2010-11-02T00:10:03 *** hellman has joined #aichallenge 2010-11-02T00:13:16 Top 10 players: bocsimacko(4346), dmj111(4264), Hazard(4187), bix0r4ever(4158), shangas(4154), FlameN(4138), george(4107), joakim_mjardner(4083), felixcoto(4074), Accoun(4070) 2010-11-02T00:15:55 *** demerzel_ has quit IRC (Ping timeout: 265 seconds) 2010-11-02T00:20:12 *** AlliedEnvy has joined #aichallenge 2010-11-02T00:39:19 *** JamesGecko has joined #aichallenge 2010-11-02T00:39:33 *** AlliedEnvy has quit IRC (Remote host closed the connection) 2010-11-02T00:41:16 *** AlliedEnvy has joined #aichallenge 2010-11-02T00:52:30 *** superflit has joined #aichallenge 2010-11-02T01:06:46 *** Azrathud has quit IRC (Ping timeout: 265 seconds) 2010-11-02T01:07:49 *** airhorns has joined #aichallenge 2010-11-02T01:10:07 *** hornairs has quit IRC (Ping timeout: 245 seconds) 2010-11-02T01:11:43 McLeopold: close match 2010-11-02T01:12:32 *** AlliedEnvy has joined #aichallenge 2010-11-02T01:13:02 this is one of the closest games i've watched 2010-11-02T01:13:16 Top 10 players: bocsimacko(4329), dmj111(4245), Hazard(4169), shangas(4146), FlameN(4125), george(4098), joakim_mjardner(4071), Accoun(4054), felixcoto(4052), rsergio(4045) 2010-11-02T01:14:15 *** Frontier has quit IRC (Ping timeout: 240 seconds) 2010-11-02T01:14:36 aww, i think i timed out 2010-11-02T01:14:43 and i had the win on ships coming 2010-11-02T01:15:11 *** JamesGecko_ has joined #aichallenge 2010-11-02T01:15:32 *** JamesGecko has quit IRC (Ping timeout: 245 seconds) 2010-11-02T01:15:49 http://72.44.46.68/canvas?game_id=816248 2010-11-02T01:15:54 interesting match anyway 2010-11-02T01:21:51 heh, suprisingly even 2010-11-02T01:22:26 I think I'm going to give up on minimax 2010-11-02T01:22:46 any other ideas? :P 2010-11-02T01:23:25 lol, so I resubmit the same bot on the official server, and it gets way worse 2010-11-02T01:23:27 dunno 2010-11-02T01:23:36 *** AlliedEnvy has quit IRC (Remote host closed the connection) 2010-11-02T01:23:41 davidjliu: why did you do that? 2010-11-02T01:23:50 antimatroid: did we just play? 2010-11-02T01:24:02 well, I submitted a different bot. And I realized that it actually sucked. So I resubmitted my original one 2010-11-02T01:24:07 yeah, link is above 2010-11-02T01:24:14 ahh 2010-11-02T01:24:16 i lost on timeout, i had you, it was a close match though 2010-11-02T01:24:22 *** AlliedEnvy has joined #aichallenge 2010-11-02T01:24:34 oh, that's my old one 2010-11-02T01:24:46 I'm running the latest greatest AntBot right now. 2010-11-02T01:24:52 It's still very much broken. 2010-11-02T01:25:08 Like, really really broken. 2010-11-02T01:25:22 But it does do some cool things sometimes. 2010-11-02T01:25:32 yeah, mines much the same i think 2010-11-02T01:26:07 I think I turned off my forwarding somehow 2010-11-02T01:27:50 That was one of the more interesting matches I've seen. 2010-11-02T01:32:06 that's what i thought 2010-11-02T01:32:34 http://72.44.46.68/canvas?game_id=816411 2010-11-02T01:32:46 watch me go for the 37 up top 2010-11-02T01:33:10 that game started out great 2010-11-02T01:33:22 but I sat on my ships in the back 2010-11-02T01:33:48 it's the boscimacko trick at it's best 2010-11-02T01:33:50 gtg 2010-11-02T01:34:08 cya 2010-11-02T01:36:34 *** McLeopold has quit IRC (Quit: Leaving) 2010-11-02T01:39:28 *** AlliedEnvy has quit IRC (Remote host closed the connection) 2010-11-02T01:40:54 *** virdo has quit IRC (Ping timeout: 252 seconds) 2010-11-02T01:53:04 *** Palmik has joined #aichallenge 2010-11-02T02:00:07 *** JamesGecko_ has quit IRC (Ping timeout: 245 seconds) 2010-11-02T02:10:40 *** superflit has quit IRC (Ping timeout: 240 seconds) 2010-11-02T02:12:41 *** davidjliu has quit IRC (Quit: Page closed) 2010-11-02T02:13:16 Top 10 players: bocsimacko(4330), dmj111(4253), shangas(4163), Hazard(4162), george(4122), FlameN(4115), joakim_mjardner(4071), Accoun(4063), rsergio(4056), felixcoto(4047) 2010-11-02T02:13:47 *** superflit has joined #aichallenge 2010-11-02T02:16:09 *** the_graid has joined #aichallenge 2010-11-02T02:17:46 *** the_graid has quit IRC (Client Quit) 2010-11-02T02:21:07 *** jhawthorn has joined #aichallenge 2010-11-02T02:32:57 *** hornairs has joined #aichallenge 2010-11-02T02:34:15 *** Itkovian has joined #aichallenge 2010-11-02T02:35:16 *** airhorns has quit IRC (Ping timeout: 276 seconds) 2010-11-02T02:41:25 *** Itkovian has quit IRC (Quit: Itkovian) 2010-11-02T02:42:55 *** fmeyer has quit IRC (Ping timeout: 240 seconds) 2010-11-02T02:45:04 amstan: you there? I need your personal intransitive verb for feelings of great joy and happiness (e.g. rejoice) :) 2010-11-02T02:45:23 intrasitive? 2010-11-02T02:46:07 rainbow! 2010-11-02T02:46:51 hmm, rainbow! doesn't quite work for what I need. It needs to be express a feeling as in "Amstan can rejoice" 2010-11-02T02:47:13 i'm not sure what you're seeking 2010-11-02T02:47:14 are you saying that amstan cannot rainbow!? 2010-11-02T02:47:20 i can rainbow: 2010-11-02T02:47:26 @yay 2010-11-02T02:47:26 Double rainbow all the way! 2010-11-02T02:48:03 ok, just sec and I suppose I can edit it. :) 2010-11-02T02:48:56 a close second to rainbow! is jubilate 2010-11-02T02:49:27 amstan is having a gay time? :) 2010-11-02T02:49:59 * antimatroid hopes those ice creams aren't just in australia 2010-11-02T02:49:59 @rainbow truth is out there 2010-11-02T02:49:59 rabidus: truth is out there 2010-11-02T02:50:48 antimatroid: it is *golden* gay time 2010-11-02T02:50:49 oh god, i now have my first vector of priority queues of moves 2010-11-02T02:50:50 get it right 2010-11-02T02:54:51 amstan: take a look in the announcement forum 2010-11-02T02:55:22 janzert: yay 2010-11-02T02:55:39 janzert: how do they add new ones though? 2010-11-02T02:55:47 still need to add new organizations manually though 2010-11-02T02:56:09 but it should hopefully cut down on the requests for changes at least 2010-11-02T02:56:19 *** CIA-38 has quit IRC (Ping timeout: 272 seconds) 2010-11-02T02:56:19 ok, cool 2010-11-02T02:57:15 janzert: can you change the names of current organisatins? 2010-11-02T02:57:33 i would like (Australia) removed off University of Tasmania if possible, but meh if it's lots of effort 2010-11-02T02:57:58 anyone with database access can manually of course, but yeah I'm trying to avoid getting into that in general 2010-11-02T02:58:21 amstan? :P 2010-11-02T02:58:33 I'll let amstan stick to editing the user information in the database and I'll stick to the code :/ 2010-11-02T02:58:42 yeah, i understand ;) 2010-11-02T03:00:11 antimatroid: seceding from Australia, are we? 2010-11-02T03:01:36 sigh: last contest universities were listed under countries (I think), so my forum post had (Australia) there, which then got tacked onto the organisation name 2010-11-02T03:01:42 i didn't actually intend for it to be there 2010-11-02T03:02:42 fair enough 2010-11-02T03:02:57 I had assumed that was there because no one knows where Tasmania is 2010-11-02T03:03:30 yeah, i don't know where half the places are on earth 2010-11-02T03:03:34 i'm terrible with that kind of thing 2010-11-02T03:03:39 :) 2010-11-02T03:04:10 you could show me an unamed map of europe and i wouldn't even be able to point out uk 2010-11-02T03:05:04 well, I keep forgetting the difference between english, uk, great britain, etc 2010-11-02T03:05:08 *england 2010-11-02T03:05:44 yeah so do i, occasionally i see some post somewhere of someone furiously explaining it, but it's usually gone again not long after 2010-11-02T03:06:12 when I need to, I just look up this venn diagram: http://qntm.org/files/uk/uk.gif 2010-11-02T03:08:03 lol, so retarded 2010-11-02T03:13:17 Top 10 players: bocsimacko(4331), dmj111(4253), shangas(4165), Hazard(4165), george(4125), FlameN(4118), joakim_mjardner(4072), Accoun(4071), rsergio(4058), GreenTea(4055) 2010-11-02T03:13:38 http://72.44.46.68/getplayer?player=hahahahahaha 2010-11-02T03:13:40 i laughed 2010-11-02T03:15:57 what's special about that account? 2010-11-02T03:17:03 the name 2010-11-02T03:17:15 it helps that they're doing reasonably well 2010-11-02T03:17:57 *** hellman has quit IRC (Remote host closed the connection) 2010-11-02T03:32:00 YYYYYYYYYYYYYYYYYYYEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAHH 2010-11-02T03:32:03 12th 2010-11-02T03:32:54 *** jhawthorn has left #aichallenge 2010-11-02T03:51:26 Error323: congrats :D 2010-11-02T04:03:46 thanks :) 2010-11-02T04:03:59 whoah XD 2010-11-02T04:08:12 it sucks when your bot wont even make moves consistently 2010-11-02T04:13:18 Top 10 players: bocsimacko(4330), dmj111(4254), shangas(4156), Hazard(4153), george(4134), FlameN(4128), Accoun(4073), joakim_mjardner(4073), rsergio(4056), felixcoto(4055) 2010-11-02T04:13:23 *** narnach has quit IRC (Quit: Leaving.) 2010-11-02T04:14:04 *** narnach has joined #aichallenge 2010-11-02T04:14:04 *** narnach is now known as narnach_mb 2010-11-02T04:21:30 mine probably won't because it has rand() 2010-11-02T04:21:32 lol 2010-11-02T04:22:06 *** Rubicon-|-Cross has quit IRC (Quit: Leaving) 2010-11-02T04:22:28 so does mine, but i don't expect it to use it very often 2010-11-02T04:22:40 *** Rubicon-|-Cross has joined #aichallenge 2010-11-02T04:23:42 *** Itkovian has joined #aichallenge 2010-11-02T04:27:43 *** perror has joined #aichallenge 2010-11-02T04:35:43 *** Florian95 has joined #aichallenge 2010-11-02T04:38:02 *** aerique has joined #aichallenge 2010-11-02T04:39:21 Mine uses rand, but I seed it with the same value every time. 2010-11-02T04:43:02 why the *bleep* would one use rand here? 2010-11-02T04:46:04 many reasons, one would be the reason one would use rand() in rock-paper-scissors 2010-11-02T04:46:20 another reason is that sometimes randomised algorithms can give better runtime 2010-11-02T04:47:04 *** perestrelka has quit IRC (Ping timeout: 276 seconds) 2010-11-02T04:47:38 mmm, indeed ur right 2010-11-02T04:48:20 im so surprised my code actually works without segfaulting :D 2010-11-02T04:48:28 lol 2010-11-02T04:48:30 now i need the bot to do something 2010-11-02T04:48:38 haha 2010-11-02T04:48:41 xD 2010-11-02T04:48:45 well it runs, it can actually play 2010-11-02T04:48:47 but it stays at home 2010-11-02T04:48:48 currently it writes "hello world" to file? :x 2010-11-02T04:48:49 :P 2010-11-02T04:49:04 heh i seee 2010-11-02T04:49:06 yep it does "hello world" in about 2000 lines of code 2010-11-02T04:49:08 :P 2010-11-02T04:49:21 2000 o_O 2010-11-02T04:49:35 xD 2010-11-02T04:49:42 ah, you were the c breather =) 2010-11-02T04:49:46 ^^ 2010-11-02T04:49:47 that explains 2010-11-02T04:50:01 i hope all the added code actually is worthwhile 2010-11-02T04:50:03 otherwise ill cry 2010-11-02T04:50:30 the code is pretty though 2010-11-02T04:50:35 spend a day refactoring/improving yesterday only to find out it had regressed 2010-11-02T04:50:39 lol 2010-11-02T04:50:46 but then i found a bug 2010-11-02T04:50:51 fixed that 2010-11-02T04:50:56 atm, im sending fleets to itself 2010-11-02T04:50:58 and it became much better ^^ 2010-11-02T04:51:05 which isn't very bright of it 2010-11-02T04:51:10 so planet 1 to planet 1 2010-11-02T04:51:12 :/ 2010-11-02T04:51:14 lol 2010-11-02T04:52:45 *** jaspervdj has joined #aichallenge 2010-11-02T04:56:23 *** CIA-36 has joined #aichallenge 2010-11-02T04:59:10 oh no, its the CIA 2010-11-02T04:59:30 oh no 2010-11-02T05:07:18 yep, when i'm indifferent between two strategyes, i if(getNum(0,1)) to see if i want to use the newer one or not 2010-11-02T05:07:32 and if i end up using mixed strategies i'll need them to implement them 2010-11-02T05:07:52 mm 2010-11-02T05:08:12 i should probably take that out for the time being 2010-11-02T05:08:17 and they all perform equally well? 2010-11-02T05:08:23 i also have the problem that it doesn't always generate the tree the same length 2010-11-02T05:08:29 oh srry misread 2010-11-02T05:08:39 :p i was going to say 2010-11-02T05:08:48 how is the gametree comming along 2010-11-02T05:09:01 i haven't even got a game tree anymore 2010-11-02T05:09:07 hehe 2010-11-02T05:09:13 2 decision trees and a zero sum simultaneous game 2010-11-02T05:12:16 *** perestrelka has joined #aichallenge 2010-11-02T05:13:17 Top 10 players: bocsimacko(4329), dmj111(4247), shangas(4146), Hazard(4142), george(4131), FlameN(4109), Accoun(4069), joakim_mjardner(4064), felixcoto(4048), rsergio(4039) 2010-11-02T05:44:05 *** Commandir has joined #aichallenge 2010-11-02T05:46:13 *** shades has quit IRC (Ping timeout: 276 seconds) 2010-11-02T05:46:18 *** shades has joined #aichallenge 2010-11-02T05:47:34 *** wh1teside_ has joined #aichallenge 2010-11-02T05:49:06 *** adu has quit IRC (Quit: adu) 2010-11-02T05:49:28 *** Naktibalda has joined #aichallenge 2010-11-02T05:50:37 *** wh1teside has quit IRC (Ping timeout: 252 seconds) 2010-11-02T05:55:16 mega1: looks like my bot decided not to even bother tyring. 2010-11-02T05:55:56 dmj111: where? It's pretty good on the official server. 2010-11-02T05:56:27 and it has stopped a long time ago on tcp 2010-11-02T05:56:29 *** perror has quit IRC (Quit: Bye all !) 2010-11-02T05:56:38 http://ai-contest.com/visualizer.php?game_id=6494422 2010-11-02T05:56:48 *** perror has joined #aichallenge 2010-11-02T05:57:24 i don't get many games with you... I hacked it. "if bocsimacko then give up!!" 2010-11-02T05:57:26 bah, this has happend to me once as well 2010-11-02T05:57:48 Gonna have to look and see if it is a legitimate crash though. Do some of these maps look new? 2010-11-02T05:57:50 that server sucks 2010-11-02T05:58:16 dunno 2010-11-02T05:59:25 *** Utkarsh has quit IRC (Ping timeout: 252 seconds) 2010-11-02T06:01:48 hahah 2010-11-02T06:03:19 *** hellman has joined #aichallenge 2010-11-02T06:03:55 *** Olathe has quit IRC (Ping timeout: 265 seconds) 2010-11-02T06:04:22 lol 2010-11-02T06:05:43 *** Utkarsh has joined #aichallenge 2010-11-02T06:06:06 wow looking at mega1 's recent games, he just swarms over the enemy 2010-11-02T06:06:09 they haven't a chance 2010-11-02T06:06:11 lol 2010-11-02T06:08:07 and the ones mega1 does lose... they seem pretty boring - looks like you can fix those? 2010-11-02T06:08:12 *** Olathe has joined #aichallenge 2010-11-02T06:08:49 not without screwing up the rest so far 2010-11-02T06:09:24 oh ok 2010-11-02T06:09:39 i think keep it as is then, it does by far a good enough percentage of wins 2010-11-02T06:09:54 mega1: that game looks like a server fluke. Here, it responds instantly. And gets destroyed by my old, old submission. Good thing it crashed instead of playing. 2010-11-02T06:10:45 dmj111: not surprised to hear that 2010-11-02T06:11:32 lol 2010-11-02T06:11:53 the embarrasment of defeat is far worse than surrendering early! 2010-11-02T06:12:26 embarrassment* 2010-11-02T06:13:18 Top 10 players: bocsimacko(4335), dmj111(4262), shangas(4158), Hazard(4153), george(4139), FlameN(4118), Accoun(4085), joakim_mjardner(4070), felixcoto(4057), medrimonia(4041) 2010-11-02T06:14:02 pfff 2010-11-02T06:14:13 #1 and #2 in the world :) 2010-11-02T06:14:18 nice 2010-11-02T06:14:23 that is true. I was disappointed that there wasn't a real game there, since bocsimacko and have had so few. Not that I think I have a chance, but I want to learn something from it. 2010-11-02T06:14:54 ("that is true" refers to the embarrassment part) 2010-11-02T06:15:25 But, I did think it was funny to imagine that my bot crapped out because it saw what was coming. 2010-11-02T06:16:05 "since bocsimacko _and I_ have had so few." I need to quit typing at 6 in the morning. 2010-11-02T06:17:08 yeah, whenever i play him it's over within 70 moves usually 2010-11-02T06:17:16 my bot always does something his bot exploits the crap out of 2010-11-02T06:18:20 not only is his far better than mine, but what ever he does is the scissors to the paper my bot does. 2010-11-02T06:18:58 I had the same thing with rsergio's last submission. our ratings were about the same, but his beat me about 10 time in a row.. 2010-11-02T06:19:04 i think i need to speed things up so i can consider more options from both players 2010-11-02T06:19:35 oef 2010-11-02T06:19:38 perfectly ranking so many bots is an impossible task, i think it's working reasonably well 2010-11-02T06:20:07 compared to most competitions, we get a pretty good go at approximating perfect rankings 2010-11-02T06:20:55 antimatroid: where are you ranked on tcp and official? 2010-11-02T06:21:04 nothing on official atm 2010-11-02T06:21:08 and tcp not well :P 2010-11-02T06:21:13 oh 2010-11-02T06:21:30 once upon a time i was in the top 10 on official 2010-11-02T06:21:45 yes, i think the ratings are working well. Just saying that there are certain strategies that dominate others, even though both are about the same strength against everyone else. 2010-11-02T06:22:15 well 2010-11-02T06:22:56 That sounds valid to me dmj111 2010-11-02T06:23:14 transitivity 2010-11-02T06:28:40 you will never get non-cyclic ordering for bots in a comp like this 2010-11-02T06:28:47 otherwise it'd be pretty easy to sort them 2010-11-02T06:42:26 *** boegel has joined #aichallenge 2010-11-02T06:48:34 wait, remind me, if i have 10 ships, and they have 15 ships arrive at a neutral with 30 ships, there are 15 left yes? 2010-11-02T06:49:07 yes. 2010-11-02T06:49:19 okay that's fine, i confused myself somehow 2010-11-02T06:49:28 that happens to me a lot. 2010-11-02T06:50:28 *** Rubicon-|-Cross has quit IRC (Quit: Leaving) 2010-11-02T06:55:23 *** Migi32 has joined #aichallenge 2010-11-02T06:55:37 antimatroid: ok sorry im confused there too 2010-11-02T06:55:58 you mean if you send 10 ships and enemy has 15 ships arrive at a neutral with 30 ships at the same time? 2010-11-02T06:56:07 yep 2010-11-02T06:56:31 erm my calculations must be wrong then... how do you calculate it? 2010-11-02T06:56:37 i thought it was 30-15 = 15 -10 = 5? 2010-11-02T06:56:46 nope 2010-11-02T06:56:58 oh ok whats it then? :P 2010-11-02T06:57:21 * raedwulf needs to fix his code 2010-11-02T06:57:28 i must have read the spec wrongly 2010-11-02T06:57:33 whois flavor on tcp 2010-11-02T06:57:43 because whatever that guy is doing, its slooooooooooooooooooooooow as hell 2010-11-02T06:58:17 find max incoming, if that's <= neutral, neutral stays as is with ships -= maxincoming 2010-11-02T06:58:18 if that's > neutral amount, if max=min incoming, neutral amount = 0, otherwise owner = max player and ships = max incoming - min(neutral amount, min incoming) 2010-11-02T06:58:26 hang one 2010-11-02T06:58:49 ah ok 2010-11-02T06:58:57 http://codepad.org/RgH7vD5x 2010-11-02T06:59:00 that's my resolution code 2010-11-02T06:59:15 max(neutral amount, min incoming) there at the end I think 2010-11-02T06:59:15 it's a bit longer than probably necessary, i was trying to minimise the number of operations performed 2010-11-02T06:59:20 ships = max incoming - max(neutral amount, min incoming) 2010-11-02T06:59:32 migi32: you're correct 2010-11-02T06:59:41 oh ok 2010-11-02T06:59:50 the posted code has it right 2010-11-02T07:00:05 so basically the smaller fleet does nothing? 2010-11-02T07:00:16 yep. just dies. 2010-11-02T07:00:19 yep 2010-11-02T07:00:23 ok 2010-11-02T07:00:37 error323: i spend my full second on moves too :P 2010-11-02T07:00:40 * raedwulf fixes his code 2010-11-02T07:02:35 antimatroid, you have a typo in your max function, it's called maxx :P 2010-11-02T07:02:57 that's in case it fucked with what's in math.h 2010-11-02T07:03:24 i have my own genreal.h file with template functions to do a bunch of crap c++ doesn't just do for me 2010-11-02T07:03:28 i pulled it from another library i have 2010-11-02T07:03:49 what's wrong with the regular max? 2010-11-02T07:04:19 antimatroid: ah its you ;), hehe grrrr :p 2010-11-02T07:04:27 i'm not flavor :P 2010-11-02T07:04:42 oh 2010-11-02T07:04:45 lol 2010-11-02T07:04:49 *** Appleman1234 has quit IRC (Ping timeout: 264 seconds) 2010-11-02T07:04:49 nothings wrong with regular max i guess, i just use that 2010-11-02T07:04:50 i should read more carefully 2010-11-02T07:04:51 -,- 2010-11-02T07:05:16 I think demerzel is flavor 2010-11-02T07:06:49 so antimatroid what functions are in there then? Stuff like toString()? 2010-11-02T07:09:25 http://codepad.org/SRXSx1ao 2010-11-02T07:09:35 my other one has other stuff for outputting vectors to streams etc. 2010-11-02T07:10:11 i think there's a comment near the bottom that is for the wrong function too :P 2010-11-02T07:12:22 lol, "//just in case the person using it is retarded" 2010-11-02T07:12:36 :p 2010-11-02T07:12:57 yea, I was confused by that comment for a while 2010-11-02T07:13:12 I was like, "am I retarded? :o" 2010-11-02T07:13:19 Top 10 players: bocsimacko(4330), dmj111(4273), shangas(4155), Hazard(4150), george(4131), FlameN(4104), Accoun(4082), felixcoto(4078), joakim_mjardner(4058), medrimonia(4030) 2010-11-02T07:13:29 lol 2010-11-02T07:15:05 http://codepad.org/g5QQVrxn fixed 2010-11-02T07:15:11 sort of.. 2010-11-02T07:15:17 :P 2010-11-02T07:18:19 antimatroid, most of what you have there is already in standard c++ though, std::max and std::accumulate (for the sum or the product) 2010-11-02T07:18:29 but I like the operator== for vectors 2010-11-02T07:18:35 not that I needed it yet, but still 2010-11-02T07:18:39 thanks for sharing :P 2010-11-02T07:19:01 my favourite functions are for input/output and [] indexing objects (vectors inside them) 2010-11-02T07:19:16 do you know how to do those? 2010-11-02T07:20:07 i can't work out how to do []...[] indexing of tensors though 2010-11-02T07:20:15 anyone know how to do that? 2010-11-02T07:20:27 when i have them stored in a 1 dimensional vector 2010-11-02T07:21:21 and i really wish i could overload the order in which operators were performed on different objects 2010-11-02T07:22:12 *** lord_shadow has joined #aichallenge 2010-11-02T07:22:13 ah, no sorry 2010-11-02T07:22:26 I use operator() 2010-11-02T07:22:31 for my Matrix class 2010-11-02T07:23:41 http://codepad.org/LTwmtgyT 2010-11-02T07:24:13 i got that little hax off a friend, it's wonderful 2010-11-02T07:25:13 hmm... why is the cast required? 2010-11-02T07:25:28 not sure, whenever i try to change it from that it breaks 2010-11-02T07:25:36 heh 2010-11-02T07:25:51 i tend to overcast things anyway 2010-11-02T07:27:33 I will be quiet for a while now, the greatest match in all StarCraft 2 history is going on atm 2010-11-02T07:27:50 meh, i don't go much on video gmaes 2010-11-02T07:29:05 except minesweeper :P 2010-11-02T07:38:18 Error323: having said that, i just reverted to heuristic bot for testing :P 2010-11-02T07:52:05 *** Appleman1234 has joined #aichallenge 2010-11-02T07:54:07 *** lord_shadow has quit IRC (Quit: Ex-Chat) 2010-11-02T07:54:44 *** Appleman1234 has quit IRC (Remote host closed the connection) 2010-11-02T07:55:07 *** Appleman1234 has joined #aichallenge 2010-11-02T07:56:22 *** JamesGecko has joined #aichallenge 2010-11-02T08:01:25 *** JamesGecko has quit IRC (Ping timeout: 276 seconds) 2010-11-02T08:02:42 *** dmj111 has quit IRC (Read error: No route to host) 2010-11-02T08:08:57 *** boegel has quit IRC (Quit: This computer has gone to sleep) 2010-11-02T08:13:20 Top 10 players: bocsimacko(4322), dmj111(4275), Hazard(4153), shangas(4131), george(4119), FlameN(4086), Accoun(4071), felixcoto(4058), joakim_mjardner(4057), rsergio(4013) 2010-11-02T08:15:01 Migi32: boxer vs nada? 2010-11-02T08:15:12 yep, brx 2010-11-02T08:15:34 2-1 for boxer so far 2010-11-02T08:16:43 *** sigh has quit IRC (Remote host closed the connection) 2010-11-02T08:19:49 *** yasith has quit IRC (Read error: Connection reset by peer) 2010-11-02T08:36:26 *** yasith has joined #aichallenge 2010-11-02T08:47:44 *** mceier has joined #aichallenge 2010-11-02T08:59:40 *** JamesGecko has joined #aichallenge 2010-11-02T08:59:43 morning 2010-11-02T09:04:55 howdy 2010-11-02T09:11:12 how goes it 2010-11-02T09:11:24 i see you have 2 new versions up on TCP 2010-11-02T09:12:37 *** JamesGecko has quit IRC (Ping timeout: 245 seconds) 2010-11-02T09:13:21 Top 10 players: bocsimacko(4317), dmj111(4264), Hazard(4152), shangas(4143), george(4122), felixcoto(4079), FlameN(4072), davidjliu(4071), Accoun(4060), joakim_mjardner(4051) 2010-11-02T09:14:25 don't you love it when you execute a good snipe 2010-11-02T09:14:26 http://72.44.46.68/canvas?game_id=821018 2010-11-02T09:14:58 though i still have a bug i guess with support after snipe ^ 2010-11-02T09:28:15 *** jezuz has quit IRC (Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com )) 2010-11-02T09:28:45 interesting game, bhasker. Full of sniping indeed. 2010-11-02T09:29:27 yup been working on fixing bugs in my sniping code 2010-11-02T09:29:32 so its better but not perfect 2010-11-02T09:29:48 it still makes wrong calculations when there are fleets in lockstep 2010-11-02T09:29:57 mine and enemies coming right after one another 2010-11-02T09:33:09 bhasker: indeed 2010-11-02T09:34:16 *** lavalamp_ has quit IRC (Quit: Page closed) 2010-11-02T09:35:04 I think my sniping is working well atm 2010-11-02T09:37:48 bhasker: have you tried 15.3 against urs? 2010-11-02T09:38:07 not yet, never got your mail? 2010-11-02T09:38:49 o_O 2010-11-02T09:39:01 F. Huizinga 2010-11-02T09:39:37 subject: google ai challenge 2010-11-02T09:39:49 "I have recently discovered that you have a relative xxx that has recently died. You have inherited..." 2010-11-02T09:40:14 ? 2010-11-02T09:40:24 bhasker: check spam? :P 2010-11-02T09:40:34 hmm 2010-11-02T09:40:34 its a joke, based on a lot of spam with the same stuff 2010-11-02T09:40:56 <--embarrasing if it went to spam ( works for y! mail delivery :) ) 2010-11-02T09:41:01 *** Appleman1234 has quit IRC (Ping timeout: 255 seconds) 2010-11-02T09:41:18 it did goto spam :-\ 2010-11-02T09:41:30 let me check why that happened 2010-11-02T09:41:32 ah if its yahoo, good chance it goes to spam 2010-11-02T09:41:45 they are really strict 2010-11-02T09:42:11 will have to look into why it went to spam 2010-11-02T09:42:45 ahw 2010-11-02T09:42:49 perhaps due to the binary? 2010-11-02T09:43:13 :( 2010-11-02T09:43:32 probably 2010-11-02T09:43:37 is it a windows executable? 2010-11-02T09:43:42 no, linux 2010-11-02T09:43:44 hmmm 2010-11-02T09:43:46 strange 2010-11-02T09:43:59 normally they don't check those 2010-11-02T09:44:03 indd 2010-11-02T09:44:17 zip or tar.gz? 2010-11-02T09:44:23 neither 2010-11-02T09:44:26 ah 2010-11-02T09:44:39 maybe they see "unrecognized glob" 2010-11-02T09:44:50 and put it into spam? 2010-11-02T09:46:51 mja, maybe 2010-11-02T09:46:59 it must be the binary 2010-11-02T09:50:54 *** bhasker has quit IRC (Ping timeout: 252 seconds) 2010-11-02T09:51:11 Error323: see what your code did, it fried his computer :D :D :D 2010-11-02T09:51:21 lol 2010-11-02T09:51:28 my plan is working >:} 2010-11-02T09:51:38 * raedwulf gasps 2010-11-02T09:55:26 lol 2010-11-02T09:56:17 *** bhasker has joined #aichallenge 2010-11-02T09:58:23 antimatroid, still here? 2010-11-02T09:59:45 *** Sarin has joined #aichallenge 2010-11-02T10:04:18 bhasker: currently ranked 42, was 12th this morning :P 2010-11-02T10:04:20 xD 2010-11-02T10:04:29 nice grats:D 2010-11-02T10:04:40 #34 now for me 2010-11-02T10:05:27 yeah, i'm still as stable as two stacked marbles 2010-11-02T10:05:51 need moar games 2010-11-02T10:06:22 watch this snipe of the planet with 70 ships 2010-11-02T10:06:22 http://72.44.46.68/canvas?game_id=821572 2010-11-02T10:07:06 i get a warm and fuzzy feeling when my bot does that 2010-11-02T10:07:11 but then i watch games like these 2010-11-02T10:07:19 hahaha 2010-11-02T10:07:28 http://72.44.46.68/canvas?game_id=821533 2010-11-02T10:07:33 watch Turn50 2010-11-02T10:07:39 and i feel like shooting my bot 2010-11-02T10:07:45 it committed suicide practically 2010-11-02T10:08:11 lol 2010-11-02T10:08:17 yeahh :( 2010-11-02T10:08:43 but you kind of had to 2010-11-02T10:08:48 in order to get back on top 2010-11-02T10:09:02 your expansion was required earlier 2010-11-02T10:09:14 yea 2010-11-02T10:09:23 his bot's trick gets him early expansion much faster 2010-11-02T10:09:32 he calculates neutrals that he will take in future into his launches 2010-11-02T10:09:36 i need to work that trick out 2010-11-02T10:09:53 so he can can acquire planets faster than my bot can 2010-11-02T10:10:04 yeah that is sweet 2010-11-02T10:10:27 mmm 2010-11-02T10:13:00 anyway gotta run gym time now 2010-11-02T10:13:01 l8rs 2010-11-02T10:13:20 Top 10 players: bocsimacko(4330), dmj111(4266), Hazard(4166), shangas(4155), george(4130), felixcoto(4092), FlameN(4089), Accoun(4078), joakim_mjardner(4055), jimrogerz(4049) 2010-11-02T10:34:15 *** JamesMG has joined #aichallenge 2010-11-02T10:35:28 *** matmat_ has joined #aichallenge 2010-11-02T10:43:37 *** superflit has quit IRC (Quit: superflit) 2010-11-02T10:49:52 *** fmeyer has joined #aichallenge 2010-11-02T10:58:30 *** demerzel_ has joined #aichallenge 2010-11-02T10:59:58 *** Utkarsh has quit IRC (Ping timeout: 250 seconds) 2010-11-02T11:00:25 *** Utkarsh has joined #aichallenge 2010-11-02T11:11:47 mega1 i shall get you one day 2010-11-02T11:12:07 25 left 2010-11-02T11:12:51 25? 2010-11-02T11:12:55 days ? 2010-11-02T11:13:09 your bot has been destroying my bot's self esteem 2010-11-02T11:13:11 :-\ 2010-11-02T11:13:22 Top 10 players: bocsimacko(4326), dmj111(4246), Hazard(4162), shangas(4134), george(4119), felixcoto(4101), FlameN(4079), davidjliu(4065), Accoun(4063), medrimonia(4045) 2010-11-02T11:13:31 finished refactoring all my code now i have perfect state and zero logic 2010-11-02T11:13:32 woop 2010-11-02T11:13:47 wow medrimonia is in the top 10? 2010-11-02T11:13:50 wtf :-\ 2010-11-02T11:13:58 my bot was beating his bot quite regularly the other day 2010-11-02T11:14:00 :( 2010-11-02T11:14:29 btw whose bots are these as-* and Monty series 2010-11-02T11:14:38 Migi32: i am not flavor lol 2010-11-02T11:14:50 the only name i've used that isn't mine is testBot and they varied widely across like 5 diff bots 2010-11-02T11:17:41 *** superflit has joined #aichallenge 2010-11-02T11:19:40 ugh i have like 3 layers to my io 2010-11-02T11:19:58 i don't know how to simplify it any more and be able to drop classes as necessary 2010-11-02T11:20:02 mega1 's bot transcends and becomes all-knowing... be afraid.... very afraid... 2010-11-02T11:20:31 how do you guys encapsulate your logic? pass a thinker func the gamestate or what? 2010-11-02T11:20:44 demerzel_: yea 2010-11-02T11:20:51 but mine is far more complex than that 2010-11-02T11:21:02 but at the most basic level, it does that 2010-11-02T11:24:52 "one of the early advantages flavor had was nobody could calc stream attacks right for a while lol" 2010-11-02T11:24:59 this made me assume you were flavor :P 2010-11-02T11:25:36 oh no i used to play him early on 2010-11-02T11:25:41 he was a great test for my defenses 2010-11-02T11:26:00 took me a while to get the calc for minimum needed to not lose my planets lol 2010-11-02T11:26:45 do you treat stream attacks different than normal attacks, btw? 2010-11-02T11:28:54 i don't really treat attacks as anything other than deductions to my planet population 2010-11-02T11:29:20 but to calculate how much you need to stop it from losing is more difficult when there's 20 turns to account for 2010-11-02T11:29:34 or rather 20 turns with incoming 2010-11-02T11:31:57 *** lavalamp_ has joined #aichallenge 2010-11-02T11:34:20 *** superflit has quit IRC (Remote host closed the connection) 2010-11-02T11:34:38 dammit google - i almost had my boss convinced to let me make a c++ socket stream for google data but they had to go and make a .net library 2010-11-02T11:37:00 google data 2010-11-02T11:37:10 as in search results? 2010-11-02T11:37:18 nah analytics etc 2010-11-02T11:37:22 oh ok 2010-11-02T11:37:45 they have a java api but my work is convinced java is the devil 2010-11-02T11:38:16 I need a little bit of that kind of forward thinking here sometimes 2010-11-02T11:38:20 *** McLeopold has joined #aichallenge 2010-11-02T11:38:29 that java is the devil? 2010-11-02T11:38:40 yes 2010-11-02T11:38:43 lol 2010-11-02T11:38:46 java is good for many things, esp. large projects 2010-11-02T11:38:50 i am starting to respect it a lot more 2010-11-02T11:39:03 this company basically runs on it, and it's great 95% of the time 2010-11-02T11:39:06 next thing you know i'll have a 'linux box' 2010-11-02T11:39:14 *** aerique has quit IRC (Quit: ...) 2010-11-02T11:39:26 then i'll start using m$ jokingly at first then seriously 2010-11-02T11:39:27 then it's over 2010-11-02T11:39:39 then you are basically RMS, sorry 2010-11-02T11:39:45 that's the progression 2010-11-02T11:39:45 they'r right that java is the devil though. http://simpleminded.hostwq.net/8 2010-11-02T11:39:58 :) 2010-11-02T11:42:57 *** iNtERrUpT has joined #aichallenge 2010-11-02T11:44:39 RMS? Root Mean Square? 2010-11-02T11:44:46 Rate Monotonic Scheduling? 2010-11-02T11:45:02 ai-contest is running games much faster now 2010-11-02T11:45:19 Richard M Stallman 2010-11-02T11:45:30 but it's only matching my bot against lower ranked bots for some reason http://ai-contest.com/profile.php?user_id=7268 2010-11-02T11:49:43 yea my c# bot has been winning and dropping rapidly in rank lol 2010-11-02T11:54:07 i think the problem with minimax is that the game doesn't start off as zerosum 2010-11-02T11:54:25 i think it can and usually does turn zero-sum 2010-11-02T11:54:30 but if you don't start right it won't get there ;) 2010-11-02T11:57:23 wow it's kinda scary that I look at this and think "oh that's map 11": http://ai-contest.com/visualizer.php?game_id=6508561 2010-11-02T11:59:05 map 11 gives my bot lots of problems, I'm surprised it won that one 2010-11-02T11:59:46 that is a little scary 2010-11-02T12:01:09 I'm impressed you got your tree bot to value moving ships up so well 2010-11-02T12:03:39 Who, me? 2010-11-02T12:04:56 yeah 2010-11-02T12:05:13 is that part of your tree or a separate function? 2010-11-02T12:05:17 defense/dist were separate for me 2010-11-02T12:05:21 It's mandatory, it's the last thing the move generator does for every move 2010-11-02T12:05:43 ah, I see 2010-11-02T12:05:54 my goals are in levels and the tree searcher is allowed to mix and match some levels but others are mandatory 2010-11-02T12:06:24 Hm, 0badc0de needs to work on winning won games: http://ai-contest.com/visualizer.php?game_id=6507392 2010-11-02T12:08:23 And actually what I've got running there atm is just the move generator trying execute all its goals without any tree search 2010-11-02T12:08:53 search should help it a lot if I can ever get it working and bug free :( 2010-11-02T12:09:06 mine is ridiculously buggy right now 2010-11-02T12:09:12 actually, both of my bots are 2010-11-02T12:09:46 does anyone have a simple way of figuring out what the front-line planets are? 2010-11-02T12:09:55 *** Itkovian has quit IRC (Ping timeout: 240 seconds) 2010-11-02T12:10:39 my best working code uses the planet closest to it's nearest enemy planet, and any planets that are 2010-11-02T12:10:46 close enough for it to get to before the enemy supply it 2010-11-02T12:11:03 of course, that code is bugged right now 2010-11-02T12:11:24 lavalamp_: closest planets to enemy planets 2010-11-02T12:12:37 wow, this map is interesting: http://72.44.46.68/canvas?game_id=822669 2010-11-02T12:12:41 McLeopold: hm, I'm trying to remember why I didn't do it that way 2010-11-02T12:13:10 JamesMG: Wow, nice 2010-11-02T12:13:22 Top 10 players: bocsimacko(4302), dmj111(4243), Hazard(4145), jimrogerz(4125), shangas(4122), davidjliu(4121), george(4105), felixcoto(4080), FlameN(4068), Accoun(4045) 2010-11-02T12:13:42 *** bhasker has quit IRC (Quit: bhasker) 2010-11-02T12:14:13 JamesMG: that's a serious lack of defense... 2010-11-02T12:14:25 you can see the problems my pathing code has in this one though: http://72.44.46.68/canvas?game_id=822723 2010-11-02T12:15:00 *** iNtERrUpT has quit IRC (Remote host closed the connection) 2010-11-02T12:15:57 you lost me McLeopold, where? 2010-11-02T12:16:19 JamesMG: by bocsimacko 2010-11-02T12:16:26 oh 2010-11-02T12:16:42 It's surprising he's number one and yet plays that type of game. 2010-11-02T12:16:52 It's like defense doesn't matter. 2010-11-02T12:17:14 He just assumes he will steamroll the opponent? 2010-11-02T12:17:25 I guess 2010-11-02T12:17:43 he didn't have to let himself get sniped is the weird thing 2010-11-02T12:17:53 he had enough ships to counter snipe 2010-11-02T12:18:03 exactly 2010-11-02T12:18:15 we would've traded those planets for a turn I think 2010-11-02T12:23:17 I'll look into it :-) 2010-11-02T12:23:45 oh, no, I think it's an enduring bug. No need to fix it. 2010-11-02T12:24:46 great, now we all have even less chance of winning this contest :) 2010-11-02T12:26:18 *** AlliedEnvy has joined #aichallenge 2010-11-02T12:26:19 *** Migi32 has quit IRC (Quit: http://boinc.berkeley.edu) 2010-11-02T12:28:22 *** p4p4 has joined #aichallenge 2010-11-02T12:29:39 *** Azrathud has joined #aichallenge 2010-11-02T12:30:36 *** superflit has joined #aichallenge 2010-11-02T12:35:30 *** bhasker_ has joined #aichallenge 2010-11-02T12:37:47 so question 2010-11-02T12:37:59 lets say enemy launches 100 ships at your planet 2010-11-02T12:38:27 and you have 20 right now, but not enough to counter it, but your growth will exceed the 100 before his ships hit 2010-11-02T12:38:35 i mean the supporting planets 2010-11-02T12:38:42 so you could keep the planet if you stream ships to it 2010-11-02T12:38:55 the question is how do you compute for something like that 2010-11-02T12:39:00 my bot is rather dumb that way 2010-11-02T12:39:19 or will it never work,since enemy could counter the streamed ships? 2010-11-02T12:39:36 my bot doesn't launch unless it has ships to counter the attack 2010-11-02T12:39:37 *** Rubicon-|-Cross has joined #aichallenge 2010-11-02T12:40:01 anyone any ideas? 2010-11-02T12:40:59 bhasker: if the growth exceeds the ships then i would do nothing 2010-11-02T12:41:03 my bot looks into the future and calculates how many ships it will have at what point 2010-11-02T12:41:12 no not the growth of the planet being attacked 2010-11-02T12:41:15 and sends ships at the last possible time 2010-11-02T12:41:17 but probably the supporting planets 2010-11-02T12:41:36 bhasker: then i would have already calc'd that and queued a future order ;) 2010-11-02T12:41:49 *** narnach has joined #aichallenge 2010-11-02T12:41:55 hmm my bot doesn't do that 2010-11-02T12:42:05 queueing orders is tough 2010-11-02T12:42:08 i'm still working out priorities 2010-11-02T12:42:19 I don't queue, I recalc every turn 2010-11-02T12:42:25 yea same here 2010-11-02T12:42:30 i never queue across turns 2010-11-02T12:42:32 but if you send at the last possible time it amounts to the same thing 2010-11-02T12:42:39 sortof 2010-11-02T12:42:41 not really there is a travel time 2010-11-02T12:42:52 I account for travel time 2010-11-02T12:43:18 hmm 2010-11-02T12:43:44 only problem with what I do is that it can change its mind halfway through 2010-11-02T12:45:10 I'm working out how to deal with mind changing right now too 2010-11-02T12:45:40 I figure I'm gonna concentrate on making good decisions initially 2010-11-02T12:45:54 *** Azrathud has quit IRC (Ping timeout: 240 seconds) 2010-11-02T12:46:02 then mind changing will be a feature and not a bug 2010-11-02T12:46:33 well i just force it to consider planets which have attacks underway before other planets 2010-11-02T12:46:42 so if it can complete the order then it will 2010-11-02T12:46:47 otherwise it will go through the rest 2010-11-02T12:47:39 mine looks ahead for planets that will be less expensive at some point in the future, which handles sniping and syncing up with my prior orders 2010-11-02T12:48:07 bhasker_: your weakness there is if there's a higher priority eg defense that you are launching out anyway for lesser gain 2010-11-02T12:49:36 i tried that already 2010-11-02T12:49:45 but you are correct in the fact that followthrough is difficult 2010-11-02T12:51:50 well i tried various strategies 2010-11-02T12:51:57 defense first has its drawbacks 2010-11-02T12:52:06 especially when you launch a massive attack 2010-11-02T12:52:14 the way i structure it 2010-11-02T12:52:30 i consider my planets and enemy planets at same priority as long as they have the same growth rate 2010-11-02T12:52:41 only neutrals get prioritized lower 2010-11-02T12:52:45 unless they are under attack 2010-11-02T12:52:54 in which case i prioritize them a little higher than other neutrals 2010-11-02T12:53:13 this seems to work reasonably well to ensure that i complete launches 2010-11-02T12:59:22 mm 2010-11-02T12:59:38 so you attack before defense 2010-11-02T13:00:15 nope both are same priority 2010-11-02T13:00:24 as long as they are equal growth planets 2010-11-02T13:00:42 i think i would defend first actually 2010-11-02T13:00:47 for equal growth 2010-11-02T13:01:36 wuuut 2010-11-02T13:01:41 new valuefunction is doing nice 2010-11-02T13:01:54 playing with that function is very, very usefull =) 2010-11-02T13:02:31 ah ok 2010-11-02T13:02:48 mmmm 2010-11-02T13:02:52 i should do that aswell 2010-11-02T13:02:59 currently all defense is before all attack 2010-11-02T13:03:21 instead of using priorities 2010-11-02T13:03:23 all defense is not useful 2010-11-02T13:03:29 you may end up defending a 1 growth 2010-11-02T13:03:37 when you could have attacked a 5 growth enemy 2010-11-02T13:03:40 i still maintain there are so few circumstances where defense is not useful it's more of a pain to implement it as optional 2010-11-02T13:03:51 assuming of course you aren't taking planets behind enemy lines 2010-11-02T13:04:14 indd 2010-11-02T13:04:44 the problem with defending first also happens when you launch massive attacks 2010-11-02T13:04:51 and enemy counters with attacks 2010-11-02T13:04:57 now if you always defend first 2010-11-02T13:05:04 you would end up sending all the ships bac 2010-11-02T13:05:06 in order to make a defense not useful it must not expose currently covered planets, have so little growth that you have the ability to counter the loss by taking another even losing travle time 2010-11-02T13:05:08 when they reach the end 2010-11-02T13:05:19 instead of swapping bases 2010-11-02T13:05:35 and all defense losses are 1:1 2010-11-02T13:05:43 so it's a rare circumstance that is not optimal 2010-11-02T13:06:13 yaeh 2010-11-02T13:06:22 i knew it was stupid >_> 2010-11-02T13:06:28 this is what i see in a lot of games 2010-11-02T13:06:30 im winning 2010-11-02T13:06:42 and then i start defending crap insttead of keep going 2010-11-02T13:06:45 another reason why i removed all preference for enemy planets closer to mine 2010-11-02T13:06:52 i go for whichever planet i think i can acquire 2010-11-02T13:07:11 even if its in the middle of the enemy base 2010-11-02T13:07:13 Error323: if you're winning how is this a bad thing 2010-11-02T13:07:19 cause you lose travel time 2010-11-02T13:07:25 and growth 2010-11-02T13:07:31 while enemy is busy taking over your base 2010-11-02T13:07:42 your ships are hanging in mid air doing nothing 2010-11-02T13:07:55 because eventually i lose 2010-11-02T13:08:04 well maybe you should defend them instead of launching all your ships ;) 2010-11-02T13:08:09 the point is 2010-11-02T13:08:18 you launched a big attack to take on an enemy planet 2010-11-02T13:08:24 but enemy instead of defending 2010-11-02T13:08:27 counter attacks all your planets 2010-11-02T13:08:29 or a bunch of them 2010-11-02T13:08:31 what do you do? 2010-11-02T13:08:53 nothing because i already calc'd how much i need to hold back and didn't launc those 2010-11-02T13:10:10 then you would never launch big attacks in close matches 2010-11-02T13:10:16 that's true 2010-11-02T13:10:26 that's how you lose 2010-11-02T13:10:30 not really 2010-11-02T13:10:33 i have won a bunch of them 2010-11-02T13:10:45 well it depends on a lot of factors obv 2010-11-02T13:10:50 but if the enemy holds back enough to defend at all times 2010-11-02T13:10:54 your attack will be ineffective 2010-11-02T13:11:07 if you both kamikaze each other then whoever has the better attack algorithm will win 2010-11-02T13:11:32 *** JamesGecko_ has joined #aichallenge 2010-11-02T13:11:32 most bots will attack each other 2010-11-02T13:11:39 because suddenly there is a vaccum in your base 2010-11-02T13:11:48 and enemy will automatically launch at all your planets 2010-11-02T13:12:00 at least most bots that i have seen do that 2010-11-02T13:12:00 yea it depends 2010-11-02T13:12:08 very few bots defend a massive attack 2010-11-02T13:12:11 usually they counter 2010-11-02T13:12:24 i wonder how many people calc enemy reinforcement possibility 2010-11-02T13:12:28 or just send a flat % on top or what 2010-11-02T13:12:39 i tried that 2010-11-02T13:12:42 but it horribly fails 2010-11-02T13:12:50 i used to send over the maximum possible enemy reinforcement 2010-11-02T13:12:55 which works but i hardly think it's optimal 2010-11-02T13:13:09 *** phreeza has joined #aichallenge 2010-11-02T13:13:11 normally what i calculate is 2010-11-02T13:13:13 i do that too now 2010-11-02T13:13:15 # of ships required == 2010-11-02T13:13:23 Top 10 players: bocsimacko(4307), dmj111(4243), Hazard(4171), shangas(4134), jimrogerz(4130), george(4104), felixcoto(4075), FlameN(4059), joakim_mjardner(4045), Accoun(4040) 2010-11-02T13:13:23 all planets that can reinforce target*dist 2010-11-02T13:13:29 can anyone provide with a link to download complete games? 2010-11-02T13:13:32 + # of ships required to attck all of them 2010-11-02T13:13:35 ah so you do calc enemy reinf possibility 2010-11-02T13:13:35 from the official site? 2010-11-02T13:13:46 so i usuall launch enough to take over most of the planets nearby 2010-11-02T13:13:49 otherwise i don' 2010-11-02T13:14:01 though i don't account very well for fleets in flight 2010-11-02T13:14:03 between the planets 2010-11-02T13:14:09 so i end up underestimating it a bit 2010-11-02T13:14:14 and attacks can fail 2010-11-02T13:14:21 my attacks never fail 2010-11-02T13:14:24 but they are quite few in quantity 2010-11-02T13:14:43 i am sure there is a medium between 1+ pop and 1+pop + all enemy reinf possibility 2010-11-02T13:14:59 i have yet to find it though 2010-11-02T13:15:13 yea 2010-11-02T13:15:31 so one thing i changed is i used to double count enemy reinforcements when launching from multiple planets 2010-11-02T13:15:35 i stopped doing that 2010-11-02T13:15:48 so its a little closer to optimal 2010-11-02T13:15:50 but not there yet 2010-11-02T13:16:06 though most of my losses are not due to bad attacks 2010-11-02T13:16:10 my c# bot only launches exactly 1 over any possibility: http://ai-contest.com/visualizer.php?game_id=6509730 2010-11-02T13:16:11 but usually due to bad starts 2010-11-02T13:16:23 doesn't launch unless it can overage every possible enemy ship 2010-11-02T13:16:34 i am trying to figure out if i want to do that for this one or not 2010-11-02T13:17:05 the problem i have with launching exact ships 2010-11-02T13:17:08 is that once they get there 2010-11-02T13:17:10 they just sit 2010-11-02T13:17:16 since they are covered with enemy planets 2010-11-02T13:17:19 on all sides 2010-11-02T13:17:23 and get wiped 2010-11-02T13:17:47 sending a larger than expected force allows me to continue my attacks 2010-11-02T13:18:38 bockismacko is killing my bot :-\ 2010-11-02T13:18:40 :( 2010-11-02T13:20:17 lol 2010-11-02T13:20:27 hah, my bot is too low to be paired with bocksimacko! 2010-11-02T13:20:44 *** Bobng has joined #aichallenge 2010-11-02T13:21:02 this is on tcp 2010-11-02T13:21:05 ah ok 2010-11-02T13:21:13 i've not tried my bot yet on tcp 2010-11-02T13:21:29 but the one on the main server is only working on a rage-bot algorithm 2010-11-02T13:21:33 its embarrasing , its like he gives my bot a schooling 2010-11-02T13:21:38 lol 2010-11-02T13:24:07 lol 2010-11-02T13:24:28 i need to write a special function "beat_bocksimacko()" 2010-11-02T13:25:34 but wait... what about the 2000 others above me 2010-11-02T13:26:17 well when you got that function 2010-11-02T13:26:24 don't worrry bout the other 2000 2010-11-02T13:27:47 ^^ 2010-11-02T13:28:01 learning some haskell at the moment 2010-11-02T13:28:13 if i knew enough haskell might have been idea to write my bot in that 2010-11-02T13:28:31 i never touched that 2010-11-02T13:28:43 might know enough by the time the competition completes 2010-11-02T13:28:45 lol 2010-11-02T13:29:01 its a functional language, its pretty awesome 2010-11-02T13:29:48 *** McLeopold_ has joined #aichallenge 2010-11-02T13:32:18 alright complete encapsulated gamestate! that i do nothing with! 2010-11-02T13:32:31 someone write some logic for me i'm only good at the state part 2010-11-02T13:33:01 *** McLeopold has quit IRC (Ping timeout: 264 seconds) 2010-11-02T13:34:24 attack the enemy only 2010-11-02T13:34:53 what i don't have enough ships 2010-11-02T13:34:57 what if* 2010-11-02T13:35:41 *** MrG_ has joined #aichallenge 2010-11-02T13:36:09 *** decipher has joined #aichallenge 2010-11-02T13:37:07 my deterministic bot just beat itself. O_o 2010-11-02T13:37:44 hey guys i have a quick question: if i have an enemy fleet heading towards my planet or any other neutral planet how can i figure out how many turns he has left to reach his destination,(aka distance of fleet to destination planet) i used pw.getFleet(a).turnsRemaining() function returns 0? 2010-11-02T13:38:08 decipher: it shouldn't return zero 2010-11-02T13:38:25 what language? 2010-11-02T13:38:28 java 2010-11-02T13:39:07 hmm 2010-11-02T13:39:19 i'm storing the turns remaining into a separate wrapper function that has dest,source,ships,and turns remaining 2010-11-02T13:39:19 Did you introduce a bug in the game start parser? 2010-11-02T13:39:55 decipher: then you have to debug your own code 2010-11-02T13:40:24 the turn remaining comes to you as a string in the game state and the beginning of the turn 2010-11-02T13:40:25 sorry for the paste 2010-11-02T13:40:29 else if (pw.GetFleet(a).Owner() == 2){ Attack att = new Attack(pw.GetFleet(a).SourcePlanet(),pw.GetFleet(a).DestinationPlanet(),pw.GetFleet(a).NumShips(),pw.GetFleet(a).TurnsRemaining()); 2010-11-02T13:40:37 It's a matter of parsing and storing 2010-11-02T13:40:38 decipher: pastebin.com 2010-11-02T13:40:41 kk 2010-11-02T13:40:42 Should be easy 2010-11-02T13:40:46 ill use paste bin 2010-11-02T13:41:03 http://pastebin.com/t68J7nBm 2010-11-02T13:41:04 yeah, then we get some indentation as well 2010-11-02T13:41:49 that's a lot of ()s 2010-11-02T13:41:55 gotta love the starter pack 2010-11-02T13:42:26 TurnsRemaining should never be 0, as the engine would have applied it already 2010-11-02T13:42:52 i don't see anything wrong with it but i don't remember how the starter pack classes work 2010-11-02T13:43:09 your problem is in turns remaining anyway not that 2010-11-02T13:43:26 paste in your parsing function 2010-11-02T13:44:00 not parse - your wrapper function you mentioned 2010-11-02T13:44:32 unless you're taking it directly from parse lol 2010-11-02T13:44:33 no what maybe im writing to my file stupidly 2010-11-02T13:44:39 no what 2010-11-02T13:44:42 *** JamesGecko_ has quit IRC (Ping timeout: 245 seconds) 2010-11-02T13:44:42 know what 2010-11-02T13:44:47 i must be parsing it wrong 2010-11-02T13:44:49 lol 2010-11-02T13:45:21 write parse functions once and never touch again is my advice lol 2010-11-02T13:45:29 i have a function that calculates exactly how many ships you need to leave on if you have multiple planets attacking you i needed to store each fleet turns remaning 2010-11-02T13:45:50 let me fix it up and ill get back to you 2010-11-02T13:45:56 *** Itkovian has joined #aichallenge 2010-11-02T13:46:00 ooh that sounds like fun in a stateless system that recreates itself every turn hehe 2010-11-02T13:47:19 decipher: thats what im doing 2010-11-02T13:47:32 but i've not tested it yet 2010-11-02T13:47:40 well its "vaguely" what im doing 2010-11-02T13:47:45 you didn't rewrite it raed? 2010-11-02T13:47:58 rewrite it? 2010-11-02T13:48:05 i have this white board you should see this algorithmn 2010-11-02T13:48:08 you are modifying the starter pack or your own code 2010-11-02T13:48:18 my own code, there's no starter pack for C ^^ 2010-11-02T13:48:25 well c++ i guess 2010-11-02T13:48:39 but yea i did too 2010-11-02T13:48:40 yea, but you have to rewrite the C++ code because it uses the c++ stdlib 2010-11-02T13:48:50 no classes at all 2010-11-02T13:48:52 :P 2010-11-02T13:49:04 just a bunch of structs in my code 2010-11-02T13:49:04 i have an individual leave behind percentage for every planet i own, if it can defend it self then it leaves the min amount of planets on and uses teh rest to attack else it attacks some other planet and scatters leaving only a few ships to grow and absorb their fleets 2010-11-02T13:49:20 :D 2010-11-02T13:49:22 decipher: if you're that serious i'd create your own planet structs 2010-11-02T13:49:26 yea, mine is more precisely defined than that 2010-11-02T13:49:29 im hoping it will work ok 2010-11-02T13:49:49 dem i did but in java there objects not structs 2010-11-02T13:50:00 decipher: classes objects structs whatever you wanna call them ;) 2010-11-02T13:50:05 ahahaha 2010-11-02T13:50:09 :P 2010-11-02T13:50:31 its killing my brain thou because so much lil tweaking 2010-11-02T13:50:41 the starter pack is torture to modify for me 2010-11-02T13:50:47 clean slate every time 2010-11-02T13:51:13 test out your theories on starter then rewrite with the ones that work :) 2010-11-02T13:51:24 btw since im here i might as well ask you guys, whats the best way of testing a java jar my way is the worse im individually going map by map bot by bot investigating the results i wanna play a 1000 matches so i can get better statistics 2010-11-02T13:51:48 take the source for one of the game engines and attach your bot as a class 2010-11-02T13:51:53 the java game engines obv 2010-11-02T13:52:01 just call it directly instead of sandboxing 2010-11-02T13:52:20 oh ok modify the game engine it self 2010-11-02T13:52:26 i would since it's already java 2010-11-02T13:52:30 shouldn't be that hard 2010-11-02T13:52:51 the way i did it to avoid changing parsing functions is just make a static string for input 2010-11-02T13:52:57 what kinda ai heuristic you doing or meta heurisitic 2010-11-02T13:53:03 and have the sandbox change that 2010-11-02T13:53:04 yeah 2010-11-02T13:53:17 the parser gets the string and what put it there is irrelevant 2010-11-02T13:54:31 decipher: they're al heuristic afaik 2010-11-02T13:54:34 all* 2010-11-02T13:55:16 afaik? 2010-11-02T13:55:35 as far as i know 2010-11-02T13:55:44 oh 2010-11-02T13:55:48 some stubby attempts at minimax with heuristic eval 2010-11-02T13:55:50 hmm new lingo i didn't know ahha 2010-11-02T13:56:05 the new #1 guy might have something better tho ;) 2010-11-02T13:56:06 lol :D sounds like my prog kinda 2010-11-02T13:56:32 my turns is still giving zero let me post little more 2010-11-02T13:57:25 just paste all relevant bits into pastebin 2010-11-02T13:57:33 i guarantee you nobody is gonna steal it here hehe 2010-11-02T13:58:57 http://pastebin.com/Tss629GK 2010-11-02T13:59:04 the bottom is where i write to file 2010-11-02T13:59:25 test is to just grab one of my objects values and print it to file to see what turns returns 2010-11-02T14:01:20 so what is it doing 2010-11-02T14:01:27 i don't even see turnsremaining now lol 2010-11-02T14:01:51 remind us what the problem is again? 2010-11-02T14:03:12 ahha 2010-11-02T14:03:13 ok 2010-11-02T14:03:27 so ignore anythign that says owner == 1 this is all bout the enemy right now 2010-11-02T14:04:18 im going through all teh enemy fleets storing a new object an array with its desitinatin,source,ships,and remaining turns, then i calculate a leave behind percentage that should leave exactly how many ships to defend my self, but turns when i print it and use it comes out as zero 2010-11-02T14:04:34 decipher: I would suggest focussing on individual maps and tweaking for perfect play on them. 2010-11-02T14:04:53 you must make sure to pick 4 or 5 very different maps. 2010-11-02T14:05:14 I use map 1, 5, 7, 30 adn 64. 2010-11-02T14:05:28 decipher: so you're storing an Attack class object in enemyDest[enemyCount] = att right? 2010-11-02T14:05:35 yeah thats why i need the turns function based on size of map if the player rushing me i need ot know how far the fleet away from me so i can leave a certain amount of ships on and send the rest 2010-11-02T14:05:43 yeah dem 2010-11-02T14:05:43 i don't know what an Attack object is is that standard starter? 2010-11-02T14:06:00 its a new class i made ill paste 2010-11-02T14:06:25 http://pastebin.com/z5fttfyZ 2010-11-02T14:06:46 its a wrapper object allows me to have one value that stores many 2010-11-02T14:06:52 like a struct 2010-11-02T14:06:55 btw you don't need to declare Attack att = new Attack() then go enemyDest[blah] = att 2010-11-02T14:07:03 you can just declare it directly 2010-11-02T14:07:07 ie enemydest[blah] = new attack() 2010-11-02T14:07:27 yeah i do 2010-11-02T14:07:37 heh 2010-11-02T14:07:39 that was easy 2010-11-02T14:07:41 turnsLeft = 0; 2010-11-02T14:07:42 in java i need to make a new instance of the object or it will just overwrite 2010-11-02T14:07:46 should be turnsLeft = t 2010-11-02T14:07:55 ooo where 2010-11-02T14:07:58 line 16 2010-11-02T14:08:20 wtf 2010-11-02T14:08:26 why does my default constructor 2010-11-02T14:08:26 bhasker_: ur there? 2010-11-02T14:08:28 have 0 2010-11-02T14:08:33 and so does my main constructor 2010-11-02T14:08:39 EVERYBODY GIVE DEM A AROUND APPLAUSE 2010-11-02T14:08:42 did you apply my bot yet? 2010-11-02T14:08:44 yup 2010-11-02T14:09:07 thanks man you might of just saved me stupid amount of hours and writing to files debugging shit lol 2010-11-02T14:09:09 default constructor is fine but i don't know why it's necessary 2010-11-02T14:09:15 its not 2010-11-02T14:09:15 ok someone pointed out line 16, lol 2010-11-02T14:09:17 java defaults ints to 0 anyway doesn't it 2010-11-02T14:09:19 just good programming practice 2010-11-02T14:09:31 lol i cant believe i had line 16 2010-11-02T14:09:33 omg 2010-11-02T14:09:37 thank god i posted that 2010-11-02T14:09:46 i wrote that class at 6 in the morning 2010-11-02T14:10:08 np try figuring out why one number goes wrong only on turn 47 in c++ structs with internal arrays 2010-11-02T14:10:12 void beat_bocksimacko(void) { disable(nuclear_disarmament); prepare_nuke(22); int i; for(i = 1; i < 23; i++) nuke_planet(i); } 2010-11-02T14:10:31 *** Rubicon-|-Cross has quit IRC (Quit: Leaving) 2010-11-02T14:10:41 hmm you just nuked your own planets? 2010-11-02T14:10:46 unless you doing that at start 2010-11-02T14:10:51 in which case he doesn't have 22 planets 2010-11-02T14:10:53 im assuming planet 0 is my planet 2010-11-02T14:10:55 :P 2010-11-02T14:11:09 bhasker_: whats the outcome? :) 2010-11-02T14:11:25 uncertain 2010-11-02T14:11:26 bhasker_: i get rid of the neutrals too 2010-11-02T14:11:29 :P 2010-11-02T14:11:34 he could launch and take over planet 0 2010-11-02T14:11:37 in which case you still lose 2010-11-02T14:11:42 so you have to nuke everything 2010-11-02T14:11:43 ah, but i still have 100 fleets there 2010-11-02T14:11:46 so that ships have no where to go 2010-11-02T14:11:55 only i have the nuclear arsenal on my planet 2010-11-02T14:11:56 but he owned 22 2010-11-02T14:11:57 :P 2010-11-02T14:12:03 this is from the start 2010-11-02T14:12:05 :P 2010-11-02T14:12:22 *** fmeyer has quit IRC (Ping timeout: 255 seconds) 2010-11-02T14:12:23 heh 2010-11-02T14:12:30 there might be a debate about unfair starting advantage, but i guess its justified against bocksimacko 2010-11-02T14:12:34 ^^ 2010-11-02T14:12:38 *** fawek has joined #aichallenge 2010-11-02T14:13:07 ok i should go home now 2010-11-02T14:13:09 cya 2010-11-02T14:13:23 Top 10 players: bocsimacko(4304), jimrogerz(4243), dmj111(4238), Hazard(4170), shangas(4128), george(4110), felixcoto(4071), FlameN(4059), Boojum(4057), Accoun(4056) 2010-11-02T14:13:35 *** Itkovian has quit IRC (Quit: Itkovian) 2010-11-02T14:13:39 heh 2010-11-02T14:19:18 *** fmeyer has joined #aichallenge 2010-11-02T14:20:28 *** MrG_ has quit IRC (Quit: Page closed) 2010-11-02T14:24:01 quiz time: http://72.44.46.68/canvas?game_id=824026 2010-11-02T14:24:15 was going for the 55+5 in the center the right thing to do? 2010-11-02T14:24:33 disregarding my lack of funnel code... 2010-11-02T14:29:06 *** decipher has quit IRC (Quit: Page closed) 2010-11-02T14:31:23 *** Naktibalda has quit IRC (Quit: ChatZilla 0.9.86 [Firefox 3.6.8/20100722155716]) 2010-11-02T14:32:12 *** rebelxt has joined #aichallenge 2010-11-02T14:33:17 mega1: ping 2010-11-02T14:33:24 hi 2010-11-02T14:33:52 mega1: hey there 2010-11-02T14:34:01 wonder if you ever been asked why your bot starts sending fleets in somewhat random direction towards the end of the game 2010-11-02T14:34:37 wondering if it's fireworks mode or a side effect of ur algorithm ;-) 2010-11-02T14:34:57 because the evaluation function doesn't care anymore 2010-11-02T14:35:33 haha, got it 2010-11-02T14:35:39 McLeopold, without knowing the exact distances, my gut feeling is that it's a bad move. 2010-11-02T14:35:53 yea and you never sent any further ships to it 2010-11-02T14:36:02 i think if you had been streaming to it 2010-11-02T14:36:05 i have something similar, except for i don't take any action if score is below certain threshold.. 2010-11-02T14:36:05 you might have held it 2010-11-02T14:37:20 *** Florian95 has quit IRC (Quit: Leaving...) 2010-11-02T14:38:56 yeah, I think I could have held if I was streaming 2010-11-02T14:40:50 *** fmeyer has quit IRC (Ping timeout: 252 seconds) 2010-11-02T14:41:19 I think the best first move, is to move all 100 to the 2 planet, to position yourself for countering whatever the opponent does 2010-11-02T14:41:52 but difficult move for a bot to find probably 2010-11-02T14:42:26 HEI KEBABILLE NYT 2010-11-02T14:42:30 oops, sry 2010-11-02T14:42:33 wrong channel 2010-11-02T14:43:52 The middle might be too expensive for that. 2010-11-02T14:44:03 Now, I'm thinking I couldn't have held. 2010-11-02T14:46:03 Nope, I couldn't. And I now know why my bot made the mistake... 2010-11-02T14:47:59 *** rebelxt has left #aichallenge 2010-11-02T15:04:01 *** Florian95 has joined #aichallenge 2010-11-02T15:09:01 *** Florian95 has quit IRC (Ping timeout: 264 seconds) 2010-11-02T15:13:23 Top 10 players: bocsimacko(4330), jimrogerz(4243), dmj111(4227), Hazard(4166), shangas(4120), george(4096), felixcoto(4068), Boojum(4061), Accoun(4054), FlameN(4050) 2010-11-02T15:16:18 moving to 2 planet will have spent 2 turns in flight in which you can't counter 2010-11-02T15:16:57 oh cool, I'm no longer the only go bot: http://ai-contest.com/language_profile.php?lang=Go 2010-11-02T15:18:06 *** hornairs has quit IRC (Read error: No route to host) 2010-11-02T15:19:43 *** perror has quit IRC (Quit: Bye all !) 2010-11-02T15:20:12 *** RainCT has joined #aichallenge 2010-11-02T15:22:46 lavalamp_, thought of releasing a go - starter pack, somewhere ? 2010-11-02T15:24:04 p4p4, http://bitbucket.org/lavalamp/pw_go_starter_package 2010-11-02T15:24:24 ;) 2010-11-02T15:25:02 It's pretty much a direct port of the C++ one, warts and all 2010-11-02T15:28:59 *** Bobng has quit IRC (Quit: Leaving) 2010-11-02T15:29:42 *** Bobng has joined #aichallenge 2010-11-02T15:34:47 *** boegel has joined #aichallenge 2010-11-02T15:36:55 *** rebelxt has joined #aichallenge 2010-11-02T15:37:24 haha that page can't do C++ 2010-11-02T15:37:36 or C# either 2010-11-02T15:41:32 it is tempting to do the full conversion to C to be in the elite C club lol 2010-11-02T15:41:36 not hard for me lol 2010-11-02T15:42:09 *** narnach_mb has quit IRC (Quit: Leaving.) 2010-11-02T15:44:42 *** javagamer has quit IRC (Ping timeout: 245 seconds) 2010-11-02T15:47:27 *** Appleman1234 has joined #aichallenge 2010-11-02T15:47:55 *** fmeyer has joined #aichallenge 2010-11-02T15:51:33 *** boegel has quit IRC (Quit: This computer has gone to sleep) 2010-11-02T15:52:25 hmm why is 'this' in C++ a const* instead of a & 2010-11-02T15:53:08 oh nm: Early versions of C++ did not include references and it has been suggested that had they been so in C++ from the beginning, this would have been a reference, not a pointer 2010-11-02T15:54:03 *** Palmik has quit IRC (Remote host closed the connection) 2010-11-02T15:55:24 demerzel_: C++ is really really badly designed, its got feature tack-on 2010-11-02T15:56:40 *** yasith has quit IRC (Read error: Connection reset by peer) 2010-11-02T15:57:03 *** rebelxt has left #aichallenge 2010-11-02T15:57:03 raedwulf: you need a paradigm shift hahahaha 2010-11-02T15:57:08 i am really starting to hate that site 2010-11-02T15:57:32 q:why use x? a: because it is the C++ WAY 2010-11-02T15:59:00 demerzel_: i like functional programming paradigm 2010-11-02T15:59:19 i do like oop but it gets ridiculous sometimes 2010-11-02T15:59:25 if you can oop without performance issues do it 2010-11-02T15:59:25 too much bloat 2010-11-02T15:59:29 yea 2010-11-02T15:59:41 i was planning that with this but i wanted to inline too much 2010-11-02T15:59:58 and have backreferences to parent classes 2010-11-02T16:00:09 probably i should have skipped adding back references 2010-11-02T16:00:17 *** hornairs has joined #aichallenge 2010-11-02T16:00:29 it would have worked, but the amount of boilerplate code was too much 2010-11-02T16:00:31 lol 2010-11-02T16:01:06 *** p4p4 has quit IRC (Quit: ChatZilla 0.9.84 [SeaMonkey 2.0a3/20090223135443]) 2010-11-02T16:01:11 my C is usually very concise, because i know quite a few tricks and workarounds with it :P 2010-11-02T16:01:24 *** wh1teside_ has quit IRC (Ping timeout: 265 seconds) 2010-11-02T16:01:45 i've grown found of C# though 2010-11-02T16:01:50 fond* 2010-11-02T16:02:27 yea i like it too actually ;) 2010-11-02T16:02:40 hey can you do sprintf without and intermediate char* 2010-11-02T16:02:51 ie fprintf(magicFunc("%d",intval)) 2010-11-02T16:04:07 *** Florian95 has joined #aichallenge 2010-11-02T16:04:14 erm, you need to stick it in a buffer somewhere 2010-11-02T16:04:30 lol nm got my mind all twisted around 2010-11-02T16:04:31 im confused with your code there 2010-11-02T16:04:38 fprintf is fine all by itself 2010-11-02T16:05:04 yea, fprintf doesn't need an intermediate buffer, sprintf does 2010-11-02T16:05:08 thats the whole point of it... 2010-11-02T16:05:10 lol 2010-11-02T16:07:39 man mono has really come along 2010-11-02T16:07:46 its performance is not half bad either 2010-11-02T16:08:08 definite candidate to have a look more at 2010-11-02T16:10:04 but im also following D, thats a nice language 2010-11-02T16:10:22 *** Florian95 has quit IRC (Remote host closed the connection) 2010-11-02T16:13:24 Top 10 players: bocsimacko(4323), dmj111(4202), jimrogerz(4197), Hazard(4141), shangas(4106), george(4087), Boojum(4071), felixcoto(4053), FlameN(4040), Accoun(4038) 2010-11-02T16:14:54 *** fmeyer has quit IRC (Ping timeout: 260 seconds) 2010-11-02T16:16:39 *** sigh has joined #aichallenge 2010-11-02T16:20:59 *** airhorns has joined #aichallenge 2010-11-02T16:21:14 *** Accoun has quit IRC () 2010-11-02T16:22:15 *** hornairs has quit IRC (Ping timeout: 240 seconds) 2010-11-02T16:28:16 *** Bobng has quit IRC (Ping timeout: 255 seconds) 2010-11-02T16:30:06 *** hellman has quit IRC (Remote host closed the connection) 2010-11-02T16:35:15 *** Bobng has joined #aichallenge 2010-11-02T16:36:18 *** Accoun has joined #aichallenge 2010-11-02T16:39:59 *** Frontier has joined #aichallenge 2010-11-02T16:41:13 *** fmeyer has joined #aichallenge 2010-11-02T16:43:59 *** Itkovian has joined #aichallenge 2010-11-02T16:44:56 *** Appleman1234 has quit IRC (Read error: Connection reset by peer) 2010-11-02T16:50:14 *** Appleman1234 has joined #aichallenge 2010-11-02T16:54:03 *** sigh has quit IRC (Remote host closed the connection) 2010-11-02T17:01:31 *** bhasker has joined #aichallenge 2010-11-02T17:02:37 *** Mathnerd314 has joined #aichallenge 2010-11-02T17:03:27 *** bhasker__ has joined #aichallenge 2010-11-02T17:03:40 *** bhasker_ has quit IRC (Ping timeout: 265 seconds) 2010-11-02T17:05:37 *** bhasker has quit IRC (Ping timeout: 240 seconds) 2010-11-02T17:12:49 *** Bobng has quit IRC (Ping timeout: 255 seconds) 2010-11-02T17:13:24 Top 10 players: bocsimacko(4339), dmj111(4235), jimrogerz(4191), Hazard(4156), shangas(4116), george(4091), felixcoto(4068), Accoun(4066), Boojum(4061), FlameN(4056) 2010-11-02T17:14:10 *** narnach has quit IRC (Quit: Leaving.) 2010-11-02T17:18:08 *** narnach has joined #aichallenge 2010-11-02T17:26:03 *** Bobng has joined #aichallenge 2010-11-02T17:39:23 *** wh1teside has joined #aichallenge 2010-11-02T17:39:56 *** Naktibalda has joined #aichallenge 2010-11-02T17:41:20 *** TravisD has joined #aichallenge 2010-11-02T17:41:44 *** Cold-Phoenix has joined #aichallenge 2010-11-02T17:49:54 *** airhorns has quit IRC (Ping timeout: 240 seconds) 2010-11-02T17:56:06 \j #math 2010-11-02T17:56:17 *** Appleman1234 has quit IRC (Ping timeout: 272 seconds) 2010-11-02T17:56:18 sorry 2010-11-02T17:59:32 *** demerzel_ has quit IRC (Quit: Page closed) 2010-11-02T17:59:42 *** sigh has joined #aichallenge 2010-11-02T18:00:19 *** Azrathud has joined #aichallenge 2010-11-02T18:00:21 *** demerzel_ has joined #aichallenge 2010-11-02T18:00:37 *** Itkovian has quit IRC (Quit: Itkovian) 2010-11-02T18:00:55 *** sigh has quit IRC (Read error: Connection reset by peer) 2010-11-02T18:01:15 *** sigh has joined #aichallenge 2010-11-02T18:10:10 *** JamesMG has quit IRC (Quit: Leaving.) 2010-11-02T18:12:32 *** lavalamp_ has quit IRC (Quit: Page closed) 2010-11-02T18:13:04 *** Sarin has quit IRC (Quit: Verlassend) 2010-11-02T18:13:24 Top 10 players: bocsimacko(4344), dmj111(4221), jimrogerz(4171), Hazard(4150), shangas(4121), _Astek_(4118), george(4086), felixcoto(4062), Accoun(4059), FlameN(4053) 2010-11-02T18:13:41 *** jukkerknott has joined #aichallenge 2010-11-02T18:18:17 *** Appleman1234 has joined #aichallenge 2010-11-02T18:19:46 *** wh1teside has quit IRC (Read error: Connection reset by peer) 2010-11-02T18:21:42 *** Itkovian has joined #aichallenge 2010-11-02T18:23:59 *** wh1teside has joined #aichallenge 2010-11-02T18:32:40 *** jukkerknott has left #aichallenge ("Leaving") 2010-11-02T18:34:08 *** Mathnerd314 has quit IRC (Disconnected by services) 2010-11-02T18:34:51 *** Mathnerd314_ has joined #aichallenge 2010-11-02T18:34:53 *** Mathnerd314_ is now known as Mathnerd314 2010-11-02T18:35:01 *** mightybyte has joined #aichallenge 2010-11-02T18:40:34 *** Commandir has left #aichallenge 2010-11-02T18:42:28 @predict 2010-11-02T18:42:28 I predict that fawek will win! 2010-11-02T18:43:01 @roulette 2010-11-02T18:43:01 TravisD: *click* 2010-11-02T18:48:40 lol my sniping code was waaaaaay to optimistic it seems 2010-11-02T18:49:39 *** Appleman1234 has quit IRC (Quit: Leaving) 2010-11-02T18:49:40 *** hornairs has joined #aichallenge 2010-11-02T18:50:51 *** rebelxt has joined #aichallenge 2010-11-02T18:52:12 *** demerzel_ has quit IRC (Quit: Page closed) 2010-11-02T18:53:04 *** airhorns has joined #aichallenge 2010-11-02T18:56:14 *** hornairs has quit IRC (Ping timeout: 240 seconds) 2010-11-02T18:59:23 *** RainCT has quit IRC (Remote host closed the connection) 2010-11-02T19:00:34 *** mceier has quit IRC (Quit: leaving) 2010-11-02T19:00:53 *** bhasker has joined #aichallenge 2010-11-02T19:03:39 *** bhasker__ has quit IRC (Ping timeout: 252 seconds) 2010-11-02T19:04:28 *** Bobng has quit IRC (Quit: Leaving) 2010-11-02T19:08:50 *** rebelxt has left #aichallenge 2010-11-02T19:09:56 *** zerd has quit IRC (Remote host closed the connection) 2010-11-02T19:10:04 *** zerd has joined #aichallenge 2010-11-02T19:12:40 *** perestrelka has quit IRC (Ping timeout: 250 seconds) 2010-11-02T19:13:25 Top 10 players: bocsimacko(4361), dmj111(4250), Hazard(4162), jimrogerz(4150), shangas(4140), _Astek_(4134), george(4120), Accoun(4090), FlameN(4081), felixcoto(4076) 2010-11-02T19:15:32 *** perestrelka has joined #aichallenge 2010-11-02T19:15:34 *** fawek has quit IRC (Ping timeout: 240 seconds) 2010-11-02T19:43:33 *** JamesGecko_ has joined #aichallenge 2010-11-02T19:44:41 *** Cold-Phoenix has quit IRC () 2010-11-02T19:47:55 *** davidjliu has joined #aichallenge 2010-11-02T19:48:02 *** Mathnerd314 has quit IRC (Disconnected by services) 2010-11-02T19:48:49 *** Mathnerd314_ has joined #aichallenge 2010-11-02T19:48:51 *** Mathnerd314_ is now known as Mathnerd314 2010-11-02T19:50:22 @roulette 2010-11-02T19:50:22 McLeopold_: *click* 2010-11-02T19:50:28 @roulette 2010-11-02T19:50:28 McLeopold_: *click* 2010-11-02T19:50:32 @roulette 2010-11-02T19:50:32 *** McLeopold_ was kicked by contestbot (BANG!) 2010-11-02T19:50:33 * contestbot reloads and spins the chambers. 2010-11-02T19:55:38 *** demerzel_ has joined #aichallenge 2010-11-02T19:56:28 *** hornairs has joined #aichallenge 2010-11-02T19:58:26 *** airhorns has quit IRC (Ping timeout: 245 seconds) 2010-11-02T19:59:28 lol 2010-11-02T19:59:45 *** lavalamp_ has joined #aichallenge 2010-11-02T20:02:50 *** airhorns has joined #aichallenge 2010-11-02T20:05:16 @roulette 2010-11-02T20:05:16 TravisD: *click* 2010-11-02T20:05:34 @roulette 2010-11-02T20:05:34 *** Mathnerd314 was kicked by contestbot (BANG!) 2010-11-02T20:05:35 * contestbot reloads and spins the chambers. 2010-11-02T20:05:37 *** Mathnerd314 has joined #aichallenge 2010-11-02T20:05:43 * Mathnerd314 celebrates 2010-11-02T20:05:57 *** hornairs has quit IRC (Ping timeout: 240 seconds) 2010-11-02T20:06:45 @roulette 2010-11-02T20:06:45 Mathnerd314: *click* 2010-11-02T20:09:09 @roulette 2010-11-02T20:09:09 amstan: *click* 2010-11-02T20:09:22 @roulette 2010-11-02T20:09:22 *** TravisD was kicked by contestbot (BANG!) 2010-11-02T20:09:23 * contestbot reloads and spins the chambers. 2010-11-02T20:09:27 *** TravisD has joined #aichallenge 2010-11-02T20:09:28 :( 2010-11-02T20:09:38 lol 2010-11-02T20:09:57 fun fun fun :) 2010-11-02T20:10:08 !static const void blah(void*, (blah2&)*); 2010-11-02T20:10:17 @roulette 2010-11-02T20:10:17 phreeza: *click* 2010-11-02T20:10:17 i broke it 2010-11-02T20:10:46 lol a pointer to a ref :) 2010-11-02T20:10:55 i am having a contest with myself to see how many levels of indirection and keywords i can preface for functions 2010-11-02T20:11:13 i am winning 2010-11-02T20:11:42 actually i just redefined a type cuz of compiler error 2010-11-02T20:11:49 turned out i just missed a semicolon after a struct 2010-11-02T20:11:55 amazing how many errors that made 2010-11-02T20:12:06 ah yes... 2010-11-02T20:12:29 clang is supposed to give really good compiler errors 2010-11-02T20:12:34 anyone tried that? 2010-11-02T20:13:26 Top 10 players: bocsimacko(4356), dmj111(4241), jimrogerz(4174), Hazard(4142), shangas(4126), george(4111), bix0r4ever(4081), Accoun(4080), FlameN(4071), felixcoto(4065) 2010-11-02T20:13:47 demerzel_: try this for inticacy ;) void (*signal(int, void (*fp)(int)))(int) 2010-11-02T20:14:04 email the guy at parashift.com 2010-11-02T20:14:09 *** jaspervdj has quit IRC (Quit: NEVER GONNA GIVE YOU UP NEVER GONNA LET YOU DOWN) 2010-11-02T20:14:12 you'll probably get a 5-page rant about your shitty coding style 2010-11-02T20:14:29 wish i could exclude sites from search results lol 2010-11-02T20:14:31 signal is a function passing an int and a pointer to a function passing an int returning nothing (void) returning a pointer to a function passing an int returning nothing (void) 2010-11-02T20:14:37 not my code ;) 2010-11-02T20:14:54 i understand pointer indirection to n levels 2010-11-02T20:14:59 i get lost at references 2010-11-02T20:15:00 *** superflit has quit IRC (Quit: superflit) 2010-11-02T20:15:14 *** mightybyte has left #aichallenge 2010-11-02T20:15:14 its from a page explaining how to actually read shit like that 2010-11-02T20:15:16 http://c-faq.com/decl/spiral.anderson.htm 2010-11-02T20:15:43 i think when i have a running bot i'll fully convert it to C 2010-11-02T20:15:49 only like a dozen people on that list lol 2010-11-02T20:16:06 only thing i use from c++ is references and classes - easily changed to pointers and structs 2010-11-02T20:17:17 can c structs have member functions? i'm not even sure 2010-11-02T20:17:20 been a while 2010-11-02T20:17:27 yup 2010-11-02T20:18:00 i get that c++ is easier on devs but the reason i love it is because i can use oop easier while still using all the same c shit 2010-11-02T20:18:04 sure? 2010-11-02T20:18:07 sites like parashift annoy me 2010-11-02T20:18:26 yea i am sure 2010-11-02T20:18:34 don't yoiu have to do something slightly contrived like functiuon pointers? 2010-11-02T20:18:54 oh wait no you're right 2010-11-02T20:19:11 namespaces and then the "functions" are pointers to globals in namespaces 2010-11-02T20:19:16 been a while 2010-11-02T20:19:21 :) 2010-11-02T20:19:35 ah yea that sounds right 2010-11-02T20:20:30 i obv prefer member functions but the chance to be esoteric for a few function pointers is sounds worthwhile lol 2010-11-02T20:21:39 i was gonna try lisp but my bot sucks too bad to go running off to a completely new lang 2010-11-02T20:34:56 left hand brewing co's milk stout = very good 2010-11-02T20:38:25 *** janzert has left #aichallenge 2010-11-02T20:39:29 *** janzert has joined #aichallenge 2010-11-02T20:39:45 *** javagamer has joined #aichallenge 2010-11-02T20:41:27 *** bhasker_ has joined #aichallenge 2010-11-02T20:41:54 *** phreeza has quit IRC (Quit: Computer has gone to sleep) 2010-11-02T20:42:42 *** bhasker has quit IRC (Ping timeout: 240 seconds) 2010-11-02T20:42:47 uh oh: moves available: 0 2010-11-02T20:42:53 guess it doesn't work haha 2010-11-02T20:43:15 *** fmeyer has quit IRC (Ping timeout: 255 seconds) 2010-11-02T20:47:22 *** bhasker has joined #aichallenge 2010-11-02T20:49:15 *** bhasker_ has quit IRC (Ping timeout: 240 seconds) 2010-11-02T21:11:53 *** Gudradain has joined #aichallenge 2010-11-02T21:12:39 *** phreeza has joined #aichallenge 2010-11-02T21:13:25 Top 10 players: bocsimacko(4356), dmj111(4230), jimrogerz(4206), Hazard(4152), shangas(4134), george(4133), bix0r4ever(4084), Accoun(4082), felixcoto(4077), FlameN(4073) 2010-11-02T21:15:01 *** Mathnerd314 has quit IRC (Ping timeout: 264 seconds) 2010-11-02T21:16:53 *** bhasker has quit IRC (Quit: bhasker) 2010-11-02T21:18:46 *** alch has quit IRC (Ping timeout: 250 seconds) 2010-11-02T21:22:05 just a question like that : if your opponent is doing nothing. Is there an optimal strategy to take all the neutral on the map? 2010-11-02T21:24:58 0 35 261934 is a perfectly valid move damnit! 2010-11-02T21:25:17 Gudradain: it depends on if you're winning ;) 2010-11-02T21:35:33 if your opponent is doing nothing why do you want to take all neutrals? 2010-11-02T21:35:37 take them out and be done with it 2010-11-02T21:53:49 *** TravisD has quit IRC (Quit: Leaving.) 2010-11-02T22:07:23 *** wh1teside_ has joined #aichallenge 2010-11-02T22:08:05 *** wh1teside has quit IRC (Quit: Lost terminal) 2010-11-02T22:08:14 *** wh1teside_ is now known as wh1teside 2010-11-02T22:13:26 Top 10 players: bocsimacko(4350), dmj111(4220), jimrogerz(4200), Hazard(4135), george(4128), shangas(4119), bix0r4ever(4113), Accoun(4076), felixcoto(4069), FlameN(4068) 2010-11-02T22:16:39 *** Mathnerd314 has joined #aichallenge 2010-11-02T22:20:53 *** airhorns has quit IRC (Quit: airhorns) 2010-11-02T22:25:29 *** bhasker has joined #aichallenge 2010-11-02T22:26:20 evening 2010-11-02T22:34:28 On 3 of 8 test maps my current bot totally schools my old one 2010-11-02T22:34:41 on the other 5 it plays horribly... 2010-11-02T22:37:16 heh, I've had that before 2010-11-02T22:38:38 really hard to find strategies that work all the time 2010-11-02T22:39:42 more like not all maps have neutrals that trigger my horrible horrible bugs 2010-11-02T22:40:00 0 growth neutrals? :P 2010-11-02T22:40:03 more like time to make a hybrid bot 2010-11-02T22:40:22 Zannick: that just introduces new problems :P 2010-11-02T22:40:32 so...? 2010-11-02T22:40:34 >.> 2010-11-02T22:40:43 :) 2010-11-02T22:41:06 btw, lavalamp_, is dailyjoseki.com your site? 2010-11-02T22:41:12 no, I have a line to ignore +0 neutrals 2010-11-02T22:41:17 sigh: yes 2010-11-02T22:41:45 are you on the go (lang) mailing list or the go (game) forum? :) 2010-11-02T22:42:33 nah, I saw the post in the forum about the go language and had a look at the starter pack and saw your email :P 2010-11-02T22:42:41 I am totally neglecting my website for this, hehe 2010-11-02T22:42:42 ah ok 2010-11-02T22:42:53 so is your site like a joseki dictionary? 2010-11-02T22:43:13 yes, I build a tree from pro games 2010-11-02T22:43:49 cool, have you seen kombilo, it can do something similar 2010-11-02T22:45:25 sigh: I explain a little bit about it here: http://www.lifein19x19.com/forum/viewtopic.php?p=22211 2010-11-02T22:45:48 My site is cooler than kombilo ^^ 2010-11-02T22:46:15 it answers the question "when pros play this what does the rest of the board look like" at a glance 2010-11-02T22:46:25 yeah, that's cool :) 2010-11-02T22:46:38 I take it you must play go? 2010-11-02T22:47:03 yeah, although I've not played much recently >.> 2010-11-02T22:50:30 how does your site decide when to stop displaying moves? 2010-11-02T22:50:45 When no two pros play the same thing 2010-11-02T22:51:03 hmm... but that's not necessarily the end of the line 2010-11-02T22:51:26 no... and I have some features scheduled to show a few more moves after that 2010-11-02T22:51:46 ah, cool 2010-11-02T22:51:51 but not implemented yet and definitely won't be this month... 2010-11-02T22:52:10 hm, I think I broke something 2010-11-02T22:53:36 move selection is tricky business :( 2010-11-02T22:54:10 there's too many of them :( 2010-11-02T22:56:09 that's not my issue 2010-11-02T22:56:18 it's so hard to even pick out moves and play them against each other 2010-11-02T22:56:39 what do you mean, exactly? 2010-11-02T22:56:46 do i pick a move for a destination planet entirely into the future, or just to when i'd want to take it, and then it's just plain messy to work out claiming it into the future 2010-11-02T22:57:06 that's if the ownership is currently expected to change multiple times 2010-11-02T22:57:16 that sounds confusing 2010-11-02T22:57:52 i have no idea how to do 2 way combat resolution for future states and still get end states that evaluate as i'd like 2010-11-02T22:57:54 lavalamp: are you lavalamp on KGS too? 2010-11-02T22:58:17 i can just stop sending voided fleets, but then the game state isn't really showing hte result of the moves to get there 2010-11-02T22:58:56 *** Olathe has quit IRC (Ping timeout: 240 seconds) 2010-11-02T22:59:06 what's hard about combat resolution? 2010-11-02T22:59:23 hmm i have no idea why my bot streams now lol 2010-11-02T22:59:43 that's what happens when you abstract too much haha 2010-11-02T22:59:56 demerzel_: streams? 2010-11-02T23:00:13 http://72.44.46.68/canvas?game_id=830091 2010-11-02T23:00:39 sigh: if have a tree making future moves from both players, one of their future moves might render one of my other future moves useless, so do i just not send them anymore and still consider the "move to have been made" inside that tree 2010-11-02T23:00:57 and "moves" might be made at different times and from different sources, it just has one destination 2010-11-02T23:01:16 inside that node**/branch 2010-11-02T23:01:18 ...that sounds complicated :( 2010-11-02T23:01:31 hence me having no idea how to do it :P 2010-11-02T23:01:49 it's very hard to do future moves from both players and resolve it properly 2010-11-02T23:03:03 I just recalculate the future states for a planet when I add a new fleet... if it becomes a bottleneck then I'll think about more efficient updates 2010-11-02T23:03:43 *recalcuate all future states 2010-11-02T23:04:08 yeah, so do i, but you arne't doing moves from them yet are you? 2010-11-02T23:04:49 in my first layer i might tell myself to send 30 ships from planet 1 at time 14, but then they might send a fleet to planet 1 causing it to have less than 30 ships there at time 14, so i can't actually send that anymore 2010-11-02T23:05:01 antimatroid: I have delayed moves, so sure 2010-11-02T23:05:12 I don't have any lookahead, though 2010-11-02T23:05:42 i'm talking about strictly inside one turn too 2010-11-02T23:05:50 i'm not trying to keep information between turns anymore 2010-11-02T23:05:57 i didn't trust myself not to fuck it up 2010-11-02T23:06:08 I don't keep info between turns either 2010-11-02T23:06:39 but if I decide to make a delayed move, my current state is updated so processing down the track knows about it 2010-11-02T23:07:06 i don't want to know in the future what i planned to do in the past, the game state will be different because of the moves they make 2010-11-02T23:07:14 what i do want to do is model both players when making my decisions 2010-11-02T23:07:49 yeah, that's quite complex 2010-11-02T23:07:52 and i want both players being able to send ships they currently have from any time point 2010-11-02T23:07:59 *** farbe has quit IRC (Remote host closed the connection) 2010-11-02T23:08:22 well, i don't try to take anything past maxD 2010-11-02T23:08:31 (max distance between planets) 2010-11-02T23:09:01 i see 2010-11-02T23:09:02 *** Appleman1234 has joined #aichallenge 2010-11-02T23:09:40 the idea is sound, doing it without bugs is just stupidly hard 2010-11-02T23:09:47 haha 2010-11-02T23:10:36 my current way around it is to pick moves for each player individually, then when i make my zero sum game only implement new moves and run the funal states before evaluating them 2010-11-02T23:10:41 but it's over the top and bleh 2010-11-02T23:11:21 my functions are way too inefficient too, i need to speed things up if i want to look at enough things to make smart moves 2010-11-02T23:13:27 Top 10 players: bocsimacko(4368), jimrogerz(4240), dmj111(4238), Hazard(4153), george(4141), shangas(4135), bix0r4ever(4119), Accoun(4091), FlameN(4085), felixcoto(4079) 2010-11-02T23:15:55 sigh: Yeah, I'm lavalamp on KGS 2010-11-02T23:17:46 *** TravisD has joined #aichallenge 2010-11-02T23:18:46 lavalamp_: cool, I'm "sigh" and "pulsar" on there 2010-11-02T23:19:17 I'll keep an eye out for you if I ever stop working on my bot long enough to play ;) 2010-11-02T23:19:32 if I ever bother to log in :D 2010-11-02T23:20:29 *** Naktibalda has quit IRC (Remote host closed the connection) 2010-11-02T23:21:39 *** virdo has joined #aichallenge 2010-11-02T23:33:13 when looking at the ships the enemy has, do you assume they perfectly predicted your move or responded? 2010-11-02T23:41:30 *** fmeyer has joined #aichallenge 2010-11-02T23:42:23 *** superflit has joined #aichallenge 2010-11-02T23:48:41 antimatroid: not me, I keep trying but t makes my bot crap 2010-11-02T23:49:57 i have some weird weird bugs 2010-11-02T23:50:25 *** fmeyer has quit IRC (Ping timeout: 276 seconds) 2010-11-02T23:50:40 how long do you guys pause between tcp games 2010-11-02T23:50:49 i'm pausing 5 seconds but it doesn't seem to work 2010-11-02T23:50:53 *all the time 2010-11-02T23:52:06 pausing doesn't make much difference 2010-11-02T23:52:19 since the server waits to match you against other folks of similar rating 2010-11-02T23:52:25 yeah, you shouldn't need to pause 2010-11-02T23:52:29 well if i didn't pause it never let me back in immediately 2010-11-02T23:52:36 hmm i have never had that problme 2010-11-02T23:52:40 would hang in the receive state 2010-11-02T23:52:43 i almost always get a game after a few seconds 2010-11-02T23:53:00 oh oh you are using the tcp game manager 2010-11-02T23:53:06 i coded socket directly into the bot 2010-11-02T23:56:33 i am just using the tcp client 2010-11-02T23:56:44 didnt code it into the bot 2010-11-02T23:58:20 hah one attack turns the game in my bot favour 2010-11-02T23:58:20 http://72.44.46.68/canvas?game_id=830846