2011-11-08T00:02:48 What's a pretty basic enemy avoidance? Just stay 3 squares away from other ants? 2011-11-08T00:03:56 yeah 2011-11-08T00:04:00 however that's kind of basic 2011-11-08T00:04:16 I just implemented a system where when an ant is in combat with another one 2011-11-08T00:04:20 it counts nearby allies and nearby enemis 2011-11-08T00:04:25 if allies > enemies, attack 2011-11-08T00:04:29 if enemies > allies, run away 2011-11-08T00:04:38 *** easystar has quit IRC (Quit: Page closed) 2011-11-08T00:04:41 uses dfs with a max cap of 8 distance 2011-11-08T00:04:45 to find allies and enemies 2011-11-08T00:05:11 seems to work well :P 2011-11-08T00:05:25 *** LouisMartin has joined #aichallenge 2011-11-08T00:05:27 What the performance on your dfs? 2011-11-08T00:05:53 I just implemented one for a basic 2d array and i'm getting about 25ms to dfs a 200x200 grid with no obstacles 2011-11-08T00:06:14 not sure 2011-11-08T00:06:19 but it's just a 5 step dfs 2011-11-08T00:06:23 so it's very minimal 2011-11-08T00:06:37 datachomper1: what language is that? 2011-11-08T00:06:38 err I mean bfs 2011-11-08T00:06:42 yea my bfs is fine 2011-11-08T00:06:46 Extrarius: python 2011-11-08T00:06:50 hey, do you know if we can have folders in the source we upload? 2011-11-08T00:06:53 takes maybe about 8 ms per bfs 2011-11-08T00:06:58 I use a lot of bfs maps 2011-11-08T00:07:04 roflmao: for how many squares? 2011-11-08T00:07:11 all map 2011-11-08T00:07:14 *** smiley1983 has quit IRC (Ping timeout: 276 seconds) 2011-11-08T00:07:50 I'm using java which is faster than py 2011-11-08T00:08:06 datachomper1: I have a whole-map bfs that takes about 1ms in python, so there is plenty of room for optimization. 2011-11-08T00:08:34 datachomper1: but even 25 ms is fine if you run it the minimal number of times (once for food, once for enemies, once for hills, once for unseen) 2011-11-08T00:08:36 LouisMartin: it depends on the language 2011-11-08T00:08:39 Extrarius: I thought so, but i'm unsure how I could optimize further 2011-11-08T00:08:49 you can always try it and see if the server can handle it 2011-11-08T00:09:53 janzert: I'm in C# 2011-11-08T00:10:23 really don't know for C# 2011-11-08T00:11:19 janzert: okay, I'll try something 2011-11-08T00:11:32 datachomper1: One thing to realize is that for whole-map bfs, you don't need to do a "real" bfs, you just need to set each cell to the minimum of itself or 1+any of it's neighbors 2011-11-08T00:12:24 I set all cells to some max value (255), then set each cell with food to 0, then iteratively set every cell to min(self, min(neighbors)+1) 2011-11-08T00:12:42 then any ant can pick the minimum nearest cell to go toward food 2011-11-08T00:13:10 Extrarius: But you still have to set your cells in a bfs pattern, right? 2011-11-08T00:13:26 define bfs pattern? 2011-11-08T00:13:38 radiating out from the target cell 2011-11-08T00:14:13 datachomper1: It happens automatically if you set each cell to min(self, min(neighbors)+1) over the whole map a few times 2011-11-08T00:14:40 Ah so you iterate, interesting 2011-11-08T00:15:25 I have a function expandmap that basically does "while bChanged: bChanged = false; for row; for col; value = min(self, min(neighbors)+1); if value < cell: bchanged = true; cell = value 2011-11-08T00:16:08 *** Areks has joined #aichallenge 2011-11-08T00:17:50 and actually, I have a further optimization: I added code to keep track of land (default turn update does't) and walls in a map called 'terrain' then during each turn before doing BFS, i expand that map by pretending the outline of 'unseen' tiles is really 'land'. Then, I only iterate over cells that are marked as 'land' 2011-11-08T00:18:54 *** tmandry has quit IRC (Ping timeout: 240 seconds) 2011-11-08T00:19:56 *** Jak_o_Shadows has quit IRC (Remote host closed the connection) 2011-11-08T00:21:56 *** califax has quit IRC (Ping timeout: 265 seconds) 2011-11-08T00:23:20 Extrarius: I'll have to try that out, what's the max iterations you go through? 2011-11-08T00:30:57 *** caution has quit IRC (Quit: caution) 2011-11-08T00:31:13 http://aichallenge.org/visualizer.php?game=72056&user=284 2011-11-08T00:31:26 that game clearly shows a bug with "rank stabilized" 2011-11-08T00:33:43 uh nevermind 2011-11-08T00:35:59 datachomper1: I think around 15 is the max once I fixed all the bugs, but the usual is more like 6-8 2011-11-08T00:37:28 *** Antimony has quit IRC (Ping timeout: 258 seconds) 2011-11-08T00:39:22 *** GeorgeSebastian has quit IRC (Quit: Leaving) 2011-11-08T00:39:54 *** GeorgeSebastian has joined #aichallenge 2011-11-08T00:42:50 *** ltriant has quit IRC (Quit: Computer has gone to sleep) 2011-11-08T00:43:16 *** roflmao1 has joined #aichallenge 2011-11-08T00:43:17 *** roflmao has quit IRC (Read error: Connection reset by peer) 2011-11-08T00:44:36 A is such a beast 2011-11-08T00:44:43 *** tmandry has joined #aichallenge 2011-11-08T00:44:43 *** tmandry has joined #aichallenge 2011-11-08T00:48:31 *** smiley1983 has joined #aichallenge 2011-11-08T00:54:13 *** smiley1983 has quit IRC (Ping timeout: 252 seconds) 2011-11-08T00:54:34 *** master_ninja has joined #aichallenge 2011-11-08T01:00:02 *** theskumar has joined #aichallenge 2011-11-08T01:00:09 Extrarius: Are you using python? 2011-11-08T01:06:05 yes 2011-11-08T01:06:54 interesting, I just implemented your iterative diffuser with disasterous results :OP 2011-11-08T01:07:09 It worked but a 40x40 grid took 39 iterations and 350ms to complete 2011-11-08T01:07:44 weird that it took so many iterations 2011-11-08T01:08:24 *** thestinger has quit IRC (Quit: sleep) 2011-11-08T01:08:26 I'm probably doing something boneheaded 2011-11-08T01:08:28 when I had some major bugs, I once saw up to like 68 iterations, but since I've worked it all out, I haven't seen more than 15 about oncce/game 2011-11-08T01:09:14 make it print the number of changed cells each iteration - it should get smaller almost every iteration 2011-11-08T01:10:31 ah, 1 2011-11-08T01:10:49 wait that can't be right 2011-11-08T01:11:35 1599, 79, 78, 77, 76 and on down 2011-11-08T01:12:37 larget number of iterations from a sim I just ran: 16; [280, 280, 707, 389, 46, 162, 146, 119, 35, 4, 1, 1, 1, 3, 3, 0] 2011-11-08T01:12:44 That seems about right 40x40 is 1600 2011-11-08T01:14:12 that was for a 60*90 map 2011-11-08T01:14:27 with the fog of war considerations? 2011-11-08T01:14:43 *** Twistedlogic has joined #aichallenge 2011-11-08T01:15:25 I expand to all cells that have ever been seen plus 3 tiles outside that (treated as land) 2011-11-08T01:16:15 for that iteration, about ~36% of the map had been seen 2011-11-08T01:17:13 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has quit IRC (Read error: Operation timed out) 2011-11-08T01:17:29 with the whole map visible, one map took 8 iterations; [1736, 2129, 1049, 435, 120, 43, 1, 0] 2011-11-08T01:18:21 How long did that take? 2011-11-08T01:18:51 I have a hunch there's a lot I don't know about optimizing python 2011-11-08T01:19:07 *** pairofdice has joined #aichallenge 2011-11-08T01:19:32 hrmm, that one actually took 22 ms, so when I said 1ms earlier I was probably off by an order of magnitude (ie 10 ms instead) 2011-11-08T01:19:52 still 22ms is way shorter than i'm seeing 2011-11-08T01:21:41 *** kiv has quit IRC (Quit: Leaving) 2011-11-08T01:24:35 *hopes new bot with tons of dfs does not time out on official server* 2011-11-08T01:26:14 Extrarius: Are you using numpy or array.array? 2011-11-08T01:29:05 nope, just using normal lists. I tried array.array and bytearray and both were slower 2011-11-08T01:29:26 Yea numpy was an order of magnitude slower for me 2011-11-08T01:30:50 *** smiley1983 has joined #aichallenge 2011-11-08T01:36:26 i think the map sizes are still to small to make a big difference 2011-11-08T01:36:37 for numpy and array 2011-11-08T01:36:42 *** JorgeB has joined #aichallenge 2011-11-08T01:38:38 ah 2011-11-08T01:46:28 *** Fandekasp has joined #aichallenge 2011-11-08T01:47:48 *** roflmao1 has quit IRC (Quit: Leaving.) 2011-11-08T02:07:00 *** treeform has quit IRC (Remote host closed the connection) 2011-11-08T02:08:45 *** Zaphus has quit IRC (Ping timeout: 265 seconds) 2011-11-08T02:12:06 *** amstan has quit IRC (Ping timeout: 240 seconds) 2011-11-08T02:12:06 *** Mooloo has quit IRC (Ping timeout: 240 seconds) 2011-11-08T02:13:21 *** kilae has joined #aichallenge 2011-11-08T02:16:32 whoa. i think this guy might displace xathis: http://aichallenge.org/profile.php?user=589 2011-11-08T02:17:41 in fact i'm certain he will 2011-11-08T02:17:54 *** delt0r_ has quit IRC (Ping timeout: 258 seconds) 2011-11-08T02:18:47 Yeah, he has risen fast 2011-11-08T02:19:11 check this out: http://aichallenge.org/visualizer.php?game=68778&user=589 2011-11-08T02:19:39 he utterly mops up xathis in a situation where i thought he could hold out forever 2011-11-08T02:20:09 What's weird is that he lets many of his single ants get destroyed but still pwoers through 2011-11-08T02:20:15 powers* 2011-11-08T02:20:35 at the end? 2011-11-08T02:20:47 yeah but it's an even exchange. he doesn't lose an ant without killing an ant 2011-11-08T02:20:54 In the beginning 2011-11-08T02:21:05 I guess he switches gears 2011-11-08T02:21:08 well 2011-11-08T02:21:22 he loses ants to kill ants coming toward his base to avoid being discovered 2011-11-08T02:21:31 Ahh, might be it 2011-11-08T02:21:36 totally worth it 2011-11-08T02:22:50 *** djr_ has quit IRC (Quit: Leaving) 2011-11-08T02:24:18 *** datachomper1 has left #aichallenge 2011-11-08T02:25:55 *** Mooloo has joined #aichallenge 2011-11-08T02:27:00 *** Jak_o_Shadows has joined #aichallenge 2011-11-08T02:28:23 *** bobbydroptable has joined #aichallenge 2011-11-08T02:29:38 *** hacklash has quit IRC (Quit: hacklash) 2011-11-08T02:30:12 hi there. there was a discussion recently about numpy for python. has that been installed yet? 2011-11-08T02:31:06 *** delt0r_ has joined #aichallenge 2011-11-08T02:32:04 *** Palmik has joined #aichallenge 2011-11-08T02:32:52 *** mleise has joined #aichallenge 2011-11-08T02:35:09 *** theskumar has left #aichallenge 2011-11-08T02:49:35 *** JorgeB has quit IRC (Quit: Computer has gone to sleep.) 2011-11-08T03:02:21 *** hexren has joined #aichallenge 2011-11-08T03:03:09 *** Akranis has joined #aichallenge 2011-11-08T03:07:21 *** Jak_o_Shadows1 has joined #aichallenge 2011-11-08T03:09:25 *** Jak_o_Shadows has quit IRC (Ping timeout: 240 seconds) 2011-11-08T03:18:18 *** tmandry has quit IRC (Ping timeout: 245 seconds) 2011-11-08T03:21:49 *** ikaros has joined #aichallenge 2011-11-08T03:31:49 *** epicmonkey has joined #aichallenge 2011-11-08T03:34:29 *** Blkt has joined #aichallenge 2011-11-08T03:35:06 *** Devko has joined #aichallenge 2011-11-08T03:35:44 *** bobbydroptable has quit IRC (Quit: This computer has gone to sleep) 2011-11-08T03:37:12 good morning everyone 2011-11-08T03:37:24 *** sigh has joined #aichallenge 2011-11-08T03:38:54 *** epicmonkey has quit IRC (Ping timeout: 240 seconds) 2011-11-08T03:41:26 *** mviel has joined #aichallenge 2011-11-08T03:41:33 *** praveen has joined #aichallenge 2011-11-08T03:42:25 Quick question : Is it possible to run BFS every turn, from every ant to every possible food location. Will this run in time or should I cache the path? 2011-11-08T03:42:53 *** NotABug has joined #aichallenge 2011-11-08T03:47:46 *** cyphase has quit IRC (Ping timeout: 255 seconds) 2011-11-08T03:47:53 *** Devko has quit IRC (Quit: Page closed) 2011-11-08T03:48:24 *** GarfTop has joined #aichallenge 2011-11-08T03:49:49 *** Garf has quit IRC (Ping timeout: 240 seconds) 2011-11-08T03:50:15 *** ikaros has quit IRC (Remote host closed the connection) 2011-11-08T03:54:20 *** cyphase has joined #aichallenge 2011-11-08T03:56:43 *** Jak_o_Shadows1 has quit IRC (Remote host closed the connection) 2011-11-08T03:58:59 praveen: depends on your language, implementation and number of ants/food 2011-11-08T03:59:18 you can also just not collect all the food and only spend .x proportion of your time collecting food for example 2011-11-08T03:59:35 or cache paths if you want etc. 2011-11-08T04:00:09 *** protonh has joined #aichallenge 2011-11-08T04:00:47 How big is a submission allowed to be? I'm thinking of using machine learning for some local stuff, and I guess the only way to do it is to include the data in the source code for the bot? 2011-11-08T04:00:51 *** Akranis has quit IRC (Quit: Lmnar) 2011-11-08T04:02:03 antimatroid: yes, am implementing in Java, so I think its not that fast enough 2011-11-08T04:02:46 praveen: I use something similar, in Go, with absolutely no caching, and it's fast enough 2011-11-08T04:02:53 I think it's doable in Java, with care 2011-11-08T04:03:31 BFS from all foods to all ants? That's fast and simple ;) 2011-11-08T04:03:36 *** boegel has joined #aichallenge 2011-11-08T04:03:38 retybok: Ok let me try. To my knowledge, c++ is 2 times faster than Java 2011-11-08T04:03:55 praveen: that depends on the way you write your code 2011-11-08T04:04:00 pairofdice: thanks that will remove the necessity to track 2011-11-08T04:04:33 retybok: Exact same implementation, c++ is faster than java - Do you agree? 2011-11-08T04:04:53 praveen: you can't have the same implementation since the languages are different 2011-11-08T04:05:00 in particular, the GC makes a difference 2011-11-08T04:05:08 retybok: I mean the same algorithm 2011-11-08T04:05:17 but yes, usually C++ ends up being faster than Java 2011-11-08T04:06:12 have participated in some programming competitions and have observed that java is very slow. for instance check spoj , every problem has java solution but its no way close to the c++ one 2011-11-08T04:06:30 Ok if it times out, then I should move to some other language may be 2011-11-08T04:06:51 *** protonh has quit IRC (Quit: Page closed) 2011-11-08T04:10:49 *** replore has quit IRC (Read error: Connection reset by peer) 2011-11-08T04:17:24 *** ikaros has joined #aichallenge 2011-11-08T04:27:45 *** alc has joined #aichallenge 2011-11-08T04:31:13 *** liberforce has joined #aichallenge 2011-11-08T04:38:35 *** mviel_ has joined #aichallenge 2011-11-08T04:38:38 I'll try asking my question once more 2011-11-08T04:38:42 How big is a submission allowed to be? I'm thinking of using machine learning for some local stuff, and I guess the only way to do it is to include the data in the source code for the bot? 2011-11-08T04:41:30 *** Zaphus has joined #aichallenge 2011-11-08T04:42:14 30 minutes until my next game (until the website) - do I dare upload a new version ? 2011-11-08T04:42:35 (until the website) should read (according to the website)... sigh 2011-11-08T04:42:35 *** mviel has quit IRC (Ping timeout: 260 seconds) 2011-11-08T04:46:40 30 minutes according to the website can well be over an hour 2011-11-08T04:46:47 how many players are ahead of you? 2011-11-08T04:46:50 *** aerique has joined #aichallenge 2011-11-08T04:56:18 126 players ahead, still 25 minutes to go... 2011-11-08T04:56:34 *** mviel__ has joined #aichallenge 2011-11-08T04:57:08 the question is more like, am I happy enough with my new version to pop it onto the main site, or do I wait another day... it takes 5 days to figure out where you sit on the main site and I only uploaded the current version last night! 2011-11-08T04:59:43 *** Zaphus has quit IRC (Quit: Page closed) 2011-11-08T05:00:15 *** mviel_ has quit IRC (Ping timeout: 248 seconds) 2011-11-08T05:04:55 *** boegel has quit IRC (Read error: Operation timed out) 2011-11-08T05:06:30 *** hacklash has joined #aichallenge 2011-11-08T05:07:06 *** Twistedlogic has quit IRC (Ping timeout: 265 seconds) 2011-11-08T05:14:37 *** Califax has joined #aichallenge 2011-11-08T05:14:49 if I get Test Error: compiled, but failed test cases after submitting code 2011-11-08T05:14:57 will it still run, or not 2011-11-08T05:15:12 Califax: I don't think so. Did you try to run test_bot.sh? 2011-11-08T05:15:21 I did not 2011-11-08T05:16:02 It should tell you what is wrong 2011-11-08T05:19:56 *** boegel has joined #aichallenge 2011-11-08T05:21:59 *** g0llum has joined #aichallenge 2011-11-08T05:22:28 Got it, didn't like the logging 2011-11-08T05:22:29 thanks 2011-11-08T05:22:51 *** hacklash has left #aichallenge 2011-11-08T05:23:40 *** Califax has quit IRC (Quit: Page closed) 2011-11-08T05:23:56 *** olexs has joined #aichallenge 2011-11-08T05:24:05 *** LouisMartin has quit IRC (Quit: Page closed) 2011-11-08T05:28:16 *** pedrosorio has joined #aichallenge 2011-11-08T05:30:31 *** pedrosorio has quit IRC (Client Quit) 2011-11-08T05:32:43 *** Jak_o_Shadows has joined #aichallenge 2011-11-08T05:37:36 np 2011-11-08T05:38:44 *** hexren has quit IRC (Quit: Leaving.) 2011-11-08T05:45:19 *** Jak_o_Shadows has quit IRC (Remote host closed the connection) 2011-11-08T05:47:53 *** ikaros has quit IRC (Ping timeout: 258 seconds) 2011-11-08T05:49:28 *** grwip has joined #aichallenge 2011-11-08T05:53:52 pguillory on 2nd place now :> 2011-11-08T05:55:24 and i fell to 8th :) 2011-11-08T06:00:51 *** kilae has quit IRC (Quit: ChatZilla 0.9.87 [Firefox 7.0.1/20110928134238]) 2011-11-08T06:00:52 *** smiley1983 has quit IRC (Ping timeout: 260 seconds) 2011-11-08T06:02:29 *** smiley1983 has joined #aichallenge 2011-11-08T06:10:21 *** alc has quit IRC (Remote host closed the connection) 2011-11-08T06:22:30 *** nickjohnson has quit IRC (Ping timeout: 240 seconds) 2011-11-08T06:23:05 *** modafinil has quit IRC (Ping timeout: 252 seconds) 2011-11-08T06:32:26 *** MoleM has joined #aichallenge 2011-11-08T06:33:03 hiya 2011-11-08T06:33:15 hmmm...never received my activation mail 2011-11-08T06:33:19 can anyone resend it? 2011-11-08T06:35:18 *** cyphase has quit IRC (Ping timeout: 240 seconds) 2011-11-08T06:40:24 *** smiley1983 has quit IRC (Ping timeout: 258 seconds) 2011-11-08T06:41:34 *** modafinil has joined #aichallenge 2011-11-08T06:44:56 *** xgetc25 has joined #aichallenge 2011-11-08T06:49:26 *** cyphase has joined #aichallenge 2011-11-08T06:49:36 *** praveen has quit IRC (Quit: Page closed) 2011-11-08T06:52:45 how to change default names in visializer? 2011-11-08T06:52:55 "python sample_bots/python/HunterBot.py" \ 2011-11-08T06:53:13 but its shown as player 1 2011-11-08T06:53:25 how do I set up my name for it? 2011-11-08T06:54:00 hmmm, mine shows as HunterBot.py 2011-11-08T06:54:11 in the visualizer 2011-11-08T06:54:35 maybe one of the command line arguments?> 2011-11-08T06:54:44 can u help? 2011-11-08T06:54:50 i run on mac os lion 2011-11-08T06:54:58 #!/usr/bin/env sh ./playgame.py -So --player_seed 42 --end_wait=0.25 --verbose --log_dir game_logs --turns 300 --map_file maps/walk/random_walk_01.map "$@" \ "ruby sample_bots/ruby/zavtrak.rb" \ "python sample_bots/python/HunterBot.py" \ "python sample_bots/python/HunterBot.py" \ "python sample_bots/python/HunterBot.py" | java -jar visualizer.jar 2011-11-08T06:55:05 this is my input 2011-11-08T06:55:20 and i got Game #player 1 ... etc 2011-11-08T06:55:26 in visulizer ( 2011-11-08T06:55:55 hmm, weird, I'm on WInXP, perhaps that's why there's a display difference? 2011-11-08T06:56:49 *** smiley1983 has joined #aichallenge 2011-11-08T06:57:36 *** ikaros has joined #aichallenge 2011-11-08T06:58:47 *** zaphus has joined #aichallenge 2011-11-08T07:05:47 *** smiley1983 has quit IRC (Ping timeout: 258 seconds) 2011-11-08T07:10:33 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has joined #aichallenge 2011-11-08T07:11:12 *** amstan has joined #aichallenge 2011-11-08T07:11:12 *** ChanServ sets mode: +o amstan 2011-11-08T07:16:34 *** mviel_ has joined #aichallenge 2011-11-08T07:19:18 *** amstan has quit IRC (Ping timeout: 240 seconds) 2011-11-08T07:20:23 *** mviel__ has quit IRC (Ping timeout: 252 seconds) 2011-11-08T07:21:42 *** jcdny is now known as bugnuts2 2011-11-08T07:22:08 *** smiley1983 has joined #aichallenge 2011-11-08T07:23:27 *** zaphus has quit IRC (Quit: Page closed) 2011-11-08T07:27:37 *** kilae has joined #aichallenge 2011-11-08T07:37:17 *** mcstar has joined #aichallenge 2011-11-08T07:37:56 *** twymer has joined #aichallenge 2011-11-08T07:38:47 *** xgetc25 has quit IRC (Quit: Page closed) 2011-11-08T07:43:31 *** bobbydroptable has joined #aichallenge 2011-11-08T07:43:34 *** twymer has quit IRC (Ping timeout: 255 seconds) 2011-11-08T07:44:13 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has quit IRC (Ping timeout: 240 seconds) 2011-11-08T07:46:03 *** u_ has joined #aichallenge 2011-11-08T07:46:50 *** u__ has joined #aichallenge 2011-11-08T07:50:23 *** u_ has quit IRC (Ping timeout: 248 seconds) 2011-11-08T07:50:24 *** u__ is now known as u_ 2011-11-08T07:59:21 *** jl1990 has joined #aichallenge 2011-11-08T08:03:40 *** boegel has quit IRC (Ping timeout: 258 seconds) 2011-11-08T08:15:09 *** g0llum has quit IRC (Read error: Connection reset by peer) 2011-11-08T08:18:06 *** GeorgeSebastian has quit IRC (Read error: Connection reset by peer) 2011-11-08T08:18:44 *** boegel has joined #aichallenge 2011-11-08T08:19:00 *** bergmark has quit IRC (Ping timeout: 260 seconds) 2011-11-08T08:23:34 *** praveen has joined #aichallenge 2011-11-08T08:26:15 *** praveen has quit IRC (Client Quit) 2011-11-08T08:34:52 *** merkas has joined #aichallenge 2011-11-08T08:34:59 hallo guys 2011-11-08T08:35:04 i finaly 2011-11-08T08:35:17 compiles the bot 2011-11-08T08:35:20 in ubuntu 2011-11-08T08:35:37 but i have a proble with test_bot.sh 2011-11-08T08:35:47 i get a traceback error 2011-11-08T08:36:00 *** master_ninja has left #aichallenge 2011-11-08T08:36:01 can someone help me/ 2011-11-08T08:36:03 ? 2011-11-08T08:36:16 what's the error 2011-11-08T08:36:43 and what arguments are you passing to test_bot.sh 2011-11-08T08:36:59 i edited the 2011-11-08T08:37:06 test_bot.sh 2011-11-08T08:37:31 and where it says the ... i write 2011-11-08T08:37:31 "python submission_test/MyBot.exe" 2011-11-08T08:37:40 *** oliv3 has joined #aichallenge 2011-11-08T08:37:41 annd i run it like\ 2011-11-08T08:37:49 ./test-bot.sh 2011-11-08T08:37:57 ./test_bot.sh 2011-11-08T08:38:07 *** hkraal has joined #aichallenge 2011-11-08T08:38:45 i dont pass any rguments 2011-11-08T08:41:26 what i have to do 2011-11-08T08:42:15 whoa whoa 2011-11-08T08:42:16 what ...? 2011-11-08T08:42:22 you've got MyBot.exe? 2011-11-08T08:42:26 what language are you coding it in? 2011-11-08T08:42:30 c++ 2011-11-08T08:42:33 lol 2011-11-08T08:42:36 hallo mcstar 2011-11-08T08:42:41 hi 2011-11-08T08:42:56 why do you want python to execute your bot 2011-11-08T08:43:05 i want to test 2011-11-08T08:43:07 it 2011-11-08T08:43:11 try this 2011-11-08T08:43:11 ./test-bot.sh "submission_test/MyBot.exe" 2011-11-08T08:43:14 just give the exe as the argument to that script 2011-11-08T08:43:28 Hi all, does anyone know any means to contact the organisation directly? 2011-11-08T08:43:29 btw, you dont need to use .exe 2011-11-08T08:43:45 what i have to do/? 2011-11-08T08:43:59 boobydroptable already told you 2011-11-08T08:44:07 would ubuntu recognise .exe as a binary ? 2011-11-08T08:44:13 ok 2011-11-08T08:44:13 i type it now 2011-11-08T08:44:20 bobbydroptable: no matter what the filename is 2011-11-08T08:44:28 *** rumland has joined #aichallenge 2011-11-08T08:44:38 as long as its executable and an ELF binary it will run 2011-11-08T08:44:38 oh you mean he doesn't need to put .exe for the filename 2011-11-08T08:44:43 yep 2011-11-08T08:44:50 no need to compile it that way 2011-11-08T08:44:55 we're never on the same wavelengths :) 2011-11-08T08:46:36 SandboxError: Failed to start ['submission_test/MyBot.exe'] 2011-11-08T08:46:51 merkas: can you run your bot at all? 2011-11-08T08:47:01 does it have execute flag? 2011-11-08T08:47:16 what happens when you type "submission_test/MyBot.exe" which should run your bot 2011-11-08T08:47:33 the same error 2011-11-08T08:47:37 bobbydroptable: now he's gonna quote it 2011-11-08T08:47:52 mcstar i can run the bot 2011-11-08T08:47:54 typing 2011-11-08T08:48:01 ./mybot.exe 2011-11-08T08:48:07 *facepalm* 2011-11-08T08:48:08 really? 2011-11-08T08:48:12 come on 2011-11-08T08:48:27 ./test-bot.sh "./MyBot.exe" 2011-11-08T08:48:29 merkas: usually linux fiilesystems are case sensitive 2011-11-08T08:48:53 i type 2011-11-08T08:48:59 ./MyBot 2011-11-08T08:49:18 merkas that's 3 different stories you've told us you need to be more specific and accurate if you want the help you need 2011-11-08T08:49:53 look 2011-11-08T08:49:57 i compiles ti 2011-11-08T08:49:59 *** yoden has joined #aichallenge 2011-11-08T08:50:00 with no problem 2011-11-08T08:50:01 bobbydroptable: you, he is kind of inonsistent 2011-11-08T08:50:07 you->yeah 2011-11-08T08:50:16 then i type ./MyBot 2011-11-08T08:50:23 and it executes 2011-11-08T08:50:28 merkas: read the scripts source, you'll understand what you need to pass it 2011-11-08T08:50:33 *** rumland has quit IRC (Quit: Page closed) 2011-11-08T08:50:36 in that case, when you run the test bot, pass ./MyBot as the argument to run your bot 2011-11-08T08:50:39 ./test_bot.sh path/to/bot/MyBot 2011-11-08T08:50:56 retybok: :S 2011-11-08T08:50:58 to test 2011-11-08T08:50:59 wait 2011-11-08T08:50:59 it 2011-11-08T08:51:04 mcstar: don't you need quotes around path to bot? 2011-11-08T08:51:09 no 2011-11-08T08:51:10 no 2011-11-08T08:51:13 it's a four-line shell script... 2011-11-08T08:51:15 haha 2011-11-08T08:51:25 or is it because i'm using python 2011-11-08T08:51:29 retybok: 1-liner 2011-11-08T08:52:07 any admin around? 2011-11-08T08:52:11 bobbydroptable: yes, the script takes 1 parameter, $1, and you need to give your script WITH your interpreter 2011-11-08T08:52:18 thats why you need quotes there 2011-11-08T08:52:24 understood 2011-11-08T08:52:28 otherwise $1 would by python, $2 your bot 2011-11-08T08:52:47 merkas: is it working? 2011-11-08T08:52:58 i think he was waiting for us to give him the commandline 2011-11-08T08:53:22 i gave it to him 2011-11-08T08:53:30 he said wait 2011-11-08T08:53:35 than he laughs 2011-11-08T08:53:41 then i scratch my head 2011-11-08T08:53:48 you gave him path/to/bot/MyBot 2011-11-08T08:53:57 ./test_bot.sh path/to/bot/MyBot 2011-11-08T08:54:02 thats all he needs 2011-11-08T08:54:06 *** smiley1993 has joined #aichallenge 2011-11-08T08:54:09 bash: path/to/bot/MyBot: No such file or directory 2011-11-08T08:54:15 ;-) 2011-11-08T08:54:15 :D 2011-11-08T08:54:16 now come on 2011-11-08T08:54:18 XD lol that's what i was thinking 2011-11-08T08:55:41 now we just wait ... 2011-11-08T08:56:16 i cant test and TestBot.py 2011-11-08T08:56:18 ubuntu@ubuntu:~/Downloads/tools$ ./test_bot.sh "submission_test/TestBot.py" Traceback (most recent call last): File "/home/ubuntu/Downloads/tools/engine.py", line 93, in run_game sandbox.start(bot_cmd) File "/home/ubuntu/Downloads/tools/sandbox.py", line 385, in start raise SandboxError('Failed to start {0}'.format(shell_command)) SandboxError: Failed to start ['submission_test/TestBot.py'] waiting 0.25 seconds for bot 2011-11-08T08:56:27 look this 2011-11-08T08:57:11 TestBot.py? 2011-11-08T08:57:26 whatever happened to MyBot? 2011-11-08T08:57:35 the same 2011-11-08T08:57:51 *** capa has joined #aichallenge 2011-11-08T08:57:52 from exe to py? 2011-11-08T08:57:59 okay merkas : do us a favour 2011-11-08T08:58:06 type the following 2011-11-08T08:58:07 ubuntu@ubuntu:~/Downloads/tools$ ./test_bot.sh "submission_test/MyBot" Traceback (most recent call last): File "/home/ubuntu/Downloads/tools/engine.py", line 93, in run_game sandbox.start(bot_cmd) File "/home/ubuntu/Downloads/tools/sandbox.py", line 385, in start raise SandboxError('Failed to start {0}'.format(shell_command)) SandboxError: Failed to start ['submission_test/MyBot'] waiting 0.25 seconds for bots to proce 2011-11-08T08:58:07 pwd 2011-11-08T08:58:10 ls 2011-11-08T08:58:15 paste the results here 2011-11-08T08:58:20 ./test_bot.sh "python submission_test/MyBot" 2011-11-08T08:58:22 merkas: 2011-11-08T08:58:34 merkas: do you know what an interpreter is? 2011-11-08T08:58:42 ubuntu@ubuntu:~/Downloads/tools$ pwd /home/ubuntu/Downloads/tools ubuntu@ubuntu:~/Downloads/tools$ ls ants.py game.py play_one_game_live.sh submission_test visualizer.jar ants.pyc game.pyc play_one_game.sh test_bot.sh engine.py mapgen sample_bots test_bot.sh~ engine.pyc maps sandbox.py VERSION game_logs playgame.py sandbox.pyc visualizer 2011-11-08T08:58:42 or virtual machine? 2011-11-08T08:58:49 ye 2011-11-08T08:58:49 s 2011-11-08T08:58:55 *** bergmark has joined #aichallenge 2011-11-08T08:59:03 ok, well scripts and bytecodes need it 2011-11-08T08:59:06 your bot doesnt 2011-11-08T08:59:38 merkas: ls submission_test 2011-11-08T08:59:47 ./test_bot.sh submission_test/MyBot 2011-11-08T08:59:56 if ls submission_test lists MyBot 2011-11-08T09:00:14 ubuntu@ubuntu:~/Downloads/tools$ ls submission_test ants.py Bot.o Makefile MyBot.o State.cc TestBot.py VERSION Bot.cc Bug.h MyBot README.txt State.h test.map Bot.h Location.h MyBot.cc Square.h State.o Timer.h 2011-11-08T09:00:23 congrats 2011-11-08T09:00:29 everything is perfect 2011-11-08T09:00:33 yay we've established it is c++ lol 2011-11-08T09:00:43 merkas: follow mcstar's instruction 2011-11-08T09:00:49 it is c++ 2011-11-08T09:00:51 boobydrooltable: he already said that 2011-11-08T09:00:53 i code in c++ 2011-11-08T09:01:10 but i test the already bot in py 2011-11-08T09:01:10 mcstar: well, after he kept trying to run it in python, i had to be sure... 2011-11-08T09:01:14 *** g0llum has joined #aichallenge 2011-11-08T09:01:37 mcstar wht to do now 2011-11-08T09:01:41 > 2011-11-08T09:01:48 ./test_bot.sh submission_test/MyBot 2011-11-08T09:01:54 why do i have to repeat myself? 2011-11-08T09:01:57 merkas: what do you want to do exactly? 2011-11-08T09:02:01 sorry 2011-11-08T09:02:07 to test the bot 2011-11-08T09:02:18 merkas: your bot, or the example bot? 2011-11-08T09:02:26 retybok: dont confuse him now 2011-11-08T09:02:27 mybot 2011-11-08T09:02:30 he is on the right path 2011-11-08T09:02:47 but i tested and the example bot to see if the problem was mine 2011-11-08T09:02:49 good/ 2011-11-08T09:02:50 > 2011-11-08T09:02:52 ubuntu@ubuntu:~/Downloads/tools$ ./test_bot.sh submission_test/MyBot Traceback (most recent call last): File "/home/ubuntu/Downloads/tools/engine.py", line 93, in run_game sandbox.start(bot_cmd) File "/home/ubuntu/Downloads/tools/sandbox.py", line 385, in start raise SandboxError('Failed to start {0}'.format(shell_command)) SandboxError: Failed to start ['submission_test/MyBot'] waiting 0.25 seconds for bots to process 2011-11-08T09:03:27 can you run it from here? 2011-11-08T09:03:31 submission_test/MyBot 2011-11-08T09:03:38 no 2011-11-08T09:03:44 thats a problem 2011-11-08T09:03:49 you should be able to do it 2011-11-08T09:03:51 wait to test it 2011-11-08T09:03:53 chmod u+x submission_test/MyBot ? 2011-11-08T09:04:03 retybok: he could already runit 2011-11-08T09:04:07 from the same dir 2011-11-08T09:04:26 Is there a maximum map size? And if not what's the largest map currently used in battles? 2011-11-08T09:04:35 merkas: what error do you get when you run submission_test/MyBot? 2011-11-08T09:04:44 userjjb: from what I've heard, it is 200x200 2011-11-08T09:04:45 no error 2011-11-08T09:04:51 *** smiley1993 has quit IRC (Read error: Connection reset by peer) 2011-11-08T09:05:06 but you said "no" when mcstar asked if you could run it? 2011-11-08T09:05:13 sorry 2011-11-08T09:05:19 i get no error 2011-11-08T09:05:25 the progams 2011-11-08T09:05:25 runs 2011-11-08T09:05:29 good 2011-11-08T09:05:54 *** hjgjhg has joined #aichallenge 2011-11-08T09:05:57 gftyufufuy 2011-11-08T09:06:26 *** hjgjhg has quit IRC (Client Quit) 2011-11-08T09:06:40 ./test_bot.sh /home/ubuntu/Downloads/tools/submission_test/MyBot 2011-11-08T09:06:47 try a full path 2011-11-08T09:06:50 ok 2011-11-08T09:06:54 i try it now 2011-11-08T09:07:20 goof 2011-11-08T09:07:22 good 2011-11-08T09:07:24 it runs smooth 2011-11-08T09:07:31 thanks a lot mcstar 2011-11-08T09:07:41 ./test_bot.sh ./submission_test/MyBot 2011-11-08T09:07:44 try this to 2011-11-08T09:07:46 o 2011-11-08T09:07:48 *** capa has quit IRC (Ping timeout: 265 seconds) 2011-11-08T09:08:17 the same error 2011-11-08T09:08:21 mcstar: do you understand what the problem was? 2011-11-08T09:08:30 that's really weird 2011-11-08T09:08:40 the engine turns paths to shit again 2011-11-08T09:08:50 i run a live cd 2011-11-08T09:08:54 thats why 2011-11-08T09:08:59 no 2011-11-08T09:09:07 *** sigh has quit IRC (Remote host closed the connection) 2011-11-08T09:09:20 I don't think that's the problem either 2011-11-08T09:09:30 ok 2011-11-08T09:09:34 the problem solve 2011-11-08T09:09:35 *** smiley1993 has joined #aichallenge 2011-11-08T09:09:35 d 2011-11-08T09:09:39 thanks guys 2011-11-08T09:09:43 well, half-solved 2011-11-08T09:09:43 good luck 2011-11-08T09:09:47 god saves you 2011-11-08T09:09:48 its working, but we dont know why 2011-11-08T09:10:09 merkas: lets leave god out of this 2011-11-08T09:10:27 haha 2011-11-08T09:10:34 does anyone have a good tutorial 2011-11-08T09:10:36 on gcc 2011-11-08T09:10:44 because i used dev-cpp 2011-11-08T09:10:53 mcstar 2011-11-08T09:11:01 merkas: c++ is a standard 2011-11-08T09:11:02 if a want to change my code 2011-11-08T09:11:06 dev-cpp uses GCC 2011-11-08T09:11:09 you dont need to know the compiler 2011-11-08T09:11:13 ok 2011-11-08T09:11:16 *** Antimony has joined #aichallenge 2011-11-08T09:11:28 devcpp uses whatever you give it to 2011-11-08T09:11:40 are the server specs available ? 2011-11-08T09:11:41 its just an ide 2011-11-08T09:11:42 if i want to make some changes in mybots code 2011-11-08T09:11:48 I'm concerned with turn times 2011-11-08T09:11:50 what i type 2011-11-08T09:11:54 only to compile it 2011-11-08T09:11:56 merkas: use any text editor 2011-11-08T09:12:01 try "gedit" for instance 2011-11-08T09:12:03 and see if i have anny error 2011-11-08T09:12:03 500ms alone doesn't mean anything 2011-11-08T09:12:08 oprs: yes, you get it when your bot starts, and they are on the site too 2011-11-08T09:12:08 merkas: to compile or to get an executable? 2011-11-08T09:12:18 ionly compile 2011-11-08T09:12:24 cc -c file.c 2011-11-08T09:12:25 oh, I must have missed that then, thanks mcstar 2011-11-08T09:12:26 to see if it good the code syntax 2011-11-08T09:12:31 and the link it 2011-11-08T09:12:42 or better, cc -Wall -pedantic -c file.c 2011-11-08T09:12:52 that will show you all warnings and such 2011-11-08T09:12:53 g++ -c 2011-11-08T09:13:04 good 2011-11-08T09:13:06 thanks 2011-11-08T09:13:18 and the n i compile the three files 2011-11-08T09:13:19 merkas: do yuo know what makefiles are? 2011-11-08T09:13:24 merkas: if your file has .cpp extension, GCC will automatically switch on C++ mode 2011-11-08T09:13:30 yes 2011-11-08T09:13:31 or you can call it by c++ 2011-11-08T09:13:36 (options stay the same) 2011-11-08T09:13:39 makefiles do all the job 2011-11-08T09:13:42 p_l: cpp is c++11 on the server 2011-11-08T09:13:47 compiling and then linking 2011-11-08T09:13:52 *** TheLinker has joined #aichallenge 2011-11-08T09:14:00 merkas: use .cc extension for c++03 2011-11-08T09:14:02 mcstar: oh? no 'c++' as well? 2011-11-08T09:14:20 oh well, C++ is a mess I am not touching :) 2011-11-08T09:14:51 mcstar:can you tell me the procedure in the manual compiling and linking 2011-11-08T09:14:53 the file 2011-11-08T09:14:54 s 2011-11-08T09:14:55 i didnt either, but after couple of weeks im wuite confortable with it 2011-11-08T09:14:57 in mybot 2011-11-08T09:15:29 g++ MyBot.cc -o MyBot 2011-11-08T09:16:05 merkas: you must have a MyBot.cc file, for the sever's compiler to recognize your language 2011-11-08T09:16:27 otherwise, it just compiles all *.cc into *.o and than links them together into MyBot 2011-11-08T09:16:43 ok 2011-11-08T09:16:47 thanks 2011-11-08T09:16:56 and what i upload to server 2011-11-08T09:16:57 ? 2011-11-08T09:17:12 okay, let me put it another way ;) 500ms on a Pentium II with 128MB or RAM and 1000 processes running != 500ms on a core i7 with 16GB and only 20 processes running 2011-11-08T09:17:13 zip sub.zip *.cc *.hh 2011-11-08T09:17:23 the time unit should be CPU cycles, not ms 2011-11-08T09:17:42 or something even remotely consistent across architectures 2011-11-08T09:17:44 oprs: just gotta wing it. lol 2011-11-08T09:17:50 oprs: why not micor-lightyears? 2011-11-08T09:17:56 micro* 2011-11-08T09:18:01 *** merkas has quit IRC (Quit: Page closed) 2011-11-08T09:18:21 well, I think you missed my point 2011-11-08T09:18:23 btw 500ms == 500ms 2011-11-08T09:18:31 no matter on what machine you measure it 2011-11-08T09:18:37 :> 2011-11-08T09:18:53 I can do a whole lot more in 500ms on a core i7 than on a i386 2011-11-08T09:18:56 don't mind mcstar he's very very pedantic 2011-11-08T09:18:58 ^^ 2011-11-08T09:19:18 oprs: why do you think it matters? 2011-11-08T09:19:32 you always know what the time is, so you can see how much you have done 2011-11-08T09:20:10 all the bots run on the same specs 2011-11-08T09:20:17 you are not in a worse position than others 2011-11-08T09:21:01 aichallenge runs on EC2, i believe? 2011-11-08T09:21:21 yes 2011-11-08T09:21:39 Though I would like an approximation of how much faster it runs on the server than on my pos computer 2011-11-08T09:21:51 so are the "cores" allocated to each bot actual cores, or virtual cores? 2011-11-08T09:21:54 pairofdice: that's the point 2011-11-08T09:22:36 I'm running wildly liberal timelimits locally 2011-11-08T09:22:49 if it was cloud computing, it may be virtually impossible to tell the spec of a "virtual core" in that case. 2011-11-08T09:23:22 though i'm not familiar with the nitty gritty details of how the architecture works. 2011-11-08T09:23:40 ok, so we do need to track time 2011-11-08T09:24:00 because the target architecture/environement could be virtually anything 2011-11-08T09:24:13 Yeah 2011-11-08T09:24:37 thanks. that answers my question. 2011-11-08T09:25:17 i'm not sure if that was an answer at all, but if it leads you in the right direction... lol =) 2011-11-08T09:25:28 bedtime for me. night 2011-11-08T09:25:32 *** bobbydroptable has quit IRC (Quit: Leaving) 2011-11-08T09:25:43 *** Joris_ has joined #aichallenge 2011-11-08T09:25:52 Well my bot doesn't time out, it just starts to stutter :() 2011-11-08T09:27:30 Hi all 2011-11-08T09:28:32 I have a annoying problem with my C# code and was wondering whether someone could help me 2011-11-08T09:28:38 *an 2011-11-08T09:36:30 *** Antimony has quit IRC (Ping timeout: 240 seconds) 2011-11-08T09:43:17 *** Joris_ has quit IRC (Quit: Page closed) 2011-11-08T09:45:21 I just can't stop tinkering with my bot 2011-11-08T09:47:26 *** Hexren has joined #aichallenge 2011-11-08T09:49:01 *** Areks has quit IRC (Ping timeout: 240 seconds) 2011-11-08T09:49:53 this is interesting, a couple weeks ago, i didnt know what stl is(almost), and now all my code depend on it 2011-11-08T09:50:27 I haven't coded this much in all my life 2011-11-08T09:50:39 Possibly. 2011-11-08T09:51:34 well, i definitely havent spent this much time on code any other projects of mine 2011-11-08T09:51:35 I am hitting on my head harder atm :-) 2011-11-08T09:51:40 on coding* 2011-11-08T09:51:58 avdg: stop hurting yourslef 2011-11-08T09:52:07 I'm not punishing myself 2011-11-08T09:52:23 my brain just likes solvable frustrations 2011-11-08T09:53:29 *** Belerafon_L has joined #aichallenge 2011-11-08T09:54:17 *** Saurabh has joined #aichallenge 2011-11-08T09:55:25 *** tmandry has joined #aichallenge 2011-11-08T09:55:32 *** tmandry has joined #aichallenge 2011-11-08T09:57:07 *** NotABug has quit IRC (Ping timeout: 265 seconds) 2011-11-08T09:58:54 meh, we probably need a count down clock, only 38 days left atm 2011-11-08T09:59:07 Hi. There are question about attack. I think that the result of the battle depends only on the position of the ants on the map. There are no other factors? 2011-11-08T09:59:34 It is very simple 2011-11-08T09:59:46 *** DeGi has joined #aichallenge 2011-11-08T10:00:13 look at the http://aichallenge.org/visualizer.php?game=71133&user=3062 turn 96-98, point 56,53 2011-11-08T10:00:23 Visualiser is prophetic 2011-11-08T10:01:08 *** Huun has joined #aichallenge 2011-11-08T10:01:40 Hi 2011-11-08T10:01:41 And how to treat it? Can you give me comments about this visualizing? 2011-11-08T10:01:49 Umm? 2011-11-08T10:02:02 2 ants kill eachother? Or am I missing something 2011-11-08T10:02:26 an ant with enough coverage support won't die, an ant with equal opponents as coverage support will do suicide, an ant with not enough coverage support from his own team dies 2011-11-08T10:02:39 Yes, but on 96 turn position of the ants is the same turn 98 2011-11-08T10:02:46 (suicide, but still have a kill) 2011-11-08T10:02:50 Belerafon_L: the visualizer shows the line before the ants move to the position where they died 2011-11-08T10:03:07 If you click on the arrow that takes you step by step you notice this 2011-11-08T10:03:28 http://aichallenge.org/specification_battle.php 2011-11-08T10:03:35 *** twymer has joined #aichallenge 2011-11-08T10:04:02 *** mceier has joined #aichallenge 2011-11-08T10:04:37 pairofdice, thanks 2011-11-08T10:04:57 all clear now 2011-11-08T10:05:05 *** Belerafon_L has left #aichallenge 2011-11-08T10:05:30 *** GeorgeSebastian has joined #aichallenge 2011-11-08T10:05:33 *** GeorgeSebastian has joined #aichallenge 2011-11-08T10:05:48 finallly 2011-11-08T10:05:59 *** Belerafon_L has joined #aichallenge 2011-11-08T10:06:01 *** Belerafon_L has left #aichallenge 2011-11-08T10:06:02 with 1-ply lookahead, my bot can beat itself 2011-11-08T10:06:03 *** Belerafon_L has joined #aichallenge 2011-11-08T10:06:11 *** Belerafon_L has left #aichallenge 2011-11-08T10:07:06 haha 2011-11-08T10:07:25 the organizers should limit the amount a developer can spend developing his bot 2011-11-08T10:07:34 amount of time* 2011-11-08T10:07:43 a day 2011-11-08T10:08:05 or not 2011-11-08T10:10:50 the global loss of productivity among programmers is pretty terrifying :) 2011-11-08T10:11:13 hehe 2011-11-08T10:11:17 surely this can be spun to increase productivity 2011-11-08T10:11:26 lets get an MBA to spin it ;) 2011-11-08T10:11:28 In the long run, maybe :) 2011-11-08T10:11:32 Well, they will most likely gain some experience 2011-11-08T10:12:07 total loc/day output +2000% 2011-11-08T10:12:29 work-related loc/day output -60% 2011-11-08T10:12:34 I don't think most of us will be able to sustain that when they get back to their usual work :) 2011-11-08T10:12:35 :) 2011-11-08T10:17:36 *** delt0r_ has quit IRC (Ping timeout: 256 seconds) 2011-11-08T10:21:39 *** Harpyon has joined #aichallenge 2011-11-08T10:23:30 *** mviel__ has joined #aichallenge 2011-11-08T10:26:11 *** ikaros has quit IRC (Quit: Ex-Chat) 2011-11-08T10:26:37 *** mviel_ has quit IRC (Ping timeout: 240 seconds) 2011-11-08T10:30:10 *** delt0r_ has joined #aichallenge 2011-11-08T10:34:34 *** theskumar has joined #aichallenge 2011-11-08T10:36:03 Soon my CombatAnts achieve fusion! Or atleast slightly better combat abilities than a brick. 2011-11-08T10:36:33 I think a brick could be quite effective against ants if used right 2011-11-08T10:36:58 Tactical nuclear brick 2011-11-08T10:37:40 oh no, we don't want radioactive mut-ants 2011-11-08T10:38:10 *** HaraKiri has joined #aichallenge 2011-11-08T10:40:37 whats the official turn time limit? 2011-11-08T10:40:57 http://aichallenge.org/game_settings.php 2011-11-08T10:41:05 thx 2011-11-08T10:41:44 do you know what the tools pack's default value is? 50ms? 2011-11-08T10:42:12 No idea, you can define it yourself 2011-11-08T10:42:37 --turntime=2000 2011-11-08T10:43:30 Minthos: arent all ants mute-ants? 2011-11-08T10:44:33 *** GeorgeSebastian has quit IRC (Read error: Connection reset by peer) 2011-11-08T10:44:58 pairofdice: isn't it --turntime 5000 ? 2011-11-08T10:45:15 doesnt matter 2011-11-08T10:46:33 hm, my battle alg. sometimes doest kick in 2011-11-08T10:46:42 stpuid brainless ant 2011-11-08T10:48:41 *** olexs has quit IRC (Read error: Connection reset by peer) 2011-11-08T10:49:30 isbric: ? You can define it as whatever you want, default is 500 2011-11-08T10:50:14 Well, on the servers anyway 2011-11-08T10:50:33 pairofdice: he meant the "=" 2011-11-08T10:51:06 Oh, right 2011-11-08T10:51:33 *** datachomper has joined #aichallenge 2011-11-08T10:53:37 *** lknix has joined #aichallenge 2011-11-08T10:54:38 *** ikaros has joined #aichallenge 2011-11-08T10:56:03 *** kilae has quit IRC (Quit: ChatZilla 0.9.87 [Firefox 7.0.1/20110928134238]) 2011-11-08T10:57:01 *** Antimony has joined #aichallenge 2011-11-08T11:02:21 *** DeGi has quit IRC (Ping timeout: 265 seconds) 2011-11-08T11:05:12 *** GeorgeSebastian has joined #aichallenge 2011-11-08T11:06:32 *** Rinum has joined #aichallenge 2011-11-08T11:07:00 *** xathis has joined #aichallenge 2011-11-08T11:07:07 *** RobotCaleb has joined #aichallenge 2011-11-08T11:07:27 hmmm... I wonder who pguillory is 2011-11-08T11:09:06 his bot is really good 2011-11-08T11:09:22 *** treeform has joined #aichallenge 2011-11-08T11:09:42 i called it last night 2011-11-08T11:09:50 *** olexs1 has joined #aichallenge 2011-11-08T11:09:53 he'll be #1 within a few games 2011-11-08T11:10:22 so i am gunna enter... and *not* do the JBotAntManager. 2011-11-08T11:10:41 the current tools this time around make it somewhat redundant anyway 2011-11-08T11:10:51 esp the viewer... its very good 2011-11-08T11:11:03 and js is fast enough these days 2011-11-08T11:11:13 a1k0n: definitely! 2011-11-08T11:11:36 figured even without much time i should do tooo badly 2011-11-08T11:11:44 depending how you define badly 2011-11-08T11:11:56 should not do too badly 2011-11-08T11:12:33 C++ question: I have a class whose constuctor creates a dynamic array, and destructor deletes it. I create an instance, foo, of this class inside a function, Setup(). This function is in turn in my main(). I still would like to be able to manipulate the array in fo with it's foo's functions after my Setup() function is finished. I suspect that because foo is created inside Setup() that 2011-11-08T11:12:33 after it ends, foo will be destructed, right? If so, is there a way to preserve foo outside the scope of Setup()? 2011-11-08T11:13:18 userjjb: ? i don't quite follow 2011-11-08T11:13:35 *** tmandry has quit IRC (Ping timeout: 248 seconds) 2011-11-08T11:13:53 userjbb, make it a member of Bot, or State ? 2011-11-08T11:14:05 what is fo? 2011-11-08T11:14:16 typo: should be foo 2011-11-08T11:14:46 you are correct userjjb you can't without doing what g0llum said 2011-11-08T11:15:10 since you are *not* only using in setup then having outside is the "right way" 2011-11-08T11:15:29 Your scopes should make sense... 2011-11-08T11:15:40 ok, I figured as much 2011-11-08T11:16:32 *** roflmao has joined #aichallenge 2011-11-08T11:16:40 *** JorgeB has joined #aichallenge 2011-11-08T11:18:08 since we are all here, i have a c++ question 2011-11-08T11:18:24 lets say theres a nice struct from stl called pair 2011-11-08T11:18:35 that have fields first, second 2011-11-08T11:18:42 we're not all here, you're there .. 2011-11-08T11:18:53 now i want my own pair, with fields row, col 2011-11-08T11:19:02 but it must be compatible with pair 2011-11-08T11:19:12 i.e. must be accepted everywhere pair is 2011-11-08T11:19:47 now, how can i alias pair's members when i superclass it in my pair? 2011-11-08T11:21:00 *** aerique has quit IRC (Quit: ...) 2011-11-08T11:21:46 mcstar: getting all fancy there 2011-11-08T11:21:55 i think this is trivial 2011-11-08T11:21:58 *** Rinum has quit IRC (Quit: Page closed) 2011-11-08T11:22:00 i would just use pair, tuple etc 2011-11-08T11:22:03 sounds like it anyway 2011-11-08T11:22:26 and not bother renaming 2011-11-08T11:22:31 delt0r_: but i dont want to use .first .second 2011-11-08T11:23:01 meh... do the C thing, use a macro >: 2011-11-08T11:23:04 hehe 2011-11-08T11:23:09 just kidding btw 2011-11-08T11:23:16 is it possible that by using int& row, int& col members in my class 2011-11-08T11:23:25 that superclasses std::pair 2011-11-08T11:23:37 the compiler will actually not make 2 other fields? 2011-11-08T11:24:03 I just don't remember --there should be some way.... but the last years my C++ is working on others code. don't get to do anything that fancy 2011-11-08T11:24:20 :( 2011-11-08T11:24:28 a1k0n: maybe you can help me? 2011-11-08T11:24:39 *** boegel has quit IRC (Quit: *poof!*) 2011-11-08T11:24:54 antimatroid may know, he likes doing that kind of stuff 2011-11-08T11:25:20 he likes overloading operators 2011-11-08T11:25:29 hehe 2011-11-08T11:25:34 but i dont think he really is an OOP person 2011-11-08T11:25:38 i did too when i was young 2011-11-08T11:25:52 why not just write an own tuple class? 2011-11-08T11:26:06 ikaros: because stl thrives on the std::pair 2011-11-08T11:26:14 no but variable renaming is just what a mathematician would want to do 2011-11-08T11:26:16 will i be able to use its templated goodness? 2011-11-08T11:26:46 can you specify that mcstar ? 2011-11-08T11:27:42 mcstar: http://stackoverflow.com/questions/494597/c-member-variable-aliases 2011-11-08T11:27:45 may help 2011-11-08T11:27:50 tl;dr 2011-11-08T11:28:32 i dont see how you profit from having a stl::pair instead of an own container for 2 values.. with template or without 2011-11-08T11:29:08 but i dont see any need for this anyways :) 2011-11-08T11:29:33 ikaros: which is not the same as there is no need :) 2011-11-08T11:29:43 but meh whatever 2011-11-08T11:29:53 hehe 2011-11-08T11:29:55 I am glad to less C++ these days... 2011-11-08T11:30:06 mind you i did lots of CORBA with telecos 2011-11-08T11:30:20 not a fan 2011-11-08T11:30:36 ikaros: youre right, i could probably just change every std::pair to my pair 2011-11-08T11:30:54 the thingis that i did a hashing function for an std::pair 2011-11-08T11:31:05 probably i just need to change that to apply to my pair 2011-11-08T11:31:27 yea sounds easiest at least 2011-11-08T11:31:29 hashing functions are not a lot of work---are they 2011-11-08T11:31:33 ? 2011-11-08T11:31:49 return hash()(((long long)key.first) << 32 ^ ((long long) key.second)); 2011-11-08T11:31:56 this is mine 2011-11-08T11:31:59 for a pair 2011-11-08T11:32:00 delt0r_: A bunch of bitwise operations, maybe some arith. 2011-11-08T11:32:13 well for row col you can just do row*width + col 2011-11-08T11:32:16 should be good 2011-11-08T11:32:26 yea.. easily duplicated 2011-11-08T11:32:39 ikaros: why? 2011-11-08T11:32:42 oh god... forgot about long vers long long 2011-11-08T11:32:48 wait 2011-11-08T11:32:51 shesh what decade are we in? 2011-11-08T11:32:51 mcstar: Why an xor over or? 2011-11-08T11:32:56 more than you ever need to know: http://burtleburtle.net/bob/hash/evahash.html 2011-11-08T11:33:23 Zao: good question 2011-11-08T11:33:44 yep, since those are all zero, OR is good 2011-11-08T11:33:53 or +, which might allow slightly better instruction encodings 2011-11-08T11:34:05 (or might not) 2011-11-08T11:34:15 mcstar: Also, why a signed quantity? 2011-11-08T11:34:29 Zao: good question 2011-11-08T11:34:44 Zao: probably because i did it ~3am 2011-11-08T11:34:51 does stl do some more hashing on the hash value? 2011-11-08T11:34:52 *** amstan has joined #aichallenge 2011-11-08T11:34:52 *** ChanServ sets mode: +o amstan 2011-11-08T11:35:05 if not won't you want a lot more "randomness" 2011-11-08T11:35:23 the hashing is done by hash 2011-11-08T11:35:43 oh right 2011-11-08T11:35:46 didn't see that 2011-11-08T11:35:47 what im not sure about, is weather need to provide an == predicate 2011-11-08T11:35:57 wheather I* 2011-11-08T11:36:07 typically yes i would expect 2011-11-08T11:36:11 mcstar: You do, as hashes have a tendency to collide. 2011-11-08T11:36:19 if the hashed values match, stl needs a way to distinguish them 2011-11-08T11:36:23 mcstar: If you do not use the element-extraction parts of the container, you might be good wihtout. 2011-11-08T11:36:30 (that is, just grab buckets) 2011-11-08T11:36:30 if you do a hash function you need a == 2011-11-08T11:36:43 I'd have to hit n3290 to be sure though. 2011-11-08T11:37:40 a1k0n: thx for the link, pretty elaborate 2011-11-08T11:37:44 I always do it even if its not needed now, since i may come back to the code and expect == to work on the same fields as hash 2011-11-08T11:38:10 mcstar: as to your first question... just use first and second. :) 2011-11-08T11:38:21 I'm pretty sure that EqualityComparable is a required concept, just not if it's for the whole container or just individual member functions. 2011-11-08T11:38:36 a1k0n: im gonna make a new class, that has accessors to first and second, called row() col() 2011-11-08T11:39:08 why does it have to be a pair<> then? 2011-11-08T11:39:13 mcstar: You require std::equal_to or an alternative Pred template parameter. 2011-11-08T11:39:17 *** jasox has joined #aichallenge 2011-11-08T11:39:30 just define operator<, == 2011-11-08T11:39:59 a1k0n: yeah, when ill correct my hash ill be able to drop std::pair, and use mine with .ro .col 2011-11-08T11:40:58 Zao: a1k0n so if i define == for my data structure that i am about to hash, stl will use that, and i dont need to supply an == predicate when i make unsorted_map's for example 2011-11-08T11:41:12 *** MoleM has quit IRC (Quit: Page closed) 2011-11-08T11:41:35 (no ' there) 2011-11-08T11:42:00 ok, thx for the help 2011-11-08T11:42:01 mcstar: right 2011-11-08T11:42:38 define a global bool operator==(const blah& a, const blah& b) 2011-11-08T11:43:03 mcstar: std::equal_to's default implementation is using op==, yes. 2011-11-08T11:43:24 So defining a reachable op== will make std::equal_to instantiable, and unordered_map usable. 2011-11-08T11:43:42 k 2011-11-08T11:43:53 and is a generally useful thing on its own 2011-11-08T11:44:20 ant[i].pos == ant[j].pos, etc 2011-11-08T11:45:33 i have my ants in a linked list, and on a 2d array, pointers to them, that is 2011-11-08T11:45:39 *** Savaron has joined #aichallenge 2011-11-08T11:46:06 *** kilae has joined #aichallenge 2011-11-08T11:46:23 finally i did it properly so that they can recognize themselfes from previous turns 2011-11-08T11:52:15 *** xbelt has joined #aichallenge 2011-11-08T11:54:15 *** Antimony has quit IRC (Ping timeout: 260 seconds) 2011-11-08T11:57:08 *** ChristianK has joined #aichallenge 2011-11-08T11:57:23 *** spp has joined #aichallenge 2011-11-08T11:58:26 Wes: He wants the impossible! 2011-11-08T11:58:35 Geordi: Thats the short definition of captain. 2011-11-08T11:59:25 *** treeform has quit IRC (Remote host closed the connection) 2011-11-08T11:59:52 mcstar, you'll like #geordi 2011-11-08T12:00:26 my rank has fallen the time I've been away 2011-11-08T12:00:34 by 300 2011-11-08T12:00:35 thats bad ;) 2011-11-08T12:00:44 *** cyphase has quit IRC (Ping timeout: 256 seconds) 2011-11-08T12:00:50 "Deactivated: not in queue to play game"? 2011-11-08T12:00:58 is that because it has played 136 games? 2011-11-08T12:01:13 or played every bot? 2011-11-08T12:01:19 g0llum: :D 2011-11-08T12:01:39 aif: no its just after 3 days ;) you can activate it again 2011-11-08T12:01:44 ah 2011-11-08T12:01:54 on the bottom of the page 2011-11-08T12:02:01 man I just realised 2011-11-08T12:02:10 we're doing AI in school... 2011-11-08T12:02:20 I just wanna know if anyone is using scala? 2011-11-08T12:02:24 I should be building a tree to search 2011-11-08T12:02:44 Your tree will span the entire known universe 2011-11-08T12:02:47 *** Antimony has joined #aichallenge 2011-11-08T12:03:59 of course 2011-11-08T12:04:06 I'll just do a limited depth tree 2011-11-08T12:04:17 and ignore previous moves 2011-11-08T12:04:44 pairofdice: it wouldn't span entire known universe though, ants have limited field of view 2011-11-08T12:06:20 limited field of view of 241?... Maybe only quarter universe 2011-11-08T12:06:22 ajf: even the possible moves you have in an average single turn are more than you'll ever be able to copmute 2011-11-08T12:07:22 *** Blkt has quit IRC (Quit: gogogo) 2011-11-08T12:07:54 jix: you dont have to treat everything at once 2011-11-08T12:08:11 My combat scope is very narrow 2011-11-08T12:08:18 And pretty dumb 2011-11-08T12:08:24 jix: yes 2011-11-08T12:08:39 but a limited search I mean 2011-11-08T12:09:59 mcstar: yeah but just limiting the depth of the game tree doesn't help... 2011-11-08T12:10:25 jix: the allotted time limits it to 1-2 ply 2011-11-08T12:10:32 very sad 2011-11-08T12:10:46 *** tmandry has joined #aichallenge 2011-11-08T12:10:46 *** tmandry has joined #aichallenge 2011-11-08T12:10:52 mcstar: yeah but even to get that you have to look at isolated groups of ants 2011-11-08T12:10:56 im debugging my such tree right now 2011-11-08T12:11:02 thats right 2011-11-08T12:11:15 thats was my first point 2011-11-08T12:11:17 -s 2011-11-08T12:11:42 That's probably fine, because not all ants can affect each other within 1-2 ply :p 2011-11-08T12:12:29 oh, i thought we will have tachyon-ants 2011-11-08T12:13:04 oh only in the next contest. did noone tell you? 2011-11-08T12:13:39 together with quantum effect wormholes 2011-11-08T12:13:54 oh also 2011-11-08T12:14:05 the next contest will use quantum randomness 2011-11-08T12:14:12 mleise: can i have time-loops also? 2011-11-08T12:14:17 and will confirm to actual quantum mechanics 2011-11-08T12:14:23 but locally uses string theory 2011-11-08T12:14:31 *** ajf is now known as ajf|offline 2011-11-08T12:14:46 mcstar: I think there *was* even a real time strategy game with time-loops in it. Pretty weird for the AI as well. 2011-11-08T12:15:21 i dont even know what that can be :D 2011-11-08T12:15:43 or rather "cant even imagine" 2011-11-08T12:17:07 *** cyphase has joined #aichallenge 2011-11-08T12:17:12 mcstar: here it is: http://www.achrongame.com/site/ 2011-11-08T12:17:44 *** tmandry has quit IRC (Ping timeout: 256 seconds) 2011-11-08T12:18:04 How would a time-loop work in a real time game 2011-11-08T12:18:10 mind-blowing 2011-11-08T12:18:51 so i can fight again past battles? 2011-11-08T12:19:02 how would they impact my future? i.e. the present? 2011-11-08T12:19:12 wtf anyway? 2011-11-08T12:19:53 *** smiley1983 has quit IRC (Ping timeout: 260 seconds) 2011-11-08T12:19:53 Dr. Who is the only instance of time travel I don't hate 2011-11-08T12:20:15 *** smiley1993 has quit IRC (Ping timeout: 248 seconds) 2011-11-08T12:20:29 1969? 2011-11-08T12:20:37 the best time-travels episode ever 2011-11-08T12:20:41 The new one 2011-11-08T12:20:41 of course its stargate 2011-11-08T12:20:48 Time travel blows... it never works 2011-11-08T12:21:00 grandfathers and all that 2011-11-08T12:21:26 mleise: im trying to buy their game, but their captcha is too complicated, i cant figure it out 2011-11-08T12:21:33 lol 2011-11-08T12:21:33 lol 2011-11-08T12:21:36 *** kaemo has joined #aichallenge 2011-11-08T12:21:41 you are a machine mcstar 2011-11-08T12:21:48 a poor one 2011-11-08T12:21:58 i always had my suspicion's. 2011-11-08T12:22:06 *** smiley1983 has joined #aichallenge 2011-11-08T12:22:14 I liked the video where they produced mechs in the present and sent them to the past to assist in an assault 2011-11-08T12:22:15 i just wanted to know how much that costs 2011-11-08T12:22:23 I bet this game would give me headaches 2011-11-08T12:23:00 what are the conservation laws in that game? 2011-11-08T12:23:11 where are my troops? oh the factory was destroyed by my enemy 20 minutes in the past with troops he just produced 2011-11-08T12:23:14 is matter a conserved quantity in time? 2011-11-08T12:23:18 energy? momentum? 2011-11-08T12:23:27 in time->across time 2011-11-08T12:23:57 I have no idea 2011-11-08T12:24:10 if not, you can have positive gain loops 2011-11-08T12:24:20 feedback loops 2011-11-08T12:25:49 maybe that's what created big bang 2011-11-08T12:26:13 *** smiley1983 has quit IRC (Ping timeout: 245 seconds) 2011-11-08T12:26:24 someone created an infinite loop that just sent more and more matter into the same point in the past 2011-11-08T12:27:08 until timespace broke down and reality unraveled 2011-11-08T12:27:18 *** smiley1983 has joined #aichallenge 2011-11-08T12:27:24 Mcstar-Mintos creaton myth 2011-11-08T12:28:05 Minthos: actually this mechanism might not be as stupid as it first sounds 2011-11-08T12:28:15 do you want the jimi hendeix, or the stravinsky soundtrack ? 2011-11-08T12:28:54 the from 2001 2011-11-08T12:29:00 r something 2011-11-08T12:30:29 http://www.youtube.com/watch?v=ys8HiGObdHI&feature=related 2011-11-08T12:30:56 strauss waltzer, fits for a hungarian ! 2011-11-08T12:31:09 strauss 2011-11-08T12:31:13 zarathrustra 2011-11-08T12:31:15 wtf 2011-11-08T12:31:18 im soo bad with names 2011-11-08T12:31:31 *** smiley1993 has joined #aichallenge 2011-11-08T12:31:52 *** thestinger has joined #aichallenge 2011-11-08T12:32:47 Hi. I was wondering: does the engine informs us about the hive amount we have or do we have to maintain our hive level ? 2011-11-08T12:33:28 mleise: i like that game, i saw a trailer 2011-11-08T12:33:30 you have to keep track 2011-11-08T12:34:06 *** liberforce has left #aichallenge 2011-11-08T12:36:23 I wonder what happens if in the past the enemy destroyed my mech factory, but in the present it is still there (until the timewave arrives) and I produce a mech there, and send it back in time. Would it have disappeared if I kept it in the present? Will it stay alive in the past? 2011-11-08T12:36:47 *mind blown* \o/ 2011-11-08T12:37:00 *** Hexren has quit IRC (Quit: Leaving.) 2011-11-08T12:37:20 mleise: you get to keep it 2011-11-08T12:37:29 Bluedgis: the game tells you all the visible hives at any time 2011-11-08T12:37:47 Bluedgis: so you need to track them yourself if you want more info 2011-11-08T12:38:07 what about your own hives... are they always visible? 2011-11-08T12:38:11 nope 2011-11-08T12:38:36 *** peyton has quit IRC (Quit: peyton) 2011-11-08T12:38:45 do you see that a hive has been "capped" or whatever its called? 2011-11-08T12:39:00 destroyed 2011-11-08T12:39:03 cus i don't see that in the spec... could have just missed it however 2011-11-08T12:39:08 you can find out if you track them 2011-11-08T12:39:27 How... if you don't see then get destroyed by others 2011-11-08T12:39:39 or do you just not see them anymore after that? 2011-11-08T12:39:42 you need to scout the location before and after the destruction 2011-11-08T12:41:16 https://github.com/aichallenge/aichallenge/blob/epsilon/worker/sandbox.py 383 2011-11-08T12:41:46 https://github.com/aichallenge/aichallenge/blob/epsilon/worker/sandbox.py 383 str, on wondows it not worck, complaty ignored 2011-11-08T12:41:49 *** peyton has joined #aichallenge 2011-11-08T12:42:33 i know that my ants dont try to attack hives once they are destroyed, so i gess they arent sent anymoar 2011-11-08T12:42:57 but it might just be that i programmed into that behavior wo knowing it 2011-11-08T12:43:15 Accoun: submit a bug report 2011-11-08T12:43:17 I've got to program A's ant zapping technology 2011-11-08T12:43:19 I WILL HAVE IT 2011-11-08T12:43:26 cluster two ants together and attack lone ant 2011-11-08T12:43:29 you save yours heloses his 2011-11-08T12:44:01 i wonder when ant-pairs become superconducting 2011-11-08T12:44:23 hehe 2011-11-08T12:44:42 phonos are not strong enough with ants 2011-11-08T12:45:28 we just need a Cu2o2 lattice instead of water 2011-11-08T12:45:36 *** ajf|offline is now known as ajf 2011-11-08T12:45:59 maybe O4? im not sure about bonding 2011-11-08T12:46:16 na... carbon nanotubes, very fast phonon velocity 2011-11-08T12:47:28 *** peyton has quit IRC (Quit: peyton) 2011-11-08T12:47:32 *** xbelt has quit IRC (Remote host closed the connection) 2011-11-08T12:47:39 *** xbelt has joined #aichallenge 2011-11-08T12:47:42 *** GeorgeSebastian has quit IRC (Ping timeout: 252 seconds) 2011-11-08T12:47:47 really? 2011-11-08T12:47:49 why? 2011-11-08T12:48:01 the more strongly coupled somethig is the faster are the phonons 2011-11-08T12:48:09 i want all my leptones to be spin-coupled over large distance 2011-11-08T12:48:12 nanotubes are essentiall 2d 2011-11-08T12:48:45 *** peyton has joined #aichallenge 2011-11-08T12:50:09 leptones? just don't stare at the chrystal skull 2011-11-08T12:51:43 *** kaemo has quit IRC (Ping timeout: 260 seconds) 2011-11-08T12:51:47 so ants has turned into quasi quantum particles? intresting 2011-11-08T12:51:56 lol, yesterday there was someone named yorick, or similar here 2011-11-08T12:52:16 poor yorick 2011-11-08T12:52:28 I knew him, a fellow of infinite jest 2011-11-08T12:52:58 i wonder how lep-tones might sound 2011-11-08T12:53:26 i watch Crystall Skull every time as if i were seening it the first time 2011-11-08T12:53:30 timeless epsiode 2011-11-08T12:53:43 episode of what? 2011-11-08T12:53:50 stargate of course 2011-11-08T12:54:05 prof. nick ballard 2011-11-08T12:54:07 "of course" 2011-11-08T12:54:15 and the smoke giants 2011-11-08T12:54:25 i forget.. what did the crystal skull do 2011-11-08T12:54:27 ? 2011-11-08T12:55:18 transported the one who stare into its eyes to gigantic ziccurat 2011-11-08T12:55:27 to another planet, where smoke giants lived 2011-11-08T12:55:33 *** xbelt has quit IRC (Remote host closed the connection) 2011-11-08T12:55:35 and they were opposed to the goauld 2011-11-08T12:55:41 *** xbelt has joined #aichallenge 2011-11-08T12:56:17 oh 2011-11-08T12:56:19 yep 2011-11-08T12:56:50 jacks great joke on the neutrinos: nintendos 2011-11-08T12:56:56 hehe 2011-11-08T12:56:59 but im not sure if ti was in this episode 2011-11-08T12:57:21 I love mcgyver in sg1 2011-11-08T12:57:26 great lines 2011-11-08T12:57:53 without him, the show couldnt have reached 10 seasons 2011-11-08T12:58:09 or two 2011-11-08T12:58:16 maybe 2 2011-11-08T12:58:25 http://www.youtube.com/watch?v=HzJwttD4jr4 2011-11-08T12:58:38 it's... different from how I rememer it 2011-11-08T12:58:38 the show increasingly becomes better, sort of 2011-11-08T12:58:44 something is wrong with their voices... 2011-11-08T12:59:02 but it needed richard to kickstart it, and keep it running in the first seasons 2011-11-08T12:59:18 emilk: dont watch the german dub 2011-11-08T12:59:27 =) 2011-11-08T12:59:33 the only one I could find on youtube! 2011-11-08T12:59:52 fuckme 2011-11-08T12:59:56 emilk: i was joking 2011-11-08T13:00:02 didnt know it was really german 2011-11-08T13:00:20 you said that before clicking the link? 2011-11-08T13:00:32 i cliked the link, but my flashplugin crashed 2011-11-08T13:00:38 awesome 2011-11-08T13:00:39 and couldnt see what you linked 2011-11-08T13:00:45 points to you 2011-11-08T13:08:47 *** xbelt has quit IRC (Read error: Connection reset by peer) 2011-11-08T13:08:48 Is there any differences between the game engine in the tools pack, and the one used when a bot is submitted? 2011-11-08T13:09:07 *** peyton has quit IRC (Quit: peyton) 2011-11-08T13:10:09 Huun, no. 2011-11-08T13:10:58 *** xbelt has joined #aichallenge 2011-11-08T13:11:44 Because when i submit, my bot just crash executing the test. And it never crashed in all any recent test i did (50-100 last tests) 2011-11-08T13:13:10 Huun, there's alot of other factors, differernt maps, different opponents 2011-11-08T13:14:50 aww, got that wrong. ignore me 2011-11-08T13:14:54 *** peyton has joined #aichallenge 2011-11-08T13:16:09 *** xbelt has quit IRC (Remote host closed the connection) 2011-11-08T13:16:11 *** mviel__ has quit IRC (Remote host closed the connection) 2011-11-08T13:16:17 *** xbelt has joined #aichallenge 2011-11-08T13:16:18 *** caution has joined #aichallenge 2011-11-08T13:16:21 *** xathis has quit IRC (Read error: Connection reset by peer) 2011-11-08T13:16:53 Huun: have you tried test_bot.sh? 2011-11-08T13:17:26 *** treeform has joined #aichallenge 2011-11-08T13:18:29 No 2011-11-08T13:19:20 I'll do it 2011-11-08T13:21:50 *** kaemo has joined #aichallenge 2011-11-08T13:23:36 *** xbelt has quit IRC (Remote host closed the connection) 2011-11-08T13:23:44 *** xbelt has joined #aichallenge 2011-11-08T13:26:27 Well there is no problem, the game doesn't crash when i run test_bot.cmd 2011-11-08T13:28:23 *** peyton has quit IRC (Quit: peyton) 2011-11-08T13:29:50 *** Antimony has quit IRC (Ping timeout: 260 seconds) 2011-11-08T13:34:47 *** Antimony has joined #aichallenge 2011-11-08T13:35:07 How could i debug a crash that occurs only when i submit a bot. And the error appens on a check that occurs 100-200 by turn. 2011-11-08T13:35:21 *** peyton has joined #aichallenge 2011-11-08T13:36:04 *** Stivo has joined #aichallenge 2011-11-08T13:36:07 It should be reproducible. That's what i don't understand. WHen i submit it's systematic. 2011-11-08T13:36:55 Can I use code written by someone else found on the internet? 2011-11-08T13:37:36 use the same map and seed as on the server and repeat until you experience it 2011-11-08T13:38:09 compare the differences between your environment and the server environment 2011-11-08T13:42:30 *** marijnfs has joined #aichallenge 2011-11-08T13:44:31 *** amstan has quit IRC (Ping timeout: 248 seconds) 2011-11-08T13:44:32 *** ztfw has joined #aichallenge 2011-11-08T13:44:49 *** xbelt has quit IRC (Remote host closed the connection) 2011-11-08T13:44:56 *** xbelt has joined #aichallenge 2011-11-08T13:45:14 *** buq2_ has joined #aichallenge 2011-11-08T13:45:20 *** kilae has quit IRC (*.net *.split) 2011-11-08T13:45:21 *** Palmik has quit IRC (*.net *.split) 2011-11-08T13:45:22 *** luizribeiro has quit IRC (*.net *.split) 2011-11-08T13:45:23 *** Bluedgis has quit IRC (*.net *.split) 2011-11-08T13:45:23 *** Sir_Ragnarok has quit IRC (*.net *.split) 2011-11-08T13:45:23 *** aarossig_ has quit IRC (*.net *.split) 2011-11-08T13:45:24 *** buq2 has quit IRC (*.net *.split) 2011-11-08T13:45:24 *** knyppeldynan has quit IRC (*.net *.split) 2011-11-08T13:45:25 *** knyppeld1nan has joined #aichallenge 2011-11-08T13:45:27 *** kaemo has quit IRC (Quit: Lost terminal) 2011-11-08T13:45:49 *** Antimony has quit IRC (Ping timeout: 255 seconds) 2011-11-08T13:46:17 *** kilae has joined #aichallenge 2011-11-08T13:46:17 *** Palmik has joined #aichallenge 2011-11-08T13:46:17 *** luizribeiro has joined #aichallenge 2011-11-08T13:46:17 *** Bluedgis has joined #aichallenge 2011-11-08T13:46:17 *** Sir_Ragnarok has joined #aichallenge 2011-11-08T13:46:17 *** aarossig_ has joined #aichallenge 2011-11-08T13:46:34 *** aarossig_ has quit IRC (Ping timeout: 252 seconds) 2011-11-08T13:46:47 *** aarossig has joined #aichallenge 2011-11-08T13:48:55 .py3 is the extension for python 3.x? 2011-11-08T13:49:46 *** spp has quit IRC (Quit: Page closed) 2011-11-08T13:50:08 *** xbelt has quit IRC (Remote host closed the connection) 2011-11-08T13:50:17 *** xbelt has joined #aichallenge 2011-11-08T13:50:27 *** sh4wn has joined #aichallenge 2011-11-08T13:51:20 Yeah, though my py3 bot runs fine as a py2 bot.. 2011-11-08T13:52:09 oO 2011-11-08T13:52:28 My py3 bot don't crash as a py bot :s 2011-11-08T13:52:46 When i submit it. 2011-11-08T13:53:05 Mmh... my modules should be named .py3 as well? 2011-11-08T13:55:11 *** lu4 has joined #aichallenge 2011-11-08T13:55:40 *** xbelt has quit IRC (Remote host closed the connection) 2011-11-08T13:55:49 *** xbelt has joined #aichallenge 2011-11-08T13:57:03 *** Joris_ has joined #aichallenge 2011-11-08T13:58:16 *** conor_f has joined #aichallenge 2011-11-08T13:58:48 *** xbelt has quit IRC (Remote host closed the connection) 2011-11-08T13:58:51 im soo stupid 2011-11-08T13:58:56 *** xbelt has joined #aichallenge 2011-11-08T13:59:12 should have use a pos struct waay before, and have [] overloaded too 2011-11-08T13:59:16 *** epicmonkey has joined #aichallenge 2011-11-08T13:59:17 used* 2011-11-08T14:00:08 anyone knows if I can use java code in a scala project? 2011-11-08T14:02:06 *** xbelt has quit IRC (Remote host closed the connection) 2011-11-08T14:02:16 *** xbelt has joined #aichallenge 2011-11-08T14:04:27 I have a C# library with implementations of binary heaps and other things, is it not possible to use them for this project? 2011-11-08T14:04:35 *** ChristianK has quit IRC (Remote host closed the connection) 2011-11-08T14:04:56 *** Antimony has joined #aichallenge 2011-11-08T14:05:04 Savaron: I doubt google will let you run pre-compiled binarys... 2011-11-08T14:05:35 Google? 2011-11-08T14:05:55 I thought google were running this? 2011-11-08T14:06:11 they're a sponsor 2011-11-08T14:06:23 it's apparently "google sponsored" but that does not mean very much 2011-11-08T14:06:33 Savaron, source code, or it didn't happen 2011-11-08T14:06:40 Well, i only have python 3.2 on my computer and my bot runs fine. But when i upload my bot, it fails test if it's named MyBot.py3 but runs fine as a .py... Any idea? 2011-11-08T14:06:42 k, thanks 2011-11-08T14:06:45 oh, I've been thinking wrong this whole time :/ 2011-11-08T14:06:52 just found the source and hope it's usable ;) 2011-11-08T14:07:16 google sponsored, not google run 2011-11-08T14:07:27 d 2011-11-08T14:07:41 *** links234 has joined #aichallenge 2011-11-08T14:07:42 Huun type 'python --version' 2011-11-08T14:07:55 what test does it fail? 2011-11-08T14:08:42 actually, you can try using binaries 2011-11-08T14:08:52 if it was compiled on a similar system it could run 2011-11-08T14:09:03 a c# binary might even be more compatible 2011-11-08T14:09:07 since its clr 2011-11-08T14:09:10 Ok, holy crap. my python compiler is still 2.6 i tought i got rid of it times ago :(. 2011-11-08T14:09:21 Thanks TheLinker 2011-11-08T14:09:25 yw 2011-11-08T14:09:27 and if scala compiles to java, im sure you can use any java code 2011-11-08T14:10:49 *** tmandry has joined #aichallenge 2011-11-08T14:10:49 *** tmandry has joined #aichallenge 2011-11-08T14:11:24 *** yassya has joined #aichallenge 2011-11-08T14:12:02 Is there a way to rename my account 2011-11-08T14:12:04 ? 2011-11-08T14:13:37 anybody? 2011-11-08T14:14:08 not afaik 2011-11-08T14:15:38 I'm not allowed to create second account, and not allowed to change my name 2011-11-08T14:15:48 this seems pretty dummy 2011-11-08T14:16:03 (it) seems... 2011-11-08T14:16:17 lu4: what's your old account? 2011-11-08T14:16:27 iam-lu4 2011-11-08T14:16:34 think of how a database works, and be a good looser 2011-11-08T14:16:36 link / user number 2011-11-08T14:17:05 what do you mean by be a good looser? 2011-11-08T14:17:23 i can disable your account and you can sign up for a new one, if you really want 2011-11-08T14:17:36 stick with your 1st name 2011-11-08T14:17:37 I would appreciate that 2011-11-08T14:17:47 why do you want a new username anyways? 2011-11-08T14:18:09 At first it didn't mattered 2011-11-08T14:18:18 but i'm getting annoyed 2011-11-08T14:18:27 with my current account name 2011-11-08T14:18:30 it doesn't matter that much, really 2011-11-08T14:18:31 it really bugs me 2011-11-08T14:18:36 ))) 2011-11-08T14:18:36 Deal with it 2011-11-08T14:18:37 i know 2011-11-08T14:18:38 lol, whatever :P 2011-11-08T14:18:53 *** ztfw has quit IRC (Read error: Connection reset by peer) 2011-11-08T14:18:54 but give me a link to your profile :P 2011-11-08T14:19:16 http://aichallenge.org/profile.php?user=4902 2011-11-08T14:19:45 so can you disable it? 2011-11-08T14:20:00 *** ztfw has joined #aichallenge 2011-11-08T14:20:19 *** ztfw has quit IRC (Remote host closed the connection) 2011-11-08T14:20:34 *** Redgis has joined #aichallenge 2011-11-08T14:20:39 i bet this thing will reach the 5-digit user nr ;) 2011-11-08T14:21:55 Agh, it started to timeout at 2000, I bet it would still work on the servers but... I'm probably doing many things horribly wrong 2011-11-08T14:22:38 pairofdice: same thing was happening to me. then I wrote down exactly what I wanted to do and started from scratch 2011-11-08T14:22:55 lu4: go ahead and make your new account 2011-11-08T14:23:02 thanks!!! 2011-11-08T14:23:12 a lot!! 2011-11-08T14:23:23 but i won't do that again, so choose wisely 2011-11-08T14:23:31 as a matter of intrest, whats your new name going to be? :P 2011-11-08T14:24:12 hmm it doesn't seems that my old account is disabled... 2011-11-08T14:25:00 are you logging into it? 2011-11-08T14:25:04 yep 2011-11-08T14:25:08 without a problem 2011-11-08T14:25:16 it appears disabled to me 2011-11-08T14:25:28 are you able to upload anything to it? 2011-11-08T14:25:39 I'm able to 2011-11-08T14:25:42 log in 2011-11-08T14:25:57 though the rank says "Not Ranked" 2011-11-08T14:26:05 and skill "No Skillz" 2011-11-08T14:26:43 it looks like your most recent submission was deactivated 2011-11-08T14:26:53 so the bot is off 2011-11-08T14:27:06 you're probably fine 2011-11-08T14:27:17 probably? :))) 2011-11-08T14:27:17 amstan probably knows more about this part of the system 2011-11-08T14:27:51 does this means that I'm allowed to create new account 2011-11-08T14:27:51 ? 2011-11-08T14:28:17 I don't want to be banned forever 2011-11-08T14:29:36 yes, your old account is disabled, unless you can upload to it 2011-11-08T14:29:58 gtg 2011-11-08T14:30:23 did you authenticate lu4 as the real account owner? 2011-11-08T14:30:27 lu4: whats your new name going to be? 2011-11-08T14:31:17 thank you Zannick 2011-11-08T14:32:03 who? 2011-11-08T14:32:19 I mean 2011-11-08T14:32:48 I mean who were you asking about authentication? 2011-11-08T14:33:04 I am the account owner 2011-11-08T14:34:41 poor guy will be mad tomorrow when he find out his account is disabled 2011-11-08T14:34:47 s* 2011-11-08T14:34:52 no 2011-11-08T14:34:56 it's my 2011-11-08T14:35:03 it's my account 2011-11-08T14:35:09 it was my account 2011-11-08T14:35:14 :D 2011-11-08T14:35:24 i'm usually sign in as lu4 2011-11-08T14:35:46 though the aichallenge.org has a limitation on account name length 2011-11-08T14:35:57 so you changed from iam-lu4 to? 2011-11-08T14:36:08 cholera... 2011-11-08T14:36:09 )) 2011-11-08T14:36:28 lol, ok :P 2011-11-08T14:37:04 lol 2011-11-08T14:37:13 after changing from std::pair and _pair to pos, ive got a 90k compilation error 2011-11-08T14:37:13 :P 2011-11-08T14:37:22 lol 2011-11-08T14:37:31 g++ ftw :P 2011-11-08T14:37:37 :) 2011-11-08T14:37:38 bit verbose 2011-11-08T14:37:57 words I hate seeing: "candidates are:: >_< 2011-11-08T14:38:07 *"candidates are:" 2011-11-08T14:38:11 helpers.hh:1:24: error: #include nested too deeply 2011-11-08T14:38:30 never saw that one.. lol 2011-11-08T14:38:47 neither :P 2011-11-08T14:40:00 *** timmaxw has joined #aichallenge 2011-11-08T14:40:15 *** retybok_ has joined #aichallenge 2011-11-08T14:41:24 *** foRei has joined #aichallenge 2011-11-08T14:41:26 bah days wasted 2011-11-08T14:41:42 *** NoxiaZ^ has joined #aichallenge 2011-11-08T14:41:59 wrote a completely new bot with a different strategy to discover that it behaves worse than my first one :) 2011-11-08T14:42:16 you learned something 2011-11-08T14:42:21 true :) 2011-11-08T14:42:41 *** theskumar has quit IRC (Remote host closed the connection) 2011-11-08T14:42:42 *** Saurabh has quit IRC (Read error: Connection reset by peer) 2011-11-08T14:42:51 guess im giving it two days to try and tune it 2011-11-08T14:42:53 Does the contest specify what machine architecture the bots will be run on? I'm considering attempting some low-level assembly magic but it might be platform-dependent. 2011-11-08T14:43:10 why assembly? =) 2011-11-08T14:43:22 Specifically, I'm considering using coroutines. 2011-11-08T14:43:29 (In C++) 2011-11-08T14:43:48 embed lua.h and use those coroutines :) 2011-11-08T14:43:53 lu4: actually if you're still logged in to your old account, could you edit your about me to have a link to your new account 2011-11-08T14:44:29 at least this would be a less painful way i guess 2011-11-08T14:44:34 Does Lua actually use multiple OS-level stacks? I thought it just swapped out the Lua stack. 2011-11-08T14:44:56 I guess if it did that it couldn't switch coroutines from within a cfunction... 2011-11-08T14:45:23 timmaxw: setcontext, makecontext, etc.? 2011-11-08T14:45:53 a destructer doesn't get run after the constructer, does it? 2011-11-08T14:46:26 strictly after it 2011-11-08T14:46:31 I thought what happened was a new class instance and the constructer is run. then when it's deleted, the destructer? 2011-11-08T14:46:34 amirite? 2011-11-08T14:47:01 *** xbelt has quit IRC (Ping timeout: 255 seconds) 2011-11-08T14:47:01 is that a smei-precious stone? amirite? 2011-11-08T14:47:09 timmaxw, not sure about this but i would google it up before going assembly 2011-11-08T14:47:15 funny, mcstar, funny indeed 2011-11-08T14:47:17 thestinger: Something like that 2011-11-08T14:47:39 this is strange then 2011-11-08T14:47:57 *** tmandry has quit IRC (Ping timeout: 260 seconds) 2011-11-08T14:48:02 the destructer is being run before the class instance is being destroyed 2011-11-08T14:48:29 probably because you assign an object with = 2011-11-08T14:48:33 *** MoleM has joined #aichallenge 2011-11-08T14:48:38 and the standard constructor was called before 2011-11-08T14:48:41 ikaros: elaborate? 2011-11-08T14:48:48 but you didnt implement it and so you cant see? 2011-11-08T14:48:53 know what i mean? 2011-11-08T14:48:55 any admin here? 2011-11-08T14:49:13 like Bot b; b = Bot(500); 2011-11-08T14:49:20 ikaros: not really... I'm pushing_back()/eraseing() them 2011-11-08T14:49:38 I suppose something similar in the push_back: 2011-11-08T14:49:40 ikaros: I can't find any information online about the platform that the code will be run on. That's why I'm asking here. Maybe I should email the contest admins. 2011-11-08T14:49:53 yea 2011-11-08T14:50:05 i guess i read it was ubuntu 11.04 64 bit 2011-11-08T14:50:14 but better ask 2011-11-08T14:50:43 yea conor_f do you have the standard constructor implemented? 2011-11-08T14:50:44 grmbl....never received an activation mail 2011-11-08T14:50:47 and am pretty sure i used the right email adress 2011-11-08T14:50:50 state.myAnts.push_back(Ant(state.myAnts_on_board[i], state)); 2011-11-08T14:50:53 if not do it and put a cout in 2011-11-08T14:51:07 conor_f: use emplace_back :P 2011-11-08T14:51:18 where do you init that vector? 2011-11-08T14:51:33 in the State struct, way before 2011-11-08T14:51:40 thestinger: private? 2011-11-08T14:51:40 well 2011-11-08T14:51:47 once a turn actually ikaros 2011-11-08T14:51:54 it gets updated 2011-11-08T14:51:59 *** pairofdice has quit IRC (Quit: Just keep breathing) 2011-11-08T14:52:01 *** MoleM has quit IRC (Client Quit) 2011-11-08T14:52:13 hm 2011-11-08T14:52:14 thestinger: wut?? 2011-11-08T14:52:51 *** hacklash has joined #aichallenge 2011-11-08T14:52:52 i'd try a print in the standard constructor first.. to see if it's called (it should be if a destructor is called) =) 2011-11-08T14:53:03 the constructer is getting called 2011-11-08T14:53:05 for sure 2011-11-08T14:54:05 and so is the destructer, every turn :/ 2011-11-08T14:54:33 you said you assign a new vector every turn right? 2011-11-08T14:54:34 *** modafinil is now known as number_prophet 2011-11-08T14:54:49 ikaros: well, I actually just update the vector every turn 2011-11-08T14:55:01 loop through it and erase/push_back as fit 2011-11-08T14:55:19 *** lu4 has quit IRC (Ping timeout: 265 seconds) 2011-11-08T14:55:57 *** Skinny has joined #aichallenge 2011-11-08T14:56:01 then it's no wonder a destructor is called right? if you erase elements? or i just dont get it 2011-11-08T14:56:02 =) 2011-11-08T14:56:33 the destructer is called even though nothing is being erased is the problem :P 2011-11-08T14:56:36 hi all 2011-11-08T14:56:40 lol ok 2011-11-08T14:57:30 *** curious__ has joined #aichallenge 2011-11-08T14:57:38 i am currently testing my bot using the tools on windows, but what could be wrong if I order my bot to go east on turn X, it comes back south on the next turn ? 2011-11-08T14:58:27 no water, other ants or food in range btw.. 2011-11-08T14:59:28 I tried to rule out programming errors from my side, but when I replay the scenario step by step, it comes back south when ordered to go east 2011-11-08T14:59:43 *** Antimony has quit IRC (Ping timeout: 252 seconds) 2011-11-08T15:00:07 *** Accoun has quit IRC () 2011-11-08T15:01:42 http://ants.fluxid.pl/replay.26408 2011-11-08T15:01:47 my bot knows how to flank <3 (end) 2011-11-08T15:05:52 *** timmaxw has quit IRC (Quit: Page closed) 2011-11-08T15:06:34 does this make any sort of sense to anyone? 2011-11-08T15:06:51 global scope, I have a variable which is defined to = 0 2011-11-08T15:07:13 in the constructer (which is getting run) I tell it to increment this variable 2011-11-08T15:07:25 in the destructer, I tell it to decrement 2011-11-08T15:07:39 it gets decremented even though the constructer is run... 2011-11-08T15:08:03 you mean its < 0? 2011-11-08T15:08:15 yeah 2011-11-08T15:08:20 even though it starts at 0 2011-11-08T15:08:26 *** Antimony has joined #aichallenge 2011-11-08T15:08:27 and you have checked the standard constructor? 2011-11-08T15:08:31 I end up having -7 ants/// 2011-11-08T15:08:33 *... 2011-11-08T15:08:49 standard constructor is like Foo::Foo(), yeah? 2011-11-08T15:08:54 conor_f: c++11? 2011-11-08T15:08:54 yep 2011-11-08T15:09:00 mcstar: yeah 2011-11-08T15:09:03 Anti-ants :).. might give you a show when they bump into regular ant :) 2011-11-08T15:09:03 *** lorill has joined #aichallenge 2011-11-08T15:09:26 ikaros: yeah, chekced it 2011-11-08T15:09:30 *** zaphus has joined #aichallenge 2011-11-08T15:09:35 it's incrementing in ther (well, should be) 2011-11-08T15:09:51 hmm 2011-11-08T15:11:08 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has joined #aichallenge 2011-11-08T15:15:05 *** Joris_ has quit IRC (Quit: Page closed) 2011-11-08T15:18:08 *** curious__ has quit IRC (Quit: Page closed) 2011-11-08T15:19:11 *** mceier has quit IRC (Ping timeout: 276 seconds) 2011-11-08T15:20:30 *** mceier has joined #aichallenge 2011-11-08T15:21:55 *** g0llum has quit IRC (Read error: Connection reset by peer) 2011-11-08T15:23:14 *** Antimony_ has joined #aichallenge 2011-11-08T15:23:45 *** Accoun has joined #aichallenge 2011-11-08T15:25:29 *** Antimony has quit IRC (Ping timeout: 260 seconds) 2011-11-08T15:25:38 *** Antimony_ is now known as Antimony 2011-11-08T15:26:31 *** zaphus has quit IRC (Quit: Page closed) 2011-11-08T15:27:00 *** Relax has joined #aichallenge 2011-11-08T15:29:13 this ranking system is stupid 2011-11-08T15:29:45 I'm playing people ranked 500 places ahead of me and since I lose, I get my skill reduced 2011-11-08T15:30:19 what do you expect? gain points for loosing? :-) 2011-11-08T15:30:41 I beat someone ranked above me and still go down 2011-11-08T15:30:55 how that makes any sort of sense, I dunno 2011-11-08T15:31:28 that has something to do with mean and sigma of a distribution if i understood it correctly 2011-11-08T15:31:49 yeah I read about trueskill 2011-11-08T15:31:51 well, i'm ranked 1000 something 2011-11-08T15:31:53 still stupid though 2011-11-08T15:31:56 beat all players ranked 100 2011-11-08T15:32:02 the system did not expect you to beat that guy, so its uncertainty in the capabilty of you increases 2011-11-08T15:32:03 and end un being 700 for now 2011-11-08T15:32:03 *** links234 has quit IRC (Ping timeout: 265 seconds) 2011-11-08T15:32:12 still have to wait at least 3 games to see anything 2011-11-08T15:32:18 if you keep beating them you should go up 2011-11-08T15:32:19 this sucks at each new release :( 2011-11-08T15:32:20 yea, so? 2011-11-08T15:32:36 whats wrong with saying if you beat someone higher then you, you move one place above them? 2011-11-08T15:32:39 so i have to wait 3 days after each release to see if it's better than the previous 2011-11-08T15:32:40 this is not a flaw of the system in my oppinion :) 2011-11-08T15:32:45 and if you lose, you go down 2011-11-08T15:33:03 it ends up correctly, but it's soooo long 2011-11-08T15:33:09 how do you decide if you win because of skill or because of luck? 2011-11-08T15:33:18 with more games 2011-11-08T15:33:21 doesn't matter 2011-11-08T15:33:24 you still won 2011-11-08T15:33:30 if i by chance beat #3 should i be placed above them? 2011-11-08T15:33:42 all the time? 2011-11-08T15:33:48 *** peyton has quit IRC (Quit: peyton) 2011-11-08T15:33:58 ye, the problem is more that going up again takes forever :) 2011-11-08T15:33:59 no, i don't think so 2011-11-08T15:34:06 in an exam, you're not graded on whether or not you know the stuff, you're graded on how the exam went 2011-11-08T15:34:10 given luck and everything 2011-11-08T15:34:13 but still... with more games, it would be ok i guess 2011-11-08T15:34:14 so? 2011-11-08T15:34:15 no-one has an issue with that... 2011-11-08T15:34:23 so? 2011-11-08T15:34:28 so why not use it here? 2011-11-08T15:34:45 because 1 game isn't meaningful 2011-11-08T15:34:51 well, you are 2011-11-08T15:35:01 you should gain mu for each win 2011-11-08T15:35:15 seems excessive 2011-11-08T15:35:19 but the sigma also might increase. 2011-11-08T15:35:32 if it is because of skill you will go up eventually 2011-11-08T15:35:42 hm. trueskill as an exam grading mechanism 2011-11-08T15:35:48 lol Zannick 2011-11-08T15:35:51 :D 2011-11-08T15:36:32 lol I have absolutely no idea whats going wrong in my code :P 2011-11-08T15:36:55 a variable is being incremented, then decremented without ever being told to decrement 2011-11-08T15:38:04 damn variables. stop using them 2011-11-08T15:38:11 they cant be trusted 2011-11-08T15:38:32 always changing and stuff. so unreliable :P 2011-11-08T15:38:42 constants are the way to go :) 2011-11-08T15:38:44 conor_f: Any possibility its a scoping issue? 2011-11-08T15:38:59 don't think so tbh 2011-11-08T15:39:11 it's global so it can't be really 2011-11-08T15:39:29 and the only other scope is within a class which has only 1 instance 2011-11-08T15:39:33 so no, not scoping 2011-11-08T15:39:55 maybe it's with push_back/erase 2011-11-08T15:40:25 *** peyton has joined #aichallenge 2011-11-08T15:40:50 *** amstan has joined #aichallenge 2011-11-08T15:40:50 *** ChanServ sets mode: +o amstan 2011-11-08T15:40:50 crazy question 2011-11-08T15:41:13 when erase() is called, the destructer is called. I know that 2011-11-08T15:41:34 but when it's resizing, do the objects get redeclared? 2011-11-08T15:42:18 so stupid, but I just don't know :/ 2011-11-08T15:42:46 destructers are getting called randomly lol and variables are decrementing when they should be incrementing 2011-11-08T15:43:41 it sounds like you aren't using destructors properly :P 2011-11-08T15:43:58 possible, always possible :P 2011-11-08T15:44:12 the destructer is simple though, 3 ifs and thats it 2011-11-08T15:44:28 i am so glad i don't have to deal with distractors... 2011-11-08T15:44:33 *** Antimony has quit IRC (Ping timeout: 256 seconds) 2011-11-08T15:44:59 :P treeform 2011-11-08T15:46:19 this is a global variable 2011-11-08T15:46:22 ? 2011-11-08T15:46:31 and you are doing "b = b + 1" or something? 2011-11-08T15:46:35 and in python? 2011-11-08T15:46:50 c++ 2011-11-08T15:46:55 ++var; 2011-11-08T15:46:58 that's it 2011-11-08T15:47:03 *in the constructor 2011-11-08T15:47:09 ah 2011-11-08T15:47:29 conor_f: so, what are you doing with these objects? 2011-11-08T15:47:46 I have a vector of them 2011-11-08T15:47:46 *** ltriant has joined #aichallenge 2011-11-08T15:47:46 putting them in a std::vector? 2011-11-08T15:47:53 I push_back when theres a new ant 2011-11-08T15:47:59 erase() when one dies 2011-11-08T15:48:06 and skip it when it's still alive 2011-11-08T15:48:07 is the vector a local variable? 2011-11-08T15:48:17 vector copies objects when you add them, doesn't it? 2011-11-08T15:48:18 it's local to the State struct 2011-11-08T15:48:35 and/or destroys one 2011-11-08T15:48:49 been awhile, sorry 2011-11-08T15:48:57 Zannick: when you add them? 2011-11-08T15:49:03 I don't understand this 2011-11-08T15:49:05 or push_back 2011-11-08T15:49:14 it creates a new one 2011-11-08T15:49:29 yeah 2011-11-08T15:50:16 conor_f: it sounds like you're using the wrong data type :P 2011-11-08T15:50:19 and when you erase it calls the destructor 2011-11-08T15:50:38 conor_f: that's not what's causing the problem though ofc 2011-11-08T15:51:11 thestinger: you say I should use what data type? 2011-11-08T15:51:20 a vector is a dynamic array, it's laid out as a single block of memory 2011-11-08T15:51:22 Zannick: yes 2011-11-08T15:51:38 if you delete something in the middle it basically has to do memmove() under the hood and move everything that's after it 2011-11-08T15:51:53 *** olexs1 has left #aichallenge 2011-11-08T15:52:28 thestinger: ok, not really bothered by all that, as long as they all get moved and their constructers/destructors aren't called (which they shouldn't be I think) 2011-11-08T15:52:49 *** AndrewBC_ has joined #aichallenge 2011-11-08T15:52:56 bool operator==(const pos)const; 2011-11-08T15:53:01 constructors/destructors are supposed to be used for allocating/deallocating resources 2011-11-08T15:53:18 if you have random side effects in there, you're going to have random side effects when stuff is being constructed/deconstructed 2011-11-08T15:53:22 thestinger: no reason I can't use them for vars though... 2011-11-08T15:54:03 *** peyton has quit IRC (Quit: peyton) 2011-11-08T15:54:16 conor_f: you can use erase in for example in a vector, but you must increment the iterator before it is invalidated 2011-11-08T15:54:28 myvec.erase(++it) 2011-11-08T15:54:35 aichallenge: janzert epsilon * rd589d7d / website/starter_packages.php : Update wording on starter package page - http://git.io/trINuA 2011-11-08T15:54:48 no 2011-11-08T15:54:53 (it++) 2011-11-08T15:55:11 mcstar: that would be an issue... 2011-11-08T15:55:17 since erase will get it, but after it will have it+1 2011-11-08T15:55:39 but as far as maps and sets are concerned i could delete from them wo problems 2011-11-08T15:55:54 so a for loop with an iter being incremented 2011-11-08T15:55:55 like in a simple for loop 2011-11-08T15:56:11 with an erase inside, needs vec.erase(it++); ? 2011-11-08T15:56:19 *** AndrewBC has quit IRC (Ping timeout: 258 seconds) 2011-11-08T15:56:22 for(auto it=..; it!=endIt; it++) 2011-11-08T15:56:25 should it not be --, then it gets incremented again? 2011-11-08T15:56:29 thats not going to work with vectors 2011-11-08T15:56:35 grr :/ 2011-11-08T15:56:37 ok 2011-11-08T15:56:39 erase returns the iterator to the next element 2011-11-08T15:56:41 *** tmandry has joined #aichallenge 2011-11-08T15:56:41 *** tmandry has joined #aichallenge 2011-11-08T15:56:44 how should it be done? 2011-11-08T15:56:48 in a while 2011-11-08T15:56:58 or in a for 2011-11-08T15:57:04 but than you must break; also 2011-11-08T15:57:23 let me think 2011-11-08T15:57:33 if i break, does it++ get executed? 2011-11-08T15:57:39 no 2011-11-08T15:57:47 Zannick: i meant continue 2011-11-08T15:57:50 :D 2011-11-08T15:57:53 yes it does 2011-11-08T15:57:54 if you continue, yes. 2011-11-08T15:58:00 m| 2011-11-08T15:58:05 :D 2011-11-08T15:58:12 im sort of silly now 2011-11-08T15:58:32 *** lorill has quit IRC (Quit: Page closed) 2011-11-08T15:58:35 in any case, vectors are not really meant for random-access element removal 2011-11-08T15:58:46 *** MoleM has joined #aichallenge 2011-11-08T15:58:51 hellow 2011-11-08T15:59:14 *** JorgeB_ has joined #aichallenge 2011-11-08T15:59:30 Zannick: no reason why I couldn't do it though? :/ 2011-11-08T15:59:40 no, not really 2011-11-08T15:59:46 any admin around who can fix something for me? 2011-11-08T15:59:48 just algorithmically costly 2011-11-08T15:59:51 conor_f: http://paste.pocoo.org/show/504832/ you probably want something like that (but seriously, use a list/forward_list or a set/unordered_set for what you want) 2011-11-08T16:00:01 *** Kamachi has joined #aichallenge 2011-11-08T16:00:07 *** JorgeB has quit IRC (Read error: Operation timed out) 2011-11-08T16:00:07 *** JorgeB_ is now known as JorgeB 2011-11-08T16:00:16 MoleM: fix what? 2011-11-08T16:00:16 hello all 2011-11-08T16:00:30 thestinger: we'll see 2011-11-08T16:00:30 fix me a martini!!! 2011-11-08T16:00:33 ty for the link though 2011-11-08T16:00:36 Zannick: never received an activation mail 2011-11-08T16:00:50 that i have remarkably little control over 2011-11-08T16:00:56 check your spam folder? 2011-11-08T16:01:09 no spam received 2011-11-08T16:01:14 :/ 2011-11-08T16:01:28 MoleM: what's the username? 2011-11-08T16:01:32 only thing i can imagine is typo in email adress 2011-11-08T16:01:35 molem is my username 2011-11-08T16:01:39 *** Antimony has joined #aichallenge 2011-11-08T16:02:18 http://pastebin.com/qqBm6wPp 2011-11-08T16:02:24 conor_f: i think that should work 2011-11-08T16:02:40 its from memory, but couple of days ago i run into this 2011-11-08T16:03:12 MoleM: is your address supposed to be a @google.com? 2011-11-08T16:03:19 ehr 2011-11-08T16:03:22 gmail.com 2011-11-08T16:03:24 oopsie 2011-11-08T16:03:26 :) 2011-11-08T16:03:27 lololol 2011-11-08T16:03:36 mcstar: yep, that's basically the same as what I posted :) 2011-11-08T16:03:36 I'll fix and resend 2011-11-08T16:03:42 thnxie 2011-11-08T16:03:56 thestinger: k, didnt see it 2011-11-08T16:04:01 mcstar: ty 2011-11-08T16:05:48 MoleM: You should receive it shortly, it'll be from janzert@aichallenge.org instead of staff@aichallenge.org though 2011-11-08T16:06:05 yeah...my phone made some noise...so i think i got it 2011-11-08T16:06:10 great :) 2011-11-08T16:06:34 got it...thnx again 2011-11-08T16:06:36 *** yassya has quit IRC (Quit: Leaving) 2011-11-08T16:06:49 now i got no excuse to not start coding :/ 2011-11-08T16:09:59 to check if its the last loop using a vector, will this work? 2011-11-08T16:10:16 iter == --vec.end(); 2011-11-08T16:10:54 --vec.end(); 2011-11-08T16:11:14 this destructively modifies its end-pointer 2011-11-08T16:11:25 huh? 2011-11-08T16:11:30 what does that mean? 2011-11-08T16:11:35 you definitely shouldnt do that 2011-11-08T16:11:42 it might not be modifiable at all 2011-11-08T16:11:51 i dont think its an lvalue 2011-11-08T16:11:52 (iter+1) == vec.end(); 2011-11-08T16:11:55 maybe 2011-11-08T16:12:03 no 2011-11-08T16:12:09 1 is not an iterator 2011-11-08T16:12:20 only ++ and -- are overloaded for them 2011-11-08T16:12:28 .size() is O(1) for the standard library containers, use that 2011-11-08T16:12:47 conor_f: make a copy of .end() 2011-11-08T16:12:50 and decrement that 2011-11-08T16:13:00 is that not basically what I'm doing? 2011-11-08T16:13:08 btw what do you want? 2011-11-08T16:13:16 .end() points over the end actually 2011-11-08T16:13:19 not at the last element 2011-11-08T16:13:24 what container 2011-11-08T16:13:27 it!=vec.end() 2011-11-08T16:13:29 see? 2011-11-08T16:13:29 the last element 2011-11-08T16:13:31 yeah 2011-11-08T16:13:37 I want the last element though 2011-11-08T16:13:49 .last()? 2011-11-08T16:13:55 auto beforetheend=myvec.end(); 2011-11-08T16:14:01 before..--; 2011-11-08T16:14:03 before..--; 2011-11-08T16:14:18 two decrements? 2011-11-08T16:14:22 conor_f: http://en.cppreference.com/w/cpp/container/vector .back() 2011-11-08T16:14:24 yep 2011-11-08T16:14:34 ah 2011-11-08T16:14:36 ok 2011-11-08T16:14:40 last element, than only one 2011-11-08T16:14:52 thestinger: :) 2011-11-08T16:14:59 that is a lot better lol 2011-11-08T16:15:44 Returns reference to the last element in the container. 2011-11-08T16:15:51 thats not an iterator watch out 2011-11-08T16:16:29 ach 2011-11-08T16:16:30 yeah 2011-11-08T16:16:31 :/ 2011-11-08T16:16:40 *** marijnfs has left #aichallenge 2011-11-08T16:16:45 no easy way to get around that? 2011-11-08T16:17:24 http://www.cplusplus.com/reference/stl/vector/rbegin/ 2011-11-08T16:17:26 lol this? :P 2011-11-08T16:18:09 conor_f: you want to delete the last element? 2011-11-08T16:18:15 thats why you need the iterator? 2011-11-08T16:18:30 *** Fandekasp has quit IRC (Ping timeout: 260 seconds) 2011-11-08T16:18:34 no 2011-11-08T16:18:41 need to check if it's the last loop 2011-11-08T16:19:04 rbegin won't work so :/ 2011-11-08T16:19:11 just saw what a reverse iterator is 2011-11-08T16:19:37 *** Warrows has joined #aichallenge 2011-11-08T16:20:12 how about doing what i proposed? 2011-11-08T16:20:16 i think that should work 2011-11-08T16:20:28 ok 2011-11-08T16:20:31 I'll try 2011-11-08T16:20:46 *** ikaros_ has joined #aichallenge 2011-11-08T16:21:01 one thing im not sure is, if you delete from this container, will be a previously copied end() pointer be invalidated? 2011-11-08T16:21:03 *** AndrewBC_ is now known as AndrewBC 2011-11-08T16:21:08 yeah 2011-11-08T16:21:09 should be 2011-11-08T16:21:26 but theres a break if theres an erase so it's skipped 2011-11-08T16:22:04 *** Kamachi has quit IRC (Quit: Page closed) 2011-11-08T16:22:47 timeout with that :/ 2011-11-08T16:22:49 strange 2011-11-08T16:23:49 what are you trying to do? 2011-11-08T16:23:58 what mcstar said 2011-11-08T16:24:07 decrementing the final iter twice 2011-11-08T16:24:21 no, i corrected myself, just decrement once 2011-11-08T16:24:32 :P 2011-11-08T16:24:34 kk 2011-11-08T16:25:04 *** Warrows has quit IRC (Quit: Page closed) 2011-11-08T16:25:06 times out again :/ 2011-11-08T16:25:16 ah, see why >_< 2011-11-08T16:25:17 nvm 2011-11-08T16:27:04 still the same error of decrementing instead of incrementing the global var lol 2011-11-08T16:28:30 *** cyphase has quit IRC (Ping timeout: 240 seconds) 2011-11-08T16:29:40 *** Johan-kun has joined #aichallenge 2011-11-08T16:30:57 Hey all, I'm testing my bot localy with the test_bot.cmd file that works well, but Now I want to have a more challenging map to test on and when I change the map in test_bot.cmd It just crash... Can someone help me 2011-11-08T16:31:09 I think It's because there is no food around on the other maps 2011-11-08T16:32:18 *** kilae has quit IRC (Quit: ChatZilla 0.9.87 [Firefox 7.0.1/20110928134238]) 2011-11-08T16:33:37 take --scenario off Johan-kun 2011-11-08T16:34:04 Okay I'll try that 2011-11-08T16:35:15 Oh thanks ^^ I finally works :) 2011-11-08T16:35:24 np 2011-11-08T16:39:22 *** cyphase has joined #aichallenge 2011-11-08T16:42:08 *** Johan-kun has quit IRC (Ping timeout: 265 seconds) 2011-11-08T16:42:29 *** Migi32 has joined #aichallenge 2011-11-08T16:44:42 *** treeform has quit IRC (Remote host closed the connection) 2011-11-08T16:45:17 *** treeform has joined #aichallenge 2011-11-08T16:46:31 Question about uploading code: How is it compiled? I currently have my project based on autotools so running ./configure and make should do the job, but can't find much information about how it's done on the aichallende server 2011-11-08T16:47:00 your Makefile isn't used (neither are autotools) 2011-11-08T16:47:16 sh4wn: https://github.com/aichallenge/aichallenge/blob/epsilon/worker/compiler.py 2011-11-08T16:47:38 ah ok 2011-11-08T16:47:40 let's view 2011-11-08T16:49:13 *** retybok_ has quit IRC (Ping timeout: 258 seconds) 2011-11-08T16:50:49 *** cyphase has quit IRC (Quit: http://www.cyphase.com/) 2011-11-08T16:50:56 *** ajf is now known as ajf|offline 2011-11-08T16:51:28 *** Palmik has quit IRC (Remote host closed the connection) 2011-11-08T16:52:20 *** Jak_o_Shadows has joined #aichallenge 2011-11-08T16:52:37 *** mcstar has quit IRC (Quit: WeeChat 0.3.6) 2011-11-08T16:52:45 *** hacklash has quit IRC (Quit: hacklash) 2011-11-08T16:56:58 *** cyphase has joined #aichallenge 2011-11-08T17:04:35 Why U No Explore ants? 2011-11-08T17:04:51 They're lé tired. 2011-11-08T17:04:57 *** Jak_o_Shadows has quit IRC (Remote host closed the connection) 2011-11-08T17:05:09 *** JorgeB has quit IRC (Quit: Textual IRC Client: http://www.textualapp.com/) 2011-11-08T17:05:25 *** JorgeB has joined #aichallenge 2011-11-08T17:05:30 ... I had sudden urge to add "Challenge Accepted" somehow into that 2011-11-08T17:08:04 I think they may actually be trying to explore, but they only do one square at a time, and then choose a square on the other side of the map 2011-11-08T17:08:10 so they mostly go back and forth 2011-11-08T17:09:08 * p_l so far only did some planning. Things like formation movement, influence map, etc. 2011-11-08T17:09:26 p_l: that's a lot :O 2011-11-08T17:09:50 conor_f: triangle formation was obvious to me after seeing some games 2011-11-08T17:10:07 but implementing it, well done :) 2011-11-08T17:10:22 not yet 2011-11-08T17:10:26 I'm still stuck on defense system, which I'm going to need a formation of some sort for 2011-11-08T17:10:27 will see if I'll have enough time 2011-11-08T17:10:35 oh, well gl 2011-11-08T17:10:41 you've got plenty of time 2011-11-08T17:10:54 conor_f: I gave up on defense with a static formation :P 2011-11-08T17:10:56 conor_f: things happen outside of AI challenge and take time :) 2011-11-08T17:11:22 thestinger: static defense > no defense :P 2011-11-08T17:11:27 no, it was worse 2011-11-08T17:11:36 p_l: that is true, have until dec 18th though 2011-11-08T17:11:46 it slowed down food gathering enough that it was actually an inferior defense :P 2011-11-08T17:12:07 I'm thinking of food gatherers forming reserve defense network 2011-11-08T17:12:12 I'm going to just make a certain number of ants stay between the hill and enemy ants 2011-11-08T17:12:43 I dunno 2011-11-08T17:13:07 *** mcstar has joined #aichallenge 2011-11-08T17:13:07 I still need to get the ant tracking right too :/ 2011-11-08T17:13:51 *** foRei has quit IRC (Quit: Bye) 2011-11-08T17:13:58 when you see a single ant kill a hill in a game, it's misleading 2011-11-08T17:14:11 they've found your hill, so they'll probably be sending an army anyway if that ant dies 2011-11-08T17:14:37 and they usually have enough ants nearby (they only need 2 ants to kill any static unmoving defense) 2011-11-08T17:15:06 your defense will always have corners 2011-11-08T17:15:31 it needs to move around at least a bit or it's too easy to kill off 2011-11-08T17:15:51 *** mceier has quit IRC (Quit: leaving) 2011-11-08T17:19:08 *** twymer has quit IRC (Ping timeout: 245 seconds) 2011-11-08T17:19:43 thestinger: did you know there are precompiled headers? 2011-11-08T17:20:18 fluxid.pl's turnlimit is way too high... 2011-11-08T17:21:30 How do you handle exploration? 2011-11-08T17:21:46 btw do they update the starter packages? the scala starter package has some weird bugs 2011-11-08T17:22:06 Stivo: feel free to submit pull requests 2011-11-08T17:22:18 *** peyton has joined #aichallenge 2011-11-08T17:22:27 Antimony: http://aichallenge.org/forums/viewtopic.php?f=24&t=1728 , i use the 2nd one 2011-11-08T17:22:50 amstan: Where is the repo? 2011-11-08T17:22:54 see topic 2011-11-08T17:23:18 thanks 2011-11-08T17:24:06 *** mnstrspeed has joined #aichallenge 2011-11-08T17:24:48 hey guys 2011-11-08T17:25:03 hi 2011-11-08T17:26:00 quick question: the C# starter package implements an update with the format "r " indicating the removal of a food, but this isn't listed in the game specification? 2011-11-08T17:26:41 i think we took out those a long time ago 2011-11-08T17:26:44 If you want to fix a starterbot, how do you submit the changes? 2011-11-08T17:26:45 *** lknix has quit IRC (Ping timeout: 260 seconds) 2011-11-08T17:26:56 Antimony: github pull requests 2011-11-08T17:27:07 see their help pages, they're useful 2011-11-08T17:29:03 @amstan thanks; I guess I'll have to keep track of food myself then... 2011-11-08T17:29:04 mnstrspeed: I'm sorry Dave, err mnstrspeed; I cannot 'amstan'. 2011-11-08T17:29:22 *** mnstrspeed has quit IRC (Quit: Page closed) 2011-11-08T17:30:37 how does food rate work? 2011-11-08T17:30:37 *** peyton has quit IRC (Quit: peyton) 2011-11-08T17:33:21 isn't it something like the amount of food that spawns on average per second per square? 2011-11-08T17:33:43 the real question is though: does anyone estimate the map's food rate and use that in his bot in any way? 2011-11-08T17:35:23 mcstar: yeah, I haven't tried using them though 2011-11-08T17:35:29 *** Relax has quit IRC (Quit: Leaving) 2011-11-08T17:36:45 I'm thnking about trying to estimate it 2011-11-08T17:38:30 *** peyton has joined #aichallenge 2011-11-08T17:38:53 *** lknix has joined #aichallenge 2011-11-08T17:38:59 *** amstan_ has joined #aichallenge 2011-11-08T17:38:59 *** ChanServ sets mode: +o amstan_ 2011-11-08T17:41:38 *** bergmark has quit IRC (Ping timeout: 245 seconds) 2011-11-08T17:42:59 *** ikaros_ has quit IRC (Quit: Ex-Chat) 2011-11-08T17:45:56 *** Antimony has quit IRC (Ping timeout: 260 seconds) 2011-11-08T17:46:27 *** mj41 has joined #aichallenge 2011-11-08T17:50:34 *** GeneralMandible has joined #aichallenge 2011-11-08T17:50:54 *** conor_f has quit IRC (Quit: Lost terminal) 2011-11-08T17:51:14 *** GeneralMandible has quit IRC (Client Quit) 2011-11-08T17:52:21 dman, ive got 21 source files 2011-11-08T17:53:51 how many lines total? 2011-11-08T17:54:38 *** Stivo has quit IRC (Ping timeout: 265 seconds) 2011-11-08T17:57:15 dman! 2011-11-08T17:58:01 *** MoleM has quit IRC (Ping timeout: 265 seconds) 2011-11-08T17:59:21 wc says 1461 2011-11-08T18:01:42 wow 2011-11-08T18:01:51 I only have one ~650 loc file 2011-11-08T18:02:02 luizribeiro: what language is it? 2011-11-08T18:02:09 C++ 2011-11-08T18:02:27 did you start with the starter bot or from scratch? 2011-11-08T18:02:50 my biggest project is 3270 lines of python, according to sloccount 2011-11-08T18:03:14 and my second biggest project is 2800! 2011-11-08T18:03:32 with the starter bot, but I stripped some stuff from it 2011-11-08T18:03:48 i might be insane and/or use lots of documentation 2011-11-08T18:04:20 Hi. I started Perl bot base module from scratch .... https://github.com/mj41/AIAnts ... not done yet ... join #perl-ai to discuss. 2011-11-08T18:04:26 I'm up to ~1300 but it's mostly b/c I've added a bunch of half-baked stuff and it needs refactoring 2011-11-08T18:04:54 *** epicmonkey has quit IRC (Ping timeout: 240 seconds) 2011-11-08T18:05:31 thestinger:i didnt say that my code is final 2011-11-08T18:05:46 i dont feel like its too much 2011-11-08T18:05:50 aichallenge: Alexandru Stan epsilon * rcca5fce / ants/dist/starter_bots/clojure/ants.clj : 2011-11-08T18:05:50 aichallenge: Merge pull request #354 from dlowe/epsilon 2011-11-08T18:05:50 aichallenge: Clojure bots are crashing in Math/abs (called in 'direction') because - http://git.io/MBib1A 2011-11-08T18:05:52 well, you might have a lot more logic and stuff than me 2011-11-08T18:05:55 I still need to do a lot more 2011-11-08T18:06:00 i mean its very transparent 2011-11-08T18:06:42 lol 2011-11-08T18:06:53 i just imagined what would i do without a compiler 2011-11-08T18:07:01 yes, mine does not have much logic 2011-11-08T18:07:02 i would shit myself 2011-11-08T18:07:22 I only have food gathering and path finding stuff, to be honest 2011-11-08T18:07:27 *** ajf|offline has quit IRC (Remote host closed the connection) 2011-11-08T18:08:08 *** Antimony has joined #aichallenge 2011-11-08T18:08:30 it's more about the balance than logic 2011-11-08T18:08:32 getting the right balance 2011-11-08T18:08:52 Wow, using Psyco slowed down my bfs by an order of 3 *scratch head* 2011-11-08T18:09:28 D 2011-11-08T18:09:30 :D 2011-11-08T18:09:33 psychotic 2011-11-08T18:10:06 *** amstan_ has quit IRC (Ping timeout: 240 seconds) 2011-11-08T18:10:10 datachomper: does it contain a if declared("psycho"): timefactor/=3 ? 2011-11-08T18:10:36 wait, what does "order" mean? 2011-11-08T18:10:45 * p_l simply starts with a proper compiled language :D 2011-11-08T18:10:47 factor of 3 or order of 3? 2011-11-08T18:11:00 *3 2011-11-08T18:11:13 bfs's went from 10ms average to 30 2011-11-08T18:11:31 k, 3 orders (of magnitude) is 1000x factor 2011-11-08T18:11:49 datachomper: but maybe its much more reliable now? 2011-11-08T18:11:56 haha 2011-11-08T18:12:07 also... are you using BFS for pathfinding? Why not A*? 2011-11-08T18:12:27 *** paulwal has joined #aichallenge 2011-11-08T18:12:41 *** Rinum has joined #aichallenge 2011-11-08T18:12:56 I love how pguillory is using the contest for self-promotion lol 2011-11-08T18:13:02 p_l: I'm building costmaps and adding them together 2011-11-08T18:13:08 see here: http://aichallenge.org/profile.php?user=589 2011-11-08T18:13:22 I wonder if it'll work? 2011-11-08T18:13:25 haha 2011-11-08T18:13:45 p_l: Just trying a few different approaches. My A* algo was timing out on me after about 80 ants, even with route caching 2011-11-08T18:13:48 too bad that AI skills don't exactly correspond to web development 2011-11-08T18:14:47 a self-modifying site based on feedback from the audience/customers 2011-11-08T18:15:00 :P 2011-11-08T18:15:04 *** maysam has joined #aichallenge 2011-11-08T18:15:18 lol 2011-11-08T18:15:19 you could use all those fancy web metrics as feedback 2011-11-08T18:15:34 that's actually a sweet idea 2011-11-08T18:15:39 move stuff around on the page to where it gets you the most purchases, etc. 2011-11-08T18:15:42 a website that "evolves" 2011-11-08T18:16:12 *** paulwal has quit IRC (Client Quit) 2011-11-08T18:16:18 I don't think I've ever seen anything like that... at least not one that does so automatically 2011-11-08T18:17:25 it's usually done through A/B testing 2011-11-08T18:17:54 *** datachomper has left #aichallenge 2011-11-08T18:18:04 *** delt0r_ has quit IRC (Ping timeout: 258 seconds) 2011-11-08T18:18:08 i want black-white static pages 2011-11-08T18:18:15 with keyboard navigation 2011-11-08T18:18:22 and no sounds 2011-11-08T18:18:47 but with A/B testing, people create the A and B... what if new versions were created automatically 2011-11-08T18:19:36 you can do that, but you could end up overfitting 2011-11-08T18:20:27 *** Skinny has quit IRC (Quit: Page closed) 2011-11-08T18:20:32 *** Rinum_ has joined #aichallenge 2011-11-08T18:23:38 *** Rinum has quit IRC (Ping timeout: 265 seconds) 2011-11-08T18:25:00 no match for call to ‘(pos) (_ant*&)’ 2011-11-08T18:25:06 why? 2011-11-08T18:25:24 because i left out a ; from the previous line 2011-11-08T18:25:29 ha-hah. 2011-11-08T18:25:31 hi 2011-11-08T18:25:33 ... that's the bane of programming, C++? 2011-11-08T18:25:52 i got error loging and i know my code finishes on time, but I get timeout on first turn 2011-11-08T18:26:30 are you sure it finishes on time? 2011-11-08T18:26:34 even when i set the waiting time too long, still happens 2011-11-08T18:26:58 maybe it never sends the Go command? 2011-11-08T18:27:16 go command comes from the play, not from the bot 2011-11-08T18:27:45 also when it times out on one turn, does it get a chance to run next round or is it out? 2011-11-08T18:27:53 no, your bot sends the orders and ends with sending "go" 2011-11-08T18:28:04 it's out 2011-11-08T18:28:16 no, once you time out on one turn, your bot is done for the whole game 2011-11-08T18:28:46 have a timer in your code... if the timer is close to the turn time, send "go" 2011-11-08T18:29:27 *** ajf|offline has joined #aichallenge 2011-11-08T18:29:35 how do i sent go? 2011-11-08T18:29:41 what is the command for it? 2011-11-08T18:29:50 *** HaraKiri has quit IRC () 2011-11-08T18:29:53 *** ajf|offline is now known as Guest6456 2011-11-08T18:30:01 go() 2011-11-08T18:30:06 are you using a starter package? 2011-11-08T18:30:12 ah, wait 2011-11-08T18:30:15 gO() 2011-11-08T18:30:17 there should be some kind of end-turn function 2011-11-08T18:30:19 it is camel-case 2011-11-08T18:30:21 *** delt0r_ has joined #aichallenge 2011-11-08T18:30:22 finishTurn() 2011-11-08T18:30:38 why would you camelcase like that? 2011-11-08T18:30:39 right, in there there should be some kind of print "go" 2011-11-08T18:30:44 Zannick: pun 2011-11-08T18:30:51 amstan: I wasn't sure that clojure fix was actually correct 2011-11-08T18:30:56 that's not a pun! 2011-11-08T18:31:13 in particular will it handle the player_seed 64bit int parameter? 2011-11-08T18:31:43 a pun is like, "My dentist Dr. Kenobi said to me, Luke, trust your fillings." 2011-11-08T18:31:54 *** Rinum_ is now known as Rinum 2011-11-08T18:32:17 *** Redgis has quit IRC (Ping timeout: 244 seconds) 2011-11-08T18:32:28 go is called every time 2011-11-08T18:32:30 janzert: how's pypy coming? 2011-11-08T18:33:07 lots of people interested in using it 2011-11-08T18:33:47 (no pressure or anything) 2011-11-08T18:33:49 :P 2011-11-08T18:33:52 Zannick: that is a joke 2011-11-08T18:35:01 a pun is a funny figure of speech that is funny, beyond its meaning, something beyond it's cretor's intention 2011-11-08T18:35:16 no? 2011-11-08T18:35:26 not really, it's a type of word play 2011-11-08T18:35:40 but no always intentional 2011-11-08T18:35:45 well, yeah, mostly is 2011-11-08T18:35:57 based on a word with multiple meanings 2011-11-08T18:36:08 I guess it applies to words that sound similar too (dunno) 2011-11-08T18:36:30 *** ikaros has quit IRC (Quit: Ex-Chat) 2011-11-08T18:36:33 "very punny" is a pun 2011-11-08T18:36:46 http://lmgtfy.com/?q=define+pun 2011-11-08T18:40:02 *** maysam has quit IRC (Ping timeout: 276 seconds) 2011-11-08T18:41:31 *** ikaros has joined #aichallenge 2011-11-08T18:41:47 caution: pypy cut a new release branch today, so it seems to be coming along great ;) 2011-11-08T18:43:50 janzert: why not? 2011-11-08T18:44:15 janzert: as far as 64 bit player_seeds go, it's not really a problem if it overflows as long as it doesn't crash 2011-11-08T18:44:27 right, but that's the question 2011-11-08T18:44:34 you still have the same type of deterministic behavoir if it overflows 2011-11-08T18:44:38 if it just overflows then it's no problem 2011-11-08T18:44:57 i'm sure he would have noticed by now if that were a problem 2011-11-08T18:45:11 several other starter bots have had a problem with it causing a crash though so I just wanted to make sure it really did work 2011-11-08T18:45:17 hope so :) 2011-11-08T18:45:51 yeah, i'm a little worried when doing merges for starter packages, but this one seemed simple enough 2011-11-08T18:45:57 I've been using pypy for a while now, what is the issue? 2011-11-08T18:46:58 Extrarius: none known with the latest release it just needs to get integrated with the contest infrastructure to install and use it 2011-11-08T18:47:16 lets use paronomasia instead of pun from now on, ok? 2011-11-08T18:47:59 there was a bug fixed since the latest release that may cause problems but I don't know if it would for sure and am hoping they cut the next release in time to use that for the contest 2011-11-08T18:49:21 what was the bug? 2011-11-08T18:49:33 slow handling of stdio 2011-11-08T18:50:13 I've found the faster just-about-everything-else by far overshadows that 2011-11-08T18:50:22 amstan: I'm working on the rules update; how do you feel about the current law enforcement clause in the rules? 2011-11-08T18:50:41 janzert: most of those were made by jeff 2011-11-08T18:50:49 yeah, I know 2011-11-08T18:50:50 in a jokingly manner 2011-11-08T18:50:56 when we were hosted by csc 2011-11-08T18:51:01 ahh 2011-11-08T18:51:05 i wouldn't mind making them more serious 2011-11-08T18:51:09 ok 2011-11-08T18:51:51 since my first reaction to that is the same as I've seen repeated more than once elsewhere is that it is rather stupid and over the top 2011-11-08T18:52:09 that was a joke? 2011-11-08T18:52:29 you mean i shouldn't have called the cops on that one guy? 2011-11-08T18:52:38 (j/k) 2011-11-08T18:52:38 yeah, the official rules is probably not the best place for a somewhat non-obvious joke 2011-11-08T18:52:41 :) 2011-11-08T18:52:50 yes. 2011-11-08T18:56:50 "full extent of the law"? (or smilar) 2011-11-08T18:57:46 *** lknix has quit IRC (Quit: WeeChat 0.3.6) 2011-11-08T19:02:03 *** mj41 has quit IRC (Ping timeout: 245 seconds) 2011-11-08T19:02:08 *** h33ro has joined #aichallenge 2011-11-08T19:02:14 *** mleise has quit IRC (Ping timeout: 258 seconds) 2011-11-08T19:05:16 *** Antimony has quit IRC (Ping timeout: 259 seconds) 2011-11-08T19:06:48 *** userjjb_ has joined #aichallenge 2011-11-08T19:07:29 amstan, mleise: https://github.com/aichallenge/aichallenge/issues/344#issuecomment-2675735 2011-11-08T19:09:44 janzert: what about that cpu affinity business? 2011-11-08T19:09:54 *** Guest6456 has quit IRC (Ping timeout: 258 seconds) 2011-11-08T19:09:57 intentionally losing games 2011-11-08T19:10:13 amstan: I believe that's a technical thing outside of the scope of the rules 2011-11-08T19:10:24 janzert: yes, but are we still doing it? 2011-11-08T19:10:39 s/intentionally losing games/manipulating rankings by intentionally losing games 2011-11-08T19:10:40 *** ajf|offline has joined #aichallenge 2011-11-08T19:10:41 at the technical level I investigated it before but I'm afraid it will slow the engine down quite a bit 2011-11-08T19:11:06 *** ajf|offline is now known as Guest56998 2011-11-08T19:11:25 like that? 2011-11-08T19:12:21 janzert: it's just that some people have misunderstood that in the past, especially the conditional behavoir part 2011-11-08T19:13:26 conditional behavior part? 2011-11-08T19:13:35 oh yeah 2011-11-08T19:13:41 I never liked that part 2011-11-08T19:14:05 why I rewrote it say the information outside of the bot protocol 2011-11-08T19:14:28 I edited my comment to include the "manipulating rankings by" 2011-11-08T19:14:36 they can still encode information in the ants 2011-11-08T19:14:40 so they can talk that way 2011-11-08T19:15:30 between opponents you mean? 2011-11-08T19:15:31 sure 2011-11-08T19:15:52 although I will be very surprised if it actually catches on at all 2011-11-08T19:15:52 i'm just wondering if there's actually any point to that rule 2011-11-08T19:16:22 *** Jak_o_Shadows has joined #aichallenge 2011-11-08T19:16:56 for example I don't want them changing there behavior by trying to guess the opponent by say gather OS level information 2011-11-08T19:17:25 or guessing the engine's random seed or whatever 2011-11-08T19:17:43 guessing by looking at OS level information that is 2011-11-08T19:17:58 i've heard some funny stories about guessing seeds 2011-11-08T19:18:02 in RPS 2011-11-08T19:18:19 trying to guess by in game information is allowed but I would consider a bug in the engine if possible :) 2011-11-08T19:18:31 heh 2011-11-08T19:18:39 is our algo complicated enough that someone who detects the seed should be awarded for his troubles? 2011-11-08T19:18:52 he'd be able to predict all future food spawns 2011-11-08T19:19:01 would be pretty handy I'd say 2011-11-08T19:19:11 just use /dev/urandom or cryptrandgen for the seeds 2011-11-08T19:19:13 yes, but not everything 2011-11-08T19:19:13 how do you invert a bfs map easily? 2011-11-08T19:19:20 and that yes.. use /dev/random 2011-11-08T19:19:43 1 / cost? 2011-11-08T19:19:55 roflmao: what do you mean by "invert"? 2011-11-08T19:19:58 Extrarius: but just having a strong seed source isn't sufficient, if you're generator then leaks that seed 2011-11-08T19:20:09 so I have a map that draws attention towards enemy ants 2011-11-08T19:20:12 enemyAntsCostMap 2011-11-08T19:20:16 lower cost = enemy ants 2011-11-08T19:20:22 ants move to lower cost squares 2011-11-08T19:20:30 janzert: I don't mean use it as a source seed, but to use it as your RNG 2011-11-08T19:20:33 I was wondering if there is a simple way to just invert that 2011-11-08T19:20:38 without redoing a whole bfs 2011-11-08T19:20:49 ahh, that can be quite slow 2011-11-08T19:21:02 also means the game is non-reproducible 2011-11-08T19:21:25 which is something that we require 2011-11-08T19:21:35 ok, then use MT or CMWC4096 and use 512 bits of output hashed down for each generated number 2011-11-08T19:21:41 "On a side note, the starter kit tries to use psyco which is not supported on 64-bit systems. IIRC the game servers are all 64-bit linux machines." 2011-11-08T19:21:43 umm 2011-11-08T19:21:49 isn't that bad? 2011-11-08T19:22:01 you'll notice the engine seed is written to the replay file 2011-11-08T19:22:24 amstan: yeah I have no idea why it has the psycho import 2011-11-08T19:22:51 it certainly can't be used on the contest servers 2011-11-08T19:23:01 amstan: it just catches the import exception anyway 2011-11-08T19:23:07 if anybody running the contest wants some help working out 'secure seeds', I'd be glad to offer advice 2011-11-08T19:23:21 thestinger: yes, but it gives people a false impression about what the game servers have 2011-11-08T19:23:30 yeah, it's misleading :P 2011-11-08T19:23:31 Extrarius: I think we're good but feel free to review the code if you want 2011-11-08T19:24:08 does psyco even work with python 2.7? 2011-11-08T19:24:12 nope 2011-11-08T19:24:25 doesn't work for 2.7 or 64bit :/ 2011-11-08T19:24:37 janzert: Is there an absolute limit on map sizes? 2011-11-08T19:24:47 200x200 2011-11-08T19:24:57 ok thanks :) 2011-11-08T19:24:59 i.e. a max of 200 in either or both dimensions 2011-11-08T19:29:26 janzert: some people want octave 2011-11-08T19:29:51 janzert: should i refer them to the ada pull request and ask them to do the same? 2011-11-08T19:29:54 http://aichallenge.org/forums/viewtopic.php?f=26&t=1762#p10824 2011-11-08T19:30:05 i even got a pm about it 2011-11-08T19:30:19 yeah, I've seen that 2011-11-08T19:30:46 janzert: Is the ants\ants.py in github the official contest file that randomizes things? 2011-11-08T19:30:55 if they come up with a starter bot and the command line to run it would be sufficient 2011-11-08T19:31:10 Extrarius: i think that's the file that uses the seed 2011-11-08T19:31:24 Extrarius: worker/engine.py might be the one with the seed generator 2011-11-08T19:31:57 janzert: what about places to get it from? give them the deb preferred talk? 2011-11-08T19:32:14 Extrarius: if that's where the engine_seed is being used that would be it but I've not been involved much with that code so don't know how it is layed out off hand 2011-11-08T19:32:27 amstan: there is a natty package 2011-11-08T19:32:34 janzert: ok 2011-11-08T19:32:46 just make sure it is an acceptable version I suppose 2011-11-08T19:32:47 good job we're getting pypy to make up for the lack of psyco 2011-11-08T19:33:41 isn't pypy better? 2011-11-08T19:33:55 i rarely hear of psyco, pypy is always popular on reddit 2011-11-08T19:34:08 psyco was basically made obsolete by pypy 2011-11-08T19:34:51 except pypy has a more restricted set of modules 2011-11-08T19:34:54 *** kaemo has joined #aichallenge 2011-11-08T19:35:16 caution: you already have a restricted set of modules on the worker, so might as well 2011-11-08T19:35:36 ? 2011-11-08T19:35:39 *** Harpyon has quit IRC (Quit: Textual IRC Client: http://www.textualapp.com/) 2011-11-08T19:35:45 I want pypy support 2011-11-08T19:36:00 caution: as in.. there's not many modules you'll find on the worker anyway 2011-11-08T19:36:19 so stuff like django compatibility or stuff like that is irrelevant 2011-11-08T19:36:30 I was talking about standard library modules 2011-11-08T19:36:39 oh, nvm then 2011-11-08T19:37:47 pypy also seems to handle some modules extremely slowly (such as the array.array class), but using more basic types (eg list) is faster than python's specialized types so it's still a win 2011-11-08T19:39:37 from my tests with python is that array.array saves on memory not speed 2011-11-08T19:39:55 if your arrays are multi megabytes big then arrays are faster because of memory access 2011-11-08T19:40:26 but more importantly your programs acctually fits in memory 2011-11-08T19:40:28 "PyPy is highly compatible with existing python code. It supports ctypes and can run popular python libraries like twisted and django." 2011-11-08T19:40:41 apparently it supports django amstan :P 2011-11-08T19:40:52 caution: bad example, lol 2011-11-08T19:41:10 caution: I noticed something like a 10x increase in runtime when using pypy on a ctypes-heavy app 2011-11-08T19:41:11 caution: you're the one who said it has a restricted set of modules 2011-11-08T19:41:37 *** kaemo has quit IRC (Quit: Lost terminal) 2011-11-08T19:42:05 *** h33ro has quit IRC (Quit: Leaving) 2011-11-08T19:42:32 amstan: looks like ants\ants.py is the only py file that references engine_seed, and it seems to use python's random.randint, which is not difficult to predict. Though in this case, the bot never directly sees the values, so it might not be easy in this case 2011-11-08T19:43:17 Extrarius: feed free to make it use /dev/random for the engine_seed, just make sure it's nice code 2011-11-08T19:44:18 *** grwip has quit IRC (Quit: Leaving) 2011-11-08T19:44:43 *** bobbydroptable has joined #aichallenge 2011-11-08T19:45:25 Extrarius: it uses random.randint to get a seed so is basically using the default python seed which in python > 2.4 is from /dev/urandom if available 2011-11-08T19:46:03 Extrarius: that's quite obvious - PyPy doesn't have to manually construct the data, and I suspect PyPy includes custom compilable PyPy... 2011-11-08T19:46:29 see the explanation of the seed method in http://docs.python.org/library/random.html 2011-11-08T19:46:30 ... or it might be the lisper in me not accepting that someone in such position wouldn't get rid of indirection at compile-time 2011-11-08T19:47:09 janzert: the random generator is seeded from /dev/random, but from a sequence of output values, you can calculate the orignial seed 2011-11-08T19:47:11 ctypes is slow in CPython too, that's why people use the extension API 2011-11-08T19:47:57 Extrarius: right, but the analysis I did of it several months ago I don't think a bot can see enough output values 2011-11-08T19:48:06 ok 2011-11-08T19:48:34 I wouldn't mind seeing it reanalysed by someone else though :) 2011-11-08T19:48:35 I think I'm just stalling so I don't have to figure out how to get my bot's median score above 0 2011-11-08T19:49:03 hehe 2011-11-08T19:50:13 *** ikaros has quit IRC (Quit: Ex-Chat) 2011-11-08T19:50:25 *** replore_ has joined #aichallenge 2011-11-08T19:52:43 *** Rinum has quit IRC (Quit: Page closed) 2011-11-08T19:52:47 have any of you tried cat "nice, compressed movie" > /dev/dsp ? 2011-11-08T19:52:55 or exes 2011-11-08T19:53:01 those sound great too 2011-11-08T19:53:20 once i listened to this "music" for half an hour 2011-11-08T19:53:49 mcstar: the classic is piping your current kernel there 2011-11-08T19:54:11 but thats too short 2011-11-08T19:54:16 couple mbs 2011-11-08T19:54:22 *** RobotCaleb has quit IRC (Ping timeout: 258 seconds) 2011-11-08T19:54:24 mcstar: http://countercomplex.blogspot.com/2011/10/some-deep-analysis-of-one-line-music.html 2011-11-08T19:54:27 or you mean the sources? 2011-11-08T19:54:33 cat /dev/mem | aplay 2011-11-08T19:54:58 one-liners you run >/dev/dsp 2011-11-08T19:55:07 *** kaemo has joined #aichallenge 2011-11-08T19:55:34 similar to 8-bit cellular automatons making music 2011-11-08T19:55:55 /dev/mem? 2011-11-08T19:56:01 whats that? 2011-11-08T19:56:39 is that the base address of the memory? 2011-11-08T19:56:40 *** NoxiaZ^ has quit IRC (Ping timeout: 260 seconds) 2011-11-08T19:56:48 it's a char device for system memory 2011-11-08T19:57:04 there's also /dev/kmem, but it's turned off in most distros 2011-11-08T19:57:26 do something really access memory like that? 2011-11-08T19:57:29 does* 2011-11-08T19:57:52 a1k0n: very cool 2011-11-08T19:57:53 maybe this is a nub question, but to hell with it 2011-11-08T19:57:56 maybe 2011-11-08T19:58:09 probably just debugging stuff 2011-11-08T19:58:25 /dev/kmem can be used for all kinds of neat debugging/hackery like changing the owner of a process 2011-11-08T19:58:49 thestinger: whoa 2011-11-08T19:58:56 i hit me unprepared 2011-11-08T19:59:01 it* 2011-11-08T19:59:26 silence..... than bammmm 2011-11-08T19:59:31 screeems 2011-11-08T19:59:37 screaching 2011-11-08T19:59:48 obviously you need to be root, but that's why it was turned off by most distros (and I think by default too) 2011-11-08T20:00:38 other than debugging the kernel, the only real use case is for a rootkit :P 2011-11-08T20:00:49 oh 2011-11-08T20:01:02 is it writeable? 2011-11-08T20:01:08 yeah 2011-11-08T20:01:33 you need to recompile the kernel with it enabled though 2011-11-08T20:01:36 cat /dev/urandom > /dev/mem 2011-11-08T20:01:52 thats a short lived experiment 2011-11-08T20:01:59 http://bellard.org/jslinux/ 2011-11-08T20:02:03 do it in there :P 2011-11-08T20:02:10 *** Antimony has joined #aichallenge 2011-11-08T20:02:19 *** underflow has joined #aichallenge 2011-11-08T20:02:46 lolll 2011-11-08T20:02:53 dont tell me its a js vm? 2011-11-08T20:02:56 it is 2011-11-08T20:03:01 in my browser???? 2011-11-08T20:03:04 yeah 2011-11-08T20:03:05 yes. 2011-11-08T20:03:07 mcstar: From the creator of qemu (Fabrice Bellard). 2011-11-08T20:03:11 sweeeet 2011-11-08T20:03:12 Yes, it's a full x86 emulator. 2011-11-08T20:03:38 Others have written extensions adding network support (via ppp) and persistent local storage (via HTML5 localStorage) 2011-11-08T20:04:33 doesn't seem to have random or urandom, but it does as expected catting /dev/zero to /dev/mem =-P 2011-11-08T20:04:40 ok, now its really time for kde/gnome to become fast/useable 2011-11-08T20:05:44 Extrarius: there's a C compiler (tcc), so you could compile a random number generator 2011-11-08T20:05:58 hell, i cant type in there properly 2011-11-08T20:06:06 vimium keeps opening tabs for me 2011-11-08T20:06:09 cat: can't open '/dev/urandom': No such file or directory 2011-11-08T20:06:35 *** Jak_o_Shadows has quit IRC (Remote host closed the connection) 2011-11-08T20:06:40 there we go: http://pastebin.com/5ERBWjwD 2011-11-08T20:06:42 amstan: mknod c 1 9 /dev/urandom 2011-11-08T20:06:49 Now you have it. 2011-11-08T20:07:02 The device just wasn't included in his initrd. 2011-11-08T20:07:29 jbroman: so how do you know the quemu's creator name so offhand? 2011-11-08T20:07:41 amstan: One, I've seen jslinux before and new it from then. 2011-11-08T20:07:45 Two, his last name is in the URL. 2011-11-08T20:07:49 he also made ffmpeg 2011-11-08T20:08:37 "From the creator of qemu (Fabrice Bellard)." 2011-11-08T20:08:45 this seems like you pasted here from somewhere else 2011-11-08T20:08:56 No, I didn't paste it. 2011-11-08T20:09:01 jbroman: kidding 2011-11-08T20:10:00 *** Huun has quit IRC (Ping timeout: 258 seconds) 2011-11-08T20:12:51 Interesting 2011-11-08T20:13:05 my current bot massively out expands my old bots, but it still takes last place 2011-11-08T20:13:09 By the way, if you haven't seen it, Emscripten is similarly interesting. 2011-11-08T20:13:11 becasue they raze more hills early on 2011-11-08T20:13:15 (LLVM-to-JS compiler) 2011-11-08T20:14:00 *** treeform_ has joined #aichallenge 2011-11-08T20:15:01 i want a similar idea to build a multi-billion dollar business atop 2011-11-08T20:16:41 Anyhow, Fabrice Bellard is brilliant. His Wikipedia article (http://en.wikipedia.org/wiki/Fabrice_Bellard) has more. 2011-11-08T20:17:28 *** treeform has quit IRC (Ping timeout: 245 seconds) 2011-11-08T20:17:53 *** Antimony_ has joined #aichallenge 2011-11-08T20:18:00 *** Antimony has quit IRC (Ping timeout: 260 seconds) 2011-11-08T20:18:02 *** Antimony_ is now known as Antimony 2011-11-08T20:19:17 http://aichallenge.org/visualizer.php?game=75358&user=2772 awww raped but got the timeout shucks 2011-11-08T20:19:41 this bellard guy is a bit of a nerd 2011-11-08T20:20:40 jbroman: that pi digit calculation record is pretty impressive 2011-11-08T20:20:57 *** Migi32 has quit IRC (Remote host closed the connection) 2011-11-08T20:21:07 *** TwistedLogic has joined #aichallenge 2011-11-08T20:21:28 mcstar: And his IOCCC accomplishments. And TCCBOOT is pretty neat (boots the Linux kernel from source). 2011-11-08T20:21:58 *** caution has quit IRC (Quit: caution) 2011-11-08T20:29:15 (32 6)-(31 6)= 2011-11-08T20:29:19 *** treeform_ has quit IRC (Remote host closed the connection) 2011-11-08T20:29:22 now? 2011-11-08T20:29:26 anyone? 2011-11-08T20:29:36 ok, ill tell you 2011-11-08T20:29:39 (4294967295 0) 2011-11-08T20:29:52 *** treeform has joined #aichallenge 2011-11-08T20:30:07 i was provisional, and declared my pos members as unsgined 2011-11-08T20:30:22 (and i added a subtraction operator of course) 2011-11-08T20:31:15 „designed a system that could act as an Analog or DVB-T Digital TV transmitter by directly generating a VHF signal from a standard PC and VGA card” <--- O_O 2011-11-08T20:31:40 hah, that was him? i remember that. 2011-11-08T20:33:48 *** jasox has quit IRC (Quit: Leaving) 2011-11-08T20:33:55 *** jasox has joined #aichallenge 2011-11-08T20:34:16 :( 2011-11-08T20:34:24 I do not like it when my "new and improved bot" loses to my "old" bot 2011-11-08T20:35:13 i was playing around with some minimax scenarios and it's pretty clear that this game requires a payoff matrix approach 2011-11-08T20:35:35 in a 2v1, if the 1 knows the 2's moves in advance he can either delay indefinitely or take out one of the attackers every time 2011-11-08T20:35:41 i'm still trying to get diffusion working properly ==" 2011-11-08T20:37:29 a1k0n: A does something like that 2011-11-08T20:37:39 dunno what he does exactly but he can always get 2 ants to zap your one ant 2011-11-08T20:37:40 yeah i know he does 2011-11-08T20:37:52 so he has a huge huge edge in combat 2011-11-08T20:37:52 well, but if my one ant were smarter he couldn't necessarily. :) 2011-11-08T20:38:06 *** treeform has quit IRC (Ping timeout: 240 seconds) 2011-11-08T20:39:08 - 2011-11-08T20:39:12 this bot makes no sense ,the more ants I have the faster it runs o_o 2011-11-08T20:39:25 maybe it speeds up as you explore 2011-11-08T20:39:41 yeh do you do any bfs from unexplored regions or something 2011-11-08T20:39:49 a1k0n: what do you imply? that minimax is not "exact"? 2011-11-08T20:40:01 i thought it always leads to optimal decisions 2011-11-08T20:40:08 minimax assumes the players take turns, and each can see each other's moves 2011-11-08T20:40:18 so? 2011-11-08T20:40:22 because player 2 is minimizing the score contingent on what player 1 did 2011-11-08T20:40:30 well, in ants you take turns simultaneously. 2011-11-08T20:40:38 yes, but at the same it is happending vea-versa 2011-11-08T20:40:40 vice* 2011-11-08T20:40:40 so i don't know which way you're about to go when i have to pick. 2011-11-08T20:40:48 bobby: yes, but it never used to slow down so much 2011-11-08T20:40:54 right. so it's kind of a catch-22. you have to guess which way i'm going to guess you're going to go 2011-11-08T20:41:51 but .. yes, that is the one that's slowing down all my shiz 2011-11-08T20:42:04 so you mean if i maximize my score, than i maximize in the remaining set my enemy's score, its not the same, as if i done this the opposite way? 2011-11-08T20:42:34 how the fuck is it taking a full 300 ms 2011-11-08T20:42:59 *** rajanaresh has joined #aichallenge 2011-11-08T20:43:51 anyway, im using a simple heuristic evaluation on the leafes 2011-11-08T20:43:54 well.. right. 2011-11-08T20:44:10 when you put it that way i'm not actually sure. 2011-11-08T20:44:34 but it's definitely a different situation than simultaneous turns 2011-11-08T20:45:34 anyone here doing collaborative diffusion? 2011-11-08T20:45:48 bobbydroptable: even my granny 2011-11-08T20:45:58 mcstar: real helpful 2011-11-08T20:46:03 would she help me out then? 2011-11-08T20:46:07 lol 2011-11-08T20:46:07 yep 2011-11-08T20:46:14 ill ask her 2011-11-08T20:46:21 what you wanna know? 2011-11-08T20:46:31 granny gonna turn in soon 2011-11-08T20:46:41 is there any way to diffuse without multiple passes over the whole map? 2011-11-08T20:46:50 yes 2011-11-08T20:46:58 but its more costly/more complicated 2011-11-08T20:47:12 how could it be more costly over multiple passes? 2011-11-08T20:47:23 well, because of the access pattern 2011-11-08T20:47:26 *** Antimony has quit IRC (Ping timeout: 276 seconds) 2011-11-08T20:47:30 it wouldnt be cache friendly 2011-11-08T20:47:58 actually, if you wanna really know you have to try it 2011-11-08T20:48:14 but diffusion is both faaast and extremely simple 2011-11-08T20:48:22 i must be doing it really wrong then =) 2011-11-08T20:48:29 *** Garf_ has joined #aichallenge 2011-11-08T20:48:30 i can do several hundred iterations per turn 2011-11-08T20:48:39 but i only need 3 2011-11-08T20:48:39 or it may just be a python limitation. 2011-11-08T20:48:48 or 4 2011-11-08T20:49:24 oh, and im using 2d indexing of the array 2011-11-08T20:49:44 bobbydroptable: with numpy? 2011-11-08T20:49:51 nope don't know how to use numpy 2011-11-08T20:49:56 lol 2011-11-08T20:49:59 you must use it for this 2011-11-08T20:50:01 i'm a python noobie 2011-11-08T20:50:06 lol 2011-11-08T20:50:20 and, dont iterate over the array element-by-element 2011-11-08T20:50:24 *** rtward has left #aichallenge 2011-11-08T20:50:29 use soe cumulative functions if you can 2011-11-08T20:50:40 but others will help 2011-11-08T20:50:44 im very tired 2011-11-08T20:50:47 nite all 2011-11-08T20:50:48 thanks anyway 2011-11-08T20:50:49 *** mcstar has quit IRC (Quit: WeeChat 0.3.6) 2011-11-08T20:50:51 night granny 2011-11-08T20:50:54 *** Antimony has joined #aichallenge 2011-11-08T20:52:38 *** GarfTop has quit IRC (Ping timeout: 276 seconds) 2011-11-08T20:56:55 *** JorgeB has quit IRC (Ping timeout: 255 seconds) 2011-11-08T20:57:24 *** Antimony has quit IRC (Ping timeout: 240 seconds) 2011-11-08T20:57:58 so weird with explore map 2011-11-08T21:01:02 anyone know if there's a way to make the code attach to the debugger in visual studio? ( c++ ) 2011-11-08T21:01:14 i know of a way with C#, but i'm not sure if there is a similar solution for c++ 2011-11-08T21:02:29 hrm... mebe an assertion? 2011-11-08T21:06:34 hrm... damnt that doesn't work. for some reason if i through the VS assert in there, it just crashes when i hit retry to debug 2011-11-08T21:06:42 i guess infinite loop and attach to process it is 2011-11-08T21:12:38 hrm.... anyone know if the C++ starter pack Timer.h is broken? 2011-11-08T21:13:18 i'm getting compile errors on the server for the section of Timer.h for Mac/Linux 2011-11-08T21:15:23 *** Adam_ has joined #aichallenge 2011-11-08T21:15:30 Hi everyone 2011-11-08T21:16:02 Anyone in here using java for their ants? 2011-11-08T21:16:39 *** datachomper has joined #aichallenge 2011-11-08T21:16:46 Hello 2011-11-08T21:17:11 Adam_: ofcourse :) 2011-11-08T21:17:41 I need some guidance 2011-11-08T21:17:53 interested? 2011-11-08T21:18:07 ughhh... stupid starter pack bugs 2011-11-08T21:18:15 *** djr_ has joined #aichallenge 2011-11-08T21:18:17 seems the c++ starter pack Timer.h won't compile on the server 2011-11-08T21:18:26 you have to change #include to #include 2011-11-08T21:18:52 Does anyone know how the Aim.java class works? 2011-11-08T21:19:06 snip me some source code? 2011-11-08T21:19:18 sure, one sec 2011-11-08T21:19:39 import java.util.HashMap; import java.util.Map; /** * Represents a direction in which to move an ant. */ public enum Aim { /** North direction, or up. */ NORTH(-1, 0, 'n'), /** East direction or right. */ EAST(0, 1, 'e'), /** South direction or down. */ SOUTH(1, 0, 's'), /** West direction or left. */ WEST(0, -1, 'w'); private static final Map symbolLookup = 2011-11-08T21:19:43 shit that ugly 2011-11-08T21:19:43 ewwww 2011-11-08T21:20:09 pastebin? 2011-11-08T21:20:19 Aim(int rowDelta, int colDelta, char symbol) { this.rowDelta = rowDelta; this.colDelta = colDelta; this.symbol = symbol; } 2011-11-08T21:20:28 that's what I'm really interested in 2011-11-08T21:20:44 that looks like the initializer/constructor ? 2011-11-08T21:20:56 just passing the params to the class so it can store them locally? 2011-11-08T21:21:13 yeah, but I can't figure out where this is being initialized 2011-11-08T21:21:32 meh 2011-11-08T21:21:40 let me ask a better question 2011-11-08T21:22:04 yeah just looking at the Aim class itself 2011-11-08T21:22:05 Ants.java has a method called getTile 2011-11-08T21:22:07 doesn't really tell me much about it haha 2011-11-08T21:22:21 oh nm 2011-11-08T21:22:31 i see it now i think 2011-11-08T21:22:41 the Aim class represents an "aim" from a specific tile 2011-11-08T21:22:51 Do you have Ants.java in front of you? 2011-11-08T21:22:56 yeah i can pull it up 2011-11-08T21:22:59 i just downloaded the starter pack 2011-11-08T21:23:02 sweet. 2011-11-08T21:23:06 alright 2011-11-08T21:23:12 I'm trying to figure out 2011-11-08T21:23:15 why that method 2011-11-08T21:23:19 getTile 2011-11-08T21:23:24 is not returning the tile 2011-11-08T21:23:40 when the tile is on the right bound border of the map 2011-11-08T21:24:20 ok so getTile will get the tile from the current location in the directino you pass it 2011-11-08T21:24:33 yeah 2011-11-08T21:24:40 so I'm doing that in my method 2011-11-08T21:24:49 Aim just has the cardinal directions stored as characters AND as the delta x/y coords 2011-11-08T21:24:51 and somehow, that tile is not being added to my list 2011-11-08T21:25:09 what is the Aim you are passing to getTile? 2011-11-08T21:25:18 *** Antimony has joined #aichallenge 2011-11-08T21:25:19 east 2011-11-08T21:25:25 Aim.EAST 2011-11-08T21:26:00 So I was wondering if there is a bug when you ask for a tile on the edge 2011-11-08T21:26:09 no 2011-11-08T21:26:14 that's what the % is for 2011-11-08T21:26:17 in getTile 2011-11-08T21:26:40 so lets say the map has 100 columns 2011-11-08T21:26:43 col = 111 2011-11-08T21:26:45 your ant is at 99,0 2011-11-08T21:26:49 ok 111 cols 2011-11-08T21:26:53 ur ant is at 110,0 2011-11-08T21:26:59 east is (1,0) 2011-11-08T21:27:00 right 2011-11-08T21:27:04 add those together u get 111,0 2011-11-08T21:27:09 but 111 % 111 = 0 2011-11-08T21:27:12 so u get 0,0 2011-11-08T21:28:04 shouldn't I get 111,0 returned to me? 2011-11-08T21:28:15 no if the map has 111 columns, then 110 is the last column 2011-11-08T21:28:18 after 110 it wraps to 0 2011-11-08T21:28:26 1st columns is column 0 2011-11-08T21:28:35 right 2011-11-08T21:28:43 so this map I'm looking at has 2011-11-08T21:28:46 112 cols 2011-11-08T21:28:55 my ant is on 110 2011-11-08T21:29:02 I want the tile 111,0 2011-11-08T21:29:09 so then yeah you should be getting 111,0 2011-11-08T21:29:13 from an east move 2011-11-08T21:29:58 the only thing I can think that might be happening is Aim is wrong? 2011-11-08T21:30:12 what value do you get back from that function? 2011-11-08T21:30:16 *** thestinger has left #aichallenge ("WeeChat 0.3.6") 2011-11-08T21:30:26 Aim e = Aim.fromSymbol('e'); 2011-11-08T21:30:31 should this work? 2011-11-08T21:30:45 yeah that should work 2011-11-08T21:30:49 i would think 2011-11-08T21:31:05 what the heck does A use to be so awesome? 2011-11-08T21:31:08 but so shoudl your Aim.East 2011-11-08T21:31:17 or rather Aim.EAST 2011-11-08T21:31:23 who's A? 2011-11-08T21:31:29 I'll try that. see what happens. 2011-11-08T21:31:29 a player/bot 2011-11-08T21:31:42 http://ants.fluxid.pl/player/A 2011-11-08T21:31:50 how does he do it 2011-11-08T21:32:12 yes, but who is he? 2011-11-08T21:32:19 no idea 2011-11-08T21:32:20 magiik linked to him 2011-11-08T21:32:23 ah 2011-11-08T21:32:26 yea he's never on irc afaik 2011-11-08T21:32:30 or maybe he is someone here in secret 2011-11-08T21:32:35 maybe it's accoun 2011-11-08T21:32:42 *** thestinger has joined #aichallenge 2011-11-08T21:33:13 http://ants.fluxid.pl/replay.4124 2011-11-08T21:33:16 can't be thar great 2011-11-08T21:33:18 he timed out in this game 2011-11-08T21:33:36 well that game is ultra old 2011-11-08T21:33:41 his bot wasn't nearly as good back then 2011-11-08T21:33:43 you can tell 2011-11-08T21:33:48 aaand also sometimes tcp randomly times out 2011-11-08T21:33:54 true 2011-11-08T21:34:06 i have problems with the tcp client 2011-11-08T21:34:11 i can't run multiple games 2011-11-08T21:34:16 or eventually it will deadlock the computer 2011-11-08T21:34:20 yea 2011-11-08T21:34:21 something with threads not closing 2011-11-08T21:34:24 it eventualyl just deadlocks 2011-11-08T21:34:42 fun stuff lol 2011-11-08T21:34:46 man i'm out of ideas for my bot :( 2011-11-08T21:34:51 implemented all the key things 2011-11-08T21:34:52 man i really need to add some micromanagement for battles 2011-11-08T21:35:00 enemy avoidance, food, exploration, hill razing 2011-11-08T21:35:02 maybe i should try implementing minimax or w/e 2011-11-08T21:35:07 yet my bot still kind of sucks 2011-11-08T21:35:09 yea magiik maybe 2011-11-08T21:35:15 however I want my bot to work on official servers 2011-11-08T21:35:19 and minimax sounds very expensive 2011-11-08T21:35:28 omg i went from python to c++ 2011-11-08T21:35:37 and i can do like 1000000x more calculations 2011-11-08T21:35:37 big diff? 2011-11-08T21:35:40 oh yea 2011-11-08T21:35:43 I know what you mean 2011-11-08T21:35:47 (went from ruby to java) 2011-11-08T21:36:07 like things that i was kludging together to get by performance issues 2011-11-08T21:36:14 are not even an issue it seems on this side 2011-11-08T21:36:19 and i'm not even doing any optimizations 2011-11-08T21:36:26 Hey Magiik 2011-11-08T21:36:36 like I'm doing BFS maps, but I'm like adding 100s of BFSes together 2011-11-08T21:36:42 *** TwistedLogic has quit IRC (Quit: Page closed) 2011-11-08T21:36:45 Adam_: sup 2011-11-08T21:36:47 those changes didn't seem to do anything. 2011-11-08T21:36:57 magiik: do you use unexplored tiles as one of your bfs maps>? 2011-11-08T21:37:06 on big maps this will take 300 ms for me at start for no apparent reason 2011-11-08T21:37:08 no, because eventually i will explore them all ;) 2011-11-08T21:37:10 all my other bfs maps are ULTRA fast 2011-11-08T21:37:17 well unseen tiles :P 2011-11-08T21:37:19 hrm... i need to try that large map 2011-11-08T21:37:38 i use lastseen values of tiles loosely to generate goals 2011-11-08T21:38:09 so basically i generate explore goals for older seen tiles, and update the goals as the lastseens change 2011-11-08T21:38:13 Is there anything wrong with throwing all of the Aims into a set ? 2011-11-08T21:38:24 then iterating through them 2011-11-08T21:38:28 multiple times? 2011-11-08T21:38:36 Adam_: i think u need to debug a little bit, i doubt the Java starter pack has a bug in it's Aim or wrap calculations 2011-11-08T21:38:54 I've been debugging for a week :( 2011-11-08T21:39:00 get some debug output going somewhere, and print out things you think may be suspect, validating that what you think is going on is goign on 2011-11-08T21:39:06 or use a remote debugger and get inside the program 2011-11-08T21:39:18 are you using an IDE? 2011-11-08T21:39:26 Netbeans 2011-11-08T21:39:38 never used netbeans, but I know eclipse allows you to remote debug 2011-11-08T21:40:20 Well let's talk about some of my code then, maybe I'm using the classes wrong 2011-11-08T21:40:20 that's what i'm doing for visual studio and C++ - i put huge wait times on the playgame.py options, then put an infinite loop near the start, load it up, attach the debugger, break out of the infinite loop 2011-11-08T21:40:39 Adam_: roflmao may be able to help you a lil better, he's using java 2011-11-08T21:40:40 cool 2011-11-08T21:40:54 and i'm about to leave soon 2011-11-08T21:40:57 but i'll help as much as i can 2011-11-08T21:40:59 Okay 2011-11-08T21:41:02 well, until then 2011-11-08T21:41:29 is there anything wrong with re-using a previously iterated set of Aims? 2011-11-08T21:41:40 not that I can think of 2011-11-08T21:41:43 unless you changed the set 2011-11-08T21:41:59 nah. 2011-11-08T21:42:04 but that gives me an Idea 2011-11-08T21:42:13 thanks for your time though :) 2011-11-08T21:42:36 Adam_: i would recommend lots of debug/stderr/log outputs 2011-11-08T21:42:53 print out all kidns of crap just for debugging the one problem, then you can nuke all those outputs 2011-11-08T21:43:13 when all my debugging is turned on, my logs turn into megabytes 2011-11-08T21:43:33 hey Adam 2011-11-08T21:43:40 Hi 2011-11-08T21:43:47 Ill do that magiik, thanks 2011-11-08T21:44:08 aite my ride is here 2011-11-08T21:44:09 i gotta run 2011-11-08T21:44:10 peace 2011-11-08T21:46:46 Ciao 2011-11-08T21:46:59 Hi rofl. 2011-11-08T21:51:28 *** Adam_ has quit IRC (Ping timeout: 265 seconds) 2011-11-08T21:51:59 *** rmmh has quit IRC (Ping timeout: 248 seconds) 2011-11-08T21:53:09 *** djr_ has quit IRC (Read error: Connection reset by peer) 2011-11-08T21:53:43 *** djr_ has joined #aichallenge 2011-11-08T21:57:11 *** bergmark has joined #aichallenge 2011-11-08T22:00:17 *** rmmh has joined #aichallenge 2011-11-08T22:01:15 *** lg41 has joined #aichallenge 2011-11-08T22:15:03 *** TheDigitalNinja has joined #aichallenge 2011-11-08T22:17:54 *** goffrie has quit IRC (Remote host closed the connection) 2011-11-08T22:23:15 *** galdor has quit IRC (Read error: Operation timed out) 2011-11-08T22:23:18 *** dr- has quit IRC (Ping timeout: 240 seconds) 2011-11-08T22:23:20 *** goffrie has joined #aichallenge 2011-11-08T22:23:22 *** hjax has joined #aichallenge 2011-11-08T22:23:52 *** Clex has quit IRC (Ping timeout: 258 seconds) 2011-11-08T22:24:16 *** galdor has joined #aichallenge 2011-11-08T22:24:33 *** Clex has joined #aichallenge 2011-11-08T22:25:05 *** janzert has quit IRC (Read error: Connection reset by peer) 2011-11-08T22:25:22 *** djr_ has quit IRC (Read error: Connection reset by peer) 2011-11-08T22:26:39 *** djr_ has joined #aichallenge 2011-11-08T22:26:57 *** janzert has joined #aichallenge 2011-11-08T22:27:42 *** hjax has quit IRC (Remote host closed the connection) 2011-11-08T22:28:11 *** dr- has joined #aichallenge 2011-11-08T22:28:13 *** thestinger has quit IRC (Ping timeout: 240 seconds) 2011-11-08T22:28:40 *** andar has quit IRC (Ping timeout: 260 seconds) 2011-11-08T22:28:54 *** smiley1993 has quit IRC (Ping timeout: 240 seconds) 2011-11-08T22:29:07 *** andar has joined #aichallenge 2011-11-08T22:29:42 *** twymer has joined #aichallenge 2011-11-08T22:29:46 *** vasile has joined #aichallenge 2011-11-08T22:29:47 *** thestinger has joined #aichallenge 2011-11-08T22:29:54 hi 2011-11-08T22:29:56 *** smiley1993 has joined #aichallenge 2011-11-08T22:30:12 wow so many people doing it 2011-11-08T22:31:26 *** u_ has quit IRC (Ping timeout: 276 seconds) 2011-11-08T22:36:08 *** Antimony has quit IRC (Ping timeout: 258 seconds) 2011-11-08T22:38:31 yeh :) some doing a lot worse than they'd like XD 2011-11-08T22:42:41 *** Raimondi has joined #aichallenge 2011-11-08T22:42:54 *** dr- has quit IRC (Ping timeout: 240 seconds) 2011-11-08T22:43:15 *** u_ has joined #aichallenge 2011-11-08T22:44:21 *** userjjb_ has quit IRC (Quit: Page closed) 2011-11-08T22:45:11 *** dr- has joined #aichallenge 2011-11-08T22:46:29 *** vladimirfol has joined #aichallenge 2011-11-08T22:46:38 *** dvladim has joined #aichallenge 2011-11-08T22:49:59 :) 2011-11-08T22:50:11 bobbydroptable: how are you doing? 2011-11-08T22:52:40 *** goffrie has left #aichallenge 2011-11-08T22:53:31 not very well :( 2011-11-08T22:56:54 well to be honest, i can make an "okay" bot 2011-11-08T22:57:06 but i'm not satisfied with an "okay" bot, but I can't seem to do anything better. 2011-11-08T22:57:08 :) 2011-11-08T23:01:11 *** JorgeB has joined #aichallenge 2011-11-08T23:03:18 *** Raimondi has quit IRC (Quit: Get MacIrssi - http://www.sysctl.co.uk/projects/macirssi/) 2011-11-08T23:10:24 *** yoden has quit IRC (Quit: Leaving.) 2011-11-08T23:13:31 *** Raimondi has joined #aichallenge 2011-11-08T23:16:37 *** dvladim has quit IRC (Ping timeout: 240 seconds) 2011-11-08T23:20:12 *** Adam__ has joined #aichallenge 2011-11-08T23:20:18 Hello 2011-11-08T23:20:58 is anybody out there? Just nod if you can here me... 2011-11-08T23:21:00 *** Antimony has joined #aichallenge 2011-11-08T23:21:12 Hi 2011-11-08T23:22:20 *** vladimirfol has quit IRC (Ping timeout: 265 seconds) 2011-11-08T23:24:16 Adam__: hear* 2011-11-08T23:28:01 *** rajanaresh has quit IRC (Ping timeout: 252 seconds) 2011-11-08T23:28:54 yeah 2011-11-08T23:28:56 that 2011-11-08T23:29:24 Hey, does anyone know anything about getting the time your turn takes to process? 2011-11-08T23:29:32 ...java 2011-11-08T23:33:53 did the starter kit provide a timer? 2011-11-08T23:36:12 never mind, figured it out 2011-11-08T23:36:15 thanks though 2011-11-08T23:36:29 although...any tips on how that should be used? 2011-11-08T23:36:38 *** AVAVT has joined #aichallenge 2011-11-08T23:36:52 I know nothing of scaling techniques ;) 2011-11-08T23:37:41 well the easiest thing is to just check the time remaining 2011-11-08T23:37:55 and not do anything computationally expensive if there isn't a lot of time left 2011-11-08T23:38:51 *** roflmao has quit IRC (Quit: Leaving.) 2011-11-08T23:41:40 *** b0rder_ has joined #aichallenge 2011-11-08T23:41:45 cool 2011-11-08T23:41:47 ^ "roflmao quit im leaving" 2011-11-08T23:43:38 *** u_ has quit IRC (Quit: u_) 2011-11-08T23:44:32 My bot seems to be ignoring enemy mounds. :( 2011-11-08T23:51:59 *** Nails has joined #aichallenge 2011-11-08T23:56:31 Wow 2011-11-08T23:56:39 It turns out I never actually processed seen hills 2011-11-08T23:56:44 nice. 2011-11-08T23:56:59 no wonder it was so bad at attacking 2011-11-08T23:58:19 How do you grab a random number from the seed? 2011-11-08T23:58:21 java 2011-11-08T23:58:30 *** twymer has quit IRC (Ping timeout: 240 seconds)