) {
2010-10-02T10:21:28  <medeman> ?
2010-10-02T10:22:00  <sigh> no, it means that you need to change PlanetWars.java to have a Production method
2010-10-02T10:22:09  <sigh> hmmm... they really should fix this :S
2010-10-02T10:23:12  <medeman> how to change it?
2010-10-02T10:23:14  <medeman> sorry im new
2010-10-02T10:24:48  <sigh> open example_bots/PlanetWars.java and copy the Production method into PlanetWars.java (I was going to tell you to copy the entire file, but it seems that it is also missing some methods)
2010-10-02T10:25:24  <medeman> okay
2010-10-02T10:26:18  <Migi32> antimatroid2, I've been reading up on Nash equilibria a bit, and I'd say there are 0 Nash equilibria in PlanetWars. What do you think?
2010-10-02T10:26:35  <medeman> hmm i cant find the production method
2010-10-02T10:26:40  <sigh> Migi32: wouldn't that depend on the map?
2010-10-02T10:26:55  <Migi32> sigh, well, there's a center planet in every map, so I guess not
2010-10-02T10:26:58  <sigh> medeman: line 221-229
2010-10-02T10:27:04  <antimatroid2> Migi32: there exists at least one mixed strategy nash equilibrium
2010-10-02T10:27:22  <Migi32> ah yes mixed strategy, with probabilities and so on
2010-10-02T10:27:29  <Migi32> didn't think of that
2010-10-02T10:27:57  <antimatroid2> one approach would be to try and have some kind of numerical approximation for one of those
2010-10-02T10:28:03  <antimatroid2> but i wouldn't even know where to begin
2010-10-02T10:28:10  <antimatroid2> and doubt you'd be able to get anything worth using
2010-10-02T10:28:24  <antimatroid2> but i intend to incorporate mixed strategies into my bot in a way
2010-10-02T10:29:15  <medeman> thats ship number
2010-10-02T10:29:18  <medeman> but not production
2010-10-02T10:30:56  *** keynes <keynes!~AndChat@211.234.145.8> has quit IRC (Ping timeout: 276 seconds)
2010-10-02T10:31:03  *** boegel <boegel!~boegel@91.182.173.227> has joined #aichallenge
2010-10-02T10:31:34  <boegel> DanielVF: just noticed another weird timed-out game, see http://ai-contest.com/visualizer.php?game_id=5198425
2010-10-02T10:31:55  <DanielVF> boegel Thanks
2010-10-02T10:32:05  <DanielVF> The orginal server is still running a few agmes
2010-10-02T10:32:15  <DanielVF> It's not on the latest code, and so gives no errors
2010-10-02T10:32:18  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has joined #aichallenge
2010-10-02T10:32:22  <u1tr0n> hi all
2010-10-02T10:32:23  <boegel> DanielVF: it's a lot less frequent though
2010-10-02T10:32:40  <DanielVF> Yeah. It's only running %10 of games
2010-10-02T10:32:52  <DanielVF> And I backed it down so it's got cpu to spare
2010-10-02T10:34:13  <sigh> medeman: sorry, I'm not sure how to help you then... the starter pack I just downloaded has a Production method on line 221 in "example_bots/PlanetWars.java"
2010-10-02T10:35:12  <Migi32> what I was thinking of doing was a sort of minimax search but instead of letting the players take turns, I'd do this: at each ply you have P1's possible moves and P2's possible moves, recursively find how good each "pair" of moves (a move being ALL the player's orders for that turn) is for each player, then a miracle occurs using Nash equilibrium magic and you get the EV of the move
2010-10-02T10:35:40  <antimatroid2> good luck with that :P
2010-10-02T10:35:41  <medeman> sigh thanks anyways
2010-10-02T10:35:52  *** smellyhippy <smellyhippy!~asmodean@about/csharp/regular/smellyhippy> has quit IRC (Quit: BUT WAIT! What are those penguins up to?!)
2010-10-02T10:36:12  <antimatroid2> you could throw together pure strategies and do minimax on the strategic version of that
2010-10-02T10:36:31  <antimatroid2> you can even use linear programming to solve for a mixed strategy nash equilibrium if you keep it zero sum
2010-10-02T10:36:46  <antimatroid2> it's a pain in the fucking ass to calculus msne for non-zero sum 2 player games
2010-10-02T10:37:09  <antimatroid2> i'm intending to do my maths honours thesis next year on calculating them for n-player non-zersum games
2010-10-02T10:37:16  <antimatroid2> you got me sigh :P
2010-10-02T10:37:36  <Migi32> should be possible to make a good zero-sum evaluation heuristic
2010-10-02T10:37:39  <sigh> medeman: here, I copied the method to pastebin for you: http://pastebin.com/XisH1TZA
2010-10-02T10:37:44  <sigh> antimatroid2: :P
2010-10-02T10:37:54  <medeman> thank you
2010-10-02T10:39:29  <Migi32> if nothing else: # P1 ships - # P2 ships + turns_left * ( P1 total growth - P2 total growth )
2010-10-02T10:39:33  <Migi32> that's zero-sum, right?
2010-10-02T10:40:01  *** smellyhippy <smellyhippy!~asmodean@about/csharp/regular/smellyhippy> has joined #aichallenge
2010-10-02T10:40:40  <antimatroid2> yep
2010-10-02T10:41:16  <antimatroid2> payoff(player1, player2) =- payoff(player2, player1) just need that for all possible strategies
2010-10-02T10:41:29  *** boegel <boegel!~boegel@91.182.173.227> has quit IRC (Quit: Leaving)
2010-10-02T10:41:55  <Migi32> but if there are 2 Nash equilibria, each with a different payoff for P1, is that possible? if so, what would be the total payoff for P1?
2010-10-02T10:42:22  <antimatroid2> multiple nash equilibria are possible, you can only play one, if you find many pick the best
2010-10-02T10:42:32  <antimatroid2> you might want to pseudo randomise over multiple best choices
2010-10-02T10:42:39  <antimatroid2> lest someone try to predict you using one of them
2010-10-02T10:43:37  <Migi32> they can predict me all they want, if it's a real Nash equilibrium they can't do anything better than play out that move
2010-10-02T10:44:02  <antimatroid2> depeds on whether it's strict or not (i think)
2010-10-02T10:44:09  <medeman> gtg
2010-10-02T10:44:10  <medeman> bye
2010-10-02T10:44:18  <antimatroid2> and even then, is it a nash equilibria in which you are at least drawing?
2010-10-02T10:44:23  <Migi32> bye
2010-10-02T10:44:29  <Migi32> you mean strong instead of strict?
2010-10-02T10:44:30  <antimatroid2> if you start from a draw state, sure
2010-10-02T10:44:32  <antimatroid2> otherwise?
2010-10-02T10:44:39  *** medeman <medeman!4fc84033@gateway/web/freenode/ip.79.200.64.51> has quit IRC (Quit: Page closed)
2010-10-02T10:45:46  *** sinan <sinan!~sinan@82.212.70.90> has quit IRC (Ping timeout: 252 seconds)
2010-10-02T10:47:37  <antimatroid2> wait, i think i'm wrong there
2010-10-02T10:47:57  <antimatroid2> strong/strict is interchangeable is it not?
2010-10-02T10:48:59  <Migi32> I don't know what strict is
2010-10-02T10:49:00  <Migi32> :P
2010-10-02T10:49:14  <Migi32> that's why I asked
2010-10-02T10:49:46  <antimatroid2> well yeah
2010-10-02T10:50:16  <BtbN> Server gives me a compile-error while compiling my C# submission. No error localy using .NET 2.0. Not even a warning. And i didn't change that much.
2010-10-02T10:50:23  <antimatroid2> basically it's struct if u(si*, s-i*) > u(si', s-i*) for all si' != si*
2010-10-02T10:50:26  <antimatroid2> and weak for >=
2010-10-02T10:50:33  <antimatroid2> strict*
2010-10-02T10:51:31  *** smellyhippy <smellyhippy!~asmodean@about/csharp/regular/smellyhippy> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T10:52:08  <Migi32> well, given the fact that the values I will assign to each combination of moves will not be perfect anyway, I guess it will indeed be better to randomise a little
2010-10-02T10:54:40  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has quit IRC (Ping timeout: 255 seconds)
2010-10-02T10:55:44  <DanielVF> BtbN Server's having mono issues
2010-10-02T10:55:47  <DanielVF> ::cry::
2010-10-02T10:56:08  <BtbN> They at least tried to update it?
2010-10-02T10:56:48  <DanielVF> They need to, plan to soon
2010-10-02T10:58:45  <contestbot> Top 10 players: ruilov(3909), Hazard(3804), rsergio(3738), EBraun(3694), cfaftw(3682), felixcoto(3662), dmj111(3636), sequoh(3619), _iouri_(3611), sighbot(3575)
2010-10-02T11:00:29  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has joined #aichallenge
2010-10-02T11:02:21  <sigh> wtf, my bot is 5th on tcp
2010-10-02T11:02:29  <sigh> that doesn't make sense, it loses too much
2010-10-02T11:04:36  *** BtbN <BtbN!~BtbN@p54BB7467.dip.t-dialin.net> has quit IRC (Quit: Verlassend)
2010-10-02T11:04:36  *** fawek <fawek!~faw@chello089079023212.chello.pl> has joined #aichallenge
2010-10-02T11:05:08  *** sinan <sinan!~sinan@82.212.70.90> has joined #aichallenge
2010-10-02T11:06:32  *** Appleman1234 <Appleman1234!~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au> has joined #aichallenge
2010-10-02T11:08:09  <Sylph> :)
2010-10-02T11:08:42  <antimatroid2> oh crap, it's 2am and not 1am
2010-10-02T11:08:48  <antimatroid2> yay for daylight savings thought
2010-10-02T11:09:03  <antimatroid2> and yay for oncoming summer
2010-10-02T11:09:17  *** espes <espes!espes@123-243-221-248.tpgi.com.au> has quit IRC (Remote host closed the connection)
2010-10-02T11:09:40  <sigh> hmmm... time to sleep then
2010-10-02T11:09:52  <sigh> night
2010-10-02T11:09:55  <Sylph> I need to make a few more things to work then my bot hopefully will be a challenge. nn
2010-10-02T11:10:05  *** sigh <sigh!~sigh@137.206-50-210.dynamic.dsl.mel.iprimus.net.au> has quit IRC (Quit: Leaving...)
2010-10-02T11:13:27  *** smellyhippy <smellyhippy!~asmodean@about/csharp/regular/smellyhippy> has joined #aichallenge
2010-10-02T11:13:34  *** sinan <sinan!~sinan@82.212.70.90> has quit IRC (Ping timeout: 255 seconds)
2010-10-02T11:22:23  <antimatroid2> bam, ruilov has elo 3901
2010-10-02T11:22:31  <Migi32> wow
2010-10-02T11:22:45  <antimatroid2> and he's using java too :(
2010-10-02T11:23:46  <Migi32> and his only loss is a very weird loss: http://ai-contest.com/visualizer.php?game_id=5222302
2010-10-02T11:24:03  <Migi32> he didn't reinforce
2010-10-02T11:26:22  <delt0r_> not really
2010-10-02T11:26:33  <delt0r_> that map is tricky for "good" tactics...
2010-10-02T11:26:39  <delt0r_> outside that one map
2010-10-02T11:26:49  <antimatroid2> yeah, but he should have reinforced when the other person went for him
2010-10-02T11:28:54  <antimatroid2> lol, he's not far off 4000
2010-10-02T11:29:25  <antimatroid2> i've managed to climb to 30
2010-10-02T11:37:31  *** Mathnerd314 <Mathnerd314!~mathnerd3@dsl.6.28.53.206.cos.dyn.pcisys.net> has quit IRC (Quit: ChatZilla 0.9.86-rdmsoft [XULRunner 1.9.2.8/20100722155716])
2010-10-02T11:49:50  *** Olathe <Olathe!~Olathe@pool-98-111-156-209.phlapa.east.verizon.net> has joined #aichallenge
2010-10-02T11:50:04  *** Mathnerd314 <Mathnerd314!~mathnerd3@dsl.6.28.53.206.cos.dyn.pcisys.net> has joined #aichallenge
2010-10-02T11:51:52  *** sinan <sinan!~sinan@94.142.51.113> has joined #aichallenge
2010-10-02T11:52:01  <antimatroid2> lol, some bastard is using LOSE on tcp now
2010-10-02T11:52:07  <Sylph> lol
2010-10-02T11:52:15  <antimatroid2> i better beat them
2010-10-02T11:52:59  <Sylph> I doubt my bot can win much atm, still incomplete, hrm. :/
2010-10-02T11:53:02  * Sylph go rest more
2010-10-02T11:53:22  <antimatroid2> i'm working on building a minimax tree
2010-10-02T11:53:29  <FlagCapper> I'm the bastard
2010-10-02T11:53:39  <antimatroid2> i lost too :(
2010-10-02T11:53:43  <FlagCapper> :)
2010-10-02T11:53:51  <antimatroid2> i'ma get you next time
2010-10-02T11:54:05  <FlagCapper> Probably
2010-10-02T11:54:19  <FlagCapper> I'm only at around ~330 ELO
2010-10-02T11:55:48  <antimatroid2> i'm up to 14th on the main site
2010-10-02T11:55:56  <antimatroid2> few more wins and i should break top 10 again
2010-10-02T11:56:00  <antimatroid2> has been a while
2010-10-02T11:56:16  <Sylph> nice
2010-10-02T11:56:47  <antimatroid2> it's at 12/12 though, so one loss could make it drop a lot
2010-10-02T11:57:55  <jmcarthur> a loss against one of the top 10 probably wouldn't hurt too badly
2010-10-02T11:58:45  <contestbot> Top 10 players: ruilov(3972), Hazard(3787), rsergio(3737), EBraun(3656), cfaftw(3648), felixcoto(3644), dmj111(3621), sequoh(3611), _iouri_(3588), sighbot(3556)
2010-10-02T11:59:18  <antimatroid2> well i jurt kcm1700 twice in a row
2010-10-02T11:59:22  <antimatroid2> hurt*
2010-10-02T12:00:33  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T12:01:40  <Sylph> it got harder to think of the algorithm, when I'm able to predict much more of the future, somehow. >.>
2010-10-02T12:02:06  <antimatroid2> oh dear, everybody run
2010-10-02T12:02:14  <antimatroid2> McLeopold is playing on tcp again
2010-10-02T12:02:17  <Sylph> lol
2010-10-02T12:03:16  <Sylph> I'm just playing with unfinished bot atm, don't care about winning :p
2010-10-02T12:03:28  <Sylph> seems my attack script is bugged atm.
2010-10-02T12:03:44  <Sylph> not mentioning i still need to fix my defense and add front line stuff
2010-10-02T12:04:26  <antimatroid2> i don't think anyone will run out of things to add
2010-10-02T12:05:07  <Sylph> true
2010-10-02T12:05:18  <Sylph> but i mean i don't even got those basic stuff added in yet :|
2010-10-02T12:07:12  <antimatroid2> ha, i cracked top 10
2010-10-02T12:07:16  <antimatroid2> now i just need to not lose
2010-10-02T12:07:22  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has joined #aichallenge
2010-10-02T12:07:44  <antimatroid2> 4th* :D
2010-10-02T12:07:51  <krokokrusa> go antimatroid2 ! :)
2010-10-02T12:08:12  *** keynes <keynes!~AndChat@203.236.22.68> has joined #aichallenge
2010-10-02T12:08:35  <Sylph> Grats
2010-10-02T12:08:36  <Sylph> :)
2010-10-02T12:08:48  <antimatroid2> i'm pretty sure that bot on there will get caned if someone streams my planets well though
2010-10-02T12:09:12  <antimatroid2> i have only got it making up to 5 save moves before expanding each turn
2010-10-02T12:10:28  * Sylph nods
2010-10-02T12:12:00  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T12:13:45  <antimatroid2> it still hasn't played a current top 10 bot
2010-10-02T12:17:28  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has joined #aichallenge
2010-10-02T12:20:40  <antimatroid2> wow, hazards taken a real beating recently
2010-10-02T12:22:12  <FlagCapper> Wow, the top guy is almost at 4000
2010-10-02T12:22:55  <antimatroid2> he was in the 90
2010-10-02T12:22:57  <antimatroid2> s before
2010-10-02T12:23:02  <antimatroid2> i hate ' being next to enter
2010-10-02T12:23:03  *** sinan <sinan!~sinan@94.142.51.113> has quit IRC (Quit: WeeChat 0.2.6.3)
2010-10-02T12:23:44  <FlagCapper> 90s?
2010-10-02T12:23:49  <FlagCapper> As in 90th place?
2010-10-02T12:23:52  <antimatroid2> 3990's
2010-10-02T12:24:05  <antimatroid2> elo score wise
2010-10-02T12:24:13  <FlagCapper> Oh
2010-10-02T12:24:25  <FlagCapper> Well he hasn't lost, so someone else must've dragged him down
2010-10-02T12:24:37  <FlagCapper> At least he hasn't lost for awhile
2010-10-02T12:24:39  <antimatroid2> yeah, i suspect it just adjusted
2010-10-02T12:41:01  *** FlagCapper <FlagCapper!45a5adac@gateway/web/freenode/ip.69.165.173.172> has quit IRC (Quit: Page closed)
2010-10-02T12:44:38  <Migi32> ruilov is 4000
2010-10-02T12:44:57  <Migi32> and showing no sign of stopping
2010-10-02T12:46:01  <Migi32> Who is he? Has he ever been on this channel? Is he on TCP?
2010-10-02T12:46:42  <wh1teside> I'll put some dents in him... just as soon as I finish rewriting everything... for over two weeks now...
2010-10-02T12:47:00  <antimatroid2> the worst part is he's using java
2010-10-02T12:47:40  <wh1teside> this tells me he's a masochist
2010-10-02T12:47:45  <wh1teside> which implies that his AI is a risk taker
2010-10-02T12:48:46  <Migi32> it's incredible how fast this game is evolving though
2010-10-02T12:49:02  <antimatroid2> yeah, my approach is ever evolving
2010-10-02T12:49:08  <antimatroid2> and i expect everyone elses is too
2010-10-02T12:52:43  <wh1teside> does anyone else have their bot mostly on paper/napkins/whiteboard?
2010-10-02T12:53:08  <wh1teside> because I have indeed been evolving mine quite a bit, but haven't implemented the improvements in a week or so
2010-10-02T12:57:53  <antimatroid2> i haven't actually implemented many proper ideas
2010-10-02T12:57:57  <antimatroid2> most are on paper/in my mind
2010-10-02T12:58:45  <contestbot> Top 10 players: ruilov(4013), Hazard(3777), rsergio(3723), cfaftw(3641), EBraun(3640), felixcoto(3636), dmj111(3618), sequoh(3614), youwillfail(3612), _iouri_(3583)
2010-10-02T12:59:26  <Sylph> [12:38:58 02/10] <antimatroid2> the worst part is he's using java
2010-10-02T12:59:26  <Sylph> [12:39:38 02/10] <wh1teside> this tells me he's a masochist
2010-10-02T12:59:26  <Sylph> [12:39:43 02/10] <wh1teside> which implies that his AI is a risk taker
2010-10-02T12:59:27  <Sylph> lol
2010-10-02T12:59:28  <Sylph> silly
2010-10-02T12:59:59  <Sylph> I'm tempted to use java again eventhough I never really used java ever, because C# is broken
2010-10-02T13:00:03  <Sylph> might be the same with him
2010-10-02T13:00:12  <Sylph> broken on the server, that is
2010-10-02T13:00:25  <Sylph>  <wh1teside> does anyone else have their bot mostly on paper/napkins/whiteboard?
2010-10-02T13:00:29  * Sylph raises hand
2010-10-02T13:00:36  <Sylph> I just rewrote my bot from scratch in c#
2010-10-02T13:00:41  <Sylph> spent most time on the framework,
2010-10-02T13:01:12  <Sylph> haven't even finished fixing the bugs on attacking, and need to fix my defense, etc
2010-10-02T13:06:15  *** Bobng <Bobng!~T@host86-130-158-50.range86-130.btcentralplus.com> has joined #aichallenge
2010-10-02T13:06:32  <wh1teside> speaking of my masochism comment... I'm not sure that I'm not the pot calling the kettle black
2010-10-02T13:07:11  <Migi32> if a prize pool were added to the tournament, would this make me unable to use software that can't be used for "commercial use"?
2010-10-02T13:07:48  <Sylph> :p wh1teside
2010-10-02T13:09:36  <wh1teside> I went a little overboard with the STL on my last rewrite, and have several constructions of types like... map<unsigned int, map<unsigned int, pair<unsigned int, double> > > ... which makes for pretty nasty for-loops
2010-10-02T13:10:05  <Sylph> ow
2010-10-02T13:11:45  *** wvdschel <wvdschel!~wim@78-20-14-180.access.telenet.be> has joined #aichallenge
2010-10-02T13:17:47  *** FlagCapper <FlagCapper!45a5adac@gateway/web/freenode/ip.69.165.173.172> has joined #aichallenge
2010-10-02T13:19:31  *** Naith <Naith!~nickbarth@S0106001b38137d1b.ed.shawcable.net> has joined #aichallenge
2010-10-02T13:19:42  *** McLeopold <McLeopold!~scott@c-76-121-152-123.hsd1.wa.comcast.net> has joined #aichallenge
2010-10-02T13:21:57  *** boegel <boegel!~boegel@91.182.173.227> has joined #aichallenge
2010-10-02T13:25:46  *** Jahnson <Jahnson!44c53a8a@gateway/web/freenode/ip.68.197.58.138> has joined #aichallenge
2010-10-02T13:25:55  *** keynes <keynes!~AndChat@203.236.22.68> has quit IRC (Remote host closed the connection)
2010-10-02T13:32:19  *** keynes <keynes!~AndChat@203.236.22.68> has joined #aichallenge
2010-10-02T13:35:55  <Migi32> would this qualify as "research purposes"? :P
2010-10-02T13:37:05  <Sylph> i have no clue about the 'prize pool' thing :p
2010-10-02T13:37:10  <Sylph> but atm, there's no prize pool afaik
2010-10-02T13:37:19  <Sylph> so i don't see why it won't work hrm.
2010-10-02T13:37:35  <Jahnson> there's a prize?
2010-10-02T13:37:43  <Sylph> no
2010-10-02T13:37:49  <Jahnson> :'(
2010-10-02T13:37:59  <Sylph> there might be, if thigns goes well, but unlikely atm tbh
2010-10-02T13:38:13  <Sylph> i was just talking to migi about 'commercial use' worries if there's a prize pool
2010-10-02T13:38:18  <Sylph> not sure how the law is about that
2010-10-02T13:38:29  <Jahnson> oh
2010-10-02T13:40:53  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T13:41:12  <Migi32> it's surprisingly hard to find a linear programming / simplex library for c++
2010-10-02T13:42:04  *** jesionaj <jesionaj!~dongs@174-18-5-185.tcso.qwest.net> has joined #aichallenge
2010-10-02T13:45:10  * Sylph nods
2010-10-02T13:45:18  <Sylph> I'm still not sure what kind of algorithm I want to use
2010-10-02T13:45:30  <Sylph> the algorithm i'm thinking of using might not be good enough
2010-10-02T13:45:49  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has joined #aichallenge
2010-10-02T13:46:02  <Sylph> got all the info i need, just need to figure out how to use that info atm :|
2010-10-02T13:46:43  <Migi32> what info do you mean?
2010-10-02T13:49:59  <krokokrusa> that's the problem everybody is trying to solve
2010-10-02T13:50:13  <krokokrusa> got all info from stdin, just need to figure out what to do :)
2010-10-02T13:54:36  <jmcarthur> wh1teside: you could possibly simplify your loops a little by changing that to map<pair<unsigned int, unsigned int>, pair<unsigned int, double>> ... at least if you're allowed to do that in C++
2010-10-02T13:55:20  <jmcarthur> i've never abused C++ templates the way i use haskell type constructors
2010-10-02T13:56:06  <Sylph> <Migi32> what info do you mean?
2010-10-02T13:56:10  <Sylph> future info :|
2010-10-02T13:56:18  <Sylph> of like everything every turn :|
2010-10-02T13:57:17  <Migi32> Sylph, if you have all that information why are you not winning this game? Or do you mean info assuming the enemy captures no new planets etc?
2010-10-02T13:57:21  <Sylph> need to rewrite my attack script to work properly though, it only check up to taking the planet, not incoming after that
2010-10-02T13:57:34  <Sylph> getting info is not the same as having finished coding the bot behaviour algorithm
2010-10-02T13:57:50  <Sylph> I've barely started coding the bot behaviour yesterday
2010-10-02T13:58:21  <Sylph> the info include enemy capturing new planets, the growth of planets, etc
2010-10-02T13:58:46  <contestbot> Top 10 players: ruilov(3836), Hazard(3771), rsergio(3732), EBraun(3666), felixcoto(3643), cfaftw(3638), dmj111(3628), sequoh(3611), bix0r4ever(3577), _iouri_(3576)
2010-10-02T13:59:36  <Sylph> ships that they'll send from those new planets, etc
2010-10-02T13:59:51  <jmcarthur> speculatively, you mean
2010-10-02T14:00:14  <Sylph> yeah :p
2010-10-02T14:00:32  <McLeopold> Check this out: http://72.44.46.68/canvas?game_id=276162
2010-10-02T14:00:44  <antimatroid2> can has win?
2010-10-02T14:00:45  <McLeopold> The proper way to hold the line and increase growth.
2010-10-02T14:01:02  <Sylph> lol McLeopold
2010-10-02T14:01:08  <Sylph> yeah, nice
2010-10-02T14:01:12  <McLeopold> I just need to fix the whole "invalid orders" bug
2010-10-02T14:01:55  <McLeopold> I'm not sure when I introduced it, but I'm sending too many fleets all the time.
2010-10-02T14:02:29  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has quit IRC (Ping timeout: 265 seconds)
2010-10-02T14:02:47  <McLeopold> antimatroid2: at the end you just sit there and take it
2010-10-02T14:03:36  <antimatroid2> take what?
2010-10-02T14:04:20  <Migi32> McLeopold, well that aggressiveness seems to be working out for you
2010-10-02T14:05:01  <Sylph> unless against someone who defend precisely like antimatroid?
2010-10-02T14:05:29  <McLeopold> I don't really care if the enemy exists or not, I just wanna grow fast.
2010-10-02T14:05:58  <antimatroid2> i used to do that till i played dmj a few times
2010-10-02T14:06:05  <McLeopold> Sometimes that means taking his planet, sometimes not.
2010-10-02T14:06:09  <antimatroid2> whenever i expanded too fast he handed my ass to me
2010-10-02T14:06:11  *** BtbN <BtbN!~BtbN@p54BB7467.dip.t-dialin.net> has joined #aichallenge
2010-10-02T14:06:13  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has joined #aichallenge
2010-10-02T14:06:14  *** frontier <frontier!5847f22b@gateway/web/freenode/ip.88.71.242.43> has joined #aichallenge
2010-10-02T14:06:21  <McLeopold> Yeah, he's good at that.
2010-10-02T14:06:25  <antimatroid2> yep
2010-10-02T14:06:25  <Migi32> well, there's a fine line
2010-10-02T14:06:41  <antimatroid2> my bot is now a whole lot of fucking about to try and not lose to him imediately every game
2010-10-02T14:06:42  *** Nachtaktiv <Nachtaktiv!~Marco_Lei@dslb-088-071-242-043.pools.arcor-ip.net> has joined #aichallenge
2010-10-02T14:06:45  <McLeopold> True, I do have a couple of checks.
2010-10-02T14:06:47  <antimatroid2> i still usually lose
2010-10-02T14:06:50  <antimatroid2> but i get him sometimes
2010-10-02T14:07:21  <McLeopold> I think it depends on the map.  Sometimes he doesn't expand enough before he goes into lurk mode.
2010-10-02T14:07:41  <antimatroid2> yep, i'm trying to get the right balance
2010-10-02T14:07:48  <antimatroid2> but alas, it's 5am, i'm going to bed
2010-10-02T14:07:49  <antimatroid2> night
2010-10-02T14:07:52  <Sylph> nn
2010-10-02T14:07:55  *** fawek <fawek!~faw@chello089079023212.chello.pl> has quit IRC (Remote host closed the connection)
2010-10-02T14:07:58  * boegel starts looking into his bot again, after a week orso of not looking at it
2010-10-02T14:08:01  <boegel> antimatroid2: night!
2010-10-02T14:08:07  <McLeopold> For everyone who enjoyed SnakeBot, I should have WormBot up and running later today. :D
2010-10-02T14:08:18  <antimatroid2> i considered making drawbot
2010-10-02T14:08:30  <antimatroid2> that tries to get the lead then force a draw at the end
2010-10-02T14:08:35  <antimatroid2> would be awesome for teasing people
2010-10-02T14:08:50  <McLeopold> voidptr already does a really good job of that...
2010-10-02T14:09:11  <antimatroid2> i just had 3 wins in a row against him :P
2010-10-02T14:09:15  <McLeopold> http://72.44.46.68/canvas?game_id=276305
2010-10-02T14:09:21  <antimatroid2> gah, now i'm awake again aha
2010-10-02T14:09:26  <McLeopold> watch the fireworks
2010-10-02T14:09:27  <antimatroid2> that always happens when i consider going to bed
2010-10-02T14:09:36  <Nachtaktiv> This bot challenge got me keep up all night again. Hacked mine together in 2 days then. Funny thing is, it made it to rank 104.
2010-10-02T14:09:55  <antimatroid2> i won a game in 7 moves before
2010-10-02T14:10:02  <McLeopold> I think the top 100 is still pretty easy to crack into.
2010-10-02T14:10:06  <boegel> McLeopold: hah, that's cool! :D
2010-10-02T14:10:28  <McLeopold> tell void.  I was totally going to do stuff like that too at some point
2010-10-02T14:10:46  <McLeopold> I still my try to write an "M" in the game if I win.
2010-10-02T14:10:51  <antimatroid2> yeah, i was going to do a star
2010-10-02T14:10:55  *** frontier <frontier!5847f22b@gateway/web/freenode/ip.88.71.242.43> has quit IRC (Ping timeout: 265 seconds)
2010-10-02T14:11:08  <antimatroid2> hadn't got that far yet aha
2010-10-02T14:11:12  <McLeopold> I figured I'd do the fireworks when I lose.
2010-10-02T14:11:15  <antimatroid2> kudo to him for jumping in first though
2010-10-02T14:11:35  <boegel> that's so cool
2010-10-02T14:11:36  <McLeopold> Or send a "good game" morse code signal to their home planet. :)
2010-10-02T14:12:00  <antimatroid2> I think a big start would be nice
2010-10-02T14:12:02  <boegel> McLeopold: that's less obvious, no-one would know
2010-10-02T14:12:06  <antimatroid2> just gotta work out how to pick the planets :P
2010-10-02T14:12:10  <antimatroid2> then lets the fleets stream
2010-10-02T14:12:27  <McLeopold> boegel: I'm totally okay with sublime.
2010-10-02T14:12:56  <boegel> McLeopold: as long as you're winning
2010-10-02T14:13:38  <McLeopold> g is dash-dash-dot
2010-10-02T14:13:43  <Nachtaktiv> I thought about sending 1 ship fleets out to every planet every round when I lose ^^
2010-10-02T14:14:06  <McLeopold> so "gg" would be 2 fleet-2 fleet-1 fleet pause 2 fleet-2 fleet-1 fleet
2010-10-02T14:14:11  <Nachtaktiv> Or save up on fleets for a crushing defeat when I win.
2010-10-02T14:14:13  <boegel> Nachtaktiv: if you have enough fleets left :P
2010-10-02T14:14:18  <McLeopold> that's 7 turns and 10 fleets
2010-10-02T14:14:46  <McLeopold> I think I could detect a loss with that amount of time left.
2010-10-02T14:14:53  <Sylph> lol nice firework on voipdptr
2010-10-02T14:15:14  <antimatroid2> aye, i beat dmj
2010-10-02T14:15:28  <Nachtaktiv> boegel: http://ai-contest.com/visualizer.php?game_id=4946339  WIN :p
2010-10-02T14:15:56  *** jesionaj` <jesionaj`!~dongs@174-18-5-185.tcso.qwest.net> has joined #aichallenge
2010-10-02T14:16:36  <McLeopold> antimatroid2: that's a really good win
2010-10-02T14:16:41  *** Nachtaktiv is now known as Frontier
2010-10-02T14:17:08  *** jesionaj <jesionaj!~dongs@174-18-5-185.tcso.qwest.net> has quit IRC (Ping timeout: 276 seconds)
2010-10-02T14:17:24  <antimatroid2> it made some pretty stupid moves
2010-10-02T14:17:31  <antimatroid2> but had enough of a position advantage anyway
2010-10-02T14:17:52  <antimatroid2> oh, i beat him on the main site too haa
2010-10-02T14:17:56  <Frontier> wait.. dmj is NOT placed first!?
2010-10-02T14:18:05  <McLeopold> I don't think you did have better position...
2010-10-02T14:18:16  <boegel> Frontier: yeah, but *why*?
2010-10-02T14:18:21  <Sylph> he got a perfect-bot waiting, just letting people think they beat him already
2010-10-02T14:18:24  <Sylph> I bet :D
2010-10-02T14:18:44  <McLeopold> Same as me. :)
2010-10-02T14:18:49  <Sylph> :p
2010-10-02T14:20:23  *** McLeopold <McLeopold!~scott@c-76-121-152-123.hsd1.wa.comcast.net> has quit IRC (Quit: Leaving)
2010-10-02T14:20:42  <Frontier> Does anyone do crazy data management, like splitting fleets into smaller objects, maybe down to a one-object-per-ship model?
2010-10-02T14:22:21  <antimatroid2> my fleets are done as a vector for each planet
2010-10-02T14:22:32  <antimatroid2> i only care about the arrival time and where it lands
2010-10-02T14:22:44  <antimatroid2> all similar ones like that are merged into one from each player
2010-10-02T14:23:08  <Frontier> Me too, that was the forst thing for me: drop total_time, remaining_time and source and replace it by an ETA
2010-10-02T14:23:43  <antimatroid2> i don't even have a fleet class, it's just a vector<vector<vector<int> > > :P
2010-10-02T14:23:56  <Frontier> So you have an anti-denial-of-service bot that removes 0-fleets and merges similar 1-fleets?
2010-10-02T14:24:07  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has quit IRC (Ping timeout: 252 seconds)
2010-10-02T14:24:17  <antimatroid2> 0 fleets are removed by the engine now anyway
2010-10-02T14:24:36  <antimatroid2> but it wouldn't change my code much, it's just adding 0 to a vlue
2010-10-02T14:25:06  <Frontier> Well.. you would still have 0 fleets around wouldnt you? and maybe a division by 0 somewhere that kills your bot ^^
2010-10-02T14:25:53  *** jesionaj <jesionaj!~dongs@174-18-20-177.tcso.qwest.net> has joined #aichallenge
2010-10-02T14:25:54  <antimatroid2> don't divide by nonpositive fleetsizes
2010-10-02T14:26:08  <antimatroid2> and yeah sure there are 0 sized fleets around
2010-10-02T14:26:31  <antimatroid2> my worst problem atm is i'm deleting elements from the start of vectors, i really need to fix that
2010-10-02T14:26:33  <Sylph> i thought the server got fixed
2010-10-02T14:26:40  <Sylph> to ignore size 0 fleets
2010-10-02T14:26:49  <antimatroid2> yep
2010-10-02T14:27:08  <Frontier> Till now I only worked with a simple vector<Fleet>. I think about verifying the game state I recieve right now and ... in case of an error send it to my computer for debugging ^^.
2010-10-02T14:27:36  <antimatroid2> i' trying to set up an approximate minimax tree now
2010-10-02T14:27:46  <antimatroid2> so i have to fix everything up to calculate things for either player
2010-10-02T14:27:47  <Frontier> The question was more like "did we forsee the problem and was our bot vulnerable"
2010-10-02T14:29:08  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has joined #aichallenge
2010-10-02T14:29:20  <Frontier> I have a stupid bug in my bot. It only comes to defend own planets if an enemy fleet is closer than my own planets.
2010-10-02T14:29:48  <Sylph> I'd be happy once i fix htis bug on attacking
2010-10-02T14:30:10  <Sylph> :|
2010-10-02T14:30:10  *** jesionaj` <jesionaj`!~dongs@174-18-5-185.tcso.qwest.net> has quit IRC (Ping timeout: 272 seconds)
2010-10-02T14:30:13  <Frontier> You only attack when an enemy fleet is closer than your own planet? ^^
2010-10-02T14:30:22  <Sylph> lol
2010-10-02T14:30:43  <Sylph> it only make sure it get the planet, but shortcircuit there and not consider the future turns.
2010-10-02T14:30:52  <Sylph> need to write the attack function
2010-10-02T14:31:01  *** jaspervdj <jaspervdj!~jaspervdj@zeus.ugent.be> has quit IRC (Quit: NEVER GONNA GIVE YOU UP NEVER GONNA LET YOU DOWN)
2010-10-02T14:31:01  <Sylph> the predictions got all info on future turns
2010-10-02T14:31:32  <Sylph> just need to add a minor fix somewhere
2010-10-02T14:32:04  <Frontier> Yeah, I don't take change of ownership into account. So I send to few ships to defense, when the enemy gets the chance to capture the planet for 1 or more rounds.
2010-10-02T14:32:32  * Sylph nods
2010-10-02T14:32:50  <Sylph> my predictions take that into account, but not my attack function
2010-10-02T14:32:51  <Sylph> broken
2010-10-02T14:33:12  <Frontier> And when I plan on attacking an enemy planet I just add +5 to the planets production to be 'relatively' sure I get it - instead of tanking into account what the enemy could really send from its planets to defend.
2010-10-02T14:33:18  <Sylph> but so tired coding the framework, kinda lost my drive for the bot algorithm :p
2010-10-02T14:33:25  * Sylph nods
2010-10-02T14:33:43  <Sylph> mine take into account all that, but there are benefit or bad points on that.
2010-10-02T14:33:51  <jmcarthur> antimatroid2: a minimax tree? interesting
2010-10-02T14:34:13  <Frontier> So your bot is quite advanced. What ranking?
2010-10-02T14:34:18  <Sylph> no where yet
2010-10-02T14:34:20  <Sylph> it's a rewrite from scratch
2010-10-02T14:34:24  *** JCS^ <JCS^!jcs@201-92-156-159.dsl.telesp.net.br> has joined #aichallenge
2010-10-02T14:34:30  <Sylph> so it's not even got its attack function finished yet
2010-10-02T14:34:36  <Sylph> til that bug got fixed
2010-10-02T14:34:40  <Sylph> get
2010-10-02T14:34:43  <Sylph> get
2010-10-02T14:34:46  <Frontier> s?
2010-10-02T14:35:07  <Sylph> well, the attack function is done other than that bug
2010-10-02T14:35:19  <Sylph> then just need to work on finishing defend function, and front line stuff :/
2010-10-02T14:35:37  <Sylph> the framework is advanced, which actually makes it harder to code the bot
2010-10-02T14:35:42  <Sylph> >.<;;
2010-10-02T14:38:09  <Frontier> hmm. i am about to rewrite my bot with a great framework as well, but i guess i will write 50% of the functionality for the trash can and add code in the bot core algorithm that would have been nice in the framework, but doesnt really work with it. as i said my first try was a hackjob with one huge function, but went quick and did its job
2010-10-02T14:38:31  * Sylph nods
2010-10-02T14:38:55  <Frontier> I wonder if great frameworks are overrated. It's just a bot. How often will I use function X? Once, twice?
2010-10-02T14:39:22  <Sylph> when i enable my debugging, i get like 100k lines of log in 3-5 turns
2010-10-02T14:39:24  <krokokrusa> close one :) http://72.44.46.68/canvas?game_id=276913
2010-10-02T14:39:29  <Sylph> most of that is from the framework
2010-10-02T14:39:41  <Sylph> so i use them functions a lot
2010-10-02T14:39:51  <Sylph> i just wish i haven't lost my drive after working on the framework,:p
2010-10-02T14:41:03  <Frontier> Did you write a logger for the bot or did you have one from previous projects or did you use log4cpp?
2010-10-02T14:41:15  <Sylph> i'm on c# and wrote a simple logger
2010-10-02T14:41:19  <Frontier> Or is it just the plain cerr <<?
2010-10-02T14:41:27  <Sylph> it's a plain console.write
2010-10-02T14:41:28  <Sylph> in C#
2010-10-02T14:41:31  <Sylph> well, not console
2010-10-02T14:41:31  *** Jahnson <Jahnson!44c53a8a@gateway/web/freenode/ip.68.197.58.138> has quit IRC (Quit: Page closed)
2010-10-02T14:41:33  <Sylph> actually using textwriter
2010-10-02T14:41:41  <Sylph> log to a MyLog.txt
2010-10-02T14:42:09  <Sylph> just add one line each time i call the function  with a string
2010-10-02T14:42:31  <Sylph> it get nasty on resource requirement when i enabled verbose debugging
2010-10-02T14:42:41  <Frontier> I'm writing something more comfigurable right now, so I can reduce the debugging output for different parts of my code when I see them working as expected
2010-10-02T14:42:43  *** Zeiris <Zeiris!~Zeiris@d206-116-158-135.bchsia.telus.net> has joined #aichallenge
2010-10-02T14:42:53  * Sylph nods
2010-10-02T14:43:19  <Sylph> I just got debug/verbosedebugging atm,
2010-10-02T14:43:38  <Frontier> log[info] << "This looks cool in the source code. " << 12345 << endl;
2010-10-02T14:43:52  <Sylph> :p
2010-10-02T14:45:20  <Frontier> The operator[] works out if the debugging level info is in use and if so returns an output stream that is backed by a specialized logging buffer. If not it returns a dummy stream with no buffer. The operator << is standard C++ stuff.
2010-10-02T14:46:21  <Frontier> Learning all that stuff about streams was so time consuming that I lost my drive now, too.
2010-10-02T14:46:31  * Sylph nods
2010-10-02T14:46:34  <Sylph> ah
2010-10-02T14:46:37  <Sylph> yea :/
2010-10-02T14:47:05  <Sylph> working out exact future predictions to the T got me a bit of a headache
2010-10-02T14:47:09  <Sylph> :|
2010-10-02T14:47:16  <Frontier> what T?
2010-10-02T14:47:23  <Sylph> just a figure of speech
2010-10-02T14:47:32  <Sylph> working it out to this accurate
2010-10-02T14:47:35  <Sylph> and informative
2010-10-02T14:47:45  <Sylph> got me too tired for coding the bot :p
2010-10-02T14:47:56  <Sylph> I don't know too much about c++ stream, was neat though
2010-10-02T14:48:05  <Sylph> so you keep a file stream open at all time?
2010-10-02T14:48:22  <Sylph> and just write when it dump?
2010-10-02T14:48:27  <Sylph> or write at all time
2010-10-02T14:48:38  <Frontier> well, it isn't done yet, but I want to open the files only when I actually write something into them
2010-10-02T14:48:47  * Sylph nods
2010-10-02T14:49:27  <Frontier> i have to write on every 'endl' (end-line) in case the program crashes or gets crashed by a certain server code that recently fixed 0 fleets
2010-10-02T14:49:35  <Sylph> ah
2010-10-02T14:49:43  <Sylph> lol
2010-10-02T14:49:56  <Sylph> did you send 0 fleets out?
2010-10-02T14:50:10  <Frontier> No, the idea was too far off for me ^^
2010-10-02T14:50:17  <Sylph> :p
2010-10-02T14:50:53  <Frontier> But I know people had their bots die because of it. I don't know if their code was too slow or the input phase took too long with all these fleets
2010-10-02T14:52:21  *** Appleman1234 <Appleman1234!~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au> has quit IRC (Ping timeout: 272 seconds)
2010-10-02T14:52:21  <Frontier> Do your predictions include ... uh ... predictions? Do you make assumptions about the enemy starting n ships from planet a to b?
2010-10-02T14:52:46  <Sylph> hrm, it is a worst case predictions
2010-10-02T14:52:49  <Sylph> on everything
2010-10-02T14:53:06  <Sylph> and what's actually happening
2010-10-02T14:53:12  *** Appleman1234 <Appleman1234!~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au> has joined #aichallenge
2010-10-02T14:53:17  <Frontier> My next bot must have that as well. You talk about the maximum of ships the enemy could send to a certain planet, right?
2010-10-02T14:53:25  * Sylph nods
2010-10-02T14:53:29  <Frontier> I call it the 'ship-stream'
2010-10-02T14:53:43  <Sylph> i assume with both of those info, I can do my best
2010-10-02T14:53:44  <Frontier> not to confuse with the slip-stream
2010-10-02T14:53:52  <Sylph> predictions
2010-10-02T14:53:55  <Sylph> on what i need to do
2010-10-02T14:54:05  *** boegel <boegel!~boegel@91.182.173.227> has quit IRC (Quit: This computer has gone to sleep)
2010-10-02T14:54:06  <Sylph> if only my brain still work after putting all that in my framework
2010-10-02T14:56:56  <Frontier> I don't know yet what to do with those worst-case predictions. They don't reflect what is happening very well. If the opponent left all of its planets unguarded for an attack on a 1 growth planet, we should give it up and attack him, right?
2010-10-02T14:57:19  <Sylph> naturally
2010-10-02T14:57:52  <Sylph> my bot code will implement that when I get it going
2010-10-02T14:58:02  <Frontier> Looks like a lot of trial and error to get the numbers right. I'm not such a genius to exactly calculate the planet's value
2010-10-02T14:58:03  <Sylph> if not already
2010-10-02T14:58:46  <contestbot> Top 10 players: ruilov(3867), Hazard(3765), rsergio(3724), EBraun(3648), felixcoto(3637), cfaftw(3632), dmj111(3621), sequoh(3602), _iouri_(3565), youwillfail(3564)
2010-10-02T14:59:17  <Sylph> I just need to fix my bot functions :/
2010-10-02T15:00:53  <Frontier> You have all the time ^^. I think I need to put some thought into the definition of a 'lead in growth'. Even one ship more per round means you win if you can keep it. So how does that go into the attack/defense calculations.
2010-10-02T15:01:59  <Sylph> yea, i haven't even gotten that far.
2010-10-02T15:02:00  <Frontier> I see a lot of matches end in a stale mate where the situation of bot A gets gradually worse because it missed a +1 planet somewhere near its home planet
2010-10-02T15:02:05  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has quit IRC (Ping timeout: 252 seconds)
2010-10-02T15:02:20  <Sylph> my bot is good at deciding what planet is good to attack, just need 2 bugs to be fixed
2010-10-02T15:02:42  <Sylph> on the attack
2010-10-02T15:02:48  <Sylph> one major bug, and the other strategic bug
2010-10-02T15:03:02  <Frontier> But what does it do when it is in a stale mate with one planet small behind in production?
2010-10-02T15:03:11  <Frontier> *one small planet
2010-10-02T15:03:57  <Sylph> not sure
2010-10-02T15:04:03  <Sylph> shouldn't have been there in the first place
2010-10-02T15:04:07  <Sylph> if it is worth taking :|
2010-10-02T15:04:18  <Frontier> It is not worth taking
2010-10-02T15:04:26  * Sylph nods
2010-10-02T15:04:31  *** Frontier <Frontier!~Marco_Lei@dslb-088-071-242-043.pools.arcor-ip.net> has left #aichallenge
2010-10-02T15:04:32  <Sylph> then probably draw :/
2010-10-02T15:04:44  <Sylph> if you take it, you lose that much ship enemy can use to attack you too >.<;;
2010-10-02T15:04:50  <Sylph> no idea
2010-10-02T15:04:56  <Sylph> i wish i got my functions done so i can see
2010-10-02T15:05:03  *** Frontier <Frontier!~Marco_Lei@dslb-088-071-242-043.pools.arcor-ip.net> has joined #aichallenge
2010-10-02T15:05:06  <Frontier> sorry, closed my browser accidently
2010-10-02T15:05:11  <Sylph> [14:56:23 02/10] <Sylph> then probably draw :/
2010-10-02T15:05:11  <Sylph> [14:56:36 02/10] <Sylph> if you take it, you lose that much ship enemy can use to attack you too >.<;;
2010-10-02T15:05:11  <Sylph> [14:56:42 02/10] <Sylph> no idea
2010-10-02T15:05:11  <Sylph> [14:56:47 02/10] <Sylph> i wish i got my functions done so i can see
2010-10-02T15:05:13  <Sylph> wb
2010-10-02T15:05:20  <Sylph> I hate using those web browser irc
2010-10-02T15:05:33  <Sylph> addons, client side, server side :|
2010-10-02T15:05:41  <Sylph> especially since firefox need restart every a while
2010-10-02T15:05:48  <Frontier> No it isn't firefox, it is opera ^^
2010-10-02T15:05:52  * Sylph nods
2010-10-02T15:05:59  <Sylph> don't know how good opera is on memory leak
2010-10-02T15:07:39  <Frontier> Hmm... never had any trouble with it so far and I like how it tries to have everything inclusive. (BitTorrent, Mail, IRC, Feeds, Web-Server, ...)
2010-10-02T15:07:46  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has joined #aichallenge
2010-10-02T15:07:47  * Sylph nods
2010-10-02T15:08:11  <Sylph> i haven't read my feeds for ages :/
2010-10-02T15:08:36  <Frontier> For example look at this useless app: http://admin.home.nerdiii.operaunite.com/whiteboard/
2010-10-02T15:08:47  <Frontier> You can draw something there and I can see it.
2010-10-02T15:09:40  *** boegel <boegel!~boegel@91.182.173.227> has joined #aichallenge
2010-10-02T15:09:41  <Frontier> That is an add-on though.
2010-10-02T15:09:51  * Sylph nods
2010-10-02T15:09:58  <Sylph> that's neat
2010-10-02T15:10:12  <Sylph> in some kind of opera official site?
2010-10-02T15:10:19  <Sylph> or some large opera community site ?
2010-10-02T15:11:59  <Sylph> xD
2010-10-02T15:12:01  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T15:12:16  <Sylph> okay i stop writing there
2010-10-02T15:14:06  <Frontier> http://www.youtube.com/watch?v=BuFMacUqKsU
2010-10-02T15:14:58  <Frontier> The add-ons are listed in opera itself. They call it Opera Unite.
2010-10-02T15:15:17  *** holmar <holmar!~holmar@89-160-163-48.du.xdsl.is> has quit IRC (Ping timeout: 252 seconds)
2010-10-02T15:15:28  * Sylph nods
2010-10-02T15:15:29  *** HakanD <HakanD!~hakand@78.160.65.83> has joined #aichallenge
2010-10-02T15:15:33  <Sylph> that's cool
2010-10-02T15:15:50  <HakanD> hi
2010-10-02T15:16:28  <Frontier> hi HakanD
2010-10-02T15:16:38  <Sylph> hi :)
2010-10-02T15:18:20  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has joined #aichallenge
2010-10-02T15:19:36  <HakanD> Sylph, is your bot working?
2010-10-02T15:19:43  <Sylph> not yet
2010-10-02T15:19:47  <Sylph> well, it plays
2010-10-02T15:19:56  <Sylph> have a bug on the attack function, and defend function is only half done
2010-10-02T15:20:01  <Sylph> major bug
2010-10-02T15:20:10  <Sylph> but at least it plays
2010-10-02T15:21:56  <HakanD> heh, still seems lots of work (:
2010-10-02T15:21:59  <Sylph> yeah :/
2010-10-02T15:22:06  <Sylph> and kinda losing my drive some :p
2010-10-02T15:23:00  *** Accoun <Accoun!~1@213.88.2.78> has quit IRC ()
2010-10-02T15:24:50  <Sylph> Okay, I think I get how I need to do this >.<;;
2010-10-02T15:24:53  *** impomatic <impomatic!~chatzilla@87.115.83.45> has joined #aichallenge
2010-10-02T15:25:28  <Frontier> When I see my bot in the rankings around place 104. It seems to say "I want to be in the top 100, please work on me."
2010-10-02T15:25:33  <Sylph> lol
2010-10-02T15:25:40  <Sylph> that'll be good incentive
2010-10-02T15:25:47  <Sylph> I only  got my bot that I made on my first day
2010-10-02T15:25:53  <Sylph> with a bit of tweaking on 2nd day :/
2010-10-02T15:26:20  <Sylph> which is beating my current bot that I spent 5 days on, because I've not even half done yet with it. :p
2010-10-02T15:26:40  <Frontier> Lol, guess what RageBot is doing to me every now and then.
2010-10-02T15:26:45  <Sylph> lol
2010-10-02T15:28:18  <Sylph> lol
2010-10-02T15:28:25  <Sylph> cat jumped into a large empty box :|
2010-10-02T15:29:32  <Frontier> Is it Schrödinger's cat? Can you predict where it will be in 10 minutes?
2010-10-02T15:30:35  <Sylph> seems it jumped out, very unpredictable. xd
2010-10-02T15:31:26  <Sylph> It came into existence when I looked into the box.
2010-10-02T15:34:03  *** Accoun <Accoun!~1@213.88.2.78> has joined #aichallenge
2010-10-02T15:36:57  <Zannick> my bot was in the top ten once
2010-10-02T15:37:02  <Zannick> ...when there were only ten bots :P
2010-10-02T15:37:23  <Sylph> my bot is in the top ten
2010-10-02T15:37:24  <Sylph> on my PC
2010-10-02T15:37:25  <Sylph> :P
2010-10-02T15:37:37  *** ratatata <ratatata!58f251fc@gateway/web/freenode/ip.88.242.81.252> has joined #aichallenge
2010-10-02T15:37:57  <u1tr0n> my bot top 1500 jn my pc :D he lose even starter kit bot :D
2010-10-02T15:38:20  *** ratatata <ratatata!58f251fc@gateway/web/freenode/ip.88.242.81.252> has quit IRC (Client Quit)
2010-10-02T15:39:47  *** fawek <fawek!~faw@chello089079023212.chello.pl> has joined #aichallenge
2010-10-02T15:43:10  *** FBP <FBP!~chatzilla@207.160.210.187> has joined #aichallenge
2010-10-02T15:43:18  <Sylph> lol
2010-10-02T15:43:31  <Sylph> crashing or something ?
2010-10-02T15:44:09  <u1tr0n> yep
2010-10-02T15:44:15  <Sylph> ow
2010-10-02T15:44:17  <Sylph> what language again ?
2010-10-02T15:44:23  <u1tr0n> andd delete last & start new one
2010-10-02T15:44:39  <u1tr0n> still php
2010-10-02T15:44:43  <Sylph> ah yeah
2010-10-02T15:44:45  <Sylph> forgot
2010-10-02T15:44:52  <u1tr0n> i dont want win i want make good bot on php
2010-10-02T15:44:57  <Sylph> ah
2010-10-02T15:45:02  <Sylph> I want to win, but i'm tired coding it
2010-10-02T15:45:50  <u1tr0n> im tired of debug :( it hard on php
2010-10-02T15:46:18  <u1tr0n> i have alot ot ideas but debuging all killing me
2010-10-02T15:46:40  <Sylph> php debugging is pretty easy, just print out/write errors and stuff
2010-10-02T15:47:01  <Sylph> not sure what you mean xd
2010-10-02T15:47:18  <u1tr0n> yep my log for 200 turns is more then 2mb
2010-10-02T15:47:22  * Sylph nods
2010-10-02T15:47:43  <u1tr0n> and i need to read and calk whay or what wrong
2010-10-02T15:47:44  <Sylph> my verbose log got to >100k lines in 5 turns. I think I need to make it less verbose :/
2010-10-02T15:48:26  <u1tr0n> or write program which will read log and say you whats wrong ;)
2010-10-02T15:49:12  <Sylph> if you do, tell me :p
2010-10-02T15:49:15  <Sylph> i want one :P
2010-10-02T15:49:54  <Frontier> it is called grep "ERROR"
2010-10-02T15:50:00  <Sylph> lol
2010-10-02T15:50:08  <u1tr0n> :)
2010-10-02T15:52:58  <impomatic> Someone just added planet wars to the programming games wiki so I thought I'd investigate :-)
2010-10-02T15:54:36  <Sylph> :)
2010-10-02T15:54:43  <Sylph> welcome
2010-10-02T15:54:46  <Sylph> it's fun
2010-10-02T15:54:49  <Sylph> It's a game
2010-10-02T15:55:03  <Sylph> til you try too hard
2010-10-02T15:55:11  <Sylph> to make something perfect. :p
2010-10-02T15:55:15  <Sylph> then it can become headache :p
2010-10-02T15:57:01  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has quit IRC (Ping timeout: 265 seconds)
2010-10-02T15:57:18  <FBP> impomatic: I just found it that way to. Well, by reddit -> wiki -> here
2010-10-02T15:57:36  <Sylph> :)
2010-10-02T15:57:59  <HakanD> join before its too late (:
2010-10-02T15:58:09  <FBP> So, I'm guessing that at this point the best bots are already quite sophisticated?
2010-10-02T15:58:27  <impomatic> I need to figure what's going on first. :-)  I normally play Corewar.
2010-10-02T15:58:38  <impomatic> #corewars is on freenode too!
2010-10-02T15:58:39  <contestbot> impomatic: An error occured while trying to show the previous error.
2010-10-02T15:58:48  <contestbot> Top 10 players: ruilov(3891), Hazard(3783), rsergio(3723), felixcoto(3666), dmj111(3641), cfaftw(3638), EBraun(3637), youwillfail(3615), bix0r4ever(3614), sequoh(3590)
2010-10-02T15:58:57  <krokokrusa> how active are corewars now ?
2010-10-02T15:59:06  <Zannick> hahaha, ignore contestbot. he thinks any line starting with #@! or his name is a command
2010-10-02T15:59:14  <Zannick> contestbot: ain't that right?
2010-10-02T15:59:15  <contestbot> Zannick: An error occured while trying to show the previous error.
2010-10-02T15:59:23  <Zannick> (he means yes)
2010-10-02T15:59:32  <krokokrusa> impomatic: is it mostly genetic algorithms based bots now ?
2010-10-02T15:59:34  <impomatic> Corewar has a small active community.  Still a few new ideas even after 25 years :-)
2010-10-02T15:59:36  <Frontier> !how dumb is contestbot?
2010-10-02T15:59:37  <contestbot> Frontier: Run as fast as you can and don't look back.
2010-10-02T15:59:52  <Frontier> !You are talking to me?
2010-10-02T15:59:53  <contestbot> Frontier: An error occured while trying to show the previous error.
2010-10-02T16:00:01  <Frontier> !Thought so -.-
2010-10-02T16:00:02  <contestbot> Frontier: You have no gotten any error messages recently, so here's a random one just to let you know that we care.
2010-10-02T16:00:19  <Sylph> <FBP> So, I'm guessing that at this point the best bots are already quite sophisticated?
2010-10-02T16:00:25  <impomatic> krokokrusa: no, the GA are good making warriors for small memories (80 or 800 cells).
2010-10-02T16:00:27  <Sylph> still relatively simple from what I heard
2010-10-02T16:00:43  <Sylph> though people are working on more sophisticated bots
2010-10-02T16:00:49  <Zannick> sophistication is hard to program
2010-10-02T16:00:49  <FBP> hmm. Goodbye free time.
2010-10-02T16:00:54  <Sylph> yea
2010-10-02T16:01:12  <impomatic> But hand-written warriors are better in larger memories (and just as good in the small memories)
2010-10-02T16:01:28  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has joined #aichallenge
2010-10-02T16:01:44  <FBP> I can't focus on my physics homework because I'm to busy trying to figure out the most important algorithms and select the right data structures for good performance
2010-10-02T16:01:54  <Frontier> FBP: My bot is @104 in the ranking and took 2 days of work. Just don't try to make your first bot the perfect program ;)
2010-10-02T16:02:12  <Sylph> nice
2010-10-02T16:02:26  <Sylph> mine only got to 300 in 2 days of work, though most of those work was not coding the bot itself
2010-10-02T16:02:27  <Frontier> FBP: It performes very well CPU time wise, because it doesn't predict much.
2010-10-02T16:02:46  <FBP> I think my first bot will probably just go with the "capture as many planets as possible and then evade conflict" strategy for my first.
2010-10-02T16:03:08  <Frontier> Sylph: I tried it on the tcp server and tuned it half a day.
2010-10-02T16:03:13  * Sylph nods
2010-10-02T16:03:21  <Sylph> i didn't know how to connect to tcp server yet :/
2010-10-02T16:03:22  <FBP> Hmmm, which came first, Galcon or Eufloria (or whatever it was called when it was a prototype)?
2010-10-02T16:03:27  <Sylph> remember attacking enemy = enemy lose units, fbp
2010-10-02T16:03:29  <Sylph> Not sure
2010-10-02T16:03:33  <Sylph> i think eufloria was first?
2010-10-02T16:03:37  <Sylph> dyson
2010-10-02T16:03:41  <Sylph> was its original name
2010-10-02T16:03:57  <FBP> I need to go and beat eufloria. I love it, but it takes a while.
2010-10-02T16:04:03  * Sylph nods
2010-10-02T16:04:03  <Frontier> ok .. i did have another bot before, but that not much more than extending on the starter kit
2010-10-02T16:04:09  <Sylph> yeah
2010-10-02T16:04:18  <Sylph> that was the bot i coded in 1-2 day when i joined xd
2010-10-02T16:04:24  <Sylph> just extending starter kit
2010-10-02T16:04:38  <Sylph> only got to 300
2010-10-02T16:04:45  <Frontier> Sylph: download the tcp.c file and compile it (instructions come in the file)
2010-10-02T16:04:54  <Sylph> no need, i already got access to the tcp stuff
2010-10-02T16:05:03  <Sylph> through the tcp.exe someone compiled already, so don't need to compile more
2010-10-02T16:05:18  <greghaynes> oo, I should make a .exe for the gui
2010-10-02T16:05:28  <Sylph> that'll be nice
2010-10-02T16:05:41  <Sylph> I have a .bat that make my bot fight 10 times in a row atm
2010-10-02T16:05:53  *** delt0r___ <delt0r___!~delt0r@188-22-161-210.adsl.highway.telekom.at> has joined #aichallenge
2010-10-02T16:06:02  <Sylph> so i don't have to re-run the file :|
2010-10-02T16:06:33  <Frontier> Batching hides the fact that your opponets change from day to day. One day you play 8 times against dmj, another day against a starter ^^
2010-10-02T16:07:17  *** delt0r_ <delt0r_!~delt0r@188-22-161-210.adsl.highway.telekom.at> has quit IRC (Ping timeout: 252 seconds)
2010-10-02T16:07:26  <Frontier> Or, I just want to make up a reason why I don't use it...
2010-10-02T16:07:29  <Sylph> yea
2010-10-02T16:07:32  <Sylph> but you still get multiple fights
2010-10-02T16:07:37  <Sylph> from different people
2010-10-02T16:08:00  <Sylph> you get a few pairs of same opponents, but often different
2010-10-02T16:09:26  *** MarcAndrePC <MarcAndrePC!6015a376@gateway/web/freenode/ip.96.21.163.118> has joined #aichallenge
2010-10-02T16:09:38  <MarcAndrePC> yay my new submission finally work :)
2010-10-02T16:09:56  <MarcAndrePC> anyone had trouble with submission in java in which you have package?
2010-10-02T16:10:14  <krokokrusa> yep, packages don't work
2010-10-02T16:10:31  <MarcAndrePC> that's why my bot always timeout on the server
2010-10-02T16:10:33  * krokokrusa has to flatten his sources every time before submitting
2010-10-02T16:11:27  <krokokrusa> it's hard to maintain 3k lines of java code without packages :)
2010-10-02T16:11:28  <MarcAndrePC> any easy way to do that except moving everything to the default package and then remove package and import line?
2010-10-02T16:11:44  <MarcAndrePC> yea I think the same thing
2010-10-02T16:11:55  <krokokrusa> i'm using Eclipse, just drag'n'dropping everything to the default package
2010-10-02T16:12:25  <krokokrusa> commit to a separate branch, then go to master and continue development..
2010-10-02T16:12:49  <krokokrusa> somebody should figure the patch for compile_everything.py :)
2010-10-02T16:12:50  <boegel> when one allied and one hostile fleet arrive in the same turn with the same number of ships on a non-neutral, nothing changes, right?
2010-10-02T16:12:52  <MarcAndrePC> hmmm I delete the old package after I move them, that give me an error with the import...
2010-10-02T16:13:06  <MarcAndrePC> if I leave the import there is it a problem?
2010-10-02T16:13:10  <boegel> s/non-neutral/non-neutral planet/g
2010-10-02T16:13:42  <krokokrusa> MarcAndrePC: I'd say better remove old packages and all imports to be sure
2010-10-02T16:13:52  <MarcAndrePC> ok
2010-10-02T16:14:16  <krokokrusa> also, if you use enums, better use fully-qualified names
2010-10-02T16:14:45  <krokokrusa> I've had some problems with enum usages when everything is in the default package
2010-10-02T16:15:05  <MarcAndrePC> enum in java?
2010-10-02T16:15:08  *** yasith <yasith!~yasith@112.135.78.217> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T16:15:25  <krokokrusa> yes
2010-10-02T16:15:52  <Zannick> boegel: with the same number of ships, yes.
2010-10-02T16:16:12  <MarcAndrePC> http://download.oracle.com/javase/tutorial/java/javaOO/enum.html
2010-10-02T16:16:18  <MarcAndrePC> I guess I just never used them
2010-10-02T16:16:30  <MarcAndrePC> what rank are you krok?
2010-10-02T16:16:49  <krokokrusa> 104
2010-10-02T16:17:11  <krokokrusa> was in 100 until recently :/
2010-10-02T16:17:31  <MarcAndrePC> was in 100 before my new sub
2010-10-02T16:17:34  <krokokrusa> worst thing is that i'm almost out of ideas, doing mostly bug fixing :)
2010-10-02T16:18:02  <MarcAndrePC> 971 at the moment
2010-10-02T16:18:04  <MarcAndrePC> yea
2010-10-02T16:18:26  <krokokrusa> I'd like to move towards something iterative deepening search-like
2010-10-02T16:18:40  <krokokrusa> instead of bunch of heuristics :)
2010-10-02T16:18:48  <MarcAndrePC> it a big problem, I have plenty of "good idea" but then I run it against my old bot and it don't give good result
2010-10-02T16:18:49  <Zannick> good luck.
2010-10-02T16:19:02  <Zannick> each player has 5151 possible moves on the first turn
2010-10-02T16:19:02  <Sylph> lol
2010-10-02T16:19:15  <MarcAndrePC> only 5151...
2010-10-02T16:19:22  <krokokrusa> Zannick: obviously there is a need for some wise directed search
2010-10-02T16:19:25  <Zannick> it increases
2010-10-02T16:19:28  <Zannick> very fast
2010-10-02T16:19:36  <Sylph> exponentially :p
2010-10-02T16:19:40  <Zannick> i figured out the formula this morning
2010-10-02T16:19:56  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T16:20:03  <krokokrusa> I'm just trying to say that better bot should use full second for thinking
2010-10-02T16:20:26  <MarcAndrePC> what I realise is that's it's usually the defensive function and the taking neutral function that make a big difference
2010-10-02T16:20:41  <MarcAndrePC> any offensive move I make doesn't make much change
2010-10-02T16:20:44  <Zannick> for each player, the number of possible sets of orders that can be made is the product over each planet they own of C(planet.num_ships + num_planets - 1, num_planets - 1)
2010-10-02T16:20:48  *** javagamer <javagamer!~javagamer@ool-4356d788.dyn.optonline.net> has joined #aichallenge
2010-10-02T16:20:51  <Zannick> where C is the choose function
2010-10-02T16:21:29  <Zannick> er
2010-10-02T16:21:38  <Zannick> 5151 is for a simple 3-planet case
2010-10-02T16:21:41  <Zannick> sorry
2010-10-02T16:21:53  <Zannick> the actual number is 122 choose 22.
2010-10-02T16:22:17  <Zannick> which is 9.41388x10^23
2010-10-02T16:23:01  <Frontier> another cool number: 21 planets with 3 possible owners = 3^21 = 10,460,353,203 possible game states (ignoring ships and population)
2010-10-02T16:23:09  <Zannick> i thought that number seemed a bit low...
2010-10-02T16:23:44  <boegel> DanielVF: can you check out this game? http://ai-contest.com/visualizer.php?game_id=5257040
2010-10-02T16:24:07  <boegel> DanielVF: I suspect my adversary just crashed because it had no more planets left, but I just want to make sure... :)
2010-10-02T16:24:50  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has joined #aichallenge
2010-10-02T16:24:52  <Sylph> People prefer to crash than die maybe :p
2010-10-02T16:25:03  <boegel> MarcAndrePC: I agree, I'm ranked ~250, but my offense really sucks
2010-10-02T16:25:11  <Frontier> You can reduce your number by saying that planets are only attacked with enough ships
2010-10-02T16:25:14  <krokokrusa> boegel: I think this game is a clear win
2010-10-02T16:25:16  <boegel> MarcAndrePC: defense and conquering is decent
2010-10-02T16:25:33  <boegel> krokokrusa: yeah, but in the last turn, he should be regaining two planets, but the game ends
2010-10-02T16:25:37  <krokokrusa> 51 turn show opponent's two remaining fleets dying
2010-10-02T16:25:41  <Frontier> boegel: don't call that an opponent, it was clearly depressive and commiting suicide Oo
2010-10-02T16:25:49  <krokokrusa> no, you forgot to account for planet's growths
2010-10-02T16:25:49  <Sylph> did you count planet growth, boegel?
2010-10-02T16:25:53  <boegel> Frontier: :)
2010-10-02T16:25:59  <Sylph> :p krokokrusa
2010-10-02T16:26:00  <boegel> Sylph: ah, right, my bad :
2010-10-02T16:26:01  <boegel> :)
2010-10-02T16:26:15  <boegel> krokokrusa: yeah, right, sorry
2010-10-02T16:26:26  * boegel needs to get back into this
2010-10-02T16:27:46  *** yasith <yasith!~yasith@112.135.79.120> has joined #aichallenge
2010-10-02T16:28:21  *** Naith <Naith!~nickbarth@S0106001b38137d1b.ed.shawcable.net> has quit IRC (Quit: Naith)
2010-10-02T16:31:11  <MarcAndrePC> lol you beat my old bot 2 time krokokrusa
2010-10-02T16:32:07  <krokokrusa> let's see how the new one will work :)
2010-10-02T16:32:43  <MarcAndrePC> meh
2010-10-02T16:33:08  *** boegel <boegel!~boegel@91.182.173.227> has quit IRC (Quit: Leaving)
2010-10-02T16:33:52  <MarcAndrePC> It will take some time, I'm getting match up against people in 1000
2010-10-02T16:34:07  <Sylph> it should be quick to get back to 300+
2010-10-02T16:34:27  <Sylph> but what were your rank ?
2010-10-02T16:34:28  *** Florian95 <Florian95!~Florian@per92-9-88-160-132-90.fbx.proxad.net> has joined #aichallenge
2010-10-02T16:34:33  <Sylph> I forgot
2010-10-02T16:34:59  <MarcAndrePC> 80
2010-10-02T16:35:04  <Sylph> ah
2010-10-02T16:35:17  <Sylph> shouldn't be long if you get enough matches
2010-10-02T16:35:57  <MarcAndrePC> http://www.ai-contest.com/visualizer.php?game_id=5257854
2010-10-02T16:36:02  <MarcAndrePC> poor guy
2010-10-02T16:36:22  <Sylph> lol
2010-10-02T16:36:26  <Sylph> meanie
2010-10-02T16:36:30  <Sylph> he's new, don't beat on him
2010-10-02T16:36:48  <Sylph> he's on starter bot :|
2010-10-02T16:37:38  <MarcAndrePC> hmm maybe he just increase the number of fleet he can send at the same time
2010-10-02T16:37:50  <Sylph> lol probs
2010-10-02T16:38:44  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T16:41:01  *** krokokrusa <krokokrusa!~zvold@89.31.118.163> has left #aichallenge
2010-10-02T16:43:16  *** hellman <hellman!~hellman@nat232-248-205-109.tvoe.tv> has quit IRC (Remote host closed the connection)
2010-10-02T16:44:22  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has joined #aichallenge
2010-10-02T16:51:27  *** ebrahim <ebrahim!~ebrahim@unaffiliated/ebrahim> has joined #aichallenge
2010-10-02T16:51:57  <Frontier> They should keep your previous ranking when you upload a new bot and start from there :)
2010-10-02T16:52:05  <Sylph> many says that :p
2010-10-02T16:52:25  <Sylph> especially since you can test out your bot on the tcp and local visualizer anyway
2010-10-02T16:52:30  <Sylph> so no reason for your bot to get any worse
2010-10-02T16:52:37  <Sylph> and dumped to rank 2k again
2010-10-02T16:52:45  <jmcarthur> it can't work that way with bayeselo
2010-10-02T16:53:07  <Frontier> explain pls :)
2010-10-02T16:53:58  <jmcarthur> it runs as a batch, counting all games for all submissions at once. new submissions are (rightly) treated as distinct from old one so that old games don't affect your score positively or negatively. since it's a batch, this is really the only way to do it
2010-10-02T16:54:06  <jmcarthur> you can't gradually phase out the affects of old games over time
2010-10-02T16:54:10  <jmcarthur> *effects
2010-10-02T16:54:41  <Sylph> I guess it might cause new games find it harder to increase in rank?
2010-10-02T16:54:45  <Sylph> mm
2010-10-02T16:54:48  <Sylph> new bots i mean
2010-10-02T16:54:54  <Sylph> compared to old ones, because of previous fights
2010-10-02T16:54:57  <jmcarthur> imagine you have 100 games on an old version of your bot and then you upload a new one
2010-10-02T16:55:17  <jmcarthur> if you count all those old games too then your rank is basically meaningless
2010-10-02T16:55:57  * Sylph nods
2010-10-02T16:56:11  <Sylph> making it harder and harder to move from the old rank
2010-10-02T16:56:16  <Sylph> the more you play/upload new bots
2010-10-02T16:56:27  <jmcarthur> and since it's a batch, you can't just keep some score around between submissions. such a persistent score simply doesn't exist
2010-10-02T16:56:36  <Sylph> But any way to show the old fights so we can still watch them? Maybe separate windows, since they're not deleted
2010-10-02T16:56:41  <Sylph> separate page i mean
2010-10-02T16:56:49  <Sylph> ah
2010-10-02T16:56:51  <jmcarthur> in fact, the scores before the rank update and after it are unrelated
2010-10-02T16:57:52  <jmcarthur> if i wanted to save replays from old versions of my bot i would bookmark or download them before submitting a new bot
2010-10-02T16:58:18  <jmcarthur> it would be convenient to have those available from your profile, but i don't know that it would be worth the trouble to implement that
2010-10-02T16:58:47  <contestbot> Top 10 players: ruilov(3888), Hazard(3780), rsergio(3726), bix0r4ever(3691), felixcoto(3666), dmj111(3649), EBraun(3648), cfaftw(3639), youwillfail(3605), sequoh(3585)
2010-10-02T16:59:14  <Sylph> yeah
2010-10-02T16:59:42  <Sylph> still, not very convenient for seeing other people's previous fights
2010-10-02T17:03:17  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has quit IRC (Ping timeout: 264 seconds)
2010-10-02T17:03:38  *** Appleman1234 <Appleman1234!~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au> has quit IRC (Ping timeout: 252 seconds)
2010-10-02T17:08:08  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has joined #aichallenge
2010-10-02T17:10:13  *** nullkuhl <nullkuhl!~nullkuhl@196.221.230.20> has joined #aichallenge
2010-10-02T17:12:28  <Frontier> I lost the source code of my rank 104 bot, can I download it from the server again :D
2010-10-02T17:13:40  *** aerique <aerique!~euqirea@aerique.xs4all.nl> has joined #aichallenge
2010-10-02T17:13:46  <HakanD> ask amstan he might be able to help
2010-10-02T17:14:13  *** elijahbal_ <elijahbal_!bbear@all.your.bases.belong.to.shellium.org> has quit IRC (Ping timeout: 272 seconds)
2010-10-02T17:14:15  <Sylph> ouch
2010-10-02T17:14:28  <Sylph> not in source control ?
2010-10-02T17:15:04  <Sylph> I'm rewriting my ship needed, roi calculation, attack function because of an important bug. >.<
2010-10-02T17:15:05  <Sylph> oh well
2010-10-02T17:15:18  <HakanD> hey nice to see antimatroid2 on top on
2010-10-02T17:15:24  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has quit IRC (Ping timeout: 276 seconds)
2010-10-02T17:15:30  <Frontier> Actually, I'm just not sure I still have that version of source codes. And no - I did not use SVN or GIT or CVS.
2010-10-02T17:15:35  <Sylph> ah
2010-10-02T17:15:37  <HakanD> I just realized
2010-10-02T17:15:50  <Sylph> yeah its kinda hard to remember the exact config at certain times
2010-10-02T17:16:38  *** McLeopold <McLeopold!~scott@97-126-22-239.tukw.qwest.net> has joined #aichallenge
2010-10-02T17:16:54  *** elijahbal <elijahbal!bbear@all.your.bases.belong.to.shellium.org> has joined #aichallenge
2010-10-02T17:17:02  <Frontier> I like how people version on the tcp server: 32-bit hashes, abcde..., 1 2 3 4..., cryptical stuff
2010-10-02T17:18:06  <Frontier> Some people have 10 versions of their bot running simultaniously to test their tuning.
2010-10-02T17:18:32  <amstan> Frontier: ask tonight
2010-10-02T17:19:12  <Frontier> When is tonight in your country? I have 23:11 right now.
2010-10-02T17:19:16  *** Appleman1234 <Appleman1234!~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au> has joined #aichallenge
2010-10-02T17:19:30  <HakanD> (:
2010-10-02T17:19:49  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has joined #aichallenge
2010-10-02T17:20:03  <Frontier> don't tell me tonight is someones nick name ^^
2010-10-02T17:21:02  <Migi32> hey Frontier where're you from? It's 23:11 (well 23:13 now) here as well
2010-10-02T17:22:15  *** Zeiris <Zeiris!~Zeiris@d206-116-158-135.bchsia.telus.net> has quit IRC (Ping timeout: 264 seconds)
2010-10-02T17:22:40  <Frontier> amstan: I don't want to cause you any trouble. If it is more than just copying the file to an FTP folder or into an email I can live without the upload.zip ;)
2010-10-02T17:22:58  <amstan> Frontier: i have to lookup the id
2010-10-02T17:23:06  <amstan> Frontier: and verify who you are
2010-10-02T17:23:11  <amstan> Frontier: just wait till the night
2010-10-02T17:23:15  <amstan> i'm a little busy now
2010-10-02T17:23:22  <Frontier> Migi32: There are many countries in this timezone, from South Africa to Sweden I think. Im from Germany.
2010-10-02T17:24:02  <Migi32> yea I know but there was at least a small chance you were from Belgium/Holland
2010-10-02T17:24:17  <Frontier> amstan: the user id would be 9727.
2010-10-02T17:24:55  <amstan> Frontier: not the user id, submission id
2010-10-02T17:26:48  <u1tr0n> amstan: hey can you send me my bot too my nick is "ruilov" and i lost my email so you can send it to GiveMeHisBot@ultron.com.ua ;)
2010-10-02T17:27:00  <u1tr0n> just joke...
2010-10-02T17:28:29  *** Appleman1234 <Appleman1234!~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au> has quit IRC (Ping timeout: 264 seconds)
2010-10-02T17:29:52  *** amstan <amstan!~alex@75-119-241-68.dsl.teksavvy.com> has left #aichallenge ("Konversation terminated!")
2010-10-02T17:30:14  <Frontier> INSERT INTO annoying_people_who_lost_their_source_code_and_want_their_submission_emailed(user_id), VALUES(9727);
2010-10-02T17:34:45  <Sylph> lol
2010-10-02T17:42:03  *** MarcAndrePC <MarcAndrePC!6015a376@gateway/web/freenode/ip.96.21.163.118> has quit IRC (Quit: Page closed)
2010-10-02T17:43:03  *** jmcarthur <jmcarthur!~jake@76.73.220.85> has quit IRC (Quit: WeeChat 0.3.0)
2010-10-02T17:43:35  *** jmcarthur <jmcarthur!~jake@76.73.220.85> has joined #aichallenge
2010-10-02T17:44:01  *** jaspervdj <jaspervdj!~jaspervdj@zeus.ugent.be> has joined #aichallenge
2010-10-02T17:45:51  *** jmcarthur <jmcarthur!~jake@76.73.220.85> has joined #aichallenge
2010-10-02T17:49:41  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T17:54:24  *** u1tr0n <u1tr0n!~u1tr0n@90-131-93-178.pool.ukrtel.net> has joined #aichallenge
2010-10-02T17:55:09  *** ebrahim <ebrahim!~ebrahim@unaffiliated/ebrahim> has quit IRC (Read error: Connection reset by peer)
2010-10-02T17:55:53  *** Blkt <Blkt!~user@93-33-140-101.ip44.fastwebnet.it> has quit IRC (Quit: Error: do not makunbound t please)
2010-10-02T17:58:48  <contestbot> Top 10 players: ruilov(3884), Hazard(3786), rsergio(3740), bix0r4ever(3726), felixcoto(3674), youwillfail(3670), EBraun(3652), dmj111(3650), cfaftw(3646), sequoh(3596)
2010-10-02T18:09:04  *** jaspervdj <jaspervdj!~jaspervdj@zeus.ugent.be> has quit IRC (Quit: NEVER GONNA GIVE YOU UP NEVER GONNA LET YOU DOWN)
2010-10-02T18:10:49  *** BtbN <BtbN!~BtbN@p54BB7467.dip.t-dialin.net> has quit IRC (Quit: Verlassend)
2010-10-02T18:11:21  *** FBP <FBP!~chatzilla@207.160.210.187> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T18:15:37  *** amjka <amjka!44c53a8a@gateway/web/freenode/ip.68.197.58.138> has joined #aichallenge
2010-10-02T18:20:22  *** choas <choas!~lars@p578F6CD8.dip.t-dialin.net> has quit IRC (Quit: leaving)
2010-10-02T18:22:53  *** aerique <aerique!~euqirea@aerique.xs4all.nl> has quit IRC (Quit: ...)
2010-10-02T18:41:32  *** medice <medice!medice@xob.kapsi.fi> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T18:44:23  <Frontier> Almighty, I just finished finding a 1 hour bug hunt around a single line of code. I was deleting in a list from the current iterator position like this:
2010-10-02T18:44:23  <Frontier> it_waiting == m_waiting_entries.erase(it_waiting);
2010-10-02T18:44:49  <Frontier> Somehow it would crash on the second iteration of the loop
2010-10-02T18:45:33  <Frontier> Then I saw the debugger step into operator== and I was like "No! Did I ?!!?....."
2010-10-02T18:46:20  <u1tr0n> :D
2010-10-02T18:46:59  *** medice <medice!medice@xob.kapsi.fi> has joined #aichallenge
2010-10-02T18:48:15  *** fawek <fawek!~faw@chello089079023212.chello.pl> has quit IRC (Remote host closed the connection)
2010-10-02T18:48:34  <Frontier> Sure erase() returned me the next valid iterator position to use... and I compared it to the old invalid one. false would be the result I guess ^^.
2010-10-02T18:49:23  <nullkuhl> hello, i have made an account for myself on ai-contest.com website, however i didnt do any submission yet and i would like to create a new account for a team as i decided to join with some friends, am not sure if i must delete my own account then ( as i have never used it )
2010-10-02T18:50:24  <Frontier> nullkuhl: the admin is not online right now
2010-10-02T18:50:49  <nullkuhl> Frontier: what is his nick ? do you have any idea how can i delete my account ?
2010-10-02T18:51:15  <Frontier> I guess if you didn't make any submissions they wont rip off your head and just delete you old account. The admins nick is amstan here.
2010-10-02T18:51:47  <Frontier> You cannot delete your account or change anything about it as far as I know.
2010-10-02T18:54:33  <Sylph> may want to confirm though
2010-10-02T18:54:45  <Sylph> I read somewhere about not allowing someone to have more than one account
2010-10-02T18:55:25  <Sylph> i don't know if that extend to team, or unused account (of which they might not really think much about considering its usage)
2010-10-02T18:55:53  <Zannick> you aren't supposed to have multiple accounts
2010-10-02T18:56:10  <Sylph> yeah exactly
2010-10-02T18:56:15  <Zannick> so you can't do individual and team
2010-10-02T18:56:28  *** Naith <Naith!~nickbarth@S0106001b38137d1b.ed.shawcable.net> has joined #aichallenge
2010-10-02T18:56:34  <Sylph> so the question is getting in contact with an admin about deleting old account or somemthing maybe
2010-10-02T18:57:45  *** RainCT <RainCT!~RainCT@ubuntu/member/pdpc.student.rainct> has quit IRC (Remote host closed the connection)
2010-10-02T18:58:50  <contestbot> Top 10 players: ruilov(3906), Hazard(3778), rsergio(3733), bix0r4ever(3704), felixcoto(3672), EBraun(3657), cfaftw(3650), dmj111(3646), youwillfail(3611), sequoh(3600)
2010-10-02T19:07:50  *** bss03-AFK is now known as bss03
2010-10-02T19:09:42  <Frontier> DELETE FROM users WHERE username = nullkuhl;
2010-10-02T19:09:53  <Frontier> DELETE FROM users WHERE username = 'nullkuhl';
2010-10-02T19:10:14  <Frontier> Is the website protected against SQL-injection anyway?
2010-10-02T19:13:16  *** smellyhippy <smellyhippy!~asmodean@about/csharp/regular/smellyhippy> has quit IRC (Ping timeout: 264 seconds)
2010-10-02T19:13:19  <u1tr0n> try it ;)
2010-10-02T19:13:28  *** FBP <FBP!~chatzilla@207.160.210.187> has joined #aichallenge
2010-10-02T19:14:21  *** impomatic <impomatic!~chatzilla@87.115.83.45> has left #aichallenge
2010-10-02T19:14:57  <Frontier> Don't know... my IP doesn't change often.
2010-10-02T19:18:49  *** bss03 is now known as bss03-AFK
2010-10-02T19:28:15  *** amjka <amjka!44c53a8a@gateway/web/freenode/ip.68.197.58.138> has quit IRC (Quit: Page closed)
2010-10-02T19:28:28  *** Itkovian <Itkovian!~Itkovian@78-22-226-160.access.telenet.be> has joined #aichallenge
2010-10-02T19:31:34  *** medice <medice!medice@xob.kapsi.fi> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T19:37:01  *** medice <medice!medice@xob.kapsi.fi> has joined #aichallenge
2010-10-02T19:37:05  *** Itkovian <Itkovian!~Itkovian@78-22-226-160.access.telenet.be> has quit IRC (Quit: Itkovian)
2010-10-02T19:38:44  *** smellyhippy <smellyhippy!~asmodean@about/csharp/regular/smellyhippy> has joined #aichallenge
2010-10-02T19:39:47  <Sylph> lol Frontier
2010-10-02T19:40:02  *** Mathnerd314 <Mathnerd314!~mathnerd3@dsl.6.28.53.206.cos.dyn.pcisys.net> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T19:49:34  *** medice <medice!medice@xob.kapsi.fi> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T19:49:40  *** medice <medice!medice@xob.kapsi.fi> has joined #aichallenge
2010-10-02T19:49:42  *** davidjliu <davidjliu!473a6d0e@gateway/web/freenode/ip.71.58.109.14> has joined #aichallenge
2010-10-02T19:55:19  *** sigh <sigh!~sigh@137.206-50-210.dynamic.dsl.mel.iprimus.net.au> has joined #aichallenge
2010-10-02T19:58:49  <contestbot> Top 10 players: ruilov(3877), Hazard(3779), rsergio(3750), felixcoto(3679), cfaftw(3668), EBraun(3663), bix0r4ever(3655), dmj111(3638), sequoh(3603), youwillfail(3592)
2010-10-02T20:15:05  *** iFire <iFire!~kittens@unaffiliated/ifire> has quit IRC (Read error: Connection reset by peer)
2010-10-02T20:15:50  *** Haggis <Haggis!81d73b91@gateway/web/freenode/ip.129.215.59.145> has quit IRC (Quit: Page closed)
2010-10-02T20:19:22  *** Frontier <Frontier!~Marco_Lei@dslb-088-071-242-043.pools.arcor-ip.net> has quit IRC (Ping timeout: 276 seconds)
2010-10-02T20:22:37  *** jimmerton_ <jimmerton_!~jamespick@142.166.63.86> has joined #aichallenge
2010-10-02T20:24:54  *** jimmerton <jimmerton!~jamespick@142.166.63.87> has quit IRC (Ping timeout: 264 seconds)
2010-10-02T20:24:54  *** jimmerton_ is now known as jimmerton
2010-10-02T20:27:03  *** Migi32 <Migi32!~migi@d54C128C5.access.telenet.be> has quit IRC (Quit: http://boinc.berkeley.edu)
2010-10-02T20:28:45  *** Mathnerd314 <Mathnerd314!~mathnerd3@dsl.6.28.53.206.cos.dyn.pcisys.net> has joined #aichallenge
2010-10-02T20:33:59  *** xlorm <xlorm!45b57924@gateway/web/freenode/ip.69.181.121.36> has joined #aichallenge
2010-10-02T20:34:51  *** Mathnerd314 <Mathnerd314!~mathnerd3@dsl.6.28.53.206.cos.dyn.pcisys.net> has quit IRC (Ping timeout: 252 seconds)
2010-10-02T20:37:29  *** iFire <iFire!~kittens@unaffiliated/ifire> has joined #aichallenge
2010-10-02T20:38:01  *** Bobng <Bobng!~T@host86-130-158-50.range86-130.btcentralplus.com> has quit IRC (Quit: Leaving)
2010-10-02T20:39:18  *** Mathnerd314 <Mathnerd314!~mathnerd3@dsl.6.28.53.206.cos.dyn.pcisys.net> has joined #aichallenge
2010-10-02T20:40:18  *** Appleman1234 <Appleman1234!~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au> has joined #aichallenge
2010-10-02T20:50:35  <McLeopold> I have officially unleashed WormBot
2010-10-02T20:51:25  <Sylph> url ? :)
2010-10-02T20:51:27  <Sylph> i wanna watch
2010-10-02T20:51:33  <McLeopold> http://72.44.46.68/getplayer?player=McLeopold.WormBot.py
2010-10-02T20:51:35  <Sylph> saving the PerfectBot still?
2010-10-02T20:51:36  *** Frontier <Frontier!~Marco_Lei@dslb-088-071-242-043.pools.arcor-ip.net> has joined #aichallenge
2010-10-02T20:51:42  <McLeopold> It's not that exciting.
2010-10-02T20:51:48  <McLeopold> But it's a WIN!
2010-10-02T20:51:51  <Sylph> LOL
2010-10-02T20:52:04  <Sylph> nice :P
2010-10-02T20:52:06  <McLeopold> Dang, lose to LudaV
2010-10-02T20:52:27  <Sylph> it's being slow
2010-10-02T20:52:29  <Sylph> on updating
2010-10-02T20:52:34  <McLeopold> hmm
2010-10-02T20:52:37  <Sylph> can't see your loss
2010-10-02T20:52:44  <Sylph> there
2010-10-02T20:52:45  <Sylph> updated
2010-10-02T20:52:48  <McLeopold> me neither
2010-10-02T20:52:59  <McLeopold> 171 turns
2010-10-02T20:52:59  <Sylph> so, attack two planet at once ?
2010-10-02T20:53:04  <Sylph> or same snakebot stuff
2010-10-02T20:53:23  <McLeopold> SnakeBot with more than only planet and some defense.
2010-10-02T20:54:09  *** Umsturz <Umsturz!~Umsturz@nat-wh.rz.uni-karlsruhe.de> has left #aichallenge
2010-10-02T20:54:10  <McLeopold> I did have the lead
2010-10-02T20:54:41  <McLeopold> Maybe I need to split more...
2010-10-02T20:54:54  *** FBP <FBP!~chatzilla@207.160.210.187> has quit IRC (Ping timeout: 264 seconds)
2010-10-02T20:55:50  <McLeopold> Win against retybok
2010-10-02T20:56:35  <McLeopold> Next up will be FrogBot.
2010-10-02T20:58:08  <McLeopold> rank 181
2010-10-02T20:58:49  <contestbot> Top 10 players: ruilov(3873), Hazard(3790), rsergio(3732), felixcoto(3708), EBraun(3691), cfaftw(3671), dmj111(3659), bix0r4ever(3634), sequoh(3616), davidjliu(3586)
2010-10-02T21:04:52  *** wvdschel <wvdschel!~wim@78-20-14-180.access.telenet.be> has quit IRC (Ping timeout: 252 seconds)
2010-10-02T21:06:49  * Sylph nods
2010-10-02T21:06:53  <Sylph> lol
2010-10-02T21:07:04  <Sylph> how would frog bot works
2010-10-02T21:07:20  <McLeopold> My though is claim my half of the board and hold.
2010-10-02T21:07:33  <McLeopold> Then send a giant fleet around the opponent's side.
2010-10-02T21:07:37  <Sylph> I have to stop coding my bot for a few days :/
2010-10-02T21:07:40  <Sylph> got assignments to do :|
2010-10-02T21:07:41  <Sylph> hrm
2010-10-02T21:07:45  <Sylph> ah
2010-10-02T21:08:05  <McLeopold> I need to solve the traveling salesman problem for that.
2010-10-02T21:08:14  <Sylph> yea true
2010-10-02T21:18:48  *** coventry <coventry!~user@68.70.61.113> has joined #aichallenge
2010-10-02T21:19:17  <Sylph> gl
2010-10-02T21:20:25  <McLeopold> http://72.44.46.68/canvas?game_id=281836
2010-10-02T21:20:40  <McLeopold> That game really tells you what retybok is doing.
2010-10-02T21:20:42  *** Zeiris <Zeiris!~Zeiris@d206-116-158-135.bchsia.telus.net> has joined #aichallenge
2010-10-02T21:20:52  <McLeopold> It looks like he only cares about the stuff you care about
2010-10-02T21:21:11  <McLeopold> He must only value planets with incoming enemy fleets.
2010-10-02T21:23:05  <McLeopold> win against a1k0n
2010-10-02T21:24:50  <McLeopold> win against retybok again.
2010-10-02T21:24:55  <McLeopold> WormBot is pleasing me.
2010-10-02T21:27:36  *** McLeopold <McLeopold!~scott@97-126-22-239.tukw.qwest.net> has quit IRC (Quit: Leaving)
2010-10-02T21:28:33  <Sylph> lol
2010-10-02T21:28:40  <Sylph> He's a ganking bot then ?
2010-10-02T21:28:50  <Sylph> made a 'GankBot' yet?
2010-10-02T21:29:33  <Sylph> nah, retybok attack one by one mm
2010-10-02T21:29:38  <Sylph> not just planets you attacked
2010-10-02T21:29:47  <Sylph> but you attacking a planet might be a trigger too yeah
2010-10-02T21:32:51  *** Janzert <Janzert!~Brian@97-83-124-169.dhcp.trcy.mi.charter.com> has joined #aichallenge
2010-10-02T21:36:00  *** coventry <coventry!~user@68.70.61.113> has left #aichallenge ("ERC Version 5.3 (IRC client for Emacs)")
2010-10-02T21:36:23  *** FlagCapper_ <FlagCapper_!45a5adac@gateway/web/freenode/ip.69.165.173.172> has joined #aichallenge
2010-10-02T21:38:01  *** FlagCapper <FlagCapper!45a5adac@gateway/web/freenode/ip.69.165.173.172> has quit IRC (Ping timeout: 265 seconds)
2010-10-02T21:40:29  *** FlagCapper_ <FlagCapper_!45a5adac@gateway/web/freenode/ip.69.165.173.172> has quit IRC (Client Quit)
2010-10-02T21:40:44  *** FlagCapper <FlagCapper!45a5adac@gateway/web/freenode/ip.69.165.173.172> has joined #aichallenge
2010-10-02T21:49:02  *** ichti <ichti!~ichti@x1-6-00-24-b2-9f-a2-1a.k429.webspeed.dk> has quit IRC (Ping timeout: 240 seconds)
2010-10-02T21:49:22  *** jahnson <jahnson!44c53a8a@gateway/web/freenode/ip.68.197.58.138> has joined #aichallenge
2010-10-02T21:55:37  *** Naith <Naith!~nickbarth@S0106001b38137d1b.ed.shawcable.net> has quit IRC (Quit: Naith)
2010-10-02T21:58:49  <contestbot> Top 10 players: ruilov(3884), Hazard(3804), davidjliu(3751), rsergio(3735), felixcoto(3718), EBraun(3708), cfaftw(3689), bix0r4ever(3685), dmj111(3676), sequoh(3612)
2010-10-02T22:13:48  *** FBP <FBP!~chatzilla@207.160.210.187> has joined #aichallenge
2010-10-02T22:22:23  *** pjreddie <pjreddie!8ce9ce82@gateway/web/freenode/ip.140.233.206.130> has joined #aichallenge
2010-10-02T22:25:56  *** FlagCapper <FlagCapper!45a5adac@gateway/web/freenode/ip.69.165.173.172> has quit IRC (Quit: Page closed)
2010-10-02T22:31:40  *** jahnson <jahnson!44c53a8a@gateway/web/freenode/ip.68.197.58.138> has quit IRC (Quit: Page closed)
2010-10-02T22:55:23  <antimatroid2> is anyone taking note of the moves their opponent makes?
2010-10-02T22:55:36  <antimatroid2> and if so what are you doing about the possibility of moves being from 1 distance away?
2010-10-02T22:58:50  <contestbot> Top 10 players: ruilov(3890), Hazard(3809), davidjliu(3757), rsergio(3756), felixcoto(3739), EBraun(3708), cfaftw(3704), dmj111(3695), bix0r4ever(3670), sequoh(3633)
2010-10-02T23:01:15  *** MarcAndrePC <MarcAndrePC!6015a376@gateway/web/freenode/ip.96.21.163.118> has joined #aichallenge
2010-10-02T23:06:17  *** FlagCapper <FlagCapper!45a5adac@gateway/web/freenode/ip.69.165.173.172> has joined #aichallenge
2010-10-02T23:14:32  <MarcAndrePC> nice nice, new sub = rank 28 :)
2010-10-02T23:15:16  <antimatroid2> nice
2010-10-02T23:15:31  <antimatroid2> it's getting more competitive every damn day
2010-10-02T23:17:22  <MarcAndrePC> yup
2010-10-02T23:17:32  <MarcAndrePC> mostly out of idea right now :(
2010-10-02T23:18:01  <antimatroid2> really? i have too many
2010-10-02T23:18:15  <MarcAndrePC> well I had a lot
2010-10-02T23:18:27  <MarcAndrePC> but I tried some and it didn't give good result at all
2010-10-02T23:18:32  <antimatroid2> i'm currently trying to set up to sort of approximate a game tree to run minimax on
2010-10-02T23:18:43  <MarcAndrePC> so now I'm rethinking a lot of good idea...
2010-10-02T23:18:47  <antimatroid2> no idea how that'll go, i think it'll largely depend on how well one picks possible combinations of moves
2010-10-02T23:19:10  <antimatroid2> i tried dynamic programming, but there's just too many possible moves
2010-10-02T23:19:21  <MarcAndrePC> I guess, if you manage to reduce the number of possible move you can work it out
2010-10-02T23:19:38  <MarcAndrePC> what's your rank?
2010-10-02T23:19:39  <antimatroid2> yeah, but it's still hard to do that
2010-10-02T23:19:48  <antimatroid2> "youwillfail"
2010-10-02T23:19:52  <antimatroid2> not sure where it's at now
2010-10-02T23:20:00  <antimatroid2> 11
2010-10-02T23:20:33  <antimatroid2> i beat you earlier it seems :P
2010-10-02T23:20:41  <MarcAndrePC> yea yea gj
2010-10-02T23:20:58  <MarcAndrePC> that's one mistake I need to change
2010-10-02T23:21:08  <MarcAndrePC> dont send too many ship at the begin
2010-10-02T23:21:22  <antimatroid2> i basically just counter for rage bot like tactics
2010-10-02T23:21:28  <antimatroid2> in that if planets all attacked me i have enough left
2010-10-02T23:21:37  <antimatroid2> i do this on any move for planets that need it
2010-10-02T23:21:42  <MarcAndrePC> hmmm and it work?
2010-10-02T23:21:45  <antimatroid2> if possible anyway
2010-10-02T23:22:24  <MarcAndrePC> I just have a simple protection home vs home when we both have only 1 planet, it protects against rage bot but I guess it needs to be more general
2010-10-02T23:22:51  <antimatroid2> yeah, depends on how much the planet is worth to you i guess
2010-10-02T23:22:54  *** rwa <rwa!~rwa@99.178.115.235> has joined #aichallenge
2010-10-02T23:23:09  <MarcAndrePC> so for each planet you have you check that you have enough ship on that planet to counter all the ship from the enemy planet?
2010-10-02T23:23:14  <MarcAndrePC> hmmm
2010-10-02T23:23:20  <antimatroid2> not every planet
2010-10-02T23:23:26  <MarcAndrePC> as I said it I realise it doesnt make sense
2010-10-02T23:23:27  <antimatroid2> only those that can't be backed up in time
2010-10-02T23:23:46  <MarcAndrePC> oh, but it's hard to find those that can't be backed up in time
2010-10-02T23:24:53  <antimatroid2> i'm hardly using any time at all so far
2010-10-02T23:25:48  <MarcAndrePC> like your 1 second limit?
2010-10-02T23:26:20  <antimatroid2> yep
2010-10-02T23:26:25  <MarcAndrePC> me neither, some turn are slower than other but I guess it's just strange bug in java like always (or me doing bad programming :) )
2010-10-02T23:26:27  <antimatroid2> sometimes it'll use over 0.01
2010-10-02T23:26:36  <antimatroid2> i'm using c++ :)
2010-10-02T23:26:52  <MarcAndrePC> I blame the garbage collector :)
2010-10-02T23:27:13  <MarcAndrePC> you did a lot of submission, any fail strat in those?
2010-10-02T23:27:37  <antimatroid2> i have had bugs with it running locally but not on the main server
2010-10-02T23:27:43  <antimatroid2> i'm also shocking at just generally spamming
2010-10-02T23:27:51  <antimatroid2> both have made it quite horrid aha
2010-10-02T23:29:13  <MarcAndrePC> oh I have another week point, I try to take neutral closer to enemy home than my home at the begin, not very wise...
2010-10-02T23:29:23  <MarcAndrePC> weak*
2010-10-02T23:30:10  <MarcAndrePC> enemy just counter with fewer ship and take it...
2010-10-02T23:30:36  <antimatroid2> aha yeah, shouldn't do that :P
2010-10-02T23:31:03  *** davidjliu <davidjliu!473a6d0e@gateway/web/freenode/ip.71.58.109.14> has quit IRC (Quit: Page closed)
2010-10-02T23:31:16  <MarcAndrePC> the problem is when I shouldn't do that and when I should
2010-10-02T23:31:47  <MarcAndrePC> like farther in the game, if I have a good position I should take neutral around me even if they are close to enemy home
2010-10-02T23:32:47  <MarcAndrePC> any easy way to find out if a neutral planet is ''secure'' to attack?
2010-10-02T23:33:36  <antimatroid2> if you have a good enough position to do that it probably doesn't matter if you don't :P
2010-10-02T23:33:43  <antimatroid2> just go in for the kill already
2010-10-02T23:34:30  <antimatroid2> i don't tend to really care about tactics for when i'm winning or losing
2010-10-02T23:34:40  <antimatroid2> i act like i'm fighting on equal terms always
2010-10-02T23:34:53  <antimatroid2> under the assumption that if it's winning it's going to win now anyway and if i'm losing wel fuck it
2010-10-02T23:35:28  <antimatroid2> although it would be nice to have a function that executes a win as soon as it's pretty sure it can
2010-10-02T23:36:42  <u1tr0n> if (your_distance_for_neutral_planet < enimy_distance_for_neutral_planet) send neutral+1 ships and check all enimy fleets if fleed->DestinationPlanet == your_netral_planet send there same count of ships
2010-10-02T23:37:39  <MarcAndrePC> hmmm not bad ultr0n
2010-10-02T23:38:03  <u1tr0n> if (your_distance_for_neutral_planet > enimy_distance_for_neutral_planet) then count all ships on planets where distanse less and send neutralships+sum(enimy_nears_ships)
2010-10-02T23:38:58  <antimatroid2> i try to estiamte an amount needed to claim and protect any neutral that is between me and them
2010-10-02T23:39:08  <antimatroid2> same with claiming them
2010-10-02T23:39:27  <MarcAndrePC> yup I guess that could work
2010-10-02T23:51:56  *** iFire <iFire!~kittens@unaffiliated/ifire> has quit IRC (Read error: Connection reset by peer)
2010-10-02T23:54:10  *** iFire <iFire!~kittens@unaffiliated/ifire> has joined #aichallenge
2010-10-02T23:55:44  *** MarcAndrePC <MarcAndrePC!6015a376@gateway/web/freenode/ip.96.21.163.118> has quit IRC (Quit: Page closed)
2010-10-02T23:58:23  <j3camero> yo yo
2010-10-02T23:58:56  <contestbot> Top 10 players: ruilov(3876), davidjliu(3847), Hazard(3809), rsergio(3766), felixcoto(3730), EBraun(3714), cfaftw(3704), dmj111(3693), bix0r4ever(3661), sequoh(3648)
2010-10-02T23:58:58  <j3camero> So I noticed there has been a bit of turbulence at the top of leaderboard as of late.
2010-10-02T23:59:25  <j3camero> I guess people have been working hard on their bots!
2010-10-02T23:59:45  <antimatroid2> hello
2010-10-02T23:59:52  <antimatroid2> top 10 is still hard to stay in