2010-11-16T00:04:06  *** bhasker <bhasker!~bhasker@adsl-99-25-118-235.dsl.pltn13.sbcglobal.net> has quit IRC (Read error: Connection reset by peer)
2010-11-16T00:04:44  *** bhasker <bhasker!~bhasker@adsl-99-25-118-235.dsl.pltn13.sbcglobal.net> has joined #aichallenge
2010-11-16T00:06:16  *** McLeopold <McLeopold!~scott@c-76-121-152-123.hsd1.wa.comcast.net> has joined #aichallenge
2010-11-16T00:06:50  *** hellman <hellman!~hellman@nat232-248-205-109.tvoe.tv> has joined #aichallenge
2010-11-16T00:08:08  <McLeopold> lost power, server back up
2010-11-16T00:08:14  <antimatroid> cheers
2010-11-16T00:13:05  *** McLeopold <McLeopold!~scott@c-76-121-152-123.hsd1.wa.comcast.net> has quit IRC (Ping timeout: 240 seconds)
2010-11-16T00:17:17  *** delt0r <delt0r!~delt0r@188-22-165-195.adsl.highway.telekom.at> has joined #aichallenge
2010-11-16T00:17:28  <contestbot> Top 10 players: bocsimacko(3740), bix0r4ever(3562), davidjliu(3553), dmj111(3548), _iouri_(3545), protocolocon(3477), shangas(3474), george(3466), Xrillo(3451), SUA.khb(3431)
2010-11-16T00:17:53  *** delt0r___ <delt0r___!~delt0r@188-22-20-47.adsl.highway.telekom.at> has quit IRC (Ping timeout: 240 seconds)
2010-11-16T00:18:44  *** TheMacca <TheMacca!~Macca@adsl-75-27-235-203.dsl.irvnca.sbcglobal.net> has left #aichallenge
2010-11-16T00:23:59  *** Utkarsh <Utkarsh!~admin@117.201.80.228> has quit IRC (Ping timeout: 264 seconds)
2010-11-16T00:24:00  *** bhasker <bhasker!~bhasker@adsl-99-25-118-235.dsl.pltn13.sbcglobal.net> has quit IRC (Read error: Connection reset by peer)
2010-11-16T00:24:39  *** genericbob <genericbob!~patm@c-67-190-73-32.hsd1.co.comcast.net> has quit IRC (Ping timeout: 276 seconds)
2010-11-16T00:24:41  *** bhasker <bhasker!~bhasker@adsl-99-25-118-235.dsl.pltn13.sbcglobal.net> has joined #aichallenge
2010-11-16T00:25:07  *** genericbob <genericbob!~patm@c-67-190-73-32.hsd1.co.comcast.net> has joined #aichallenge
2010-11-16T00:28:10  *** Utkarsh <Utkarsh!~admin@117.201.87.11> has joined #aichallenge
2010-11-16T00:36:27  <antimatroid> the main competition seems to be for second place
2010-11-16T00:36:41  <antimatroid> it's completely random
2010-11-16T00:36:47  <antimatroid> whereas 1st is securely held
2010-11-16T00:40:44  *** rebelxt_ <rebelxt_!181c0b18@gateway/web/freenode/ip.24.28.11.24> has quit IRC (Quit: Page closed)
2010-11-16T00:45:51  *** bhasker_ <bhasker_!~bhasker@adsl-99-25-118-235.dsl.pltn13.sbcglobal.net> has joined #aichallenge
2010-11-16T00:47:40  <bhasker_> hmm down to 3-4ms per state update
2010-11-16T00:47:44  <bhasker_> with tuples
2010-11-16T00:47:49  <bhasker_> and no debugging
2010-11-16T00:47:54  <bhasker_> and no debug logging rather
2010-11-16T00:49:26  *** bhasker <bhasker!~bhasker@adsl-99-25-118-235.dsl.pltn13.sbcglobal.net> has quit IRC (Ping timeout: 272 seconds)
2010-11-16T00:52:47  <bhasker_> not bad down to 2.74ms by removing all repeated object variable lookups
2010-11-16T00:54:38  <bhasker_> down to 2ms
2010-11-16T00:54:49  <bhasker_> goddamn object lookups are SLOW as hell in python
2010-11-16T00:56:36  *** amriedle <amriedle!~amriedle@CPE0013f7f2be12-CM0013f7f2be0e.cpe.net.cable.rogers.com> has quit IRC (Quit: Lost terminal)
2010-11-16T00:56:52  *** Mjothvitnir <Mjothvitnir!432ace74@gateway/web/freenode/ip.67.42.206.116> has joined #aichallenge
2010-11-16T00:57:33  *** Utkarsh <Utkarsh!~admin@117.201.87.11> has quit IRC (Ping timeout: 245 seconds)
2010-11-16T00:59:25  *** Palmik <Palmik!~Palmik@palmikpc.sks6.muni.cz> has joined #aichallenge
2010-11-16T00:59:56  <Accoun> python SLOW as hell
2010-11-16T01:03:04  <antimatroid> Accoun: at least we agree on something
2010-11-16T01:03:06  <antimatroid> c++ <3
2010-11-16T01:03:23  *** Utkarsh <Utkarsh!~admin@117.201.83.32> has joined #aichallenge
2010-11-16T01:03:33  <bhasker_> heh yea
2010-11-16T01:03:40  <bhasker_> well at least i got my state code to update state in under 2ms
2010-11-16T01:03:53  <bhasker_> so i can at least do 100 different state updates in probably 200 ms
2010-11-16T01:03:54  <bhasker_> !
2010-11-16T01:04:00  <bhasker_> ...
2010-11-16T01:04:19  <bhasker_> c++ could do it probably in 1/4th tie
2010-11-16T01:04:25  <bhasker_> will see what more i can optimize away
2010-11-16T01:04:38  <bhasker_> just need to rethink the way i do stuff with python
2010-11-16T01:05:01  <bhasker_> did mcleopold stop his tcp server?
2010-11-16T01:17:28  <contestbot> Top 10 players: bocsimacko(3728), Accoun(3643), hutabhu(3623), davidjliu(3575), dmj111(3541), bix0r4ever(3538), _iouri_(3534), george(3477), shangas(3473), DarkEmi(3466)
2010-11-16T01:21:50  *** tapwater <tapwater!~tapwater@cpe-70-112-195-55.austin.res.rr.com> has quit IRC (Quit: tapwater)
2010-11-16T01:27:34  <antimatroid> bhasker_: he said it's windy there tonight and likely to go down
2010-11-16T01:33:05  *** sigh <sigh!~sigh@lithium.mailguard.com.au> has quit IRC (Ping timeout: 240 seconds)
2010-11-16T01:40:28  *** mega1 <mega1!~quassel@adsl-63-195-37-158.dsl.snfc21.pacbell.net> has quit IRC (Ping timeout: 245 seconds)
2010-11-16T01:43:00  *** HangFire_ <HangFire_!cb811a4b@gateway/web/freenode/ip.203.129.26.75> has quit IRC (Quit: Page closed)
2010-11-16T02:07:44  *** yasith <yasith!~yasith@112.134.125.39> has joined #aichallenge
2010-11-16T02:10:01  *** narnach <narnach!~Narnach@195-241-27-75.ip.telfort.nl> has quit IRC (Quit: Leaving.)
2010-11-16T02:10:35  *** narnach <narnach!~Narnach@195-241-27-75.ip.telfort.nl> has joined #aichallenge
2010-11-16T02:11:15  *** Mjothvitnir <Mjothvitnir!432ace74@gateway/web/freenode/ip.67.42.206.116> has quit IRC (Quit: Page closed)
2010-11-16T02:16:44  *** Ice_Harley <Ice_Harley!4d7921fb@gateway/web/freenode/ip.77.121.33.251> has joined #aichallenge
2010-11-16T02:17:29  <contestbot> Top 10 players: bocsimacko(3724), dmj111(3536), davidjliu(3536), DarkEmi(3523), bix0r4ever(3512), _iouri_(3503), shangas(3478), george(3478), hutabhu(3469), Accoun(3468)
2010-11-16T02:29:23  *** hellman <hellman!~hellman@nat232-248-205-109.tvoe.tv> has quit IRC (Ping timeout: 264 seconds)
2010-11-16T02:40:44  <Accoun> http://zeroviz.us down
2010-11-16T02:41:05  <antimatroid> Accoun: McLeopolds got bad weather
2010-11-16T02:41:08  *** mistmanov <mistmanov!4da38f1c@gateway/web/freenode/ip.77.163.143.28> has joined #aichallenge
2010-11-16T02:45:38  *** sigh <sigh!~sigh@28.250-50-210.dynamic.dsl.mel.iprimus.net.au> has joined #aichallenge
2010-11-16T02:57:08  *** j3camero <j3camero!~jeff@207.150.202.47> has quit IRC (Ping timeout: 245 seconds)
2010-11-16T02:58:23  *** j3camero <j3camero!~jeff@207.150.202.47> has joined #aichallenge
2010-11-16T03:00:29  *** phxtrack <phxtrack!phxtrack@ip24-56-47-107.ph.ph.cox.net> has quit IRC (Ping timeout: 250 seconds)
2010-11-16T03:09:07  *** phxtrack <phxtrack!phxtrack@ip24-56-47-107.ph.ph.cox.net> has joined #aichallenge
2010-11-16T03:12:54  *** jaspervdj <jaspervdj!~jaspervdj@zeus.ugent.be> has joined #aichallenge
2010-11-16T03:17:29  <contestbot> Top 10 players: bocsimacko(3720), davidjliu(3577), bix0r4ever(3551), dmj111(3522), _iouri_(3517), DarkEmi(3503), hutabhu(3492), george(3473), shangas(3470), Xrillo(3461)
2010-11-16T03:22:53  <antimatroid> right, i'm trying minimax agian
2010-11-16T03:22:56  <antimatroid> :P
2010-11-16T03:28:41  *** aerique <aerique!euqirea@xs2.xs4all.nl> has joined #aichallenge
2010-11-16T03:32:49  *** perror <perror!~fleury@aldebaran.labri.fr> has joined #aichallenge
2010-11-16T03:46:15  *** narnach1 <narnach1!~Narnach@5356EA66.cm-6-7d.dynamic.ziggo.nl> has joined #aichallenge
2010-11-16T03:46:16  *** narnach1 is now known as narnach_mb
2010-11-16T03:55:38  *** medrimonia <medrimonia!~medrimoni@ABordeaux-158-1-106-27.w90-60.abo.wanadoo.fr> has quit IRC (Ping timeout: 240 seconds)
2010-11-16T04:09:01  *** medrimonia <medrimonia!~medrimoni@ABordeaux-158-1-38-173.w90-55.abo.wanadoo.fr> has joined #aichallenge
2010-11-16T04:15:08  <Accoun> very need TCP server with new map set
2010-11-16T04:17:30  <contestbot> Top 10 players: bocsimacko(3721), bix0r4ever(3544), Accoun(3532), davidjliu(3520), dmj111(3513), _iouri_(3513), hutabhu(3506), Raschi(3477), george(3474), shangas(3464)
2010-11-16T04:18:02  <d0ugal> Accoun: Doesn't the new tcp server have that already?
2010-11-16T04:18:04  <d0ugal> @tcp
2010-11-16T04:18:05  <contestbot> d0ugal: tcp = (#1) http://www.benzedrine.cx/planetwars/, or (#2) http://72.44.46.68/, or (#3) http://zeroviz.us:8080, or (#4) 76.121.152.123:995 (zeroviz.us)
2010-11-16T04:18:41  <Accoun> http://zeroviz.us down
2010-11-16T04:19:15  <antimatroid> if you can get hold of dhartmei, he'd probably be happy to put some up
2010-11-16T04:19:19  <antimatroid> but no one knows where he is
2010-11-16T04:20:22  <Accoun> oficial server useless for testing
2010-11-16T04:20:38  <Accoun> 3-4 days to get true reiting
2010-11-16T04:22:01  <Accoun> new map set in averange more compact
2010-11-16T04:22:14  <Accoun> less averange palnet nunber
2010-11-16T04:22:23  <Accoun> and less abverange distance
2010-11-16T04:23:11  <Accoun> fast expand dont own
2010-11-16T04:23:36  <antimatroid> Accoun: the distance thing will be fixed, janzert is away
2010-11-16T04:23:57  <antimatroid> there are new maps on zeroviz with fixed distances, but it's down :P
2010-11-16T04:24:16  <Accoun> antimatroid> no any more fixes!!!!
2010-11-16T04:24:26  <Accoun> antimatroid> evil kid!
2010-11-16T04:24:41  <antimatroid> they are 2 things that are miss haps
2010-11-16T04:25:06  <antimatroid> they're not actual changes to the structure of maps (other than correcting for original maximum distances)
2010-11-16T04:25:43  <Accoun> antimatroid> no any more fixes!!!!!
2010-11-16T04:26:04  <Accoun> antimatroid> keep u evil hand avay from map_generator.py
2010-11-16T04:26:36  <Accoun> суки канадские
2010-11-16T04:33:45  <lavalamp_> Accoun: sample maps with the two fixes antimatroid (well, everyone) wants here: http://ai-contest.com/forum/viewtopic.php?f=19&t=1094&p=7353#p7353
2010-11-16T04:34:23  <lavalamp_> if McLeopold's sever was up he would have maps from the same generator running
2010-11-16T04:35:07  *** Naktibalda <Naktibalda!~Naktibald@host-81-20-37-12.eckoh.com> has joined #aichallenge
2010-11-16T04:48:07  *** mistmanov <mistmanov!4da38f1c@gateway/web/freenode/ip.77.163.143.28> has quit IRC ()
2010-11-16T04:50:39  <Accoun> lavalamp_> never be used
2010-11-16T05:10:22  <Accoun> look like oficial cerver haz bug with ELO score rescale
2010-11-16T05:10:43  <Accoun> new postings plased on top
2010-11-16T05:17:31  <contestbot> Top 10 players: Raschi(3700), bocsimacko(3698), Accoun(3619), davidjliu(3537), bix0r4ever(3528), dmj111(3510), _iouri_(3501), hutabhu(3495), george(3464), shangas(3462)
2010-11-16T05:30:43  *** mistmanov <mistmanov!4da38f1c@gateway/web/freenode/ip.77.163.143.28> has joined #aichallenge
2010-11-16T05:31:05  <Ice_Harley> bocsimacko is not first - pleasure to see. Even temporary )
2010-11-16T05:35:00  *** Froist <Froist!~Froist@88.211.17.170> has joined #aichallenge
2010-11-16T05:35:19  *** wvdschel <wvdschel!~wim@unaffiliated/yukito> has joined #aichallenge
2010-11-16T05:38:07  *** Froist <Froist!~Froist@88.211.17.170> has quit IRC (Client Quit)
2010-11-16T05:40:57  *** Froist <Froist!~Froist@88.211.17.170> has joined #aichallenge
2010-11-16T05:54:00  <antimatroid> Accoun: new postings quite often seem to overshoot their elo score, it doesn't seem to last long though and shouldn't matter with the final contest
2010-11-16T06:05:26  *** Rubicon-|-Cross <Rubicon-|-Cross!~RubiconCr@adsl-99-30-224-80.dsl.pltn13.sbcglobal.net> has joined #aichallenge
2010-11-16T06:10:32  <antimatroid> what is -2^2?
2010-11-16T06:15:56  *** wvdschel <wvdschel!~wim@unaffiliated/yukito> has quit IRC (Ping timeout: 255 seconds)
2010-11-16T06:17:31  <contestbot> Top 10 players: bocsimacko(3678), Raschi(3560), bix0r4ever(3530), Accoun(3505), davidjliu(3500), _iouri_(3496), dmj111(3488), hutabhu(3471), george(3446), DarkEmi(3441)
2010-11-16T06:17:58  *** yasith <yasith!~yasith@112.134.125.39> has quit IRC (Ping timeout: 245 seconds)
2010-11-16T06:21:30  *** Frontier <Frontier!~Marco_Lei@dslb-088-071-185-122.pools.arcor-ip.net> has joined #aichallenge
2010-11-16T06:30:50  *** yasith <yasith!~yasith@112.134.121.155> has joined #aichallenge
2010-11-16T06:34:27  <d0ugal> antimatroid: http://www.google.co.uk/search?q=-2^2
2010-11-16T06:36:19  <antimatroid> yeah, for some reason I read a reddit post and was like nah it is -4, but then i looked at it again and was like nah, i do give ^ preference
2010-11-16T06:44:53  <Frontier> @google -2^2
2010-11-16T06:44:53  <contestbot> Frontier: (google <search> [--{filter,language} <value>]) -- Searches google.com for the given string. As many results as can fit are included. --language accepts a language abbreviation; --filter accepts a filtering level ('active', 'moderate', 'off').
2010-11-16T06:45:24  *** Froist <Froist!~Froist@88.211.17.170> has quit IRC (Quit: Froist)
2010-11-16T06:50:27  *** iFire` <iFire`!~kittens@unaffiliated/ifire> has joined #aichallenge
2010-11-16T06:50:51  *** phreeza <phreeza!~phreeza@f048104213.adsl.alicedsl.de> has joined #aichallenge
2010-11-16T06:51:40  *** iFire <iFire!~kittens@unaffiliated/ifire> has quit IRC (Ping timeout: 240 seconds)
2010-11-16T06:52:22  *** Froist <Froist!~Froist@88.211.17.170> has joined #aichallenge
2010-11-16T06:52:32  *** wvdschel <wvdschel!~wim@minimusa.elis.UGent.be> has joined #aichallenge
2010-11-16T06:52:32  *** wvdschel <wvdschel!~wim@unaffiliated/yukito> has joined #aichallenge
2010-11-16T06:52:36  *** Froist <Froist!~Froist@88.211.17.170> has quit IRC (Client Quit)
2010-11-16T06:57:25  <antimatroid> "and was like nah, it's 4***" then realised i was an idiot, and was again :)
2010-11-16T07:01:31  *** mistmanov <mistmanov!4da38f1c@gateway/web/freenode/ip.77.163.143.28> has quit IRC (Quit: ragequit)
2010-11-16T07:08:45  <Accoun> @google -2^2
2010-11-16T07:08:45  <contestbot> Accoun: (google <search> [--{filter,language} <value>]) -- Searches google.com for the given string. As many results as can fit are included. --language accepts a language abbreviation; --filter accepts a filtering level ('active', 'moderate', 'off').
2010-11-16T07:08:53  <Accoun> @google gfdhdfhg
2010-11-16T07:08:54  <contestbot> Accoun: No matches found.
2010-11-16T07:09:04  <Accoun> @google "-2^2"
2010-11-16T07:09:04  <contestbot> Accoun: (google <search> [--{filter,language} <value>]) -- Searches google.com for the given string. As many results as can fit are included. --language accepts a language abbreviation; --filter accepts a filtering level ('active', 'moderate', 'off').
2010-11-16T07:09:20  <Accoun> @google привет
2010-11-16T07:09:21  <contestbot> Accoun: No matches found.
2010-11-16T07:09:29  <Accoun> @google privet
2010-11-16T07:09:30  <contestbot> Accoun: Privet - Wikipedia, the free encyclopedia: <http://en.wikipedia.org/wiki/Privet>; Russian Family in Seattle: <http://www.privet.com/>; Privet: <http://www.hybridpoplars.com/privet.htm>; Privet Hedges | Ligustrum Shrubs | Privet Shrubs: <http://landscaping.about.com/od/hedgesfences/p/privet_hedges.htm>; Privet Shrubs | Gardening Tips | Garden Guides: <http://www.gardenguides.com (1 more message)
2010-11-16T07:17:32  <contestbot> Top 10 players: bocsimacko(3685), bix0r4ever(3528), dmj111(3500), Accoun(3491), _iouri_(3491), davidjliu(3475), hutabhu(3454), DarkEmi(3450), george(3449), protocolocon(3443)
2010-11-16T07:33:09  *** wvdschel <wvdschel!~wim@unaffiliated/yukito> has left #aichallenge ("Leaving")
2010-11-16T08:17:07  *** delt0r___ <delt0r___!~delt0r@188-22-165-46.adsl.highway.telekom.at> has joined #aichallenge
2010-11-16T08:17:32  <contestbot> Top 10 players: bocsimacko(3688), bix0r4ever(3520), dmj111(3493), davidjliu(3491), Accoun(3487), _iouri_(3479), hutabhu(3458), DarkEmi(3455), shangas(3453), george(3442)
2010-11-16T08:18:19  *** delt0r <delt0r!~delt0r@188-22-165-195.adsl.highway.telekom.at> has quit IRC (Ping timeout: 240 seconds)
2010-11-16T08:19:04  <d0ugal> @google @d0ugal
2010-11-16T08:19:05  <contestbot> d0ugal: Dougal Matthews (d0ugal) on Twitter: <http://twitter.com/d0ugal>; AWS Developer Forums: User Profile for d0ugal: <https://forums.aws.amazon.com/profile.jspa?userID=109497>; AWS Developer Forums: Amazon CloudFront Signed URLs helper ...: <http://developer.amazonwebservices.com/connect/message.jspa?messageID=192521>; Amazon Web Services Developer Community : User Profile for d0ugal: (2 more messages)
2010-11-16T08:23:38  *** yasith <yasith!~yasith@112.134.121.155> has quit IRC (Ping timeout: 260 seconds)
2010-11-16T08:25:33  <antimatroid> @google @antimatroid
2010-11-16T08:25:34  <contestbot> antimatroid: Antimatroid - Wikipedia, the free encyclopedia: <http://en.wikipedia.org/wiki/Antimatroid>; Antimatroid, The: <http://antimatroid.wordpress.com/>; 11011110: How to implement an antimatroid: <http://11011110.livejournal.com/134610.html>; Correspondence between two antimatroid algorithmic characterizations: <http://www.combinatorics.org/Volume_10/PDF/v10i1r44.pdf>; [math/0307013] (2 more messages)
2010-11-16T08:25:35  *** mceier <mceier!~mceier@chello089077208118.chello.pl> has joined #aichallenge
2010-11-16T08:26:21  <antimatroid> i found an imposter
2010-11-16T08:26:26  <antimatroid> http://antimatroid.wordpress.com/ that's not me
2010-11-16T08:30:14  <d0ugal> heh
2010-11-16T08:30:28  <d0ugal> I don't think the amazon results are me, not sure
2010-11-16T08:37:03  *** yasith <yasith!~yasith@112.134.122.137> has joined #aichallenge
2010-11-16T08:39:54  <antimatroid> i
2010-11-16T08:40:03  <antimatroid> i got reminded of that 17x17 problem before
2010-11-16T08:40:10  <antimatroid> so i've taken a night out writing up a new solver for it
2010-11-16T08:40:24  <antimatroid> i'm almost done, then i'm going to leave it running over night and see if it manages to find anything
2010-11-16T08:40:37  <antimatroid> doubtful though
2010-11-16T08:43:22  *** Mathnerd314 <Mathnerd314!~mathnerd3@dsl.6.28.53.206.cos.dyn.pcisys.net> has quit IRC (Ping timeout: 240 seconds)
2010-11-16T09:17:33  <contestbot> Top 10 players: bocsimacko(3695), Raschi(3546), bix0r4ever(3512), Accoun(3512), dmj111(3496), _iouri_(3486), davidjliu(3480), shangas(3464), hutabhu(3453), DarkEmi(3452)
2010-11-16T09:17:46  *** mceier <mceier!~mceier@chello089077208118.chello.pl> has quit IRC (Quit: leaving)
2010-11-16T09:30:36  *** Rubicon-|-Cross <Rubicon-|-Cross!~RubiconCr@adsl-99-30-224-80.dsl.pltn13.sbcglobal.net> has quit IRC (Quit: Leaving)
2010-11-16T09:38:18  *** lavalamp_ <lavalamp_!180dae74@gateway/web/freenode/ip.24.13.174.116> has quit IRC (Quit: Page closed)
2010-11-16T10:06:11  *** JamesMG <JamesMG!~jamesg@38.98.50.162> has joined #aichallenge
2010-11-16T10:07:45  *** Zoost <Zoost!lestat@ihq.in> has quit IRC (Ping timeout: 250 seconds)
2010-11-16T10:08:16  *** Zoost <Zoost!lestat@ihq.in> has joined #aichallenge
2010-11-16T10:17:34  <contestbot> Top 10 players: bocsimacko(3683), davidjliu(3504), dmj111(3485), bix0r4ever(3485), _iouri_(3484), DarkEmi(3479), BetterThanBose(3460), shangas(3454), protocolocon(3453), george(3440)
2010-11-16T10:38:15  *** Migi32 <Migi32!~migi@d54C1291C.access.telenet.be> has joined #aichallenge
2010-11-16T10:43:40  *** ebrahim <ebrahim!~ebrahim@unaffiliated/ebrahim> has joined #aichallenge
2010-11-16T10:44:14  *** lavalamp_ <lavalamp_!4507d76a@gateway/web/freenode/ip.69.7.215.106> has joined #aichallenge
2010-11-16T10:53:53  *** bhasker_ <bhasker_!~bhasker@adsl-99-25-118-235.dsl.pltn13.sbcglobal.net> has quit IRC (Ping timeout: 240 seconds)
2010-11-16T10:57:54  *** McLeopold <McLeopold!~scott@67.139.105.186> has joined #aichallenge
2010-11-16T11:00:50  *** bhasker <bhasker!~bhasker@nat/yahoo/x-wdrrhtcqqzgitiwb> has joined #aichallenge
2010-11-16T11:06:41  <Error323> amstan: or other devs
2010-11-16T11:06:58  <Error323> Imo after having changed the maps all the ratings should have been resetted
2010-11-16T11:07:22  <McLeopold> I think they'll merge to the correct spot
2010-11-16T11:07:26  <Error323> now there are ppl with such stable ratings slowly but surely dropping because of the new maps that their bots are less good at
2010-11-16T11:07:38  *** Mistmanov <Mistmanov!4da38f1c@gateway/web/freenode/ip.77.163.143.28> has joined #aichallenge
2010-11-16T11:07:46  <Error323> instead they should have been resetted
2010-11-16T11:07:48  <McLeopold> It only uses 3 days of results, so it should be mostly new maps only.
2010-11-16T11:07:56  <Error323> Mistmanov: has this problem also i think
2010-11-16T11:08:18  <Error323> the point is the players still have ratings from way before
2010-11-16T11:08:19  <lavalamp_> Wow, even the barest hint of unbreaking my defense code was worth over 200 Elo on tcp...
2010-11-16T11:08:23  <Mistmanov> abuh?
2010-11-16T11:08:24  *** superflit <superflit!~superflit@c-24-9-162-197.hsd1.co.comcast.net> has quit IRC (Quit: superflit)
2010-11-16T11:08:46  <lavalamp_> Error323: the system only rates the last three days
2010-11-16T11:08:47  <Mistmanov> what problem do I have also?
2010-11-16T11:09:02  <Error323> lavalamp_: then why are all their histories still intact?
2010-11-16T11:09:17  <lavalamp_> Dunno, but that's what amstan and janzert said
2010-11-16T11:09:24  <Error323> are you saying that the history has no impact from before 3 days back?
2010-11-16T11:09:30  <Error323> because I think it does
2010-11-16T11:09:32  <lavalamp_> yeah that's what I'm saying ^^
2010-11-16T11:09:48  <lavalamp_> I could be wrong but at least my misinformation comes from the best source
2010-11-16T11:10:05  <McLeopold> Yes, janzert said 3 days.
2010-11-16T11:10:06  <Error323> then why keep the history in the first place
2010-11-16T11:10:11  <Error323> doesn't make sense
2010-11-16T11:10:17  <McLeopold> So you can review your past matches
2010-11-16T11:10:38  <McLeopold> The issue is, it takes too long to process the scores.
2010-11-16T11:11:02  <McLeopold> With BayesElo, you always process the entire batch from scratch.
2010-11-16T11:11:30  <Error323> I fear that people resubmitting now for the first time in > 10 days are in for a scare
2010-11-16T11:14:28  <Mistmanov> mcleopolds tcp server down?
2010-11-16T11:16:54  <Mistmanov> nvm, just saw on the forum you have no power
2010-11-16T11:17:34  <contestbot> Top 10 players: bocsimacko(3701), bix0r4ever(3632), davidjliu(3508), protocolocon(3497), Accoun(3492), DarkEmi(3485), dmj111(3482), _iouri_(3479), shangas(3466), george(3442)
2010-11-16T11:17:36  <Error323> McLeopold: what takes too long?
2010-11-16T11:17:37  *** Utkarsh <Utkarsh!~admin@117.201.83.32> has quit IRC (Read error: Connection reset by peer)
2010-11-16T11:17:42  <McLeopold> ?
2010-11-16T11:17:45  *** aerique <aerique!euqirea@xs2.xs4all.nl> has quit IRC (Quit: ...)
2010-11-16T11:17:51  <McLeopold> Oh, on the elo results?
2010-11-16T11:17:55  <Error323> ye
2010-11-16T11:17:59  <McLeopold> It's lots of data
2010-11-16T11:18:45  <Error323> I'm only wondering if the history has been fully reset for every contestant as if he was resubmitting when the maps changed on the server.
2010-11-16T11:18:57  <McLeopold> btw: Yes, my tcp is down.  No power at home.
2010-11-16T11:19:08  <lavalamp_> how long ago did the maps change?
2010-11-16T11:19:10  <lavalamp_> 2 days?
2010-11-16T11:19:15  <McLeopold> 3 now
2010-11-16T11:19:16  *** mega1 <mega1!~quassel@2001:470:1f05:548:215:58ff:fe7d:773b> has joined #aichallenge
2010-11-16T11:19:24  <McLeopold> saturday night I think
2010-11-16T11:19:31  <Error323> because if not.... some will have a huge problem
2010-11-16T11:19:49  <Error323> you see my point?
2010-11-16T11:20:10  <Accoun> McLeopold> obama make USE powerless?
2010-11-16T11:20:10  <lavalamp_> Well by tonight everyone will be rated only on new games
2010-11-16T11:20:16  <Accoun> McLeopold> obama make USA powerless?
2010-11-16T11:20:25  <Error323> lavalamp_: how is that?
2010-11-16T11:20:30  <McLeopold> Accoun: is that a joke?
2010-11-16T11:20:38  <lavalamp_> Accoun, yeah, Obama personally caused the storm that took out his power, he's that awesome
2010-11-16T11:20:46  * lavalamp_ rolls eyes
2010-11-16T11:20:47  <McLeopold> China makes USA powerless.
2010-11-16T11:20:48  <Accoun> McLeopold> why USA select obama?
2010-11-16T11:21:08  <lavalamp_> Error323: I think you should look up Bayeselo
2010-11-16T11:21:19  <McLeopold> Because we hated the party of the last guy 2 years ago.
2010-11-16T11:21:23  <lavalamp_> It's a whole-history method, not an incremental method
2010-11-16T11:21:24  <Mistmanov> conspiracy by candle manufactureres
2010-11-16T11:21:27  <McLeopold> But now we hate Obama's party.
2010-11-16T11:21:40  <Error323> lavalamp_: I think your misunderstanding me
2010-11-16T11:21:48  <Error323> I know how it works
2010-11-16T11:21:53  <Error323> roughly
2010-11-16T11:22:14  <lavalamp_> if you only plug three days of games into it, you'll only be rated on those three days
2010-11-16T11:22:33  <Accoun> McLeopold> obama - mr klinton - ms klinton - bad company
2010-11-16T11:22:34  <lavalamp_> the rating you had before that period effectively shouldn't matter
2010-11-16T11:23:19  <McLeopold> ok
2010-11-16T11:24:07  <Accoun> McLeopold> select respublican paty for me next time
2010-11-16T11:24:13  <McLeopold> Sure
2010-11-16T11:24:32  <Accoun> McLeopold> democrates make $ like tualet paper
2010-11-16T11:24:37  <Error323> let say contestant X is rated 20th and has a history with its latest submission of 1000 games. He submitted 20 days before the maps changed. 20 days later the maps changed and because of his massive history his rating is very stable resulting in very slow decrease of his rating. Whilest when he would resubmit, the true problem would show up as he wouldn't come anywhere near his old ranking.
2010-11-16T11:25:30  *** p4p4 <p4p4!~chatzilla@82.113.106.31> has joined #aichallenge
2010-11-16T11:25:42  <Error323> My point is everyones score should have been completely resetted (as if they all resubmitted) when the server maps had changed.
2010-11-16T11:25:47  <lavalamp_> AFAIK, bayeselo doesn't carry over stability info like that; you are rated solely on the games it's given
2010-11-16T11:26:07  <lavalamp_> it's as if everyone submitted their bot three days ago
2010-11-16T11:26:25  <Error323> well thats not true, bayeselo holds a history
2010-11-16T11:26:33  <lavalamp_> only for the games you feed it
2010-11-16T11:26:37  <Error323> losing 10 times of X is very different from losing 1 time of X
2010-11-16T11:26:47  <lavalamp_> right
2010-11-16T11:27:08  <lavalamp_> it forgets everything between rating runs though. That's why you have to give it all the games at once.
2010-11-16T11:28:25  <lavalamp_> I could be thinking of another rating algorithm but I'm pretty sure I'm not
2010-11-16T11:34:29  *** Utkarsh <Utkarsh!~admin@117.201.87.72> has joined #aichallenge
2010-11-16T11:37:05  *** narnach_mb <narnach_mb!~Narnach@5356EA66.cm-6-7d.dynamic.ziggo.nl> has quit IRC (Quit: Leaving.)
2010-11-16T11:41:26  <lavalamp_> Error323: make sense?
2010-11-16T11:41:30  *** superflit <superflit!~superflit@140.226.49.160> has joined #aichallenge
2010-11-16T11:44:03  *** AlliedEnvy <AlliedEnvy!~AlliedEnv@adsl-66-141-23-30.dsl.hstntx.swbell.net> has quit IRC (Ping timeout: 276 seconds)
2010-11-16T11:51:06  <d0ugal> Wow, that's very un cool. If you go here http://ai-contest.com/check_submit.php and your logged in, it clears your current submission thinking an upload has failed
2010-11-16T11:51:45  <Naktibalda> why would you go there?
2010-11-16T11:51:55  <phreeza> wha i almost clicked the link without fully reading
2010-11-16T11:52:02  <phreeza> bad idea
2010-11-16T11:52:50  <d0ugal> Naktibalda: I clicked back in the browser a few times and ended up there
2010-11-16T11:53:09  <d0ugal> phreeza: sorry, i probably shouldn't have psoted it :/
2010-11-16T11:55:53  *** Utkarsh <Utkarsh!~admin@117.201.87.72> has quit IRC (Ping timeout: 245 seconds)
2010-11-16T12:00:35  *** Utkarsh <Utkarsh!~admin@117.201.89.153> has joined #aichallenge
2010-11-16T12:03:24  *** tapwater <tapwater!~tapwater@cpe-70-112-195-55.austin.res.rr.com> has joined #aichallenge
2010-11-16T12:12:50  <Zannick> @issues
2010-11-16T12:12:51  <contestbot> Zannick: issues = http://code.google.com/p/ai-contest/issues/list
2010-11-16T12:13:00  <Zannick> d0ugal: please file a bug ^
2010-11-16T12:17:34  <contestbot> Top 10 players: bocsimacko(3704), bix0r4ever(3610), davidjliu(3500), protocolocon(3496), Accoun(3486), Raschi(3483), dmj111(3480), _iouri_(3473), shangas(3468), DarkEmi(3453)
2010-11-16T12:24:14  <Error323> lavalamp_: no
2010-11-16T12:24:23  *** Bobng <Bobng!~T@78.151.105.40> has joined #aichallenge
2010-11-16T12:24:29  <Error323> well, I mean, my rating is very unstable with little games
2010-11-16T12:24:35  <Error323> and becomes more stable over time
2010-11-16T12:24:46  <Error323> so I don't understand your story
2010-11-16T12:24:59  <lavalamp_> it stops getting more stable after three days
2010-11-16T12:25:19  <Error323> ahhh
2010-11-16T12:25:32  <Error323> now i understand :P
2010-11-16T12:25:36  <lavalamp_> haha ok
2010-11-16T12:27:12  <Error323> thats very weird though
2010-11-16T12:27:35  <Error323> only a few games are played for a contestant in 3 days
2010-11-16T12:28:18  <Error323> lavalamp_: are you sure thats the case?
2010-11-16T12:28:26  <lavalamp_> when I left mine running it got quite a few
2010-11-16T12:28:30  <lavalamp_> for a week that is
2010-11-16T12:28:46  <lavalamp_> it seemed to play like one an hour or so
2010-11-16T12:29:16  <Error323> yes well thats like 72 games
2010-11-16T12:29:19  <lavalamp_> hm, Error323 if my bot were a little brighter it should have won this I think:
2010-11-16T12:29:21  <lavalamp_> http://72.44.46.68/canvas?game_id=1055373
2010-11-16T12:29:49  <lavalamp_> if you're arguing not enough games are played for accurate ratings, I agree ^^
2010-11-16T12:30:10  <Error323> indeed i am
2010-11-16T12:30:31  <lavalamp_> :)
2010-11-16T12:30:36  <Error323> its one big randomfest if what you are saying is true
2010-11-16T12:30:39  <Error323> :(
2010-11-16T12:31:15  <lavalamp_> it probably depends on the level of the program to some extent
2010-11-16T12:31:40  <lavalamp_> my first submission bounced between place 120 and 270-- that's incredibly innacurate
2010-11-16T12:31:41  <Error323> no but seriously
2010-11-16T12:31:48  <Error323> the variety in the maps is so big now
2010-11-16T12:32:05  <Error323> that 3 days of play is not enough for accurate ratings
2010-11-16T12:32:14  <Error323> ~72 games
2010-11-16T12:32:24  <Error323> its highly dependent on the map
2010-11-16T12:32:32  <Error323> maps*
2010-11-16T12:32:37  *** Apophis <Apophis!~apophis@80-219-211-230.dclient.hispeed.ch> has joined #aichallenge
2010-11-16T12:32:38  <lavalamp_> I totally agree, and further, it wasn't before the new maps, either
2010-11-16T12:33:02  <Error323> no... but probably less of a problem
2010-11-16T12:33:35  <Error323> devs?
2010-11-16T12:33:44  <lavalamp_> you can double check with janzert but I'm 95% sure what I'm saying is correct
2010-11-16T12:33:58  <lavalamp_> or amstan
2010-11-16T12:34:02  <Error323> amstan: ping
2010-11-16T12:35:14  <Error323> well i hope i'm missing something
2010-11-16T12:35:25  <Error323> but from the looks of the dropping of some players I'm not sure
2010-11-16T12:36:07  <lavalamp_> awesome, my bot is so much better at close positions now: http://72.44.46.68/canvas?game_id=1055535
2010-11-16T12:42:22  *** AlliedEnvy <AlliedEnvy!~AlliedEnv@adsl-66-141-23-30.dsl.hstntx.swbell.net> has joined #aichallenge
2010-11-16T12:42:48  <Error323> mine is failing more each day
2010-11-16T12:42:49  <Error323> :/
2010-11-16T12:44:05  <medrimonia> Mine too, I'm making a big work, but I just can hope that I'll finish it before the deadline
2010-11-16T12:45:21  <lavalamp_> I have a lot left to do still, I'm worried I won't have much time to do any tuning once I get it all written :/
2010-11-16T12:49:39  <Error323> i'm having a crisis in wether to pursue my new approach
2010-11-16T12:49:46  <Error323> or optimize/tweak my old
2010-11-16T12:50:17  <Error323> new approach is more risk with higher payoff
2010-11-16T12:50:23  <Error323> old is less payoff less risk
2010-11-16T12:50:25  <Error323> :P
2010-11-16T12:51:33  <Error323> bhasker: what the hell happened to ur bot? :S
2010-11-16T12:51:44  <Error323> at tcp
2010-11-16T12:53:49  <bhasker> oh i am just trying out new stuff
2010-11-16T12:53:55  <Error323> ah ok
2010-11-16T12:53:57  <bhasker> changed the routing logic
2010-11-16T12:54:04  <bhasker> so it doesn't do the fly past stuff
2010-11-16T12:54:11  <bhasker> but its doing other stuff that isn't correct
2010-11-16T12:54:54  <Error323> i see
2010-11-16T12:55:04  <Error323> well i really need to update my attacking system
2010-11-16T12:55:08  <Error323> cuz its teh suck atm
2010-11-16T12:55:24  <lavalamp_> not as bad as my defense, I'm sure
2010-11-16T12:56:26  *** hellman <hellman!~hellman@nat232-248-205-109.tvoe.tv> has joined #aichallenge
2010-11-16T12:57:32  <bhasker> http://72.44.46.68/canvas?game_id=1055815
2010-11-16T12:57:38  <bhasker> the new routing doesn't fly past my planets
2010-11-16T12:57:38  <Error323> i wish i could predict with how much to attack an enemy planet better without heavily overestimating
2010-11-16T12:57:51  <bhasker> error i wish i could do the same
2010-11-16T12:58:00  <bhasker> but i can't get the math right for that:-\
2010-11-16T12:58:14  <Error323> thats because its not just math
2010-11-16T12:58:27  <Error323> but highly dependent on the enemy's behavior
2010-11-16T12:58:51  <Error323> nice work on the routing though, looking more like mine now :>
2010-11-16T12:59:11  <Error323> I like looking at routing :P
2010-11-16T12:59:23  <Error323> i'm going mental over this game
2010-11-16T12:59:34  <bhasker> lol
2010-11-16T12:59:42  <Error323> ^_^
2010-11-16T13:00:33  <bhasker> i fixed my state udpation code, now takes 2 ms to update all 23 planets
2010-11-16T13:00:43  <bhasker> and do all the fleet resolution for all planets
2010-11-16T13:01:04  <Error323> cool, your approach is the complement of mine
2010-11-16T13:01:09  <Error323> xD
2010-11-16T13:01:12  <bhasker> heh
2010-11-16T13:01:28  <Error323> though for some stuff urs make more sense
2010-11-16T13:01:38  <Error323> like fleetrequests or sth
2010-11-16T13:01:56  <bhasker> this bogusbrainpower keeps killing my bot
2010-11-16T13:04:47  <Error323> lavalamp_: did you last stated that taking the average was a pretty good eval function?
2010-11-16T13:07:59  <bhasker> average of what?
2010-11-16T13:08:22  <Error323> that would be my question
2010-11-16T13:09:43  *** p4p4 <p4p4!~chatzilla@82.113.106.31> has quit IRC (Quit: ChatZilla 0.9.84 [SeaMonkey 2.0a3/20090223135443])
2010-11-16T13:17:34  <contestbot> Top 10 players: bocsimacko(3707), bix0r4ever(3692), Raschi(3575), davidjliu(3518), protocolocon(3493), Accoun(3485), dmj111(3476), DarkEmi(3474), _iouri_(3466), shangas(3465)
2010-11-16T13:18:55  <bhasker> i am getting tired of this bogusbrainpower bot
2010-11-16T13:35:33  <lavalamp_> Error323: average ships count * average production (over the next max distance turns)
2010-11-16T13:36:01  <lavalamp_> that's a large part of my eval function right now and it's surprisingly hard to improve on
2010-11-16T13:36:50  <d0ugal> Zannick: Sorry, was away - filing a bug now.
2010-11-16T13:36:51  <Error323> lavalamp_: average over what?
2010-11-16T13:37:32  <lavalamp_> from the current turn until [max distance on the board] turns from now
2010-11-16T13:37:33  <Error323> planets?
2010-11-16T13:37:38  <Error323> ohhh
2010-11-16T13:37:39  <lavalamp_> sorry yes
2010-11-16T13:37:42  <Error323> over time
2010-11-16T13:37:43  <Error323> ?
2010-11-16T13:37:46  <lavalamp_> yeah
2010-11-16T13:37:51  <lavalamp_> all my planets over time
2010-11-16T13:38:36  <Error323> mmm
2010-11-16T13:38:40  <Error323> makes sense i guess
2010-11-16T13:39:12  <Error323> but why only the furthest
2010-11-16T13:39:17  <Error323> or max distance
2010-11-16T13:39:35  <lavalamp_> actually I used 20 and just changed it to max distance this morning
2010-11-16T13:39:57  <lavalamp_> I do know if I use max distance *2 it's horrible
2010-11-16T13:41:51  <lavalamp_> I need to do more experimentation to figure out what is actually going on
2010-11-16T13:42:00  <Error323> yeah
2010-11-16T13:42:12  <Error323> I don't have an explanation for it
2010-11-16T13:42:13  <lavalamp_> there's some more in my eval, now, too--
2010-11-16T13:42:15  <Error323> but i'll try it out
2010-11-16T13:42:20  <Error323> see if it works here as well
2010-11-16T13:42:56  <lavalamp_> For example, I don't include planets that I think ragebot could take
2010-11-16T13:43:05  <d0ugal> Zannick and anybody else it concerns: Issue can be found here; http://code.google.com/p/ai-contest/issues/detail?id=211
2010-11-16T13:43:52  <lavalamp_> I also have a connectedness metric, positions score better if their planets are more cohesive
2010-11-16T13:44:12  <Error323> ah
2010-11-16T13:44:22  <lavalamp_> if it gets near the end of the game I switch to just using ship count at the last turn
2010-11-16T13:44:52  <Error323> yeah its already becomming more hary :P
2010-11-16T13:44:58  <lavalamp_> but the connectedness thing doesn't really do much
2010-11-16T13:45:01  <lavalamp_> haha
2010-11-16T13:45:25  <lavalamp_> some of what I have is band-aids to fix problems that should be fixed other ways
2010-11-16T13:45:26  *** Rubicon-|-Cross <Rubicon-|-Cross!~RubiconCr@adsl-99-30-224-80.dsl.pltn13.sbcglobal.net> has joined #aichallenge
2010-11-16T13:46:33  <lavalamp_> lesson being: if your bot doesn't know how to fortify/defend itself, you can't fix that with an eval function
2010-11-16T13:49:40  <lavalamp_> because the eval function won't notice a planet is vulnerable until it actually gets taken
2010-11-16T13:49:53  *** narnach1 <narnach1!~Narnach@195-241-27-75.ip.telfort.nl> has joined #aichallenge
2010-11-16T13:49:53  *** narnach1 is now known as narnach_mb
2010-11-16T13:50:05  <Error323> yea
2010-11-16T13:52:10  <lavalamp_> I will probably remove the connectedness thing unless I can show it does something useful
2010-11-16T13:57:45  <lavalamp_> yay my bot is climbing back up on official now that I have a new version... disastrous upload yesterday, ugh
2010-11-16T13:58:33  <Error323> whats ur unofficial rank?
2010-11-16T13:59:32  <lavalamp_> sucky xP
2010-11-16T13:59:44  <Error323> ow :P
2010-11-16T13:59:47  <lavalamp_> I was in the 500's before I killed it yesterday
2010-11-16T13:59:57  <lavalamp_> shot down to 1500
2010-11-16T14:00:03  <Error323> ouch
2010-11-16T14:00:03  <lavalamp_> now it's back to 600
2010-11-16T14:00:19  <lavalamp_> hopefully it will keep going, hehe
2010-11-16T14:00:30  <lavalamp_> the bot I had before the map change was 100-200
2010-11-16T14:00:41  <lavalamp_> but I had to put up my new one early
2010-11-16T14:02:26  <lavalamp_> -150 on tcp right now but who knows what that means, tcp is brutal lately
2010-11-16T14:02:39  <Error323> true that
2010-11-16T14:02:52  <Error323> is it in anyway representive of the maps on official?
2010-11-16T14:03:40  <lavalamp_> tcp? doesn't look like it atm, the owner is AWOL
2010-11-16T14:04:34  <Error323> mmm
2010-11-16T14:04:38  <Error323> dhartmei?
2010-11-16T14:06:34  *** McLeopold <McLeopold!~scott@67.139.105.186> has quit IRC (Quit: Leaving)
2010-11-16T14:06:40  *** mceier <mceier!~mceier@chello089077208118.chello.pl> has joined #aichallenge
2010-11-16T14:06:59  <lavalamp_> yeah
2010-11-16T14:14:17  *** yasith <yasith!~yasith@112.134.122.137> has quit IRC (Ping timeout: 255 seconds)
2010-11-16T14:15:28  *** hellman <hellman!~hellman@nat232-248-205-109.tvoe.tv> has quit IRC (Remote host closed the connection)
2010-11-16T14:17:25  *** Accoun <Accoun!~1@213.88.2.78> has quit IRC ()
2010-11-16T14:17:34  <contestbot> Top 10 players: bix0r4ever(3808), bocsimacko(3686), davidjliu(3507), protocolocon(3478), shangas(3458), dmj111(3455), _iouri_(3450), DarkEmi(3449), george(3442), Xrillo(3434)
2010-11-16T14:29:08  *** Accoun <Accoun!~1@static-213-88-2-78.chebnet.ru> has joined #aichallenge
2010-11-16T14:30:31  *** McLeopold <McLeopold!~scott@c-98-247-248-39.hsd1.wa.comcast.net> has joined #aichallenge
2010-11-16T14:38:12  *** perror <perror!~fleury@aldebaran.labri.fr> has quit IRC (Quit: Bye all !)
2010-11-16T14:39:10  <McLeopold> @tcp
2010-11-16T14:39:11  <contestbot> McLeopold: tcp = (#1) http://www.benzedrine.cx/planetwars/, or (#2) http://72.44.46.68/, or (#3) http://zeroviz.us:8080, or (#4) 76.121.152.123:995 (zeroviz.us)
2010-11-16T14:39:23  <McLeopold> contestbot forget tcp 4
2010-11-16T14:39:23  <contestbot> McLeopold: As you wish.
2010-11-16T14:39:57  <McLeopold> contestbot learn tcp as 98.247.248.39:995 (zeroviz.us)
2010-11-16T14:39:57  <contestbot> McLeopold: I come to serve.
2010-11-16T14:40:01  <McLeopold> @tcp
2010-11-16T14:40:02  <contestbot> McLeopold: tcp = (#1) http://www.benzedrine.cx/planetwars/, or (#2) http://72.44.46.68/, or (#3) http://zeroviz.us:8080, or (#4) 98.247.248.39:995 (zeroviz.us)
2010-11-16T14:41:56  <McLeopold> tcp is back up
2010-11-16T14:42:15  <JamesMG> don't seem to be connecting
2010-11-16T14:42:22  <McLeopold> new ip
2010-11-16T14:42:32  <bhasker> how does this happen
2010-11-16T14:42:33  <bhasker> http://72.44.46.68/canvas?game_id=1057166
2010-11-16T14:42:38  <bhasker> and then i go lose to flavor!@#!@#!@#
2010-11-16T14:42:39  <McLeopold> umm... maybe I didn't save my firwall rules...
2010-11-16T14:42:39  <JamesMG> dns not propogated?
2010-11-16T14:42:42  *** ebrahim <ebrahim!~ebrahim@unaffiliated/ebrahim> has quit IRC (Remote host closed the connection)
2010-11-16T14:43:51  <Error323> lavalamp_: lol well its performing incredibly bad here =)
2010-11-16T14:44:14  <Error323> do you also count inflight ships of yourself?
2010-11-16T14:45:57  <McLeopold> JamesMG: can you try again?
2010-11-16T14:46:10  <McLeopold> dns not propogated yet
2010-11-16T14:46:13  <McLeopold> use ip
2010-11-16T14:47:15  <McLeopold> Nice, thx
2010-11-16T14:47:16  *** RainCT <RainCT!~RainCT@ubuntu/member/pdpc.student.rainct> has joined #aichallenge
2010-11-16T14:48:05  <McLeopold> Accoun: tcp back up
2010-11-16T14:52:36  *** AlliedEnvy <AlliedEnvy!~AlliedEnv@adsl-66-141-23-30.dsl.hstntx.swbell.net> has quit IRC (Quit: AlliedEnvy)
2010-11-16T14:59:19  <lavalamp_> Error323: let me check...
2010-11-16T14:59:36  <Error323> well i lose like everything xD
2010-11-16T14:59:36  *** AlliedEnvy <AlliedEnvy!~AlliedEnv@adsl-66-141-23-30.dsl.hstntx.swbell.net> has joined #aichallenge
2010-11-16T15:00:18  <lavalamp_> Yeah I do count ships in flight
2010-11-16T15:00:43  <lavalamp_> lol, our programs might be too different
2010-11-16T15:00:45  <Error323> mja so did i
2010-11-16T15:00:57  <Error323> hehe prolly action generation itself is
2010-11-16T15:01:29  <Mistmanov> worst game ever:
2010-11-16T15:01:30  <Mistmanov> http://72.44.46.68/canvas?game_id=1054828
2010-11-16T15:01:47  <lavalamp_> currently my code is pretty focused on expansion, if yours isn't maybe that's the difference
2010-11-16T15:01:59  <Error323> Mistmanov: omg
2010-11-16T15:02:06  <Mistmanov> really weird bug somewhere in my defensive routine causes that.. just have to track it down (first time I've it happening)
2010-11-16T15:02:14  <lavalamp_> it's also possible that yours was way better than mine already ^^
2010-11-16T15:02:18  <Error323> thats insande =)
2010-11-16T15:02:21  <Error323> insane*
2010-11-16T15:02:49  <Mistmanov> having 3x as many ships, more production, and still managing to lose
2010-11-16T15:02:51  <Mistmanov> that's skills
2010-11-16T15:02:58  <Error323> jeep
2010-11-16T15:03:00  <Error323> n1
2010-11-16T15:03:07  <lavalamp_> loooool
2010-11-16T15:03:10  <bhasker> ouch that game must hurt
2010-11-16T15:03:14  <bhasker> he just basically had a free pass
2010-11-16T15:03:24  <Error323> you were in some sort of lock
2010-11-16T15:03:42  <Mistmanov> yeah the defensive thingie wasn't looking forward for some reason
2010-11-16T15:03:58  <McLeopold> Error323 and Mistmanov: go back to zeroviz.us!
2010-11-16T15:04:25  <Mistmanov> will go back to testing once I track down the bug ;)
2010-11-16T15:04:37  <Error323> McLeopold: me too, still working on something
2010-11-16T15:04:45  <McLeopold> :)
2010-11-16T15:04:58  <Error323> btw McLeopold the reason i'm not on it is because i have no idea what the ratings mean
2010-11-16T15:05:01  <Error323> and its unstable
2010-11-16T15:05:16  <Error323> I need atleast a night of non-stop running for results
2010-11-16T15:05:31  <Error323> but then again
2010-11-16T15:05:34  <Error323> you need critical mass
2010-11-16T15:05:38  <Error323> so i'll join
2010-11-16T15:05:45  <Error323> just make it stable :P
2010-11-16T15:05:52  *** narnach_mb <narnach_mb!~Narnach@195-241-27-75.ip.telfort.nl> has quit IRC (Quit: Leaving.)
2010-11-16T15:05:57  * RainCT wonders what's the last date he can upload an improved bot
2010-11-16T15:06:01  <lavalamp_> If I were at home I'd connect mine
2010-11-16T15:06:16  *** narnach1 <narnach1!~Narnach@195-241-27-75.ip.telfort.nl> has joined #aichallenge
2010-11-16T15:06:16  *** narnach1 is now known as narnach_mb
2010-11-16T15:06:25  <McLeopold> They have no meaning at dhartmei's either
2010-11-16T15:06:35  <McLeopold> @rankings
2010-11-16T15:06:36  <contestbot> McLeopold: Top 10 players: Raschi(3825), bocsimacko(3676), bix0r4ever(3646), davidjliu(3506), protocolocon(3468), shangas(3454), dmj111(3451), _iouri_(3440), Xrillo(3439), george(3439)
2010-11-16T15:06:40  <Error323> McLeopold: well more to me :) since its been around quite some time
2010-11-16T15:06:43  <lavalamp_> yeah ratings on those are totally variable
2010-11-16T15:06:44  <Error323> holy crap
2010-11-16T15:07:02  <McLeopold> Once dmj111 and mega1 start playing, they'll have some meaning
2010-11-16T15:07:17  <Error323> McLeopold: but could you solely generate maps using http://tinyurl.com/aimapgen
2010-11-16T15:07:27  <Error323> then i'll get on it for sure
2010-11-16T15:07:27  <McLeopold> Accoun is #12 right now on official, so that has some meaning.
2010-11-16T15:07:44  <Error323> nah he recently uploaded
2010-11-16T15:08:02  <bhasker> fail defense here too
2010-11-16T15:08:02  <bhasker> http://zeroviz.us:8080/canvas?game_id=35
2010-11-16T15:08:10  <McLeopold> if that's antimatroid's code, then yes
2010-11-16T15:08:32  <Error323> its in the topic
2010-11-16T15:08:35  <Error323> it must be right?
2010-11-16T15:08:51  <Error323> just generate 500 maps or sth
2010-11-16T15:09:00  <Error323> no more experimental weirdness
2010-11-16T15:09:06  <Error323> cuz it says nothing to me
2010-11-16T15:09:17  <McLeopold> no, I removed those
2010-11-16T15:09:23  <McLeopold> I'll have a port just for fun later.
2010-11-16T15:09:26  <lavalamp_> McLeopold you used the updated script with the two changes right?
2010-11-16T15:09:30  <McLeopold> yes
2010-11-16T15:09:40  <McLeopold> but the original 30 are still there
2010-11-16T15:09:45  <McLeopold> They still fit the spec
2010-11-16T15:10:09  <lavalamp_> yeah ok
2010-11-16T15:10:13  <McLeopold> Although, I don't like the circle shape instead of a square, personally.
2010-11-16T15:10:54  <Error323> men i need to change my attackscheme but i don't know how :/
2010-11-16T15:11:00  <lavalamp_> I thought I heard antimatroid say something about changing that
2010-11-16T15:11:00  <McLeopold> dmj111's in
2010-11-16T15:11:14  <McLeopold> http://10.7.10.1:8080/canvas?game_id=40
2010-11-16T15:11:19  <McLeopold> ouch, serious defense fail
2010-11-16T15:11:56  <lavalamp_> McLeopold: local ip fail ^^
2010-11-16T15:12:01  <Error323> hehe
2010-11-16T15:12:23  <McLeopold> haha
2010-11-16T15:12:48  <McLeopold> http://zeroviz.us:8080/canvas?game_id=46
2010-11-16T15:13:29  <lavalamp_> huh wouldn't rage have beaten dmj with a start like that?
2010-11-16T15:13:31  *** AlliedEnvy <AlliedEnvy!~AlliedEnv@adsl-66-141-23-30.dsl.hstntx.swbell.net> has quit IRC (Remote host closed the connection)
2010-11-16T15:13:37  *** McLeopold <McLeopold!~scott@c-98-247-248-39.hsd1.wa.comcast.net> has quit IRC (Quit: Leaving)
2010-11-16T15:14:02  *** AlliedEnvy <AlliedEnvy!~AlliedEnv@adsl-66-141-23-30.dsl.hstntx.swbell.net> has joined #aichallenge
2010-11-16T15:14:21  <dmj111> dmj.13 was written on tequilla... gotta debug tonight...
2010-11-16T15:14:51  <lavalamp_> haha ok
2010-11-16T15:16:06  <lavalamp_> it frustrates me that I have to hold back my expansion just to beat ragebot, even though no one good will play like ragebot. So I was thinking about making my bot play rage's moves sometimes on close maps just to keep people honest
2010-11-16T15:16:26  <bhasker> who is ron?
2010-11-16T15:16:51  <Error323> dmj111: nice =)
2010-11-16T15:17:16  <JamesMG> man, I have some debugging to do :(
2010-11-16T15:17:34  <Mistmanov> don't we all, don't we all.. :)
2010-11-16T15:17:34  <contestbot> Top 10 players: bocsimacko(3655), bix0r4ever(3556), davidjliu(3493), Raschi(3477), protocolocon(3475), shangas(3457), dmj111(3455), _iouri_(3444), Xrillo(3442), DarkEmi(3440)
2010-11-16T15:17:47  <Error323> Mistmanov: stop developing
2010-11-16T15:17:48  *** shades <shades!~shades@cheshire.telephreak.org> has joined #aichallenge
2010-11-16T15:17:50  <Error323> ^_^
2010-11-16T15:18:32  <Mistmanov> All the new maps business has encouraged me to start working on my bot again actually..
2010-11-16T15:18:42  <Mistmanov> must beat the people who were in favor of new bots ;)
2010-11-16T15:18:46  <Mistmanov> eh maps
2010-11-16T15:18:49  <JamesMG> the new maps broke my most recent code more than my older code
2010-11-16T15:19:01  <JamesMG> probably a good thing though
2010-11-16T15:19:26  <Error323> Mistmanov: are you in for switching binaries?
2010-11-16T15:19:44  * Error323 needs data
2010-11-16T15:19:48  <Error323> =)
2010-11-16T15:20:04  <Mistmanov> you can have my source code :p
2010-11-16T15:20:12  <Mistmanov> really don't care that much :p
2010-11-16T15:20:32  <Mistmanov> how do you do the red letter/private message thing btw -_- (irc noob along with programming noob)
2010-11-16T15:20:53  <JamesMG> that's a client feature when your name is mentioned Mistmanov
2010-11-16T15:21:01  <Mistmanov> ow ok
2010-11-16T15:21:06  <Mistmanov> so everyone can still read it
2010-11-16T15:21:08  <Mistmanov> never mind then :p
2010-11-16T15:21:14  <Error323> ehh o_O
2010-11-16T15:21:19  <JamesMG> you can do /msg name something also though
2010-11-16T15:21:22  <JamesMG> for a real PM
2010-11-16T15:22:07  <Error323> nah you don't wanna give out ur sourcecode man
2010-11-16T15:22:13  <Error323> thats fail =)
2010-11-16T15:22:37  <Mistmanov> suit yourself =)
2010-11-16T15:22:52  <Error323> Mistmanov: are you on windows or linux?
2010-11-16T15:23:00  <Mistmanov> cygwin
2010-11-16T15:23:05  <Mistmanov> no idea how that counts
2010-11-16T15:23:38  *** Utkarsh <Utkarsh!~admin@117.201.89.153> has quit IRC (Ping timeout: 260 seconds)
2010-11-16T15:24:55  <Error323> Mistmanov: is your binary a *.exe
2010-11-16T15:24:56  <Error323> =)
2010-11-16T15:25:06  <Mistmanov> hmhm
2010-11-16T15:25:21  <Error323> mmmkej, crap
2010-11-16T15:25:35  * lavalamp_ wishes my bot were good enough that someone would want to do a binary swap
2010-11-16T15:25:56  <JamesMG> I'll swap your binary for my old minimax source
2010-11-16T15:26:06  <JamesMG> not a good deal for you though
2010-11-16T15:26:09  <Error323> lol
2010-11-16T15:26:12  <lavalamp_> lol
2010-11-16T15:27:53  *** Utkarsh <Utkarsh!~admin@117.201.82.8> has joined #aichallenge
2010-11-16T15:28:27  <brx> maybe i should start writing my bot now.
2010-11-16T15:28:33  * brx looks at the date
2010-11-16T15:28:49  <JamesMG> nah, you still got a couple weeks
2010-11-16T15:28:51  <JamesMG> plenty of time
2010-11-16T15:29:30  <Mistmanov> Error323, still interested?
2010-11-16T15:29:50  <Error323> Mistmanov: well i can't run your binary and you can't run mine
2010-11-16T15:29:55  <Mistmanov> aha
2010-11-16T15:30:07  <Mistmanov> well in that case..
2010-11-16T15:30:07  <Error323> Hostname: jarvis - OS: Linux 2.6.35-gentoo-r5/x86_64 - CPU: 4 x Intel(R) Core(TM)2 Quad  (2666.923 MHz) - Processes: 199 - Uptime: 16d 11h 42m - Load Average: 0.18 - Memory Usage: 655.84MB/2008.39MB (32.65%) - Disk Usage: 635.75GB/1768.99GB (35.94%)
2010-11-16T15:30:11  <JamesMG> ugh, ouch: http://zeroviz.us:8080/canvas?game_id=71
2010-11-16T15:30:16  <brx> lavalamp_: i noticed you linked to a modified "new map generator" in the forums. notably the change of maxRadius. what's up with that? :)
2010-11-16T15:31:09  <lavalamp_> brx: old maps could be larger than new maps. There's more discussion in the forum thread
2010-11-16T15:31:47  <brx> will this be changed officially or is this unknown?
2010-11-16T15:31:53  <lavalamp_> damn dmj stole my idea! lol
2010-11-16T15:32:14  <lavalamp_> brx I think it will be change officially but getting commitments out of them is like pulling teeth
2010-11-16T15:36:28  *** McLeopold <McLeopold!~scott@67.139.105.186> has joined #aichallenge
2010-11-16T15:37:03  <bhasker> interesting game
2010-11-16T15:37:04  <bhasker> http://zeroviz.us:8080/canvas?game_id=118
2010-11-16T15:37:42  <Error323> hahah cool
2010-11-16T15:39:19  <bhasker> dmj dmj.13 needs fixing, i think its failing rage bot
2010-11-16T15:39:26  <bhasker> as someone mentioned before
2010-11-16T15:42:43  <McLeopold> http://zeroviz.us:8080/by_Turns.html
2010-11-16T15:46:11  <bhasker> hah E323 you and me are rated 9/10 on zeroviz lol
2010-11-16T15:46:17  <bhasker> elo of 10 and 9
2010-11-16T15:46:45  <bhasker> lol http://zeroviz.us:8080/canvas?game_id=145
2010-11-16T15:46:47  <bhasker> bug
2010-11-16T15:47:15  <bhasker> we both seem to ignore neutrals with 0 left?
2010-11-16T15:47:21  <JamesMG> hahaha, wow
2010-11-16T15:47:32  <Error323> holy shit
2010-11-16T15:47:36  <JamesMG> I like that game, thanks bhasker
2010-11-16T15:47:48  <bhasker> i guess i will have to see what happened
2010-11-16T15:48:19  <Error323> this version of mine ain't good, but that is a real bug in my code somewhere
2010-11-16T15:48:32  <lavalamp_> lol that's hilarious
2010-11-16T15:48:56  <bhasker> yea
2010-11-16T15:50:02  <bhasker> heh lamo, got sniped all over http://zeroviz.us:8080/canvas?game_id=157
2010-11-16T15:50:53  <Error323> oehf
2010-11-16T15:50:55  <Error323> that hurts
2010-11-16T15:55:24  <Mistmanov> @tcp
2010-11-16T15:55:25  <contestbot> Mistmanov: tcp = (#1) http://www.benzedrine.cx/planetwars/, or (#2) http://72.44.46.68/, or (#3) http://zeroviz.us:8080, or (#4) 98.247.248.39:995 (zeroviz.us)
2010-11-16T15:56:09  <bhasker> lol error both of us in -ve elo on new server
2010-11-16T15:56:10  <bhasker> :(
2010-11-16T15:56:11  *** sigh <sigh!~sigh@28.250-50-210.dynamic.dsl.mel.iprimus.net.au> has quit IRC (Remote host closed the connection)
2010-11-16T15:56:35  <Error323> yeah
2010-11-16T15:56:36  <Error323> :(
2010-11-16T15:57:59  <Error323> i'll just put my regular bot in there
2010-11-16T15:58:06  <Error323> but it'll prolly won't do much better
2010-11-16T15:58:41  <mega1> McLeopold: where does your tcp server live now?
2010-11-16T15:58:58  <Error323> http://zeroviz.us:8080/
2010-11-16T15:59:00  <Error323> just ping it mega1
2010-11-16T15:59:03  <McLeopold> @tcp
2010-11-16T15:59:03  <contestbot> McLeopold: tcp = (#1) http://www.benzedrine.cx/planetwars/, or (#2) http://72.44.46.68/, or (#3) http://zeroviz.us:8080, or (#4) 98.247.248.39:995 (zeroviz.us)
2010-11-16T15:59:13  <Error323> McLeopold: your IP is dynamic
2010-11-16T15:59:14  <McLeopold> ^^ new ip
2010-11-16T15:59:16  <mega1> saw that
2010-11-16T15:59:24  <mega1> same home connection?
2010-11-16T15:59:28  <Error323> which is kinda annoying =)
2010-11-16T15:59:30  <McLeopold> It only changes after a power outage. :)
2010-11-16T15:59:41  <Error323> ah
2010-11-16T15:59:42  <McLeopold> So, once a year.
2010-11-16T15:59:42  <Error323> hehe
2010-11-16T15:59:47  <mega1> thought you moved it for a minute
2010-11-16T16:00:42  <Error323> http://zeroviz.us:8080/canvas?game_id=180 lol Mistmanov
2010-11-16T16:02:02  <Mistmanov> eh, the insane "supplying each other" thing still needs to be fixed ;)
2010-11-16T16:02:21  <Mistmanov> although it may help in screwing up the opponent's calculations
2010-11-16T16:03:42  <Error323> hehe
2010-11-16T16:04:21  <Error323> not in the top 50 i think
2010-11-16T16:04:31  <Error323> or 100
2010-11-16T16:04:52  <McLeopold> sweet, I'm number 1 on my own server!
2010-11-16T16:04:56  <JamesMG> hooray, my dev bot finally won something on zerovis
2010-11-16T16:05:30  <McLeopold> you've got 5 wins
2010-11-16T16:05:48  <McLeopold> oh, 4 to yourself...
2010-11-16T16:06:03  <Error323> lol i can't believe there are still user registries
2010-11-16T16:06:05  <JamesMG> it's a rousing success
2010-11-16T16:06:05  <Error323> http://ai-contest.com/game_stats.php
2010-11-16T16:07:58  <Error323> i recommend not instantiating multiple versions at the same time on a tcp servr
2010-11-16T16:08:10  <Error323> because it will bias your rating a lot
2010-11-16T16:09:11  <Mistmanov> being second to last one a server (even if there are only 14 players above you.) feels really depressing -_-
2010-11-16T16:09:36  <McLeopold> I'll turn Rage back on. :)
2010-11-16T16:09:42  <Error323> just as negative elo
2010-11-16T16:09:45  <Error323> ;_;
2010-11-16T16:09:52  <JamesMG> Mistmanov: don't feel bad, I'm last :)
2010-11-16T16:10:06  <Mistmanov> yeah it could always be worse ;)
2010-11-16T16:10:15  <Error323> not always :P
2010-11-16T16:11:23  <Error323> shit i'm such a moron
2010-11-16T16:11:29  <Error323> still testing on the standard maps >_<
2010-11-16T16:11:54  *** eburnette <eburnette!621b3c3f@gateway/web/freenode/ip.98.27.60.63> has joined #aichallenge
2010-11-16T16:13:58  <eburnette> I'm trying the zeroviz server now, it's a little mind blowing after staring at point symmetric maps for so long
2010-11-16T16:15:22  <eburnette> There must be something wrong with the server, it's printing this a lot: You LOSE against E323-17.2 :)
2010-11-16T16:15:23  *** delt0r___ <delt0r___!~delt0r@188-22-165-46.adsl.highway.telekom.at> has quit IRC (Read error: Connection reset by peer)
2010-11-16T16:15:26  <Error323> http://ai-contest.com/game_stats.php the cloud is vaporizing
2010-11-16T16:15:50  <Error323> eburnette: lol
2010-11-16T16:16:25  <JamesMG> yeah I'm pretty sure it's broken, I only win against myself
2010-11-16T16:17:06  *** delt0r___ <delt0r___!~delt0r@188-22-26-87.adsl.highway.telekom.at> has joined #aichallenge
2010-11-16T16:17:34  <contestbot> Top 10 players: bocsimacko(3656), bix0r4ever(3544), _Astek_(3535), protocolocon(3472), davidjliu(3469), xpctxpeer(3457), _iouri_(3451), dmj111(3450), shangas(3449), Xrillo(3431)
2010-11-16T16:19:44  <Error323> so guys the old tcp server doesn't have a new mapset?
2010-11-16T16:20:23  <JamesMG> correct
2010-11-16T16:20:46  <eburnette> (wishes he had better logging)
2010-11-16T16:21:06  <JamesMG> eburnette: are you electronicBurrito?
2010-11-16T16:21:20  <eburnette> funny name but no
2010-11-16T16:21:34  <eburnette> I do pokemon names, currently up to Houndoom
2010-11-16T16:21:37  <JamesMG> oh, they just appeared in the rankings on zeroviz so I assumed
2010-11-16T16:25:33  <Error323> McLeopold: you are sure that your server is using the maps generated by the topic mapgen right?
2010-11-16T16:25:45  <eburnette> Accoun is sniping me to death. My minimax is definitely not working right.
2010-11-16T16:25:47  <McLeopold> yes
2010-11-16T16:25:55  <Error323> ok good
2010-11-16T16:26:06  <Error323> McLeopold: how many maps?
2010-11-16T16:26:10  <McLeopold> But, I believe there is a bais in the generator
2010-11-16T16:26:20  <McLeopold> http://zeroviz.us:8080/by_Map.html
2010-11-16T16:26:23  <McLeopold> about 160
2010-11-16T16:26:26  <Error323> ok
2010-11-16T16:27:14  <bhasker> accoun snipes a lot
2010-11-16T16:27:19  <bhasker> if you are too aggressive against him
2010-11-16T16:27:20  <bhasker> you get owned
2010-11-16T16:29:25  <bhasker> E323 at top
2010-11-16T16:29:28  <McLeopold> nice
2010-11-16T16:29:58  <bhasker> with this asymmetrical maps its all  the more important not to go for disputed planets:-\
2010-11-16T16:30:03  <eburnette> lol that was quick
2010-11-16T16:30:06  <bhasker> i keep getting owned
2010-11-16T16:30:12  <bhasker> by going for wrong planets at start
2010-11-16T16:30:48  <eburnette> I think knapsack does not work on this type of map
2010-11-16T16:31:36  <eburnette> Is newmap27.txt valid?
2010-11-16T16:32:42  <lavalamp_> honestly I don't see how peolpe have gotten knapsack to work at all... do you factor in travel time?
2010-11-16T16:32:53  <eburnette> yeah
2010-11-16T16:32:58  <JamesMG> I have at various points lavalamp_
2010-11-16T16:33:08  <JamesMG> eburnette: I think it's valid, those planets are all 2 turns away, right?
2010-11-16T16:33:16  <eburnette> it helped my score a lot a couple of weeks ago
2010-11-16T16:33:30  <lavalamp_> do you factor in the ships that will be available the turn after you capture a new planet
2010-11-16T16:33:35  <lavalamp_> ?
2010-11-16T16:34:58  <eburnette> what do you count as 2 turns away? Launch on turn 3 and arrive on turn 4?
2010-11-16T16:35:41  <McLeopold> distance of 2.00 to 2.99999
2010-11-16T16:36:07  <JamesMG> talking about this game, right? http://zeroviz.us:8080/canvas?game_id=298
2010-11-16T16:36:36  <d0ugal> @tcp
2010-11-16T16:36:37  <contestbot> d0ugal: tcp = (#1) http://www.benzedrine.cx/planetwars/, or (#2) http://72.44.46.68/, or (#3) http://zeroviz.us:8080, or (#4) 98.247.248.39:995 (zeroviz.us)
2010-11-16T16:36:48  <eburnette> JamesMG yes
2010-11-16T16:38:07  <JamesMG> I can't actually tell, now that you mention it.  would have to step through the game state
2010-11-16T16:38:14  <eburnette> profit[planet] = (horizon - distanceFromStarter) * dest.GrowthRate()) - attackSize
2010-11-16T16:39:05  <McLeopold> what's horizon?
2010-11-16T16:39:41  <eburnette> it varies
2010-11-16T16:40:03  <eburnette> JamesMG first command was "F 2 17 2 17 2 1 /  F 1 20 1 16 2 1"
2010-11-16T16:40:18  <eburnette> first state I mean, turn 1
2010-11-16T16:41:22  <lavalamp_> eburnette that map looks fine to me
2010-11-16T16:42:42  <eburnette> ok
2010-11-16T16:42:57  <lavalamp_> yay I'm at my highest official rating since the map change
2010-11-16T16:43:31  <JamesMG> I think my bot has settled in at 250th or so
2010-11-16T16:44:07  <lavalamp_> unlike my last one I should have plenty of ways to continue to improve after I get up there
2010-11-16T16:44:16  <Error323> wuut
2010-11-16T16:44:20  <Error323> i'm second \o/
2010-11-16T16:44:38  <Error323> McLeopold: why is the rank messed up btw?
2010-11-16T16:44:57  <McLeopold> how so
2010-11-16T16:45:19  <Error323> http://zeroviz.us:8080/ very weird countings
2010-11-16T16:45:31  <McLeopold> that's only the active bots
2010-11-16T16:45:32  <Error323> what does it mean
2010-11-16T16:45:36  <Error323> oh
2010-11-16T16:47:05  <McLeopold> better?
2010-11-16T16:47:06  <Mistmanov> ugh.. there I was, all proud that my bot scored a victory against the new player "bitterandtwisted"
2010-11-16T16:47:20  <Mistmanov> looked at the game and went "oh... ragebot in disguise >_<"
2010-11-16T16:48:53  <McLeopold> we gotta get george and iouri on zeroviz...
2010-11-16T16:49:15  <JamesMG> all the top 10
2010-11-16T16:49:15  <McLeopold> and flag
2010-11-16T16:49:25  <eburnette> what's the holdup on betting benzedrine updated?
2010-11-16T16:49:39  <McLeopold> and oldman, even though his bot enrages me
2010-11-16T16:49:47  <JamesMG> haha, I hear that
2010-11-16T16:49:47  <McLeopold> dunno, can't get ahold of dhartmei
2010-11-16T16:50:27  *** Naktibalda <Naktibalda!~Naktibald@host-81-20-37-12.eckoh.com> has quit IRC (Quit: ChatZilla 0.9.86 [Firefox 3.6.8/20100722155716])
2010-11-16T16:50:54  <Frontier> Does anyone know how openmp works? I just though it could be active by default in the current gcc releases and unroll loops on multiple cpu cores transparently. That could become an oversight on the server side some time and give c(++) programs an advantage. Or is there no need to worry?
2010-11-16T16:50:59  *** nullkuhl <nullkuhl!~nullkuhl@196.221.227.142> has quit IRC (Read error: Connection reset by peer)
2010-11-16T16:51:23  <eburnette> OpenMP requires you to code #pragmas
2010-11-16T16:51:46  <Frontier> alright, thx for the info
2010-11-16T16:52:11  <eburnette> However the c compiler has options to do things with vectorization, not quite the same as multiple threads though
2010-11-16T16:53:38  <Frontier> it could use simd instructions, why not :)
2010-11-16T16:54:48  <bhasker> hmm weird
2010-11-16T16:54:49  <bhasker> http://zeroviz.us:8080/canvas?game_id=374
2010-11-16T16:54:53  <bhasker> why did i lose this one?
2010-11-16T16:55:22  <bhasker> no error message nothing nada
2010-11-16T16:55:26  <bhasker> it just stopped at 168
2010-11-16T16:55:30  <Error323> O_O
2010-11-16T16:56:30  <Error323> McLeopold: ^
2010-11-16T16:56:32  <Error323> ?
2010-11-16T16:56:49  <Mistmanov> I also didn't get an error message for a game I just lost by getting into an infinite loop..
2010-11-16T16:57:02  <McLeopold> oh, I haven't written the part to show you disconnects and timeouts yet
2010-11-16T16:57:39  *** amstan_ <amstan_!~alex@129-97-120-91.uwaterloo.ca> has joined #aichallenge
2010-11-16T16:57:39  *** ChanServ sets mode: +o amstan_
2010-11-16T16:58:01  <bhasker> aah
2010-11-16T16:58:43  <McLeopold> it was a timeout
2010-11-16T16:58:58  <bhasker> hmm
2010-11-16T17:03:15  *** sigh <sigh!~sigh@lithium.mailguard.com.au> has joined #aichallenge
2010-11-16T17:07:29  <Error323> holy fucking hell
2010-11-16T17:07:51  <Error323> oh no nvm
2010-11-16T17:07:56  <Error323> i need a break ^_^
2010-11-16T17:10:32  <McLeopold> http://zeroviz.us:8080/by_Problems.html
2010-11-16T17:11:00  <McLeopold> mistmanov: http://zeroviz.us:8080/INVALID.html
2010-11-16T17:11:28  <Mistmanov> yay, the only invalid move!
2010-11-16T17:11:29  <Mistmanov> :p
2010-11-16T17:11:52  <Mistmanov> ok, so it wasn't an infinite loop.. but the problem has still been solved =)
2010-11-16T17:12:26  <McLeopold> d0ugal: http://zeroviz.us:8080/canvas?game_id=424
2010-11-16T17:13:51  <d0ugal> Nice
2010-11-16T17:13:59  <Zannick> yeah, that map...
2010-11-16T17:14:21  <d0ugal> My bot is really bad on the maps that are that close
2010-11-16T17:14:30  <d0ugal> but then its poor compared to all you guys in general
2010-11-16T17:15:41  <McLeopold> so, try not to attack too far away
2010-11-16T17:16:00  <McLeopold> put a max distance on your attacking, and you'll do much better
2010-11-16T17:16:21  <McLeopold> There's probably a better way to go about that, but just try the easy way and see how you do.
2010-11-16T17:16:22  <d0ugal> McLeopold: Yeah, I'm just workin on that part
2010-11-16T17:16:55  <d0ugal> It's a total re-write, I only got it to a point where it finished a game two days ago, so plenty of tweaking to do
2010-11-16T17:17:08  <McLeopold> cool
2010-11-16T17:17:23  <d0ugal> Need to update the fitness function to give the distance more of a negative value
2010-11-16T17:17:36  <contestbot> Top 10 players: bocsimacko(3656), thedreamer(3512), davidjliu(3484), protocolocon(3470), _iouri_(3454), shangas(3449), dmj111(3442), bix0r4ever(3440), oldman(3433), Xrillo(3428)
2010-11-16T17:18:49  *** narnach <narnach!~Narnach@195-241-27-75.ip.telfort.nl> has quit IRC (Quit: Leaving.)
2010-11-16T17:19:14  *** narnach <narnach!~Narnach@195-241-27-75.ip.telfort.nl> has joined #aichallenge
2010-11-16T17:19:22  <d0ugal> But the handy thing about the tcp server allows me to collect data and learn from the top players ;) heh
2010-11-16T17:21:08  <eburnette> Darn, I forgot to add if (stupid(move)) break;
2010-11-16T17:21:12  <Error323> McLeopold: how are  you determining when players have disconnected?
2010-11-16T17:21:25  <Error323> i mean, to refresh the list
2010-11-16T17:21:28  <McLeopold> That part is dhartmei's code
2010-11-16T17:21:43  <Error323> oh
2010-11-16T17:21:48  <Error323> cuz tcp ain't doing that
2010-11-16T17:22:43  <McLeopold> if the tcp connection is closed
2010-11-16T17:22:56  <McLeopold> tcp is a stateful connection
2010-11-16T17:23:04  <McLeopold> so your end needs to not hang up
2010-11-16T17:25:32  <Error323> men
2010-11-16T17:25:54  <Error323> i wonder if the dev team truely realized how much they changed the game with the new mapgen
2010-11-16T17:26:01  <Error323> and how many ppl they pissed off
2010-11-16T17:26:20  <Mistmanov> who says there aren't any women here -_-
2010-11-16T17:26:37  <McLeopold> I think they don't care, since technically it's still within the original specs.
2010-11-16T17:26:38  <Mistmanov> but yeah.. the dev team itself isn't developing any bots?
2010-11-16T17:26:44  *** Mathnerd314 <Mathnerd314!~mathnerd3@128.198.97.194> has joined #aichallenge
2010-11-16T17:26:47  <JamesMG> McLeopold: any chance you could put a newline between player names and shipcount/growth?
2010-11-16T17:26:51  <bhasker> lol http://zeroviz.us:8080/canvas?game_id=496
2010-11-16T17:26:54  <JamesMG> on the canvas page
2010-11-16T17:26:59  <Mistmanov> that may influence their perspective on how they see changes to the rules
2010-11-16T17:27:02  <Zannick> i think the biggest problem with the new mapgen is that somehow a lot of the maps have very close planets that the visualizer overlaps
2010-11-16T17:27:13  <McLeopold> JamesMG: I'll try
2010-11-16T17:27:23  <McLeopold> tired of the wiggle?
2010-11-16T17:27:27  <JamesMG> yup
2010-11-16T17:27:41  <McLeopold> Zannick: I think there is a bias in the new generator
2010-11-16T17:27:51  <McLeopold> for planets to be closer to center
2010-11-16T17:28:12  <McLeopold> it's the whole sqrt() thing for radius, instead of log()
2010-11-16T17:28:35  <Zannick> hm
2010-11-16T17:29:00  <Zannick> that should weight it higher, though
2010-11-16T17:29:18  <Zannick> log(random.random()) would make it negative
2010-11-16T17:29:38  <lavalamp_> I don't understand why it does the generation in polar coords
2010-11-16T17:29:45  <McLeopold> log( random( 1 to 4) , 2)
2010-11-16T17:29:59  <McLeopold> yeah, that's kinda a waste to
2010-11-16T17:30:18  <Zannick> the radius is generated by sqrt( uniform[0,1) )
2010-11-16T17:30:26  <Zannick> times max and whatnot
2010-11-16T17:30:40  *** davidjliu <davidjliu!473a6d0e@gateway/web/freenode/ip.71.58.109.14> has joined #aichallenge
2010-11-16T17:30:51  <Zannick> the generation is in polar to get the symmetry more easily
2010-11-16T17:31:09  <McLeopold> you don't need polar for symmetry
2010-11-16T17:31:35  <McLeopold> generate x and y
2010-11-16T17:31:46  <Zannick> no you don't, but for arbitrary angle reflective symmetry it's easier
2010-11-16T17:31:47  <lavalamp_> ... it's possible to convert one to the other
2010-11-16T17:31:48  <McLeopold> for point symmetry, use -x and -y
2010-11-16T17:32:00  <McLeopold> for line symmetry, use -x, y   or  x, -y
2010-11-16T17:32:00  <Zannick> line symmetry, i mean
2010-11-16T17:32:24  <McLeopold> rotate afterwards if you need it to be astetically pleasing
2010-11-16T17:32:37  <lavalamp_> even for line symmetry  you can convert x,y to r,theta
2010-11-16T17:32:43  <eburnette> whoohoo
2010-11-16T17:32:44  <McLeopold> and translate to positive coords
2010-11-16T17:32:58  <Zannick> i agree, though, the distribution is making planets much closer
2010-11-16T17:33:26  <lavalamp_> there's also more planets now though, so you have to expect it to be denser
2010-11-16T17:33:31  <McLeopold> the other thing is, I think the old generator staying in a square shape, the new one is in a circle shape
2010-11-16T17:33:42  <lavalamp_> yes that is true
2010-11-16T17:33:46  <Zannick> yes
2010-11-16T17:33:57  <McLeopold> but if you don't hit large radii, it doesn't matter
2010-11-16T17:34:17  <Zannick> that removes less than a quarter of the original space
2010-11-16T17:34:32  <McLeopold> for radius 12, yes
2010-11-16T17:34:43  <McLeopold> but he want's to bump it up to 17, so more space
2010-11-16T17:34:46  <Zannick> it should hit large radii, though, it's less likely to hit small radii
2010-11-16T17:34:59  <McLeopold> Zannick, it should be even
2010-11-16T17:35:17  <Zannick> it should be that every point be as likely as before
2010-11-16T17:35:25  <McLeopold> radius 2 *should* have twice the probability, if he used log
2010-11-16T17:35:28  <Zannick> or at least balance
2010-11-16T17:36:47  <Zannick> probability of exact numbers are hard to specify
2010-11-16T17:37:16  <McLeopold> well, the circumference of radius 2 is twice radius 1
2010-11-16T17:37:33  <Zannick> yes, and the radius is a randomly selected "real"
2010-11-16T17:37:46  <Zannick> so yes, twice 0 is 0 :P
2010-11-16T17:37:51  <McLeopold> ?
2010-11-16T17:38:16  <Zannick> what's the probability a random real number between 0 and 12 is exactly 1?
2010-11-16T17:38:38  <McLeopold> very small
2010-11-16T17:38:39  <McLeopold> but
2010-11-16T17:38:56  <amstan_> ugh... a bunch of crybabies
2010-11-16T17:39:00  <McLeopold> what's the probability that it is closer to 1 than 2?
2010-11-16T17:39:02  <amstan_> now transforming in a flame war
2010-11-16T17:39:07  <amstan_> perhaps i should lock that topic
2010-11-16T17:39:11  <McLeopold> amstan: this is math, not flame
2010-11-16T17:39:25  <Zannick> he's talking about the forum
2010-11-16T17:39:33  <McLeopold> amstan: I think there is a bais in the map generator, which should be fixed
2010-11-16T17:41:34  <JamesMG> good old fashioned base swap: http://zeroviz.us:8080/canvas?game_id=538
2010-11-16T17:41:38  <eburnette> You WIN against dmj.12
2010-11-16T17:45:39  <bhasker> woohoo i have a +ve elo of 4 on mcleopolds server
2010-11-16T17:46:31  <amstan_> i'm talking about the random people ranting how this is all very unprofessional, and how we're lying
2010-11-16T17:46:42  <amstan_> and how we should not host this contest in the future
2010-11-16T17:46:57  <amstan_> like.. i'm doing this is my spare time
2010-11-16T17:47:20  <Error323> yeah don't pay attention to that
2010-11-16T17:47:32  <amstan_> it's really depressing
2010-11-16T17:47:56  <Error323> but you have to understand that the mapchange had a massive impact on some ppl's time which is easily just as frustrating
2010-11-16T17:49:22  <JamesMG> amstan_: I don't think you should take that to heart
2010-11-16T17:50:08  <Zannick> amstan_: are they calling you evil canadians yet?
2010-11-16T17:50:10  <eburnette> You can't make everybody happy amstan_
2010-11-16T17:50:14  <JamesMG> Error323: http://zeroviz.us:8080/canvas?game_id=558
2010-11-16T17:50:20  <amstan_> Zannick: that's mostly Accoun
2010-11-16T17:50:35  <Error323> JamesMG: heh yeah, be more carefull ;)
2010-11-16T17:50:37  <Zannick> i'm just saying :)
2010-11-16T17:50:53  <Zannick> wooooow
2010-11-16T17:50:57  <eburnette> I don't like maps like that one in game 558
2010-11-16T17:51:17  <Error323> amstan_: on a different topic, is it true that the elo rating has a history knowledge of only 3 days?
2010-11-16T17:51:19  <Zannick> okay, that looks like a serious issue with generating theta...
2010-11-16T17:51:49  <amstan_> Error323: i have no idea
2010-11-16T17:53:28  <McLeopold> amstan: I'm very grateful for all your work, and the other organizers.
2010-11-16T17:53:39  <McLeopold> This contest is awesome.
2010-11-16T17:53:47  <eburnette> +1
2010-11-16T17:53:49  <JamesMG> Agreed, I'm having fun
2010-11-16T17:54:24  <Error323> so am i
2010-11-16T17:54:26  <amstan_> thanks
2010-11-16T17:54:27  <lavalamp_> I'm not sure if I'm having fun or am obsessed... but thanks for running it
2010-11-16T17:54:33  <Mistmanov> =) a little difference of opinion over some aspects of the contests shouldn't overshadow the overall experience =)
2010-11-16T17:54:44  <Error323> indeed
2010-11-16T17:55:26  <Error323> damnit dmj111 I hate you
2010-11-16T17:55:36  <Error323> your attacks are scary
2010-11-16T17:56:20  <McLeopold> Actually, all the people ranting about how unprofessional this is, is really a testament to how close this is to really being professional.
2010-11-16T17:56:41  <amstan_> McLeopold: hmm?
2010-11-16T17:56:53  <McLeopold> Kinda like only being critical of things you care about.
2010-11-16T17:57:28  <Zannick> antimatroid: there seem to be problems with distances http://zeroviz.us:8080/canvas?game_id=558
2010-11-16T17:57:57  <McLeopold> Zannick: that map is totally within spec
2010-11-16T17:58:10  <dmj111> Error323: that hurts :)
2010-11-16T17:58:20  <Error323> ;)
2010-11-16T17:58:23  <Zannick> or random numbers
2010-11-16T17:58:32  <mega1> amstan_: which topic is that?
2010-11-16T17:58:35  <JamesMG> that map/game is kind of awesome actually
2010-11-16T17:58:38  <JamesMG> 558 I mean
2010-11-16T17:58:47  <JamesMG> round robin and brutally short
2010-11-16T17:58:47  <Zannick> McLeopold: when i see a lot of those maps, i suspect a problem is occurring
2010-11-16T17:58:50  *** dmj111 <dmj111!~user@c-98-235-91-228.hsd1.pa.comcast.net> has left #aichallenge ("ERC Version 5.3 (IRC client for Emacs)")
2010-11-16T17:59:01  <amstan_> mega1: http://www.ai-contest.com/forum/viewtopic.php?f=18&t=1094&start=40#p7360
2010-11-16T17:59:03  <McLeopold> True, there may be another bias
2010-11-16T17:59:58  <eburnette> mega1 is still on top though
2010-11-16T18:00:00  <eburnette> @ratings
2010-11-16T18:00:01  <contestbot> eburnette: An error occured while trying to show the previous error.
2010-11-16T18:00:02  *** dmj111 <dmj111!~user@c-98-235-91-228.hsd1.pa.comcast.net> has joined #aichallenge
2010-11-16T18:00:09  <eburnette> @rankings
2010-11-16T18:00:11  <contestbot> eburnette: Top 10 players: bocsimacko(3656), oldman(3566), GreenTea(3502), davidjliu(3493), bix0r4ever(3485), protocolocon(3466), shangas(3452), _iouri_(3443), dmj111(3442), george(3426)
2010-11-16T18:00:19  <mega1> eburnette:  with a smaller margin
2010-11-16T18:00:43  <amstan_> oh man.. oldman got second.. epic
2010-11-16T18:01:14  <lavalamp_> oldman only has 3 games
2010-11-16T18:01:38  <eburnette> mega1: do you have any plans to try the 98.247.248.39:995 server?
2010-11-16T18:01:59  <mega1> yes, when the new version settles down
2010-11-16T18:02:08  <amstan_> why do you guys use the ip address directly? wouldn't the hostname be more flexible?
2010-11-16T18:02:13  <mega1> maybe in about 5-6 hours
2010-11-16T18:02:22  <mega1> amstan_: doesn't work for some reason
2010-11-16T18:02:31  <amstan_> weird
2010-11-16T18:02:35  <mega1> maybe the client doesn't look it up properly
2010-11-16T18:03:08  <eburnette> Somebody's home computer using dyndns maybe?
2010-11-16T18:04:17  <amstan_> i noticed McLeopold changing the ip in the contestbot factoid
2010-11-16T18:05:05  <McLeopold> after the power outage, I got a new ip
2010-11-16T18:05:16  <McLeopold> dns has to propogate throughout the world now
2010-11-16T18:05:57  <iamluck> McLeopold: change TTL of your DNS, if you can, in case it happens again :)
2010-11-16T18:05:59  <McLeopold> mega1: it's working now
2010-11-16T18:06:05  <McLeopold> what to?
2010-11-16T18:06:19  <McLeopold> I think it's a 2 hours?
2010-11-16T18:06:21  <iamluck> 60 minutes? how low can you go?
2010-11-16T18:06:56  <McLeopold> 2 mintues
2010-11-16T18:07:42  <iamluck> Then after this change there shouldn't be many issues with "dns has to propogate throughout the world now"
2010-11-16T18:08:44  *** Tobu <Tobu!~Tobu@unaffiliated/tobu> has quit IRC (Ping timeout: 272 seconds)
2010-11-16T18:08:57  <Zannick> yeah, i'm tempted to rewrite the coordinate generation in the map generator
2010-11-16T18:09:15  <Zannick> so that each point is equally likely
2010-11-16T18:09:17  <amstan_> #topic add Turn count WILL stay at 200!
2010-11-16T18:09:22  <sigh> what's wrong with the coordinate generation?
2010-11-16T18:09:24  <amstan_> @topic add Turn count WILL stay at 200!
2010-11-16T18:09:24  *** contestbot changes topic to "Official Google AI Challenge: http://ai-contest.com/ || Channel Logs: http://contestbot.hypertriangle.com/ || No stupid questions. Don't ask to ask. || New MapGen is being used: http://tinyurl.com/aimapgen || Turn count WILL stay at 200!"
2010-11-16T18:09:27  <McLeopold> yeah!
2010-11-16T18:09:30  <McLeopold> a decision!
2010-11-16T18:10:03  <mega1> Zannick: does it generate angles and distances uniformly or what?
2010-11-16T18:10:05  <sigh> 200! is a bit high, imo
2010-11-16T18:10:10  <Zannick> the change in generating points by radius and theta seems to be throwing something off
2010-11-16T18:10:15  <amstan_> sigh: shut up, lol
2010-11-16T18:10:18  <sigh> :(
2010-11-16T18:10:29  <Zannick> or the reduced space is making a lot of very close maps
2010-11-16T18:10:33  <Error323> "Someone doesn't like having their cheese moved. Well AI is about handling moved cheese. And seriously, it hasn't been moved much at all."
2010-11-16T18:10:39  <Error323> yeah you gotta love that quote
2010-11-16T18:10:41  <Error323> :D
2010-11-16T18:11:02  <mega1> I have enough clueless quotes.
2010-11-16T18:11:10  <eburnette> dmj11 we're having a weird game now
2010-11-16T18:11:36  <mega1> now, who have I just insulted?
2010-11-16T18:11:43  <lavalamp_> decision w00t!
2010-11-16T18:11:50  <Zannick> though i told myself i would not contribute anything more until the contest was over
2010-11-16T18:11:58  <eburnette> I know why I'm not moving, why are you not moving dmj111?
2010-11-16T18:12:55  <dmj111> eburnette: I don't see it... not sure if I have a dead process or something?
2010-11-16T18:13:20  <Error323> mega1: clueless?
2010-11-16T18:13:29  *** Mistmanov <Mistmanov!4da38f1c@gateway/web/freenode/ip.77.163.143.28> has quit IRC (Quit: bedtime..)
2010-11-16T18:13:30  <eburnette> http://98.247.248.39:8080/canvas?game_id=641
2010-11-16T18:13:33  <sigh> oh, radius is a uniform distribution, yeah, that willl cluster points in the center
2010-11-16T18:13:58  <dmj111> Zannick: i think that would be a good change.
2010-11-16T18:14:13  <mega1> Error323: just a more combative way of saying that I'm not in agreement.
2010-11-16T18:14:46  <dmj111> eburnette: my bot is, and might always be, a chicken.
2010-11-16T18:15:31  <Error323> mega1: I see
2010-11-16T18:16:55  <Error323> holy crap its 00:12 already
2010-11-16T18:16:58  <Error323> ugh
2010-11-16T18:17:00  <McLeopold> mega1: Why dost thou not play on new tcp?
2010-11-16T18:17:31  <mega1> McLeopold: I'm at work and testing new stuff too locally.
2010-11-16T18:17:36  <contestbot> Top 10 players: bocsimacko(3655), oldman(3564), GreenTea(3497), davidjliu(3493), bix0r4ever(3484), protocolocon(3472), shangas(3450), _iouri_(3442), dmj111(3441), george(3424)
2010-11-16T18:17:53  <McLeopold> okay, just send me your old code and I'll get it up and running for you
2010-11-16T18:18:33  <mega1> I have very promising local tests.
2010-11-16T18:18:41  <antimatroid> i've had a read up above
2010-11-16T18:18:54  <McLeopold> use log!
2010-11-16T18:18:58  <antimatroid> 2 things, 1. they would be significantly more bunched up atm, the map size is tinier
2010-11-16T18:19:05  <McLeopold> ..was my opinion
2010-11-16T18:19:14  <antimatroid> seconly log(0,1) is between -inf and 0 or whatever
2010-11-16T18:19:14  <mega1> maybe I can catch up with the map changes after several days of hacking
2010-11-16T18:19:25  <antimatroid> secondly, using sqrt is meant to give uniform distribution inside the circle
2010-11-16T18:19:28  <antimatroid> on average
2010-11-16T18:19:44  <mega1> "uniform on average"
2010-11-16T18:19:46  <mega1> ?
2010-11-16T18:20:00  <McLeopold> Yeah, but this is math, and I can prove you wrong.
2010-11-16T18:20:20  <antimatroid> we did ask the #math channel :\
2010-11-16T18:20:25  <mega1> ah, just a terminology confusion
2010-11-16T18:20:35  <McLeopold> :)
2010-11-16T18:20:44  <McLeopold> seriously, do you want a graph?
2010-11-16T18:21:20  <Error323> mega1: if you got a new bot, make sure to run it at McLeopold 's server, the old tcp has obsolete maps and will rate you wrong
2010-11-16T18:21:31  *** nullkuhl <nullkuhl!~nullkuhl@196.221.227.142> has joined #aichallenge
2010-11-16T18:21:40  <McLeopold> Well, they won't be that off, but yes, use the new one.
2010-11-16T18:21:43  <mega1> Error323: I'll run at both
2010-11-16T18:21:59  <Error323> McLeopold: they will be that off hehe
2010-11-16T18:22:04  <mega1> they are significantly different
2010-11-16T18:22:04  <McLeopold> ( Little does everyone know that they are feeding my neutral net.  Bwahaha! )
2010-11-16T18:22:16  <Error323> how come ppl don't see how a massive impact those maps have on bots o_O
2010-11-16T18:22:20  <mega1> neutral net is a new concept
2010-11-16T18:22:32  <Zannick> naw, net neutrality has existed for a while
2010-11-16T18:22:43  <McLeopold> not for long...
2010-11-16T18:22:47  <Error323> lol neutral net :P
2010-11-16T18:22:52  <antimatroid> i love it how people say the rules just need to be clear about 200 turn limit so people don't get an advantage from guessing the rules
2010-11-16T18:23:00  <antimatroid> yet half of those people are bitching like crazy about the map change
2010-11-16T18:23:07  <JamesMG> if you needed to run a TCP server to feed a neural net I wouldn't be too worried about it
2010-11-16T18:23:22  <Zannick> to be honest my bot assumes the game will end on a multiple of 50...
2010-11-16T18:23:30  <mega1> Error323: maybe they have more basic problems
2010-11-16T18:23:31  <Error323> antimatroid: i'm not bitching, just stating facts
2010-11-16T18:23:51  <antimatroid> Error323: the old maps were never a rule
2010-11-16T18:24:05  <antimatroid> to now use them allows people who guessed they would be used to have that same advantage people are talking about
2010-11-16T18:24:21  <Error323> doesn't matter antimatroid, its what the server has been running for all these months
2010-11-16T18:24:24  <antimatroid> but i've been over that, i was just pointing out peoples double argumets :)
2010-11-16T18:24:27  <Error323> ppl overfit on them, it happens
2010-11-16T18:24:28  <Zannick> we could tell people that if they're unhappy with the change we'll refund their money
2010-11-16T18:24:31  <Zannick> ;)
2010-11-16T18:24:44  <mega1> I have many ideas, but until the more serious problems are fixed it's impossible to gauge their worth.
2010-11-16T18:24:46  *** lavalamp_ <lavalamp_!4507d76a@gateway/web/freenode/ip.69.7.215.106> has quit IRC (Quit: Page closed)
2010-11-16T18:24:50  <antimatroid> Zannick: time is money
2010-11-16T18:25:01  <Zannick> antimatroid: not to the irs
2010-11-16T18:25:41  <JamesMG> I hope not, if they started taxing me on free time I'd be broke
2010-11-16T18:26:06  <antimatroid> i don't think i'm going to like future contests as much, it looks like people will be demanding genetors be set in stone early so they can gear their "bots" to work on very specific kinds of maps
2010-11-16T18:27:02  <JamesMG> maybe make it a facet of the competition that generators can and will change
2010-11-16T18:27:11  <JamesMG> makes the "AI" aspect of it more interesting anyways
2010-11-16T18:27:24  <iamluck> antimatroid: then make the map generator not very specific to begin with? :)
2010-11-16T18:27:29  <mega1> antimatroid: then make the generator general enough
2010-11-16T18:27:37  <Zannick> no, we'll set in stone the qualities of the map
2010-11-16T18:27:43  <Zannick> not the generator
2010-11-16T18:28:02  *** JamesMG <JamesMG!~jamesg@38.98.50.162> has quit IRC (Quit: Leaving.)
2010-11-16T18:28:09  <Zannick> so like the things you said in the thread, map size, number of planets, and such
2010-11-16T18:28:22  <antimatroid> iamluck: it wasn't ever stated to be an official generator :P
2010-11-16T18:28:25  <antimatroid> re read the specs
2010-11-16T18:28:38  <Error323> its all ppl had
2010-11-16T18:28:52  <antimatroid> yes, it says make your own test maps
2010-11-16T18:29:08  <mega1> the right way is to make a general enough generator
2010-11-16T18:29:18  <Error323> yes
2010-11-16T18:29:22  *** RainCT <RainCT!~RainCT@ubuntu/member/pdpc.student.rainct> has quit IRC (Remote host closed the connection)
2010-11-16T18:29:29  <Zannick> the right way is to make three generators and release them over time
2010-11-16T18:29:33  <iamluck> I'm not trying to argue, it is what it is... Why not then create the generator that makes these maps from the get go? ;) You have to agree if we had line symmetrical maps from the get go we wouldn't be having this argument.
2010-11-16T18:30:01  <antimatroid> the first generator was never meant to last that long, nobody else made a suitable new one
2010-11-16T18:30:02  <antimatroid> so i did
2010-11-16T18:30:13  <antimatroid> i had been waiting for it to change, but couldn't do anything due to exams
2010-11-16T18:30:26  <Error323> thats not the point =)
2010-11-16T18:30:59  <antimatroid> my point is if people wanted clearer rules about maps they should have done something themselves, not made assumptions :P
2010-11-16T18:31:17  <mega1> like ask for clarification?
2010-11-16T18:31:30  <Error323> antimatroid: ugh your quite stubborn :P
2010-11-16T18:31:37  <antimatroid> i am :)
2010-11-16T18:31:44  <Error323> look, the point is not us creating our own generator or sth
2010-11-16T18:31:54  <Error323> we can *ONLY* test on official or tcp
2010-11-16T18:32:00  <antimatroid> mega1: that was bad yes, not my fault, and no answer doesn't mean go and make assumptions
2010-11-16T18:32:13  <amstan_> btw... i posted something about the next contest on the mailing list, anyone want to read it?
2010-11-16T18:32:16  <Error323> so our own generators would be worth *NOTHING*
2010-11-16T18:32:22  <Zannick> most people probably won't read everything anyway
2010-11-16T18:32:23  <antimatroid> amstan: brb
2010-11-16T18:32:45  <mega1> what you still cannot appreciate that there is no way to avoid assumptions
2010-11-16T18:32:49  <iamluck> amstan: link?
2010-11-16T18:33:02  <McLeopold> antimatroid: pm
2010-11-16T18:33:03  <amstan_> iamluck: on the mailing list, i can also pastebin it
2010-11-16T18:33:08  *** amstan <amstan!~amstan@206-248-160-127.dsl.teksavvy.com> has quit IRC (Remote host closed the connection)
2010-11-16T18:33:11  *** amstan_ is now known as amstan
2010-11-16T18:33:25  <Error323> amstan: yes plz
2010-11-16T18:33:27  <iamluck> amstan: is it on the forum? but sure, pastebin it :)
2010-11-16T18:33:31  <Error323> i'm not on the list
2010-11-16T18:33:39  <Zannick> amstan: can you get me on that list? i've been interested in helping organize
2010-11-16T18:33:43  <amstan> iamluck: it's not that forum related, more about the background stuff
2010-11-16T18:33:47  <amstan> Zannick: hmm, idk
2010-11-16T18:33:57  <Zannick> all right, i'll ask jeff later
2010-11-16T18:33:59  <sigh> yes, pastebin it please :)
2010-11-16T18:34:02  <amstan> @where repo
2010-11-16T18:34:02  <contestbot> amstan: repo = http://code.google.com/p/ai-contest/
2010-11-16T18:34:25  <antimatroid> amstan: I think the most important fixes needed are rules being VERY clear at the start of the contest, like maximum dimensions of maps etc.
2010-11-16T18:34:32  <amstan> my mail: http://pastebin.com/DPdvBwYx
2010-11-16T18:34:53  <antimatroid> general is fine, in fact I prefer it, but they need to be clear, and if the rules are general, they have to be followed that generally in the final contest, so people who made assumptions get no advantage
2010-11-16T18:35:00  <amstan> antimatroid: that's more related to the actual game, i'm interested in the code and framework
2010-11-16T18:35:54  <Error323> amstan: moving to git is an excellent idea
2010-11-16T18:36:09  <amstan> i know, it's been killing me this term
2010-11-16T18:36:17  <amstan> svn is way too complicated to setup
2010-11-16T18:36:21  <bhasker> 2010-11-16 15:36:07,432-v0.1-DEBUG- Turn:135                    doTurn() - MyShips:1,EnemyShips:1555
2010-11-16T18:36:22  <bhasker> 2010-11-16 15:36:07,432-v0.1-DEBUG- Turn:135                    doTurn() - Dead nothing to do
2010-11-16T18:36:22  <bhasker> T
2010-11-16T18:36:23  <bhasker> ./cry
2010-11-16T18:36:36  <bhasker> ./mourn
2010-11-16T18:36:37  <amstan> bhasker: do fireworks mode
2010-11-16T18:36:44  <bhasker> with 1 ship?
2010-11-16T18:36:45  <amstan> bhasker: maybe you can crash the enemy bot
2010-11-16T18:36:52  <amstan> bhasker: oh.. nvm, you're screwed
2010-11-16T18:37:02  <sigh> haha
2010-11-16T18:37:19  <amstan> bhasker: do fireworks earlier
2010-11-16T18:37:49  <bhasker> hehe
2010-11-16T18:37:51  <bhasker> should try that
2010-11-16T18:38:00  <bhasker> when losing, fire off tonne of 1 size fleets
2010-11-16T18:38:04  <bhasker> at all planets
2010-11-16T18:38:11  <bhasker> that should crash most bots!
2010-11-16T18:38:14  <bhasker> including mine
2010-11-16T18:38:21  *** Mathnerd314 <Mathnerd314!~mathnerd3@128.198.97.194> has quit IRC (Disconnected by services)
2010-11-16T18:38:23  <antimatroid> amstan: would it be psosible to do 2 contests? one with a very specific map generator for people who want to write bots on very specific maps and another when absolutely any kind of map someone can come up with (but is symmetric) is used?
2010-11-16T18:38:25  <amstan> bhasker: ask voidptr for the code
2010-11-16T18:38:35  <antimatroid> not for this, but in future
2010-11-16T18:38:35  <bhasker> heh
2010-11-16T18:39:12  <amstan> antimatroid: i'm not exactly sure about the multiple contest thing
2010-11-16T18:39:13  *** Mathnerd314_ <Mathnerd314_!~mathnerd3@128.198.97.194> has joined #aichallenge
2010-11-16T18:39:18  *** Mathnerd314_ is now known as Mathnerd314
2010-11-16T18:39:52  <antimatroid> an official tcp server wouldn't be a bad idea actually
2010-11-16T18:39:57  <antimatroid> given the mess tcp servers are now
2010-11-16T18:40:16  <amstan> antimatroid: i agree
2010-11-16T18:40:21  <Error323> me 2
2010-11-16T18:40:26  <sigh> yup
2010-11-16T18:40:38  <amstan> antimatroid: or just a package for the tcp server at least
2010-11-16T18:40:48  <iamluck> antimatroid: you don't really have to force symmetry :) - have two games for non-symmetrical maps - whomever wins faster wins the map? and obviously if they win from both starting positions, they double win :)
2010-11-16T18:40:49  <amstan> that we have control over
2010-11-16T18:41:24  <Error323> @rankings 20
2010-11-16T18:41:27  <amstan> iamluck: we had this talk before, there's not much advantage over doing that, it'll require double the games to get the same info
2010-11-16T18:41:31  <contestbot> Error323: Top 20 players: bocsimacko(3656), asavis(3521), bix0r4ever(3519), davidjliu(3496), protocolocon(3476), thedreamer(3454), shangas(3451), luismi(3443), dmj111(3441), oldman(3440), _iouri_(3437), george(3426), Xrillo(3420), Accoun(3407), DarkEmi(3395), Error323(3392), felixcoto(3381), sajmon(3367), hutabhu(3358), jimrogerz(3343)
2010-11-16T18:41:33  <sigh> iamluck: currently it only matter if you win/lose not how much or how fast
2010-11-16T18:41:34  <Error323> :D
2010-11-16T18:41:50  <iamluck> I know, but with all this talk about map changes, I'm just saying you don't have to force symmetry at all :)
2010-11-16T18:42:10  <McLeopold> iamluck: I'll have those maps on my fun server soon enough
2010-11-16T18:42:23  <iamluck> regarding tcp server: it would be awesome if you had a virtualbox for it setup!
2010-11-16T18:43:51  <bhasker> hah interesting
2010-11-16T18:43:52  <bhasker> http://zeroviz.us:8080/canvas?game_id=714
2010-11-16T18:44:45  <antimatroid> amstan: at the start of contests, you guys always make it sound like n-players is possible
2010-11-16T18:45:07  <antimatroid> while I would be down for an n-player ai contest, realise that as soon as players increases, it's a completely different game as coalitions become important
2010-11-16T18:45:08  <amstan> antimatroid: well.. it is
2010-11-16T18:45:09  <amstan> lol
2010-11-16T18:45:16  <iamluck> I was considering setting up the TCP server on something like Google App Engine before. Ofcourse, tcp.exe would have to change... Not sure how feasible that is :)  Google would be a sponsor then!
2010-11-16T18:45:52  *** Palmik <Palmik!~Palmik@palmikpc.sks6.muni.cz> has quit IRC (Remote host closed the connection)
2010-11-16T18:46:32  <amstan> i'm not sure how you guys can get subscribed, the email for the newsgroups is ai-contest-discuss@googlegroups.com
2010-11-16T18:47:20  <amstan> darn
2010-11-16T18:47:29  <amstan> there was this thing i wanted to suggest too, i forgot it now
2010-11-16T18:48:08  <amstan> oh! API
2010-11-16T18:48:52  <antimatroid> how much is one allowed to contribute while remaining eligible to compete?
2010-11-16T18:49:18  <antimatroid> i know that's a bit iffy, but roughly :P
2010-11-16T18:49:45  *** eburnette <eburnette!621b3c3f@gateway/web/freenode/ip.98.27.60.63> has quit IRC (Ping timeout: 265 seconds)
2010-11-16T18:49:51  <amstan> antimatroid: contribute all you want, i don't think we have any point now for not compeeting
2010-11-16T18:50:01  <amstan> antimatroid: as long as you don't get any admin accounts you should be fine
2010-11-16T18:50:10  <amstan> we got pretty open
2010-11-16T18:50:22  <antimatroid> yeah, the openness is great
2010-11-16T18:50:23  <amstan> and getting admin accounts requires action from us.. so..
2010-11-16T18:50:54  <antimatroid> tansparency and consistency, two of my most treasured attributes :P
2010-11-16T18:51:01  <Zannick> i contributed a lot early on
2010-11-16T18:52:47  <amstan> i'm thinking having 2 official repos, one as the main one... that anyone has push access on github
2010-11-16T18:52:55  <amstan> and one as a mirror of the servers
2010-11-16T18:53:11  <amstan> that only admins will be able to pull the changes from the main repo
2010-11-16T18:53:19  <amstan> and the servers will automatically pull it
2010-11-16T18:53:40  <Error323> sounds good
2010-11-16T18:55:32  *** Migi32 <Migi32!~migi@d54C1291C.access.telenet.be> has quit IRC (Remote host closed the connection)
2010-11-16T18:55:39  *** Frontier <Frontier!~Marco_Lei@dslb-088-071-185-122.pools.arcor-ip.net> has quit IRC (Ping timeout: 276 seconds)
2010-11-16T18:56:35  *** Bobng <Bobng!~T@78.151.105.40> has quit IRC (Ping timeout: 255 seconds)
2010-11-16T18:57:07  <Error323> i'm off to bed guys
2010-11-16T18:57:10  <Error323> cya
2010-11-16T18:58:25  <McLeopold> JamesMG: wiggle fixed
2010-11-16T18:58:26  <Zannick> amstan: or branches?
2010-11-16T18:58:48  <amstan> Zannick: i don't think you can manage push access to branches on github
2010-11-16T18:58:56  <Zannick> hm
2010-11-16T18:59:19  *** Mjothitnir <Mjothitnir!432ace74@gateway/web/freenode/ip.67.42.206.116> has joined #aichallenge
2010-11-16T18:59:19  <Zannick> i don't like the idea of having two separate repos
2010-11-16T18:59:35  <Zannick> unless you mean one is main and the other pulls from that
2010-11-16T19:00:16  <amstan> Zannick: yeah.. one is main, one pulls from main
2010-11-16T19:00:20  <Zannick> okay
2010-11-16T19:00:23  <amstan> and the servers automatically pull for that one
2010-11-16T19:00:41  <Error323> amstan: btw its not necessary
2010-11-16T19:00:51  <Error323> ppl that wanna change can just fork it from github
2010-11-16T19:01:00  <Error323> and send pull requests
2010-11-16T19:01:17  <Error323> or am i missing something
2010-11-16T19:01:20  <amstan> Error323: yeah, you're right, nvm
2010-11-16T19:01:24  <amstan> screw the 2 repos thing
2010-11-16T19:01:28  <Zannick> git does support patching from people without commit access
2010-11-16T19:01:59  <amstan> then we can get CIA to yell when people commit
2010-11-16T19:02:03  * amstan kicks CIA-37
2010-11-16T19:02:05  <CIA-37> ow
2010-11-16T19:02:22  <Error323> o_O
2010-11-16T19:03:19  <Zannick> cause it isn't doing anything now, afaict
2010-11-16T19:03:21  <Zannick> :P
2010-11-16T19:03:36  *** amriedle <amriedle!~amriedle@CPE0013f7f2be12-CM0013f7f2be0e.cpe.net.cable.rogers.com> has joined #aichallenge
2010-11-16T19:04:35  <Error323> but i was goinng to bed
2010-11-16T19:04:42  <amstan> amriedle: yo
2010-11-16T19:04:52  <amstan> Zannick: he was reporting on visualizer changes
2010-11-16T19:05:42  <McLeopold> d0ugal: wtf? http://zeroviz.us:8080/canvas?game_id=793
2010-11-16T19:05:57  <McLeopold> you went for the 100 ships 1 growth ?!?
2010-11-16T19:06:10  <d0ugal> heh
2010-11-16T19:06:20  <McLeopold> I think you have a greater than sign backwards or something?
2010-11-16T19:06:23  <McLeopold> :)
2010-11-16T19:06:27  <d0ugal> I just borked my bot, its throwing errors, I should probably stop it running on the tcp
2010-11-16T19:06:28  <Zannick> and the one further from him
2010-11-16T19:06:33  <d0ugal> I actually forgot it wstill one
2010-11-16T19:06:33  <Error323> McLeopold: Active Bots (in last 5 mintues) typo there ;:)
2010-11-16T19:06:40  <McLeopold> ahhh
2010-11-16T19:06:58  <d0ugal> but hey, maybe its keeping you all amused ;)
2010-11-16T19:07:17  <McLeopold> minute?
2010-11-16T19:07:30  <Error323> yes
2010-11-16T19:07:41  <Error323> and i think you should increase it to 15 mins or sth
2010-11-16T19:08:29  <McLeopold> how about 7.5?
2010-11-16T19:08:35  <Error323> ok
2010-11-16T19:15:54  *** eburnette <eburnette!621b3c3f@gateway/web/freenode/ip.98.27.60.63> has joined #aichallenge
2010-11-16T19:16:21  <antimatroid> i wish i knew why this mxn c-colourable program wont run as i'd like
2010-11-16T19:16:28  <antimatroid> i only wanted a small break from ai, not a full day :|
2010-11-16T19:17:37  <contestbot> Top 10 players: bocsimacko(3655), oldman(3524), bix0r4ever(3516), davidjliu(3496), asavis(3491), thedreamer(3483), protocolocon(3471), shangas(3449), dmj111(3440), _iouri_(3437)
2010-11-16T19:19:01  <McLeopold> woops
2010-11-16T19:19:05  <McLeopold> visualizer down...
2010-11-16T19:19:58  <McLeopold> whew
2010-11-16T19:30:06  <mega1> McLeopold: which map set is on your tcp server?
2010-11-16T19:30:17  <mega1> the new or the even newer one?
2010-11-16T19:30:23  <McLeopold> both
2010-11-16T19:30:33  <mega1> bah
2010-11-16T19:30:37  <McLeopold> much much difference, other than the radius from 12 to 17
2010-11-16T19:30:44  <McLeopold> what would you like on there?
2010-11-16T19:31:19  <mega1> something from the final generator
2010-11-16T19:31:23  <mega1> as final as possible
2010-11-16T19:31:24  <antimatroid> amstan: "organizators" ah what? :P
2010-11-16T19:31:40  <McLeopold> okay, I'll remove the radius 12 maps
2010-11-16T19:32:33  <mega1> that small change is more than enough to make my latest improvement ineffective
2010-11-16T19:32:52  <Mjothitnir> which is currently running on the contest servrs?
2010-11-16T19:32:54  <antimatroid> did you guys decide whether it matters that the actual area in which points can be generated actually matters?
2010-11-16T19:33:00  <antimatroid> i guess ^^ that means yes
2010-11-16T19:33:30  <Zannick> yes, yes it does
2010-11-16T19:33:37  <McLeopold> okay, newmaps only
2010-11-16T19:33:39  <Zannick> i am not an organizer
2010-11-16T19:33:42  <amstan> antimatroid: hmm? what where? lol
2010-11-16T19:33:42  <McLeopold> radius 12
2010-11-16T19:33:43  <Zannick> but it does matter
2010-11-16T19:33:49  <McLeopold> wait...radius 17
2010-11-16T19:33:52  <McLeopold> (doh)
2010-11-16T19:33:54  <mega1> what was 106/38/56 between two version the "old new" maps now stands at 54/39/41 (test in progress)
2010-11-16T19:33:55  <antimatroid> "Have a read only forum for organizators, that only a select g..."
2010-11-16T19:34:08  <antimatroid> Zannick: why? I thought only maximum distance would
2010-11-16T19:34:26  <mega1> +s, +on the
2010-11-16T19:34:56  <eburnette> McLeopold how many maps on your server now? and is there a zip of them?
2010-11-16T19:35:05  <Zannick> smaller space + same number of planets (or more) = closer planets, greater probability of "overlapping" planets
2010-11-16T19:35:09  <McLeopold> 100 and I need to update the zip
2010-11-16T19:35:15  <mega1> it's only 10 days and we are still messing with the generator
2010-11-16T19:35:48  <McLeopold> keep calm and carry on
2010-11-16T19:36:34  <antimatroid> i find it hard to believe that changing max distance has much effect on a bot :\
2010-11-16T19:36:38  <McLeopold> zip file updated
2010-11-16T19:36:50  <Zannick> it shouldn't
2010-11-16T19:36:59  <Zannick> unless you count maps with heavy draw percentages
2010-11-16T19:37:08  <McLeopold> I guess it depends on the bot's style.
2010-11-16T19:37:16  <McLeopold> I does have an effect on the density of the map
2010-11-16T19:37:23  <McLeopold> with the planet count set
2010-11-16T19:37:32  <Zannick> if the admins remove those anyway, i'm not going to worry
2010-11-16T19:38:07  <eburnette> technically the contest hasn't started yet :)
2010-11-16T19:38:16  <McLeopold> http://zeroviz.us:8080/maps.zip
2010-11-16T19:39:50  <eburnette> merci
2010-11-16T19:40:12  <antimatroid> bah fuck it, c-colourable grids can wait 10 days
2010-11-16T19:40:21  *** shades <shades!~shades@cheshire.telephreak.org> has left #aichallenge
2010-11-16T19:40:32  <antimatroid> @rankings
2010-11-16T19:40:34  <contestbot> antimatroid: Top 10 players: bocsimacko(3653), thedreamer(3519), bix0r4ever(3515), davidjliu(3505), protocolocon(3476), oldman(3461), shangas(3447), dmj111(3439), _iouri_(3430), george(3423)
2010-11-16T19:40:39  <antimatroid> who the fuck is thedreamer?
2010-11-16T19:41:50  <amstan> lol
2010-11-16T19:42:01  <amstan> it would be nice if irc had hyperlinks
2010-11-16T19:42:09  <antimatroid> http://ai-contest.com/profile.php?user_id=12435
2010-11-16T19:42:15  <antimatroid> they've had only 3 valid submissions
2010-11-16T19:42:15  <amstan> then i could make all of the rankings replies to contain links to the profiles
2010-11-16T19:42:19  <antimatroid> first being on the 30th
2010-11-16T19:42:44  *** lavalamp_ <lavalamp_!180dae74@gateway/web/freenode/ip.24.13.174.116> has joined #aichallenge
2010-11-16T19:44:49  <Mjothitnir> where can one get the current generator and server maps?
2010-11-16T19:45:26  *** superflit <superflit!~superflit@140.226.49.160> has quit IRC (Quit: superflit)
2010-11-16T19:46:11  <amstan> Mjothitnir: see topic
2010-11-16T19:46:16  <amstan> server maps, idk
2010-11-16T19:46:19  <amstan> repo!
2010-11-16T19:46:22  <amstan> @repo
2010-11-16T19:46:23  <contestbot> amstan: repo = http://code.google.com/p/ai-contest/
2010-11-16T19:48:36  <Mjothitnir> I looked there for the maps (before I looked for the generator) not findingthe new ones I stoppped. :(  So I see the new generator, but the current server maps
2010-11-16T19:49:03  <antimatroid> generate your own maps?
2010-11-16T19:49:13  <antimatroid> it shouldn't matter
2010-11-16T19:49:24  <Mjothitnir> yes. I do but it matters in the short run.
2010-11-16T19:49:40  <antimatroid> meh, rankings only matter at the end
2010-11-16T19:50:17  <Mjothitnir> which is better the bot than has more wins on the servermaps or on mymaps?
2010-11-16T19:50:37  <antimatroid> if it's the same generator, it doesn't matter
2010-11-16T19:50:41  <lavalamp_> Wow, maybe first time I beat Error323: http://zeroviz.us:8080/canvas?game_id=886
2010-11-16T19:50:46  <antimatroid> the server maps now are not the ones that will be used at the end
2010-11-16T19:51:32  <antimatroid> well it would appear dimkadimon is my new rival and not sigh
2010-11-16T19:51:41  <antimatroid> sigh: no offense :P
2010-11-16T19:51:45  <Mjothitnir> oh, so they are still the newmaps (528) and not the newnewmaps(526?+)
2010-11-16T19:51:57  <Mjothitnir> 536?+
2010-11-16T19:52:17  <sigh> antimatroid: I don't think my bot is going to get much better, anyway :P
2010-11-16T19:52:23  <mega1> Mjothitnir: is that a sarcastic version number?
2010-11-16T19:53:47  <Mjothitnir> the newmaps were "map_r528_#.txt" and the generator I was pointed to is "536" so the "?+" was allowing for future revisions . . . so hmm, I guess that could be considered sarcatic. :)
2010-11-16T19:54:59  <Mjothitnir> Oh crap gotta run
2010-11-16T20:02:57  *** Rubicon-|-Cross <Rubicon-|-Cross!~RubiconCr@adsl-99-30-224-80.dsl.pltn13.sbcglobal.net> has quit IRC (Quit: Leaving)
2010-11-16T20:04:24  *** Rubicon-|-Cross <Rubicon-|-Cross!~RubiconCr@adsl-99-30-224-80.dsl.pltn13.sbcglobal.net> has joined #aichallenge
2010-11-16T20:05:03  *** amstan <amstan!~alex@129-97-120-91.uwaterloo.ca> has quit IRC (Ping timeout: 276 seconds)
2010-11-16T20:05:48  *** mceier <mceier!~mceier@chello089077208118.chello.pl> has quit IRC (Quit: leaving)
2010-11-16T20:05:49  <eburnette> nice expansion lavalamp_ on 886
2010-11-16T20:07:10  <lavalamp_> eburnette: thanks :)
2010-11-16T20:07:10  <eburnette> Not sure I should be proud of this or not: You WIN against rangzen-stupid2.1
2010-11-16T20:08:55  <eburnette> quick poll: on point symmetric map: going for the central planet very early  - good or bad?
2010-11-16T20:09:12  <McLeopold> depends
2010-11-16T20:09:30  <lavalamp_> RPS
2010-11-16T20:09:30  <McLeopold> 0 growth 100 ships or 5 growth 1 ship?
2010-11-16T20:10:12  <McLeopold> I prefer lurking near the center waiting to snipe...
2010-11-16T20:10:14  <eburnette> how about 3 growth 4 ships
2010-11-16T20:10:39  <McLeopold> hmm...  Where's the growth 5 planets, near or far?
2010-11-16T20:10:45  <McLeopold> still depends
2010-11-16T20:11:03  <McLeopold> but it is more attractive now
2010-11-16T20:11:14  <eburnette> http://98.247.248.39:8080/canvas?game_id=931
2010-11-16T20:11:18  <antimatroid> yeah, depends on your distance to it and the enemies
2010-11-16T20:11:29  <McLeopold> totaly worth it
2010-11-16T20:12:05  <eburnette> why?
2010-11-16T20:12:24  <McLeopold> it's the jump point between the two sides...
2010-11-16T20:12:38  <McLeopold> It's probably the planet that will decide who has more growth...
2010-11-16T20:13:03  <McLeopold> Just go for the 21 and 12 near you first, then hold the center
2010-11-16T20:14:47  <McLeopold> you have a lot of disconnects
2010-11-16T20:15:07  <eburnette> disconnects?
2010-11-16T20:15:20  <McLeopold> yeah, that game didn't go 200
2010-11-16T20:15:26  <McLeopold> http://zeroviz.us:8080/Houndoom.exp.html
2010-11-16T20:15:33  <McLeopold> look at all the Loss (D)
2010-11-16T20:15:41  <eburnette> Oh, I was watching it and stopped it
2010-11-16T20:15:45  <McLeopold> okay
2010-11-16T20:15:56  <McLeopold> I just wanted to make sure it wasn't a server issue.
2010-11-16T20:16:21  <lavalamp_> heh, my bot plays decent on maps like that
2010-11-16T20:17:00  *** McLeopold <McLeopold!~scott@67.139.105.186> has quit IRC (Quit: Leaving)
2010-11-16T20:17:38  <contestbot> Top 10 players: bocsimacko(3648), thedreamer(3607), bix0r4ever(3546), davidjliu(3509), protocolocon(3472), asavis(3471), shangas(3444), dmj111(3441), _iouri_(3436), george(3418)
2010-11-16T20:22:47  <mega1> final results: on new maps: 108/35/58, on new new maps: 79/60/61 (win/draw/loss)
2010-11-16T20:23:29  <mega1> *108/35/56
2010-11-16T20:23:48  <mega1> clearly, there is a significant difference
2010-11-16T20:25:05  <antimatroid> what bot is that against?
2010-11-16T20:25:21  <antimatroid> or is this on the server?
2010-11-16T20:26:54  <mega1> my new creation against my old bot
2010-11-16T20:27:25  <mega1> I'll investigate a bit more later.
2010-11-16T20:27:54  <mega1> the difference is basically that 25 wins turned into draws
2010-11-16T20:28:20  <antimatroid> so what's making the planets more drawish?
2010-11-16T20:28:21  <mega1> may not necessarily mean much because the bots are similar
2010-11-16T20:28:43  <antimatroid> there's 2 reasons i could think of
2010-11-16T20:28:48  <mega1> no idea yet, don't have the time now to look at the individual matches
2010-11-16T20:29:13  <antimatroid> 1. larger distances make first moves a bit easier
2010-11-16T20:29:13  <antimatroid> 2. larger distances make planets more likely to be snipable?
2010-11-16T20:29:15  <antimatroid> those are just guesses
2010-11-16T20:29:17  <davidjliu> .............uhoh, I'm scared
2010-11-16T20:30:11  <antimatroid> mega1: the dreamer beat you :O
2010-11-16T20:30:19  <mega1> wow
2010-11-16T20:30:28  <antimatroid> they've come from absolutely nowhere
2010-11-16T20:31:08  <mega1> I'd bet on 1
2010-11-16T20:31:54  <antimatroid> the only person i would have beat you is me :P
2010-11-16T20:31:58  <antimatroid> otherwise i'm rooting for you
2010-11-16T20:36:01  <bhasker> so i have a basic state evaluation function now
2010-11-16T20:36:05  <bhasker> now to see if its stable
2010-11-16T20:36:05  <antimatroid> i'm amazed that george is still at 10th
2010-11-16T20:36:17  <bhasker> if i am losing i would expect it to constantly go down
2010-11-16T20:36:18  <antimatroid> that submission is half a month old
2010-11-16T20:36:26  <bhasker> and if i am winning then i would expect it to go up ^
2010-11-16T20:38:00  <bhasker> k eval function is broken
2010-11-16T20:39:14  <Mjothitnir> .
2010-11-16T20:39:22  <bhasker> doh
2010-11-16T20:39:25  *** mega1 <mega1!~quassel@2001:470:1f05:548:215:58ff:fe7d:773b> has quit IRC (Read error: Operation timed out)
2010-11-16T20:39:28  <bhasker> totalled growth instead of ships
2010-11-16T20:39:30  <bhasker> !!
2010-11-16T20:41:24  <antimatroid> :p
2010-11-16T20:43:42  <davidjliu> antimatroid: shangas is still seventh and submitted around the same time as george
2010-11-16T20:45:18  *** virdo <virdo!~virdoo@bas9-london14-1279330910.dsl.bell.ca> has joined #aichallenge
2010-11-16T20:49:03  <eburnette> what's wrong with totalling growth?
2010-11-16T20:49:41  <Mjothitnir> nothing if that is what you thought you were totalling
2010-11-16T20:51:47  <Mjothitnir> um, I thought y'all were saying that the "radius" was 17 not 12...
2010-11-16T20:52:02  <antimatroid> Mjothitnir: it needs fixing on the main site
2010-11-16T20:52:25  <Mjothitnir> OK, so the "decision" is 17, but it hasn't been updated yet. ok.
2010-11-16T20:52:28  <antimatroid> the old "radius" of the map was 12, but it was square, giving a max distance of 34
2010-11-16T20:52:47  <antimatroid> whereas the new maps are generated inside a circle, so i carried over the old radius and not the max distance
2010-11-16T20:52:56  <antimatroid> but preserving the max distance is considered more important
2010-11-16T20:53:30  <burny> I'd same preserve the surface area probably, rather than max distance
2010-11-16T20:54:13  <burny> which would be a radius of 13.5
2010-11-16T20:54:40  <antimatroid> that would preserve max distance anyway
2010-11-16T20:54:51  <antimatroid> wait, no err
2010-11-16T20:54:53  <antimatroid> hmmm
2010-11-16T20:54:57  <burny> it would preserve average max distance
2010-11-16T20:55:21  <antimatroid> another option is to just throw out any points generated outside the original square
2010-11-16T20:55:49  <antimatroid> i don't see why it matters though game wise, and people seem to care most about changes to the generator
2010-11-16T20:56:28  <burny> I'm giving up on doing the math to win.. I'm going to just have a simple huristic on how good each planet is, and try to maximize my value - opponents value
2010-11-16T20:59:01  <eburnette> so what do you think makes a planet valuable?
2010-11-16T20:59:06  <burny> guessing it'll put me around 10th-2th place
2010-11-16T20:59:21  <burny> growth obviously.. primarily
2010-11-16T21:02:46  <lavalamp_> burny, what place are you now?
2010-11-16T21:09:47  *** Mathnerd314 <Mathnerd314!~mathnerd3@128.198.97.194> has quit IRC (Ping timeout: 255 seconds)
2010-11-16T21:10:39  *** phreeza_ <phreeza_!~phreeza@f049003169.adsl.alicedsl.de> has joined #aichallenge
2010-11-16T21:11:16  *** amstan <amstan!~alex@129-97-120-91.uwaterloo.ca> has joined #aichallenge
2010-11-16T21:11:16  *** ChanServ sets mode: +o amstan
2010-11-16T21:12:40  *** phreeza <phreeza!~phreeza@f048104213.adsl.alicedsl.de> has quit IRC (Ping timeout: 240 seconds)
2010-11-16T21:13:52  *** superflit <superflit!~superflit@c-24-9-162-197.hsd1.co.comcast.net> has joined #aichallenge
2010-11-16T21:16:04  <bhasker> k fixed the eval function seems to evaluate state reasonably well
2010-11-16T21:16:21  <bhasker> now to change the code to maximize value based on the eval function
2010-11-16T21:17:38  <contestbot> Top 10 players: bocsimacko(3647), bix0r4ever(3581), thedreamer(3537), davidjliu(3517), protocolocon(3477), luismi(3452), _iouri_(3447), shangas(3441), asavis(3441), dmj111(3437)
2010-11-16T21:22:45  *** Mathnerd314 <Mathnerd314!~mathnerd3@dsl.6.28.53.206.cos.dyn.pcisys.net> has joined #aichallenge
2010-11-16T21:25:03  *** davidjliu <davidjliu!473a6d0e@gateway/web/freenode/ip.71.58.109.14> has quit IRC (Quit: Page closed)
2010-11-16T21:27:48  <amstan> anyone online
2010-11-16T21:27:50  <amstan> listen up
2010-11-16T21:28:01  <amstan> i wanna move the stuff from the mailing list to a specialized forum
2010-11-16T21:28:12  <amstan> people that will have write access will be put in a group
2010-11-16T21:28:18  <amstan> i need a cool name for that group
2010-11-16T21:28:33  <eburnette> secret-forum-you-cant-read
2010-11-16T21:28:51  <javagamer> Seconded.
2010-11-16T21:31:03  <lavalamp_> wow I'm pretty happy about my Elo on McLeopold's server right now
2010-11-16T21:31:37  <lavalamp_> malazanbot has been helping me out ^^
2010-11-16T21:34:42  <amstan> it'll be read for everyone
2010-11-16T21:34:46  <amstan> but write for only a few ppl
2010-11-16T21:35:24  <antimatroid> secret-forum-you-cant-write-on
2010-11-16T21:35:29  <antimatroid> all in favour?
2010-11-16T21:35:36  <eburnette> the shut-up-and-listen forum :)
2010-11-16T21:36:20  <eburnette> intelligent-discourse
2010-11-16T21:37:05  <sigh> cant-touch-this
2010-11-16T21:37:18  <eburnette> <cry> I was doing so well until turn 37: http://98.247.248.39:8080/canvas?game_id=1125
2010-11-16T21:38:10  <eburnette> How long are games kept on zeroviz server?
2010-11-16T21:38:37  <Mjothitnir> IfWeWantedYourOpinionYou'dBeInTheGroup
2010-11-16T21:39:10  <flag> Did the IP for McLeopold's server move?
2010-11-16T21:39:20  <eburnette> +1 @ Mjothitnir
2010-11-16T21:39:45  <eburnette> 98.247.248.39 is working for me
2010-11-16T21:40:00  <flag> Thanks
2010-11-16T21:40:44  <lavalamp_> McLeopold: it would be awesome if there was a way to combine the "by maps" and "by player" screens so that you could see how well your bot does on each map... would that be difficult?
2010-11-16T21:41:08  *** davidjliu <davidjliu!473a6d0e@gateway/web/freenode/ip.71.58.109.14> has joined #aichallenge
2010-11-16T21:41:21  <lavalamp_> oh I guess he's not here
2010-11-16T21:41:26  <sigh> lavalamp_: I suspect that there would not be enough data for meaningful results?
2010-11-16T21:42:03  <lavalamp_> After you let it run 24 hours maybe there'd be 10 games on each map, that's somewhat meaningful
2010-11-16T21:42:09  <lavalamp_> esp. if you go 0-10 on some maps
2010-11-16T21:43:26  <sigh> yeah, possibly
2010-11-16T21:45:42  <davidjliu> hello all. what's up?
2010-11-16T21:47:17  <eburnette> hey david did you try the new server yet?
2010-11-16T21:47:26  <eburnette> 98.247.248.39 :995
2010-11-16T21:52:35  <davidjliu> now I have
2010-11-16T21:52:58  <eburnette> a?
2010-11-16T21:53:28  <davidjliu> I'm myself... maybe I haven't appeared yet
2010-11-16T21:53:39  <davidjliu> check again
2010-11-16T21:53:57  <eburnette> oh somebody named a just beat me
2010-11-16T21:54:06  <flag> That would be I
2010-11-16T21:54:37  <flag> Which one are you?
2010-11-16T21:54:53  <eburnette> Houndoom
2010-11-16T21:55:26  <flag> ah
2010-11-16T21:55:54  <flag> Do you know what the extra numbers on the planets are in the game viewer?
2010-11-16T21:56:02  <eburnette> http://98.247.248.39:8080/canvas?game_id=1189 where do you think I went wrong?
2010-11-16T21:56:10  <eburnette> planet id, growth
2010-11-16T21:56:18  <flag> Oh, id
2010-11-16T21:57:06  <flag> As for going wrong, you let me expand too much
2010-11-16T21:57:17  <flag> You had a 100 ship lead at one point, you probably could've just attacked
2010-11-16T21:57:24  <flag> Around 36 turns
2010-11-16T21:59:50  <eburnette> yeah i'm having a hard time getting that 'just attack' tipping point nailed down
2010-11-16T22:00:25  <burny> the problem I expect to run into with my AI.. is going to be that it determins that stackign up 200ships beside a planet, is a real good idea, because it can attack that planet.. and that actually attackingthe planet isn't all that good of an idea
2010-11-16T22:01:53  <eburnette> Next contest there should be a '50 repeat moves = a draw' rule
2010-11-16T22:02:09  <antimatroid> my biggest problem is working out what ships i should save for potential enemy attacks
2010-11-16T22:02:21  <javagamer> How come my compiler didn't throw any errors when I had a function that was supposed to return something, but forgot the return statement?
2010-11-16T22:02:34  <burny> default return 0?
2010-11-16T22:03:48  <davidjliu> heh, apparently McLeopold is my bane
2010-11-16T22:03:50  <eburnette> Ok is there any way this game could have been won instead of drawn? http://98.247.248.39:8080/canvas?game_id=1231
2010-11-16T22:04:44  <davidjliu> wait one turn
2010-11-16T22:04:51  <davidjliu> and then go for the +4 that has 12 ships on it
2010-11-16T22:04:54  <davidjliu> (does it have 12 ships on it?)
2010-11-16T22:05:42  <davidjliu> specifically, if the enemy tries to take the 16 1-growth, you take the 12 with +4, and if he doesn't, then it'll be a draw
2010-11-16T22:06:24  <davidjliu> so it's still a draw, but in a different way...
2010-11-16T22:06:30  <eburnette> P 20.52377 10.092973 0 12 4
2010-11-16T22:07:39  <davidjliu> I just tested my bot on the map and it just sits there on the first turn
2010-11-16T22:14:06  <lavalamp_> eburnette: I have games where 50 repeat moves is the right thing to do while I slowly drain away the other guys forces until my superior production gives me enough ships to take over
2010-11-16T22:14:53  <eburnette> how about 50 repeats with identical production
2010-11-16T22:15:52  <lavalamp_> mmmmm maybe
2010-11-16T22:17:38  <contestbot> Top 10 players: bocsimacko(3635), bix0r4ever(3582), davidjliu(3501), asavis(3496), protocolocon(3482), shangas(3437), dmj111(3434), _iouri_(3430), thedreamer(3429), george(3418)
2010-11-16T22:22:00  <antimatroid> eburnette: and ship count
2010-11-16T22:22:09  <antimatroid> if you expanded to stalemate better than the other player, you ought to win
2010-11-16T22:22:41  <eburnette> sure
2010-11-16T22:25:30  <burny> more.. if you expand to paper-rock-scissors, better
2010-11-16T22:35:17  <burny> if I think that attacking planet A wins.. and you think A > B > C > A... then whichever oen of us is right wins
2010-11-16T22:36:56  *** McLeopold <McLeopold!~scott@c-98-247-248-39.hsd1.wa.comcast.net> has joined #aichallenge
2010-11-16T22:37:04  *** rebelxt_ <rebelxt_!181c0b18@gateway/web/freenode/ip.24.28.11.24> has joined #aichallenge
2010-11-16T22:39:29  *** phxtrack <phxtrack!phxtrack@ip24-56-47-107.ph.ph.cox.net> has quit IRC (Ping timeout: 240 seconds)
2010-11-16T22:46:22  <rebelxt_> lavalamp_: ping
2010-11-16T22:46:32  <lavalamp_> hm?
2010-11-16T22:47:11  <lavalamp_> rebelxt_ I'm around
2010-11-16T22:47:28  <rebelxt_> downloaded 200 maps you generated today. it looks like js visualizer is broken though. did you have the same issue?
2010-11-16T22:48:10  <rebelxt_> maps from maps_rad_17.zip
2010-11-16T22:48:31  *** fmeyer <fmeyer!~fmeyer@189.100.228.254> has joined #aichallenge
2010-11-16T22:48:32  <lavalamp_> dunno, I'm using jbotmanager
2010-11-16T22:48:45  <rebelxt_> i.e. visualizer is not showing some planets
2010-11-16T22:49:01  <lavalamp_> visualizer might be busted as they're made in a circle and not a square anymore...
2010-11-16T22:49:10  <lavalamp_> I'm not sure if that will be changed or no
2010-11-16T22:49:12  <rebelxt_> ahh ok. i think the issue is radius 17
2010-11-16T22:49:23  <lavalamp_> antimatroid probably knows more if he's around
2010-11-16T22:50:21  <rebelxt_> i'm using danielvf's visualizer. https://github.com/DanielVF/Planet-Wars-Canvas-Visualizer
2010-11-16T22:50:51  *** fmeyer <fmeyer!~fmeyer@189.100.228.254> has quit IRC (Client Quit)
2010-11-16T22:51:08  <eburnette> it still has to fit in a square
2010-11-16T22:51:11  *** fmeyer <fmeyer!~fmeyer@189.100.228.254> has joined #aichallenge
2010-11-16T22:51:19  *** phxtrack <phxtrack!phxtrack@ip24-56-47-107.ph.ph.cox.net> has joined #aichallenge
2010-11-16T22:53:05  <rebelxt_> having looked at the map files, some planets have x and y coordinates over 34. is this wrong?
2010-11-16T22:56:30  <lavalamp_> I'm not sure, I just ran the script... doesn't sound right to me though
2010-11-16T22:57:31  *** amstan <amstan!~alex@129-97-120-91.uwaterloo.ca> has quit IRC (Ping timeout: 265 seconds)
2010-11-16T22:57:57  <rebelxt_> antimatroid: does this ring a bell?
2010-11-16T23:01:00  <antimatroid> rebelxt_: maps are generated in a circle of radius 17 around the origin, then translated so that all coordinates are positive
2010-11-16T23:01:08  <antimatroid> so they can happen above 34
2010-11-16T23:01:41  <antimatroid> how is the visualiser broken? some maps aren't displayed? then yeah the visualiser is giving a display area smaller than the new possible area
2010-11-16T23:01:47  <antimatroid> some planets**
2010-11-16T23:01:54  <lavalamp_> wait how does that math work
2010-11-16T23:02:08  <lavalamp_> oh positive and != 0
2010-11-16T23:02:17  <lavalamp_> is that what you mean?
2010-11-16T23:02:36  <rebelxt_> one thing to look into is to check if visualizer.js can handle such maps. danielvf's visualizer can't without a patch, and it appears that ai-contest is using a modified version of daniel's visualizer
2010-11-16T23:02:43  <pgpaskar> does the java visualizer work at all?
2010-11-16T23:02:55  <pgpaskar> just curious, that code is prehistoric and bad at this point.
2010-11-16T23:03:01  <rebelxt_> antimatroid: it's not showing some planets (cropping)
2010-11-16T23:03:33  <rebelxt_> pgpaskar: i don't even have java viz any more.. it's so bad
2010-11-16T23:03:43  <pgpaskar> yup
2010-11-16T23:04:19  <pgpaskar> it was hacked together by a moron (me) in a day based on someone elses' code (jeff's) and never maintained
2010-11-16T23:04:27  <pgpaskar> i call it the student development cycle
2010-11-16T23:04:56  <rebelxt_> oops ;-)
2010-11-16T23:05:38  <pgpaskar> meh
2010-11-16T23:05:49  <rebelxt_> let me put it this way: java viz served us well until more light weight options showed up
2010-11-16T23:05:49  <pgpaskar> so whats all this talk of map generator broken?
2010-11-16T23:06:07  <pgpaskar> Hey; I'll have you know that the java visualizer uses less resources and less memory than the canvas one
2010-11-16T23:06:20  <pgpaskar> That is, before jeff did something to the animation
2010-11-16T23:06:23  <pgpaskar> I'm not sure what it does now
2010-11-16T23:07:41  <rebelxt_> canvas viz is resource intensive.. but what's also important to me is start-up time (i.e. chrome + canvas viz)
2010-11-16T23:07:47  *** sigh <sigh!~sigh@lithium.mailguard.com.au> has quit IRC (Read error: No route to host)
2010-11-16T23:08:11  <pgpaskar> Applet takes a bit to load the first time, decent browsers and JVMs cache it for subsequent uses
2010-11-16T23:08:49  <pgpaskar> I know this because writing my first applet before knowing this was a bitch because none of my changes would take effect until both jvm and firefox were killed
2010-11-16T23:09:20  <rebelxt_> i understand. again, i did not mean to say java one is bad per se ;) there are so many options these days.. java, canvas, jbotmanager, albert's stl player etc
2010-11-16T23:09:41  <pgpaskar> oh I know its bad
2010-11-16T23:10:40  <pgpaskar> But those complaints were kind of geared toward java as a langauge and the jvm, it seemed, which is stereotyped as being slow and bulky when it really isn't /that/ slow and bulky
2010-11-16T23:11:01  <pgpaskar> annnnnywho
2010-11-16T23:11:36  <pgpaskar> Whoa a lisp bot is slappin the current rankings
2010-11-16T23:11:37  <pgpaskar> awesome
2010-11-16T23:13:29  <rebelxt_> yeah. pretty cool. sbcl lisp is actually pretty fast
2010-11-16T23:14:12  *** sigh <sigh!~sigh@lithium.mailguard.com.au> has joined #aichallenge
2010-11-16T23:14:26  <rebelxt_> btw, i agree about java - it's pretty good for this contest.. "only" 2-3 times slower than c++.. unlike python which is 40 times slower than the same code in c++
2010-11-16T23:15:47  <antimatroid> is there anything i'm meant to be worrying about atm? or some visualiers just aren't displaying all planets?
2010-11-16T23:15:55  <antimatroid> sorry, i've just been cooking lunch :)
2010-11-16T23:17:25  <pgpaskar> i like your timezone
2010-11-16T23:17:28  <pgpaskar> i should go to sleep
2010-11-16T23:17:29  <pgpaskar> toodles
2010-11-16T23:17:37  <contestbot> Top 10 players: bocsimacko(3639), bix0r4ever(3610), protocolocon(3490), asavis(3487), davidjliu(3484), dmj111(3441), shangas(3435), _iouri_(3430), george(3417), Xrillo(3417)
2010-11-16T23:17:42  <antimatroid> i like australia more than just the time zone :P
2010-11-16T23:17:42  <antimatroid> night
2010-11-16T23:17:59  <antimatroid> we even get summer over christmas/new years :P
2010-11-16T23:18:32  <eburnette> white christmas rulez
2010-11-16T23:18:46  <antimatroid> nah, piss up christmas rules
2010-11-16T23:18:48  <sigh> still managed to have a white christmas two years ago
2010-11-16T23:18:53  <sigh> hail
2010-11-16T23:18:54  <antimatroid> sigh: :|
2010-11-16T23:19:12  <rebelxt_> antimatroid: i guess we just need to let janzert know there might be a problem with visualizer handling radius 17 maps.. worth looking into this so that we all don't get surprised if new R17 maps are used
2010-11-16T23:19:25  *** krokkrok <krokkrok!4c58a69c@gateway/web/freenode/ip.76.88.166.156> has joined #aichallenge
2010-11-16T23:19:32  <antimatroid> there was a time here when a politician made a sarcastic comment saying, sure if there's snow on the parliament lawns tomorrow morning we'll do it (that never happens) and it happened
2010-11-16T23:19:45  <antimatroid> yeah sure, but leave janzert alone atm
2010-11-16T23:19:47  <antimatroid> he's out of town
2010-11-16T23:20:05  <antimatroid> he already does craploads
2010-11-16T23:20:31  <sigh> is this the js visualiser?
2010-11-16T23:20:47  <antimatroid> i can't believe people can stand to still use it
2010-11-16T23:21:11  <antimatroid> sigh: i've reverted back to trying minimax
2010-11-16T23:21:30  <antimatroid> i think i have a solution to avoid my previous complications, and just heuristics sucked at combat
2010-11-16T23:21:50  <sigh> cool, good luck
2010-11-16T23:22:16  <antimatroid> i'm quickly running out of time though :|
2010-11-16T23:22:18  <eburnette> I'm doing minimax but my eval function sucks
2010-11-16T23:22:34  <sigh> eburnette: what depth and branching factor?
2010-11-16T23:22:44  <antimatroid> eburnette: are you just sending ships from your planets at the current point in time? or into the future?
2010-11-16T23:22:57  <antimatroid> and if the later, how are you handling players possibly rendering those moves useless?
2010-11-16T23:23:18  <antimatroid> i've decided to try and set it up so that i only actually need to send fleets from now at each layer but considers moves from all time points
2010-11-16T23:23:20  <eburnette> 9 plys if time allows (ply = 1/2 turn)
2010-11-16T23:23:39  <sigh> o_O
2010-11-16T23:23:39  <eburnette> 0..15 fanout typically
2010-11-16T23:23:47  *** mega1 <mega1!~quassel@adsl-63-195-37-158.dsl.snfc21.pacbell.net> has joined #aichallenge
2010-11-16T23:23:52  <sigh> ah ok, you've got a fairly small set
2010-11-16T23:25:23  <eburnette> and future moves. I try really hard to make future moves work but if they don't then I bail on the latter parts
2010-11-16T23:28:17  <eburnette> It's pretty when it works, it even does future moves from planets that aren't captured yet
2010-11-16T23:28:24  <antimatroid> the other thing that is hard is having them expand, but not trying to snipe their "potential expansions"
2010-11-16T23:28:40  <antimatroid> well, not with your actual moves anyway
2010-11-16T23:30:27  <rebelxt_> eburnette: cool. i just got this to work as well (scheduling fleets from to-be captured planets).. a minor boost
2010-11-16T23:31:23  <eburnette> If I could just get my evaluate function to reflect reality I could do much better.
2010-11-16T23:32:46  <rebelxt_> so tricky to take positional advantage into account..
2010-11-16T23:32:51  <McLeopold> rebelxt_: the visualizer handles radius just fine
2010-11-16T23:32:58  <McLeopold> radius 17 that is
2010-11-16T23:33:20  <rebelxt_> McLeopold: which viz are you using? danielvf's from github?
2010-11-16T23:33:32  <McLeopold> no, the one from ai-contest
2010-11-16T23:33:42  <McLeopold> danielvf's doesn't handle it
2010-11-16T23:34:06  <rebelxt_> McLeopold: ahh i should switch then.. i was using dvf's original one. thanks
2010-11-16T23:35:05  <lavalamp_> antimatroid: i do the latter of your two options... my preditiction object recognizes and appropriately cancels conflicting orders
2010-11-16T23:35:34  <lavalamp_> ...and yes that was just as hard as it sounds...
2010-11-16T23:36:15  *** amstan <amstan!~amstan@206-248-160-127.dsl.teksavvy.com> has joined #aichallenge
2010-11-16T23:36:15  *** ChanServ sets mode: +o amstan
2010-11-16T23:37:53  <antimatroid> lavalamp_: i've decided to try only sending current ships at each layer, then reassigining all future ones each turn
2010-11-16T23:38:06  <antimatroid> so the player can re-evalutate entire moves dependent on what they do
2010-11-16T23:38:11  <lavalamp_> that's probably more efficient
2010-11-16T23:38:34  <lavalamp_> I wanted to have re-evaluation but I don't have time to write it
2010-11-16T23:38:38  <antimatroid> that's the plan, and hopefully doesn't result in missing potential good moves just cause they fucked one way of doing it up
2010-11-16T23:38:42  <antimatroid> but i'll see
2010-11-16T23:38:49  <lavalamp_> haha
2010-11-16T23:38:56  <davidjliu> McLeopold: could you add a column for the percentage of games that reach 200 turns for each map?
2010-11-16T23:39:23  <McLeopold> maybe
2010-11-16T23:39:34  <davidjliu> for a map like this, perhaps:
2010-11-16T23:39:35  <davidjliu> http://zeroviz.us:8080/newmap17.txt.html
2010-11-16T23:46:41  <eburnette> I'm going to bring out a bunch of my old bots to see if they fare any better overnight
2010-11-16T23:50:58  <McLeopold> davidjliu: 200 turns stats are there
2010-11-16T23:51:46  *** janzert <janzert!~Brian@97-83-37-102.dhcp.trcy.mi.charter.com> has joined #aichallenge
2010-11-16T23:52:01  * janzert waves
2010-11-16T23:52:49  * sigh particles
2010-11-16T23:53:14  <eburnette> Hmm... Infernape seems to like these maps
2010-11-16T23:54:07  <McLeopold> Looks like bhasker goes 200 turns more than anyone
2010-11-16T23:55:25  <amstan> janzert: hey
2010-11-16T23:55:29  <amstan> janzert: k.. check that email there
2010-11-16T23:55:55  <amstan> janzert: i want to delete those custom groups, make trusted users mods, and have this one forum that only mods+admins may post in
2010-11-16T23:56:40  <janzert> I do like email better, but having everything in one spot is probably good
2010-11-16T23:58:16  <amstan> janzert: you can set email notifications and stuch
2010-11-16T23:58:41  <amstan> janzert: did i miss anyone? http://ai-contest.com/forum/adm/index.php?i=groups
2010-11-16T23:59:17  <amstan> janzert: so like.. that's what we need, move everyone from the mailing list in here
2010-11-16T23:59:24  <amstan> janzert: people that want to talk in this
2010-11-16T23:59:52  <Accoun> McLeopold> u use modufied http://code.google.com/p/ai-contest/source/browse/trunk/planet_wars/backend/map_generator.py#  ?