2011-04-04T00:01:18 *** stride_ has quit IRC (Ping timeout: 248 seconds) 2011-04-04T00:01:22 *** stride has joined #aichallenge 2011-04-04T00:11:52 amstan: got any strategy ideas for me? :P 2011-04-04T00:11:57 McLeopold: or you 2011-04-04T00:12:06 yes 2011-04-04T00:12:15 see https://github.com/aichallenge/aichallenge/wiki/Ants-Tutorial and the linked strategy guide page 2011-04-04T00:12:50 maybe suggest A* 2011-04-04T00:13:28 i did :P 2011-04-04T00:13:50 but it is only really helpful when you have a given objective square, which you don't 2011-04-04T00:13:52 maybe suggest it by name 2011-04-04T00:14:00 I did, i even linked it 2011-04-04T00:14:06 it's on the strategy guide page under path finding 2011-04-04T00:14:14 oh 2011-04-04T00:14:22 suggest 1 ant per food 2011-04-04T00:14:35 *** stride_ has joined #aichallenge 2011-04-04T00:14:38 yep, I sort of did but not explicitly 2011-04-04T00:14:43 i'll do that on the tutorial page 2011-04-04T00:14:57 *** _jmcarthur <_jmcarthur!~jake@76.73.216.201> has joined #aichallenge 2011-04-04T00:15:14 *** stride has quit IRC (*.net *.split) 2011-04-04T00:15:17 *** jmcarthur has quit IRC (*.net *.split) 2011-04-04T00:15:17 *** davidd has quit IRC (*.net *.split) 2011-04-04T00:15:42 i will also point out that you may want to send more if your closest is likely to die with an enemy before reaching the food square 2011-04-04T00:16:01 or send less if there are 2 food 2011-04-04T00:16:12 yep 2011-04-04T00:16:28 food collecting is quite untrivial 2011-04-04T00:16:34 is there an algo for a breeding traveling salesman? 2011-04-04T00:16:45 haha, not sure 2011-04-04T00:17:03 i can't really do the battle thing until we pick a resolution method 2011-04-04T00:17:07 *** Cerealklr has joined #aichallenge 2011-04-04T00:17:16 are we ready to pull jeff in and convince him of damage to lock it in? 2011-04-04T00:17:20 let's not make any suggestions there anyway 2011-04-04T00:17:32 i want to cover the idea of using game trees for battling 2011-04-04T00:17:41 I talked with him the day damage was suggestted 2011-04-04T00:17:50 and mention the difficulties people face with backward induction for general n player games 2011-04-04T00:17:52 don't get your hopes up 2011-04-04T00:18:39 ... it needs to be finalised some time anyway..? 2011-04-04T00:19:00 we need to make some decisions and just go with them, there's too many options all over the place 2011-04-04T00:19:50 so, if you are in range, all ants die 2011-04-04T00:20:11 that's worse than iterative closest 2011-04-04T00:20:45 *** davidd has joined #aichallenge 2011-04-04T01:28:23 *** phire has joined #aichallenge 2011-04-04T01:51:01 @seen bss03 2011-04-04T01:51:01 McLeopold: bss03 was last seen in #aichallenge 2 weeks, 3 days, 3 hours, 29 minutes, and 10 seconds ago: amstan: My email is everywhere; I don't depend on it being secret to keep me safe from SPAM. 2011-04-04T01:51:11 *** znutar has joined #aichallenge 2011-04-04T02:10:30 *** amstan has quit IRC (Remote host closed the connection) 2011-04-04T02:27:32 sigh, McLeopold: did the bot input format change recently? 2011-04-04T02:27:41 my starter bot is crashing :\ 2011-04-04T02:27:41 no 2011-04-04T02:36:55 all good now 2011-04-04T02:37:13 *** p4p4p5 has joined #aichallenge 2011-04-04T02:44:03 aichallenge: Nick Ham epsilon * r2b8f523 / (11 files): updated the c++ starter package - http://bit.ly/dKT9eX 2011-04-04T02:55:57 hopefully i don't need to touch that again 2011-04-04T03:22:56 *** Sunhay has quit IRC (Quit: leaving) 2011-04-04T03:25:08 *** Sunhay has joined #aichallenge 2011-04-04T03:26:12 *** boegel has joined #aichallenge 2011-04-04T03:39:59 *** sigh has quit IRC (Ping timeout: 255 seconds) 2011-04-04T04:00:24 *** delt0r___ has joined #aichallenge 2011-04-04T04:02:03 *** delt0r_ has quit IRC (Ping timeout: 255 seconds) 2011-04-04T04:11:07 *** needsch has joined #aichallenge 2011-04-04T04:18:08 *** Apophis_ has joined #aichallenge 2011-04-04T04:22:56 *** AlliedEnvy_ has joined #aichallenge 2011-04-04T04:26:11 *** shutch has joined #aichallenge 2011-04-04T04:26:24 *** slkjcna_ has joined #aichallenge 2011-04-04T04:26:25 *** Stocha has joined #aichallenge 2011-04-04T04:27:00 *** _aj_ <_aj_!~aj@66.133.136.43> has joined #aichallenge 2011-04-04T04:28:06 *** BtbN_ has joined #aichallenge 2011-04-04T04:29:11 yeah, the breeding salesman in ennemy territory is puzling to me :p 2011-04-04T04:29:16 *** _aj <_aj!~aj@66.133.136.43> has quit IRC (*.net *.split) 2011-04-04T04:29:16 *** BtbN has quit IRC (*.net *.split) 2011-04-04T04:29:16 *** shutch_ has quit IRC (*.net *.split) 2011-04-04T04:29:17 *** nann has quit IRC (*.net *.split) 2011-04-04T04:29:17 *** Chris_0076 has quit IRC (*.net *.split) 2011-04-04T04:29:17 *** javagamer has quit IRC (*.net *.split) 2011-04-04T04:29:18 *** AlliedEnvy has quit IRC (*.net *.split) 2011-04-04T04:29:18 *** slkjcna has quit IRC (*.net *.split) 2011-04-04T04:29:19 *** sixthgea1 has quit IRC (*.net *.split) 2011-04-04T04:29:19 *** dwchandl1r has quit IRC (*.net *.split) 2011-04-04T04:29:20 *** Heulsay has quit IRC (*.net *.split) 2011-04-04T04:29:21 *** JamesMG has quit IRC (*.net *.split) 2011-04-04T04:29:25 *** sixthgea1 has joined #aichallenge 2011-04-04T04:29:25 *** dwchandl1r has joined #aichallenge 2011-04-04T04:29:25 *** Heulsay has joined #aichallenge 2011-04-04T04:29:25 *** JamesMG has joined #aichallenge 2011-04-04T04:29:36 especially the : this unknown, so you guess what's there part. 2011-04-04T04:31:31 *** sigh has joined #aichallenge 2011-04-04T04:36:35 *** Chris_0076 has joined #aichallenge 2011-04-04T04:44:33 *** kaemo has joined #aichallenge 2011-04-04T04:47:02 i am in no position to test for score system at the moment. I think there are two reasons for that. 2011-04-04T04:47:18 First i guess you need to have some sort of fighting credibility in order to compare your scoring systems. 2011-04-04T04:47:58 Which is hard when you have no idea of what fighting there will be in. I still bet for Damage X=1. Because i think it is the most consensual at the moment. 2011-04-04T04:49:00 Second, that's a bit silly but well. I simply made nothing worthy, and i'm obsessed at doing something cool with ant bot programming. As i'm totally stuck at the moment( without much time to put in to be fair), i'm too frustrated to put my mind over testing scores systems :p 2011-04-04T04:49:03 However .... 2011-04-04T04:49:14 I still think you have to do it as straight as forward as possible. 2011-04-04T04:49:32 So you would prefer stateless. (inverse order of elimination is) 2011-04-04T04:49:39 (ant counting is also) 2011-04-04T04:50:01 So you could go for (inverse order of elimination /Tie if die in same turn + ant counting for remaining players 2011-04-04T04:50:23 As you told, the fog of war make it very dangerous to stay still and hide. Because you have no idea what's going on out there. 2011-04-04T04:51:02 I can't believe hiding in a small territory will do it. (although i did'nt do too much multiplayer study) 2011-04-04T04:51:18 Even if it does, it's hard to foretell what scoring system will prevent that. 2011-04-04T04:51:37 So i think the scoring should be stateless, and easy to undertand. 2011-04-04T04:52:23 Dear diary, today a guy named json knocked on my door and said we have some issues to resolve. He would have scared the hell out of me if it was Friday 13. But this time it is save to say that I just have to load up my IDE. 2011-04-04T04:58:25 *** Naktibalda has joined #aichallenge 2011-04-04T05:08:26 @later tell keth i'm afraid it was an april fools joke :'( 2011-04-04T05:08:26 Stocha: Ready to serve. 2011-04-04T05:17:09 *** dr-` has joined #aichallenge 2011-04-04T05:18:50 *** dr- has quit IRC (*.net *.split) 2011-04-04T05:28:31 *** needsch has quit IRC (Quit: Leaving.) 2011-04-04T05:46:52 *** Sunhay has quit IRC (*.net *.split) 2011-04-04T05:46:53 *** drhodes has quit IRC (*.net *.split) 2011-04-04T05:46:53 *** jmpespxoreax has quit IRC (*.net *.split) 2011-04-04T05:46:55 *** rstrong has quit IRC (*.net *.split) 2011-04-04T05:50:49 *** rstrong has joined #aichallenge 2011-04-04T05:50:55 *** drhodes has joined #aichallenge 2011-04-04T05:51:27 *** Sunhay has joined #aichallenge 2011-04-04T05:58:00 *** jmpespxoreax has joined #aichallenge 2011-04-04T06:11:05 *** Stocha has quit IRC (Quit: Page closed) 2011-04-04T06:15:36 *** javagamer has joined #aichallenge 2011-04-04T06:25:33 *** __jmcarthur <__jmcarthur!~jake@76.73.216.201> has joined #aichallenge 2011-04-04T06:26:47 *** _jmcarthur <_jmcarthur!~jake@76.73.216.201> has quit IRC (Read error: Operation timed out) 2011-04-04T06:30:36 *** needsch has joined #aichallenge 2011-04-04T06:35:24 *** needsch has quit IRC (Client Quit) 2011-04-04T06:40:56 *** nann has joined #aichallenge 2011-04-04T06:42:04 *** Apophis_ has quit IRC (Ping timeout: 252 seconds) 2011-04-04T06:47:46 *** Eruonen has joined #aichallenge 2011-04-04T06:57:07 *** Stocha has joined #aichallenge 2011-04-04T07:06:51 will there be any algorithm for choke point detection in the tutorials ? 2011-04-04T07:07:25 if someone could point me to a wiki of some sort on the subject. over manhatan distance :p 2011-04-04T07:08:16 *** boegel has quit IRC (Ping timeout: 240 seconds) 2011-04-04T07:15:44 *** Apophis_ has joined #aichallenge 2011-04-04T07:16:23 *** Naktibalda_ has joined #aichallenge 2011-04-04T07:19:00 *** javagamer1 has joined #aichallenge 2011-04-04T07:19:02 *** Naktibalda has quit IRC (Ping timeout: 246 seconds) 2011-04-04T07:19:03 *** dr-` has quit IRC (Ping timeout: 246 seconds) 2011-04-04T07:19:03 *** nann has quit IRC (Ping timeout: 246 seconds) 2011-04-04T07:19:03 *** javagamer has quit IRC (Ping timeout: 246 seconds) 2011-04-04T07:19:04 *** Chris_0076 has quit IRC (Ping timeout: 246 seconds) 2011-04-04T07:19:04 *** AlliedEnvy_ has quit IRC (Ping timeout: 246 seconds) 2011-04-04T07:19:21 *** dr- has joined #aichallenge 2011-04-04T07:20:05 *** Chris_0076 has joined #aichallenge 2011-04-04T07:20:11 *** Naktibalda_ is now known as Naktibalda 2011-04-04T07:20:42 *** phire has quit IRC (Ping timeout: 264 seconds) 2011-04-04T07:32:32 *** AlliedEnvy has joined #aichallenge 2011-04-04T07:33:18 *** boegel has joined #aichallenge 2011-04-04T07:43:20 *** AlliedEnvy_ has joined #aichallenge 2011-04-04T07:43:29 *** dr-` has joined #aichallenge 2011-04-04T07:43:57 *** AlliedEnvy has quit IRC (Ping timeout: 246 seconds) 2011-04-04T07:43:57 *** Chris_0076 has quit IRC (Ping timeout: 246 seconds) 2011-04-04T07:43:57 *** dr- has quit IRC (Ping timeout: 246 seconds) 2011-04-04T07:46:00 *** Chris_0076 has joined #aichallenge 2011-04-04T07:50:41 *** delt0r___ has quit IRC (Ping timeout: 240 seconds) 2011-04-04T07:53:54 *** jmpespxoreax has quit IRC (Changing host) 2011-04-04T07:53:54 *** jmpespxoreax has joined #aichallenge 2011-04-04T08:18:43 aichallenge: Marco Leise epsilon * r1b24fae / (6 files in 3 dirs): visualizer: support for 100% json replays - http://bit.ly/dXsxHl 2011-04-04T08:18:44 aichallenge: Marco Leise epsilon * ra06e7bb / (12 files in 2 dirs): Merge branch 'epsilon' of github.com:aichallenge/aichallenge into epsilon - http://bit.ly/gVXSht 2011-04-04T08:25:48 *** chris__0076 has joined #aichallenge 2011-04-04T08:28:22 *** Chris_0076 has quit IRC (Ping timeout: 246 seconds) 2011-04-04T08:31:16 *** chris__0076 has quit IRC (Ping timeout: 240 seconds) 2011-04-04T08:38:07 *** antimatroid has quit IRC (Ping timeout: 240 seconds) 2011-04-04T08:45:32 *** chris__0076 has joined #aichallenge 2011-04-04T08:54:30 *** antimatroid has joined #aichallenge 2011-04-04T08:57:08 *** sigh has quit IRC (Remote host closed the connection) 2011-04-04T09:13:15 *** FireFly has joined #aichallenge 2011-04-04T09:18:02 *** mceier has joined #aichallenge 2011-04-04T10:05:09 *** iFire has quit IRC (Ping timeout: 276 seconds) 2011-04-04T10:08:11 *** Eruonen has quit IRC (Read error: Connection reset by peer) 2011-04-04T10:11:40 *** FireFly has quit IRC (*.net *.split) 2011-04-04T10:11:40 *** boegel has quit IRC (*.net *.split) 2011-04-04T10:13:02 *** drhodes has quit IRC (*.net *.split) 2011-04-04T10:13:02 *** slkjcna_ has quit IRC (*.net *.split) 2011-04-04T10:13:02 *** znutar has quit IRC (*.net *.split) 2011-04-04T10:13:02 *** davidd has quit IRC (*.net *.split) 2011-04-04T10:15:01 *** iFire` has joined #aichallenge 2011-04-04T10:15:01 *** delt0r___ has joined #aichallenge 2011-04-04T10:15:01 *** Eruonen has joined #aichallenge 2011-04-04T10:15:01 Stocha: avoiding blocks is going to be tricky 2011-04-04T10:15:01 i haven't put anything like that in the tute 2011-04-04T10:15:01 *** Stocha has quit IRC (Quit: Page closed) 2011-04-04T10:15:01 *** znutar_ has joined #aichallenge 2011-04-04T10:15:01 *** FireFly has joined #aichallenge 2011-04-04T10:15:01 *** boegel has joined #aichallenge 2011-04-04T10:15:01 *** drhodes has joined #aichallenge 2011-04-04T10:15:01 *** slkjcna_ has joined #aichallenge 2011-04-04T10:15:01 *** znutar has joined #aichallenge 2011-04-04T10:15:01 *** davidd has joined #aichallenge 2011-04-04T10:15:01 *** znutar has quit IRC (Excess Flood) 2011-04-04T10:16:43 *** stocha has joined #aichallenge 2011-04-04T10:16:44 *** Eruonen has quit IRC (Client Quit) 2011-04-04T10:17:54 animatroid : i was simply asking for algorithm detecting bottlenecks statically on a known map. 2011-04-04T10:18:04 antimatroid even 2011-04-04T10:20:42 *** needsch has joined #aichallenge 2011-04-04T10:26:20 *** needsch has quit IRC (*.net *.split) 2011-04-04T10:26:20 *** FireFly has quit IRC (*.net *.split) 2011-04-04T10:26:21 *** boegel has quit IRC (*.net *.split) 2011-04-04T10:26:21 *** drhodes has quit IRC (*.net *.split) 2011-04-04T10:26:22 *** slkjcna_ has quit IRC (*.net *.split) 2011-04-04T10:26:24 *** davidd has quit IRC (*.net *.split) 2011-04-04T10:26:39 stocha; the best method i could think of would be to find articulation points 2011-04-04T10:26:41 but they wouldn't actually be articulation points, ignore that 2011-04-04T10:26:41 how do one do that ? 2011-04-04T10:26:41 ah so let's not do it :p 2011-04-04T10:26:41 any other ideas ? :p 2011-04-04T10:26:41 it's messy, i would try searching the grid with nxn blocks 2011-04-04T10:26:42 and finding where that can't traverse 2011-04-04T10:26:42 yeah that's the only idea i got. 2011-04-04T10:26:42 but it seems so much cpu time 2011-04-04T10:26:42 be careful with that though 2011-04-04T10:26:42 well ? 2011-04-04T10:26:42 be careful about it in what way ? 2011-04-04T10:26:42 consider this: 2011-04-04T10:26:42 ..% 2011-04-04T10:26:42 ..% 2011-04-04T10:26:42 %.. 2011-04-04T10:26:42 %.. 2011-04-04T10:26:42 that's one max block. 2011-04-04T10:26:42 on the north to south path 2011-04-04T10:26:42 if you painted that from all sides, you would mark every square as visited, but you can't actually walk a 2x2 block through there 2011-04-04T10:26:43 yeah. So i build a block of 2x2 ants and try to walk everywhere 2011-04-04T10:26:43 i retract my comment about food gathering being anywhere near trivial :P 2011-04-04T10:26:43 if it doesn't work, then this is a choke point. then i spawn randomly ... 2011-04-04T10:26:43 stocha: but you need to work out where it can't get though, not just the "squares it didn't touch" 2011-04-04T10:26:43 and do it for every block size. 2011-04-04T10:26:43 Then i come back 10 years later, and i have got all botlneck on the known map. 2011-04-04T10:26:45 *** ltriant has quit IRC (Ping timeout: 246 seconds) 2011-04-04T10:26:45 you don't want bottlenecks really unless the opening is quite small 2011-04-04T10:26:45 i think i can manager it this way. it's not very satisfying somehow. Though. 2011-04-04T10:26:45 you should be able to do that reasonably fast 2011-04-04T10:26:45 Yeah, but the thing is : i want the bottlenecks, even the large one. 2011-04-04T10:26:45 i could even explain why :p 2011-04-04T10:26:45 ... 2011-04-04T10:26:45 walking randomly 2x2 block doesn't sound that fast to me. 2011-04-04T10:26:45 i could even try to just spawn every possible 2x2 block that touch known ground. 2011-04-04T10:26:45 it would be the same i guess. 2011-04-04T10:26:45 write a block class, and when you move it, update 2 blocks locations and calculate 2 new ones 2011-04-04T10:26:45 where i can't i got a choke point. 2011-04-04T10:26:45 or just move the top left piece 2011-04-04T10:26:46 and make sure it's still validly positioned 2011-04-04T10:26:46 so basically you try every 2x2 block possible. 2011-04-04T10:26:46 and see if it fits or not. 2011-04-04T10:26:46 there's only one 2x2 block :P 2011-04-04T10:26:46 net split 2011-04-04T10:26:46 that was small 2011-04-04T10:26:46 (2x2block , (x,y) 2011-04-04T10:26:50 *** needsch has joined #aichallenge 2011-04-04T10:26:51 *** FireFly has joined #aichallenge 2011-04-04T10:26:51 *** boegel has joined #aichallenge 2011-04-04T10:26:51 *** drhodes has joined #aichallenge 2011-04-04T10:26:51 *** slkjcna_ has joined #aichallenge 2011-04-04T10:26:51 *** davidd has joined #aichallenge 2011-04-04T10:27:35 so i guess botlneck finding is untrivial also. like the splitting salesman :p 2011-04-04T10:27:47 Especially the spliting salesman in presence of ennemies :p 2011-04-04T10:28:08 maybe a litle bit less trivial though. 2011-04-04T10:28:14 but still probably expensive. 2011-04-04T10:28:28 i can only really think of a tree for breeding salesman (I like that term more :P) 2011-04-04T10:28:36 but i want to use a tree for battles 2011-04-04T10:35:54 food is just too important. 2011-04-04T10:35:54 we will have to solve it somehow. At least efficiently. if not perfectly (wich is probably quite hard, when you take unexplored potential food into account :p ) 2011-04-04T10:41:19 aichallenge: Marco Leise epsilon * ra377e35 / (15 files in 3 dirs): added html visualizer ant task ("ant html-visualizer") / corrected line endings in cpp starter bot - http://bit.ly/epodFQ 2011-04-04T10:43:45 Is there any way git can simply ignore line endings? i constantly get "modified" files from people who checked their code in with CR/LF instead of just LF. 2011-04-04T10:45:14 Ideally everyone on Windows would activate git's auto-crlf feature, which converts between Windows and Unix style during checkout/commit. 2011-04-04T10:46:46 *** phire has joined #aichallenge 2011-04-04T10:47:20 see: http://help.github.com/dealing-with-lineendings/ 2011-04-04T10:49:24 Frontier1: is that from me? :P 2011-04-04T10:49:44 antimatroid: did you work on the cpp bot? 2011-04-04T10:49:50 i did :P 2011-04-04T10:50:41 Well, I haven't heard anyone else complaining, so I wonder if I do something wrong. 2011-04-04T10:51:53 On the other hand github advertises unix line-endings for all files, so I think it is save to ask you to activate that auto-conversion feature :) 2011-04-04T10:53:37 oh ffs 2011-04-04T10:53:38 i left a line commented that does the vision for the starter bot 2011-04-04T10:53:39 i'm on ubuntu now anyway :P 2011-04-04T10:53:39 shouldn't it be alright anyway? 2011-04-04T10:54:22 don't you get these warnings about CR/LF? 2011-04-04T10:58:08 if you copied your files from windows and had windows-line-endings in it, I could imagine that your ide did not auto-convert the existing lines. 2011-04-04T10:58:15 nope 2011-04-04T11:00:41 yeah that wouldn't surprise me 2011-04-04T11:02:47 i use code::blocks for all things c++y 2011-04-04T11:04:27 i'll have a play 2011-04-04T11:04:27 you wan me in CR mode yes? not LR? 2011-04-04T11:04:27 LF* 2011-04-04T11:04:29 i think it's now fixed 2011-04-04T11:04:30 i'll upload it again later, working on gsoc application atm 2011-04-04T11:05:14 antimatroid: use \n (lf) for all files 2011-04-04T11:05:22 *** chris__0076 has quit IRC (Ping timeout: 252 seconds) 2011-04-04T11:05:35 Frontier1: it was already on lf :\ 2011-04-04T11:06:01 yeah, maybe that were leftover lines from your windows time. 2011-04-04T11:07:09 hmmm, well maybe now it's converted them all 2011-04-04T11:07:37 code::blocks is great, it just loads the plugins a lot faster on windows for me. on linux/gtk creating wxWidgets guis from xml descriptions seems to be very slow. what is your experience? 2011-04-04T11:07:55 i don't make guis :P 2011-04-04T11:08:10 antimatroid: the plugins come with guis 2011-04-04T11:08:21 and they take *ages* for me to load 2011-04-04T11:08:34 i haven't noticed any major differences :\ 2011-04-04T11:08:47 I have to manually delete them as root, to have the ide start up faster :p 2011-04-04T11:08:57 the only plugin i use is the source code formatter when people give me ugly code 2011-04-04T11:08:59 its wonderful 2011-04-04T11:09:23 it's********** 2011-04-04T11:09:25 hehe, yes that one and the auto-completion. 2011-04-04T11:10:04 so you have the start-fast-switch enabled. hmm. i start to hate gentoo. should have used ubuntu in the first place 2011-04-04T11:10:40 where is the "start fast switch"? 2011-04-04T11:10:59 i can't believe i didn't switch to ubuntu ages ago 2011-04-04T11:11:38 if i knew where the start fast switch was, I would use it on my computer to have the same fast code::blocks as on windows :( 2011-04-04T11:13:40 my father uses kubuntu. kde has some amazing tools, like digikam. on a gnome desktop the tools are mostly minimalistic, but digikam offers photo enhancement tools, you would find in a commercial app. k3b is also a nice dvd-burning application 2011-04-04T11:14:49 the music player amarok (version 1) is also one of the best. 2011-04-04T11:15:39 i generally just use my computer for programming, interwebs and movies 2011-04-04T11:15:40 all my uni work etc. is done on paper most of the time 2011-04-04T11:15:48 i have an oldschool radio next to me :P 2011-04-04T11:16:02 well, alarm clock/radio 2011-04-04T11:18:17 you just inspired me to update the time on that, it's been wrong for about 2 months 2011-04-04T11:18:58 *** Gorgoroth has joined #aichallenge 2011-04-04T11:19:25 *** Chris_0076 has joined #aichallenge 2011-04-04T11:19:47 Frontier1: so you use wxWidgets for guis in c++? 2011-04-04T11:19:56 i was told qt is superior to everything else? 2011-04-04T11:20:04 eventually i want to make a gui for my game theory library 2011-04-04T11:20:17 right QT 4 >> WxWidgets 2011-04-04T11:22:19 ergh, brb, my internet is on fail mode 2011-04-04T11:23:23 *** Gorgoroth has left #aichallenge 2011-04-04T11:30:17 *** antimatroid has quit IRC (Ping timeout: 240 seconds) 2011-04-04T11:32:33 *** stocha has quit IRC (Quit: Page closed) 2011-04-04T11:34:58 *** boegel has quit IRC (Read error: Operation timed out) 2011-04-04T11:40:40 *** antimatroid has joined #aichallenge 2011-04-04T11:46:05 *** chris__0076 has joined #aichallenge 2011-04-04T11:47:50 *** Frontier has joined #aichallenge 2011-04-04T11:47:53 *** Chris_0076 has quit IRC (Ping timeout: 246 seconds) 2011-04-04T11:48:31 *** Frontier1 has quit IRC (Ping timeout: 276 seconds) 2011-04-04T11:48:58 antimatroid: I don't write any gui applications in c++. It is just code::blocks that uses it. 2011-04-04T11:50:52 Well there was some OpenGL stuff, but that doesn't count. It was just a basic Window without any buttons and labels, so I used the xorg windowing functions directly. 2011-04-04T11:54:56 *** antimatroid has quit IRC (Ping timeout: 246 seconds) 2011-04-04T11:54:57 *** Zepp_ has joined #aichallenge 2011-04-04T11:55:02 *** chris__0076 has quit IRC (Ping timeout: 246 seconds) 2011-04-04T11:55:03 *** Zepp_ has quit IRC (Client Quit) 2011-04-04T11:55:30 *** Zepp_ has joined #aichallenge 2011-04-04T11:58:22 *** Zepp_ has joined #aichallenge 2011-04-04T11:59:35 *** Zepp_ has joined #aichallenge 2011-04-04T12:07:58 *** Chris_0076 has joined #aichallenge 2011-04-04T12:10:26 *** antimatroid has joined #aichallenge 2011-04-04T12:10:47 *** boegel has joined #aichallenge 2011-04-04T12:15:57 *** boegel has quit IRC (Ping timeout: 264 seconds) 2011-04-04T12:23:51 *** Chris_0076 has quit IRC (Ping timeout: 246 seconds) 2011-04-04T12:24:45 *** delt0r___ has quit IRC (Ping timeout: 250 seconds) 2011-04-04T12:26:25 Stocha is gone, i had an idea for breeding salesman 2011-04-04T12:26:29 McLeopold: you interested? 2011-04-04T12:27:25 when you move an ant towards a food square, you could also "place an ant" at that food square, indicating when it will be there, then when you scan from another food square, if you hit that, you add how many turns you can use the ant in and consider that ant again 2011-04-04T12:28:17 *** greghaynes has joined #aichallenge 2011-04-04T12:33:21 *** Eruonen has joined #aichallenge 2011-04-04T12:34:53 *** Chris_0076 has joined #aichallenge 2011-04-04T12:58:00 *** p4p4p5 has quit IRC (Read error: Connection timed out) 2011-04-04T13:00:19 *** p4p4p5 has joined #aichallenge 2011-04-04T13:16:40 *** Naktibalda has quit IRC (Quit: ChatZilla 0.9.86.1 [Firefox 3.6.16/20110319135224]) 2011-04-04T13:17:46 *** delt0r has joined #aichallenge 2011-04-04T13:33:20 *** Kingpin13 has joined #aichallenge 2011-04-04T13:53:20 *** boegel has joined #aichallenge 2011-04-04T14:01:02 yeehaw I overloaded Firefox 3's JavaScript garbage collector 2011-04-04T14:02:48 I really have to get a lot of memory allocation out of the drawing routine 2011-04-04T14:03:16 that drop from 27 to 1 fps is not tolerable ^^ 2011-04-04T14:06:07 *** stride_ is now known as stride 2011-04-04T14:14:04 *** Keth has joined #aichallenge 2011-04-04T14:14:04 *** __jmcarthur is now known as jmcarthur 2011-04-04T14:18:04 antimatroid, thats a great idea 2011-04-04T14:18:24 Anyone would like to collaborate and share their bot so we could run tests, let them fight and improve them? I know there is HunterBot but my AI has 99.99% win ratio against it so there isn't much to improve. And fighting with myself seems kind of pointless. Anyone interested just message me. 2011-04-04T14:18:46 Keth: i don't know anyone else with a bot sorry 2011-04-04T14:18:53 i guess there's probably quite a few of you now :\ 2011-04-04T14:19:07 Yeah I understand. 2011-04-04T14:19:13 But still, I'm sure there is somebody who did something. 2011-04-04T14:19:16 Keth: want are you doing? have you seen the strategy idea pages i started making? 2011-04-04T14:19:22 i'm welcome to ideas to add to it 2011-04-04T14:19:35 I'm not really looking for a final version from anyone 2011-04-04T14:19:58 yeah, i have only written the c++ starter package 2011-04-04T14:20:03 i've been busy doing other stuff 2011-04-04T14:20:40 but not exactly accurate, In the future you are going to have 2 ants there, not 1 2011-04-04T14:20:41 Oh nice, I'm actually using that c++ starter package 2011-04-04T14:20:49 and I know I should contribute but I'm not sure if I would like to share my AI strategy 2011-04-04T14:21:05 phire: true, place 2 new ones, with the information 2011-04-04T14:21:30 Keth: ahh, i just updated it recently to include vision calculation for you and changed the structure a bit :P 2011-04-04T14:21:53 is there anything you particularly hate or feel is missing from it? 2011-04-04T14:22:13 you're the first person i've come across using it 2011-04-04T14:23:29 although line 22 of bot.cc should not be commented for what's uploaded 2011-04-04T14:23:33 i need to fix that :\ 2011-04-04T14:23:58 jesus christ, stop workin on your bots. we will never catch up in two months. 2011-04-04T14:24:51 Frontier: it makes it less bad that we're competing now 2011-04-04T14:24:58 Then again, after all the time I put into the visualizer it might well be that I can't see any more ants when the contest starts. 2011-04-04T14:25:31 if i do gsoc i'll probably have to give myself an early end date anyway 2011-04-04T14:26:01 antimatroid: I rewrote quite a bit, wrote my own classes Map, Ant, Debug and so on. 2011-04-04T14:26:01 antimatroid: I'm quite fine :3 2011-04-04T14:26:02 *** Naktibalda has joined #aichallenge 2011-04-04T14:26:02 Frontier: It's part of contributing 2011-04-04T14:26:03 *** Andy__ has joined #aichallenge 2011-04-04T14:26:37 yeah, the starter packages are aimed mostly towards people who can't/wont end up rewriting it themselves 2011-04-04T14:26:38 *** Keth has quit IRC (Read error: Connection reset by peer) 2011-04-04T14:27:30 *** Keth has joined #aichallenge 2011-04-04T14:27:48 Keth: what was wrong with the bug struct i gave you? 2011-04-04T14:27:50 antimatroid: were any new algorithms added to the starter bots, like pathfinding? 2011-04-04T14:28:16 i expect you to restructure a lot of other stuff, it's mostly aimed at people who don't want to restructure, mine will be different, but the bug struct is awesome 2011-04-04T14:29:00 Frontier: i added vision to run behind the scenes, and i'm updating ant locations so that it's easier to "avoid collisions", but i'm still doing to naive strategy and only not walking onto water 2011-04-04T14:29:37 Frontier: once it reads the state each turn, it will update the squares "lastSeen" variable to the current turn, and i added Square and Location structs 2011-04-04T14:29:39 antimatroid: What's wrong with yours is that you're not able to recognize invidual ant and have it's own pointer data. The ant ID's change every turn so you need to have your own Ant class to keep track of the correct IDs. 2011-04-04T14:30:05 Keth: yeah that's fine, but you said you wrote your own Debug class? what was wrong with the Bug one? 2011-04-04T14:30:31 i don't care to differentiate which ant is which of mine 2011-04-04T14:30:38 yea I understand 2011-04-04T14:30:51 ... so why did you write a Debug class? :P 2011-04-04T14:31:36 https://github.com/aichallenge/aichallenge/blob/epsilon/ants/dist/starter_bots/cpp/Bug.h 2011-04-04T14:31:41 i really like that 2011-04-04T14:32:31 antimatroid: If I find the time I may look at your code and update my starter. 2011-04-04T14:32:56 for the debug, I just added more stuff for my own purposes, outputing json, logError(), logWarning() so then when I run 1000 tests I can see if there are any warnings, it sums up at the end of the game. 2011-04-04T14:33:26 ah okay, i thought you redid the whole passing debug to a file again 2011-04-04T14:33:33 No no, passing is not changed 2011-04-04T14:33:37 and silly functions to help me out debugging the AI: drawLine(), drawText() 2011-04-04T14:33:44 they go straight to vizualizer 2011-04-04T14:33:45 have you got a game tree yet? 2011-04-04T14:34:07 game tree? 2011-04-04T14:35:04 i'm planning to use a game tree for ants that are within a certain range of enemies for battles 2011-04-04T14:35:41 I think the highest priority and what is most important is being able to keep track of invidual ants 2011-04-04T14:35:44 I don't know how can you write AI without that 2011-04-04T14:35:47 basically "avoid battles if you will not get at least a 1:1 death/kill ratio" 2011-04-04T14:35:59 i don't see the need for tracking individual ants 2011-04-04T14:36:05 You really don't? 2011-04-04T14:36:42 Tell me how would you set an ant state to "wait" - just an example 2011-04-04T14:37:34 my basic strategy intention is something like follows "assign each food square an ant and move the ant towards it, use a game tree for deciding moves of any ants near enemies and use heuristics to spread the rest out and move them towards battles" 2011-04-04T14:37:57 your AI is pretty much grid based, not structural based - that's why you don't see a purpose of it. 2011-04-04T14:37:57 or unexplored land, not recently seen land etc. 2011-04-04T14:38:57 :P possibly 2011-04-04T14:39:11 I keep my invidual ants in seperate class, I have much more freedom with it. Ant* ant = getAntFrom(Location(5,5)); ant->setState(STATE_WAIT); 2011-04-04T14:39:24 *** dwchandl1r has left #aichallenge 2011-04-04T14:39:27 just an example of what you can't really 2011-04-04T14:39:45 i could have a list of locations that are to wait :P 2011-04-04T14:39:55 And you would keep 10 arrays for 10 variables? 2011-04-04T14:40:03 i'm considering using sets rather than grids for my actual bot 2011-04-04T14:40:24 yeah but 10 variables like STATE, isAggressive, hasFoundFood 2011-04-04T14:40:31 you would need like 10 arrays 2011-04-04T14:40:41 which is really a bad structure 2011-04-04T14:40:51 It really depends on your AI 2011-04-04T14:40:54 yeah, i don't keep information like that across states 2011-04-04T14:41:01 yeah, if you don't then it's fine 2011-04-04T14:41:07 but it defaintely gives you more freedom 2011-04-04T14:41:20 i had previously considered having an ant class, but i don't need that much information 2011-04-04T14:41:43 like I said, depends on strategy 2011-04-04T14:41:47 yep 2011-04-04T14:42:05 had you looked at vision yet? 2011-04-04T14:42:22 the updated bot has code for that, be careful trying to make that faster (i've made that clear with the comments) 2011-04-04T14:42:50 yeah, I'm not using vision at all for my strategy. I only calculate area_covered in percent and visibility_covered 2011-04-04T14:43:18 i use visibility information when searching the map 2011-04-04T14:43:36 if i have to send an ant somewhere, send it towards an unseen/not recently seen square 2011-04-04T14:43:51 I changed: 2011-04-04T14:43:57 grid = vector >(rows, vector(cols, '.')); to grid = vector >(rows, vector(cols, '?')); 2011-04-04T14:44:10 and then keep track of which are visible 2011-04-04T14:44:20 yeah, i changed that to a 2d vector of Squares now 2011-04-04T14:44:24 so then you just do grid[1][1] = "?" or isSquareVisible() 2011-04-04T14:44:29 but you don't need seperate grid 2011-04-04T14:44:39 which I guess you keep two seperate arrays 2011-04-04T14:45:50 and each square has a lastSeen variables, starting at 0 (when you output the grid, it will output ? if lastSeen is 0, ~ if lastSeen < turn), then after reading input, it'll do a breadth first search from each ant marking each visible square as lastSeen to the current turn 2011-04-04T14:46:38 It's not bad that way but I'm kind of used to writing fast code and efficient memory use 2011-04-04T14:46:51 yeah fair enough 2011-04-04T14:47:16 i'm trying to make it as usable for beginners as possible, i myself don't plan to use exactly what i provide 2011-04-04T14:47:33 As far as I can tell, fast code might actually be important in this challange 2011-04-04T14:47:45 phire: yep 2011-04-04T14:47:48 It just feels like your structure is kinda limited, you can't save variables for ants, you use many arrays. It might work for now but it might impact later on 2011-04-04T14:48:30 i agree, it's hard to know exactly how to make the starter bot though 2011-04-04T14:48:45 i might be just too familiar with physics engines and the way they are structured 2011-04-04T14:48:51 since I even added velocity for my ant 2011-04-04T14:48:52 lol 2011-04-04T14:48:53 beginners are least likely to be able to make any real changes for themselves, so they are my highest priority 2011-04-04T14:49:14 lol, i can't see how the velocity matters? surely it's always the same? 2011-04-04T14:49:19 naw 2011-04-04T14:49:20 well, for a moving ant 2011-04-04T14:49:23 my ants have physical position 2011-04-04T14:49:28 and velocity 2011-04-04T14:49:36 and they steer to that position 2011-04-04T14:49:54 interesting 2011-04-04T14:49:57 It's part mf oy strategy 2011-04-04T14:49:59 of my* 2011-04-04T14:50:05 i look forward to seeing your bot once the servers are up 2011-04-04T14:50:08 yeah 2011-04-04T14:50:24 I really want to test with someone who has already done a bot 2011-04-04T14:50:28 http://gyazo.com/6212b068ef0c00ab3b8e782894e2afbb 2011-04-04T14:50:32 W = stands for win 2011-04-04T14:50:35 against hunterbot 2011-04-04T14:50:54 *** Cyndre_ has joined #aichallenge 2011-04-04T14:51:01 what maps have you been using? 2011-04-04T14:51:09 I'm not sure what else to improve, I tested every random map, every attack type 2011-04-04T14:51:28 so I'm just looking for smarter bots 2011-04-04T14:51:29 did you do the random symmetric maps? 2011-04-04T14:51:34 yeah 2011-04-04T14:51:37 both symetric and non-symteric 2011-04-04T14:51:38 tried both 2011-04-04T14:51:53 *** Cyndre has quit IRC (Ping timeout: 240 seconds) 2011-04-04T14:51:56 what do you think of the maps? (I made them too :P) 2011-04-04T14:52:02 they're quite nice 2011-04-04T14:52:06 last year you could get away with just good planning 2011-04-04T14:52:07 I have a bot 2011-04-04T14:52:07 its just not any good yet 2011-04-04T14:52:07 It spends 60% of the game just sitting there doing nothing, and it still practically draws with hunter bot. 2011-04-04T14:52:13 what's your experience with the asymmetric ones? 2011-04-04T14:52:18 are the fair at all? 2011-04-04T14:52:27 phire: I see, I have been through it =D 2011-04-04T14:52:34 antimatroid: I would recommend to keep them 1v1 2011-04-04T14:52:43 antimatroid: I really can't imagine ants will be more than 2 players for the final matchups 2011-04-04T14:52:53 antimatroid: I modify your map to do 1v1 testing 2011-04-04T14:52:55 1v1 is a different game 2011-04-04T14:53:18 antimatroid: 1v1v1v1 is much more about luck 2011-04-04T14:53:28 not in the long run 2011-04-04T14:53:49 antimatroid: in a long run less but trust me, still lots of luck. 2011-04-04T14:54:07 antimatroid: I hope you're reading the forums 2011-04-04T14:54:14 i am 2011-04-04T14:55:29 the problem is n player is a different game and lots of people expressed wanting to do that 2011-04-04T14:55:33 basic question, if spawnradius is 1, does it access vertical and horizontal neighbours only or diagonals too? 2011-04-04T14:55:46 i was originally opposed, but this game has lots of cool cooperation aspects, so i would be quite disappointed to move back to 1v1 noq 2011-04-04T14:55:48 now* 2011-04-04T14:55:57 you should know about how 1v1v1v1 is being imbalanced compared to 1v1, even in a long run 2011-04-04T14:56:01 Naktibalda: diagonal is sqrt(2) away 2011-04-04T14:56:10 you have lots of info about that on forums, I give 95% percent you will never see 1v1v1v1 2011-04-04T14:56:54 * Naktibalda should think about map like a go board, not a chess board 2011-04-04T14:57:05 Haha 2011-04-04T14:57:25 textual presentation doesn't help to do that 2011-04-04T14:58:16 exactly, but you're asking for textual answer 2011-04-04T14:58:41 Keth: what about 1v1v1? 2011-04-04T14:58:48 antimatroid: It's not about the balance only 2011-04-04T14:58:52 we can do full symmetry including relative starting positions then 2011-04-04T14:58:52 antimatroid: It's a lot about TrueSkill 2011-04-04T14:59:36 antimatroid: relative starting positions has been already tried but it did not work out. 2011-04-04T15:00:24 antimatroid: trust me, it's not going to change to relative positions 2011-04-04T15:00:38 ? 2011-04-04T15:00:51 Which part? 2011-04-04T15:01:01 change to relative positions 2011-04-04T15:01:06 Well, it's not going to happen. 2011-04-04T15:01:10 I can guarantee you that 2011-04-04T15:01:11 oh, i wasn't referring to that 2011-04-04T15:01:30 i meant if you have more than 4 players, you can't have every player with "equal" access to every other player on the map 2011-04-04T15:01:38 but you can with 3 2011-04-04T15:01:54 more than or equal to 4 2011-04-04T15:02:00 More than 2 players, won't work. 2011-04-04T15:02:09 you can do it with 3 and wrapped edges 2011-04-04T15:02:31 Lots of issues with TrueSkill, coding it, beginners having difficulty understanding, very easy to figure out the map after few steps anyway 2011-04-04T15:02:45 I guarantee you it will be 2 players 2011-04-04T15:02:52 a beginner wouldn't understand bayeslo any more than they would trueskill 2011-04-04T15:02:58 (just my opinion) 2011-04-04T15:03:23 we'll see, there isn't much to debate 2011-04-04T15:04:07 *** p4p4p5 has quit IRC (Read error: Connection timed out) 2011-04-04T15:04:07 ranking algorithm doesn't matter for participant, unless it's unfair :) 2011-04-04T15:05:55 where are you storing the files of your bot? 2011-04-04T15:08:15 *** p4p4p5 has joined #aichallenge 2011-04-04T15:16:05 *** sixthgea1 is now known as sixthgear 2011-04-04T15:16:05 *** amstan has joined #aichallenge 2011-04-04T15:16:05 *** ChanServ sets mode: +o amstan 2011-04-04T15:18:27 *** Accoun has quit IRC () 2011-04-04T15:20:26 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has joined #aichallenge 2011-04-04T15:25:15 *** Zepp_ has joined #aichallenge 2011-04-04T15:27:43 *** Zepp has quit IRC (Quit: leaving) 2011-04-04T15:27:50 *** Zepp has joined #aichallenge 2011-04-04T15:27:55 *** Zepp_ has quit IRC (Client Quit) 2011-04-04T15:36:00 aichallenge: Andreas Lundgren epsilon * r78c332e / (3 files): 2011-04-04T15:36:00 aichallenge: An attempt to build a map generator for Ants. It's possible to make maps for 2, 4 or 8 players and the result will be a symmetric map. 2011-04-04T15:36:00 aichallenge: It works by creating a part and then copying it to generate the full map. This 2011-04-04T15:36:00 aichallenge: pastebin (pastebin.com/YLYPYvBd) displays how the part looks for a 8 player map. 2011-04-04T15:36:00 aichallenge: This generator is far from perfect but could perhaps still be helpful. - 2011-04-04T15:36:00 aichallenge: http://bit.ly/gY6Oxm 2011-04-04T15:36:04 aichallenge: Alexandru Stan epsilon * r44e58eb / (4 files): Merge remote branch 'lundgren/epsilon' into epsilon - http://bit.ly/fiL7KQ 2011-04-04T15:36:41 contestbot: later tell antimatroid see https://github.com/aichallenge/aichallenge/commit/78c332e1ee99e19867188894a0a6aefe2daef69a 2011-04-04T15:36:41 amstan: Aye, aye, sir 2011-04-04T15:36:49 i'm here 2011-04-04T15:38:34 antimatroid: that's what i wanted... make sure we have such kinds of maps in the final game 2011-04-04T15:38:43 *** rebelxt has joined #aichallenge 2011-04-04T15:39:40 they're as fair as the other ones 2011-04-04T15:40:19 yeah 2011-04-04T15:41:29 ditch the border and that map would be really cool 2011-04-04T15:41:37 probably better as a 4 player one 2011-04-04T15:44:45 *** Frontier has quit IRC (Quit: Leaving.) 2011-04-04T15:45:26 *** Frontier has joined #aichallenge 2011-04-04T15:46:10 *** dr- has joined #aichallenge 2011-04-04T15:46:11 *** Accoun has joined #aichallenge 2011-04-04T15:46:11 *** dr-` has quit IRC (Ping timeout: 258 seconds) 2011-04-04T15:48:57 aichallenge: Marco Leise epsilon * rd21e40b / (5 files in 3 dirs): visualizer: map can be scrolled / fixed two threading issues in the standalone player - http://bit.ly/g8TR0j 2011-04-04T15:49:42 now look at this awesome map scrolling: http://marco-leise.homedns.org/antgui/work/visualizer.php?game_id=2 2011-04-04T15:53:40 Frontier: if i drag it too much it starts selecting html elements 2011-04-04T15:53:40 but yes.. it is awesome 2011-04-04T15:55:02 amstan: any better idea than dragging with the left mouse button? 2011-04-04T15:55:23 Frontier: i guess arrows already are used 2011-04-04T15:58:10 *** rebelxt has left #aichallenge 2011-04-04T16:09:29 *** chris__0076 has joined #aichallenge 2011-04-04T16:12:25 *** Chris_0076 has quit IRC (Ping timeout: 246 seconds) 2011-04-04T16:16:11 aichallenge: Marco Leise epsilon * rafea6d1 / ants/visualizer/js/Application.js : (small change) - http://bit.ly/g7A1Wh 2011-04-04T16:17:01 Frontier: <3 :D 2011-04-04T16:17:29 Frontier: is zooming still too hard? :P 2011-04-04T16:18:00 why do you always want to zoom? is your screen too small? :p 2011-04-04T16:18:31 my screen is large 2011-04-04T16:18:44 i'm worried about having to restrict the max dimensions so that people can reasonably watch it 2011-04-04T16:19:03 if people could zoom, we can sort of ignore that more 2011-04-04T16:19:08 well, that is a different story 2011-04-04T16:19:16 *** chris__0076 has quit IRC (Ping timeout: 246 seconds) 2011-04-04T16:19:17 Frontier: nah.. it's ok 2011-04-04T16:19:17 i think you can do browser zoom 2011-04-04T16:19:17 Frontier: actually.. i'm getting weird behavoir with zooming 2011-04-04T16:19:19 Frontier: everything gets smaller 2011-04-04T16:19:26 instead of bigger 2011-04-04T16:19:36 well even 200x200 would be too hard to really watch now yeah? 2011-04-04T16:19:56 then again, i do feel that's too large to play a reasonable game on that doesn't take too long 2011-04-04T16:20:11 I don't know if it is good to have a game that doesn't fit on screen. 2011-04-04T16:20:46 provided you could sort of make out what's going on globally with it placed on 1 screen, i don't think it'd matter too much 2011-04-04T16:21:07 you could still zoom in and look at specific stuff more closely 2011-04-04T16:21:56 hehe i can make it smalled with ctrl scroll 2011-04-04T16:22:39 amstan: yeah i get that :\ 2011-04-04T16:22:43 it's a bit weird hey 2011-04-04T16:22:54 but not really an issue 2011-04-04T16:22:55 browser zoom works differently among browsers 2011-04-04T16:23:04 opera does what you expect, amstan 2011-04-04T16:23:47 Frontier: chrome and firefox don't 2011-04-04T16:23:55 *** boegel has quit IRC (Quit: Leaving) 2011-04-04T16:24:05 I'm simply using the whole browser window height if not told otherwise 2011-04-04T16:25:28 it would also allow LARGE crazy maps if we had zooming :P 2011-04-04T16:25:56 like mega 16 player games on 1000x1000 maps :P 2011-04-04T16:26:00 would be cool to watch the result 2011-04-04T16:26:13 maybe not quite 1000x1000 2011-04-04T16:26:51 and for the record: Safari behaves like Firefox and Chrome 2011-04-04T16:27:10 antimatroid: -.- 2011-04-04T16:27:19 Frontier: what vision radius are you using? 2011-04-04T16:27:33 don't know... let me look at the file 2011-04-04T16:27:55 sqrt(96) 2011-04-04T16:28:21 my vision function shows a different type of vision circle in debug files 2011-04-04T16:28:25 *** p4p4p5 has quit IRC (Read error: Connection timed out) 2011-04-04T16:28:29 i have 4 extra squares i think 2011-04-04T16:28:41 i think mines wrong somehow :\ 2011-04-04T16:29:04 antimatroid: of course my code is correct -.- 2011-04-04T16:29:17 *** p4p4p4 has joined #aichallenge 2011-04-04T16:29:18 i'm pretty sure it was chosen to be like yours and not mine 2011-04-04T16:29:41 do you have extra squares at the outer bounds (n, s, e, w)? 2011-04-04T16:29:49 i had that in one version 2011-04-04T16:30:10 i do 2011-04-04T16:31:43 i think i was using a ceil function to calculate the bounding box and used that value for the visibility calculation, but it was wrong. it has to be floor. 2011-04-04T16:33:08 fixed, i was using ints for viewradius etc :P 2011-04-04T16:33:17 thanks for that comment :P 2011-04-04T16:33:40 Keth: change viewradius etc. in state to doubles 2011-04-04T16:34:40 there is nothing wrong with ints, as long as you keep every distance squared 2011-04-04T16:35:28 it may be faster than floating point arithmetic 2011-04-04T16:35:35 yes, but it's the starter package, i want it to be easy to understand for beginners 2011-04-04T16:35:40 i'm going to square the metric for my own bot 2011-04-04T16:36:36 kk, next on my list is version for color blind people 2011-04-04T16:36:59 I plan to draw letters on the ants, so keep the map size small antimatroid :p 2011-04-04T16:37:26 Frontier: i think 120x120 is the best size 2011-04-04T16:38:28 that's the lcm of like (2,3,4,5,6,8), so that people can't guess how many players might be on the game if we chose them strictly from that set 2011-04-04T16:38:48 ah that is smart 2011-04-04T16:39:47 Do you have a .replay of a game on such a map? 2011-04-04T16:39:59 nope sorry, i could make you a map of that size? 2011-04-04T16:40:09 actually, i could make you the replay :P 2011-04-04T16:40:12 give me 5 2011-04-04T16:40:16 5 2011-04-04T16:40:24 minutes 2011-04-04T16:41:00 I'm always lost with the command line options for the engine ^^ 2011-04-04T16:42:04 i just use play_one_game.sh 2011-04-04T16:42:08 and change stuff as needed 2011-04-04T16:43:08 wow there was a massive access to my web server. must have been someone with an empty browser cache or IE (loading the applet) 2011-04-04T16:46:08 it's playing now 2011-04-04T16:46:15 O:-) 2011-04-04T16:47:50 Frontier: woah.. round hat! 2011-04-04T16:49:02 so it doesn't express my patience? 2011-04-04T16:49:05 Frontier: http://pastebin.com/hgycZ0Fd 2011-04-04T16:49:08 is that right? 2011-04-04T16:49:10 are the 4 bots antimatroid, McLeopold, amstan, Frontier in python? 2011-04-04T16:49:49 ok let's try game_id 3 2011-04-04T16:50:08 jesus christ, an error 2011-04-04T16:50:38 * amstan gets in the car 2011-04-04T16:50:42 :( 2011-04-04T16:50:50 : () is a duck aha 2011-04-04T16:52:37 Keth: that's just lefty both and hunterbot i think 2011-04-04T16:52:57 Frontier: that game plays with the viewer thing mcleopold made 2011-04-04T16:53:52 ok i had a regression in my code 2011-04-04T16:55:41 aichallenge: Marco Leise epsilon * rfaab9e2 / ants/visualizer/js/Replay.js : visualizer: fixed regression in replay parser - http://bit.ly/fPs38h 2011-04-04T16:55:42 antimatroid: ah I see, they don't act like one 2011-04-04T16:56:05 Frontier: i want to see the game in your visualiser already :P 2011-04-04T16:56:09 Why does my playgame.py doesn't export replay in json? I do --output_json but still same thing 2011-04-04T16:56:14 and the vizualizer cannot parse non-json 2011-04-04T16:56:16 game_id 3, http://marco-leise.homedns.org/antgui/work/visualizer.php?game_id=3 2011-04-04T16:56:42 I get same error in game_id3 2011-04-04T16:56:53 it works now 2011-04-04T16:57:17 ergh, my "net" semi died again, brb 2011-04-04T16:57:26 Frontier: How do you get json replay? D: 2011-04-04T16:57:58 I don't know. Ask antimatroid 2011-04-04T16:58:03 or sigh 2011-04-04T16:58:14 But you did the replay 2011-04-04T16:58:18 so what's the command line you usejd? 2011-04-04T16:58:29 playgame.py --something 2011-04-04T16:58:36 I'm just visualizing files someone gives me 2011-04-04T16:58:46 Oh right 2011-04-04T16:58:48 so you haven't ran the games yea? 2011-04-04T16:58:52 look for the .replay files 2011-04-04T16:59:07 I did, but the .replay files I create from my bot are not the same 2011-04-04T16:59:08 they should work in the visualizer 2011-04-04T16:59:21 my .replay files aren't json 2011-04-04T17:00:38 they *should* still work, but probably your engine is out of date? 2011-04-04T17:01:02 yeah I thought about engine being out of date 2011-04-04T17:01:08 and no, old one's not work 2011-04-04T17:01:55 antimatroid: You should try to load the replay you sent me in Opera 2011-04-04T17:02:31 *** antimatroid has quit IRC (Ping timeout: 276 seconds) 2011-04-04T17:06:29 with this map size the performance becomes an issue again :-| 2011-04-04T17:06:29 Keth: try to delete any .pyc files that were generated 2011-04-04T17:07:38 yeah that I know 2011-04-04T17:07:56 *** Cyndre_ has quit IRC (*.net *.split) 2011-04-04T17:07:58 *** Zannick has quit IRC (*.net *.split) 2011-04-04T17:07:59 *** pgpaskar1 has quit IRC (*.net *.split) 2011-04-04T17:07:59 *** computerwiz_222 has quit IRC (*.net *.split) 2011-04-04T17:07:59 *** CIA-90 has quit IRC (*.net *.split) 2011-04-04T17:08:00 *** spacebat has quit IRC (*.net *.split) 2011-04-04T17:08:01 *** amstan has quit IRC (*.net *.split) 2011-04-04T17:08:01 *** Kingpin13 has quit IRC (*.net *.split) 2011-04-04T17:08:02 *** sixthgear has quit IRC (*.net *.split) 2011-04-04T17:08:03 *** Heulsay has quit IRC (*.net *.split) 2011-04-04T17:08:04 *** JamesMG has quit IRC (*.net *.split) 2011-04-04T17:08:05 *** javagamer1 has quit IRC (*.net *.split) 2011-04-04T17:08:06 *** McLeopold has quit IRC (*.net *.split) 2011-04-04T17:08:07 *** iFire` has quit IRC (*.net *.split) 2011-04-04T17:08:08 *** shutch has quit IRC (*.net *.split) 2011-04-04T17:08:08 *** Cerealklr has quit IRC (*.net *.split) 2011-04-04T17:08:08 *** cmoylan has quit IRC (*.net *.split) 2011-04-04T17:08:09 *** Ttech has quit IRC (*.net *.split) 2011-04-04T17:08:09 *** jbroman has quit IRC (*.net *.split) 2011-04-04T17:08:10 *** janzert has quit IRC (*.net *.split) 2011-04-04T17:08:11 *** acieroid has quit IRC (*.net *.split) 2011-04-04T17:08:13 *** AlliedEnvy_ has quit IRC (*.net *.split) 2011-04-04T17:08:14 *** smellyhippy has quit IRC (*.net *.split) 2011-04-04T17:08:16 *** mcsmash has quit IRC (*.net *.split) 2011-04-04T17:08:17 *** jmcarthur has quit IRC (*.net *.split) 2011-04-04T17:08:17 *** kaemo has quit IRC (*.net *.split) 2011-04-04T17:08:18 *** _aj_ <_aj_!~aj@66.133.136.43> has quit IRC (*.net *.split) 2011-04-04T17:08:23 *** BtbN_ has quit IRC (*.net *.split) 2011-04-04T17:08:24 *** Andy__ has quit IRC (*.net *.split) 2011-04-04T17:08:24 *** Accoun has quit IRC (*.net *.split) 2011-04-04T17:08:25 *** Zepp has quit IRC (*.net *.split) 2011-04-04T17:08:25 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has quit IRC (*.net *.split) 2011-04-04T17:08:25 *** Eruonen has quit IRC (*.net *.split) 2011-04-04T17:08:25 *** greghaynes has quit IRC (*.net *.split) 2011-04-04T17:08:25 *** mceier has quit IRC (*.net *.split) 2011-04-04T17:08:26 *** rstrong has quit IRC (*.net *.split) 2011-04-04T17:08:27 *** sente has quit IRC (*.net *.split) 2011-04-04T17:08:27 *** p4p4p4 has quit IRC (*.net *.split) 2011-04-04T17:08:27 *** dr- has quit IRC (*.net *.split) 2011-04-04T17:08:27 *** Naktibalda has quit IRC (*.net *.split) 2011-04-04T17:08:28 *** delt0r has quit IRC (*.net *.split) 2011-04-04T17:08:28 *** znutar_ has quit IRC (*.net *.split) 2011-04-04T17:08:29 *** ChanServ has quit IRC (*.net *.split) 2011-04-04T17:08:44 *** znutar has joined #aichallenge 2011-04-04T17:09:11 *** Accoun has joined #aichallenge 2011-04-04T17:09:11 *** dr- has joined #aichallenge 2011-04-04T17:09:11 *** Zepp has joined #aichallenge 2011-04-04T17:09:11 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has joined #aichallenge 2011-04-04T17:09:11 *** amstan has joined #aichallenge 2011-04-04T17:09:11 *** Cyndre_ has joined #aichallenge 2011-04-04T17:09:11 *** Andy__ has joined #aichallenge 2011-04-04T17:09:11 *** Naktibalda has joined #aichallenge 2011-04-04T17:09:11 *** Kingpin13 has joined #aichallenge 2011-04-04T17:09:11 *** delt0r has joined #aichallenge 2011-04-04T17:09:11 *** Eruonen has joined #aichallenge 2011-04-04T17:09:11 *** greghaynes has joined #aichallenge 2011-04-04T17:09:11 *** znutar_ has joined #aichallenge 2011-04-04T17:09:11 *** iFire` has joined #aichallenge 2011-04-04T17:09:11 *** mceier has joined #aichallenge 2011-04-04T17:09:11 *** AlliedEnvy_ has joined #aichallenge 2011-04-04T17:09:11 *** javagamer1 has joined #aichallenge 2011-04-04T17:09:11 *** jmcarthur has joined #aichallenge 2011-04-04T17:09:11 *** rstrong has joined #aichallenge 2011-04-04T17:09:11 *** kaemo has joined #aichallenge 2011-04-04T17:09:11 *** JamesMG has joined #aichallenge 2011-04-04T17:09:11 *** Heulsay has joined #aichallenge 2011-04-04T17:09:11 *** sixthgear has joined #aichallenge 2011-04-04T17:09:11 *** BtbN_ has joined #aichallenge 2011-04-04T17:09:11 *** _aj_ <_aj_!~aj@66.133.136.43> has joined #aichallenge 2011-04-04T17:09:11 *** shutch has joined #aichallenge 2011-04-04T17:09:11 *** Cerealklr has joined #aichallenge 2011-04-04T17:09:11 *** McLeopold has joined #aichallenge 2011-04-04T17:09:11 *** cmoylan has joined #aichallenge 2011-04-04T17:09:11 *** smellyhippy has joined #aichallenge 2011-04-04T17:09:11 *** Zannick has joined #aichallenge 2011-04-04T17:09:11 *** sente has joined #aichallenge 2011-04-04T17:09:11 *** Ttech has joined #aichallenge 2011-04-04T17:09:11 *** jbroman has joined #aichallenge 2011-04-04T17:09:11 *** pgpaskar1 has joined #aichallenge 2011-04-04T17:09:11 *** computerwiz_222 has joined #aichallenge 2011-04-04T17:09:11 *** janzert has joined #aichallenge 2011-04-04T17:09:11 *** mcsmash has joined #aichallenge 2011-04-04T17:09:11 *** spacebat has joined #aichallenge 2011-04-04T17:09:11 *** CIA-90 has joined #aichallenge 2011-04-04T17:09:11 *** acieroid has joined #aichallenge 2011-04-04T17:09:11 *** ChanServ has joined #aichallenge 2011-04-04T17:09:11 *** kornbluth.freenode.net sets mode: +oo amstan ChanServ 2011-04-04T17:09:11 *** znutar has quit IRC (Excess Flood) 2011-04-04T17:09:37 *** znutar_ has quit IRC (Excess Flood) 2011-04-04T17:09:37 *** znutar has joined #aichallenge 2011-04-04T17:09:57 Is the visualizer up to date then? 2011-04-04T17:12:38 git pull, cd ants/visualizer, ant jar -> gives you a visualizer.jar that you can run with the .replay as the command-line option. But you can use the html version of course with visualize_locally.py. 2011-04-04T17:13:42 there is a new file 'copy_paste.html' now as well where you can paste replay strings. But you said the engine doesn't emit json files, so the problem must be somewhere else. 2011-04-04T17:17:27 I updated to newest everything from today 2011-04-04T17:17:31 the replay file is still not json 2011-04-04T17:17:36 (i did delete *.pyc files) 2011-04-04T17:18:06 the replay antimatroid just posted a little while ago wasn't json either was it? 2011-04-04T17:18:19 *** antimatroid has joined #aichallenge 2011-04-04T17:18:33 non-json replays give error when trying to be loaded into the new vizualizer 2011-04-04T17:18:52 http://pastebin.com/hgycZ0Fd 2011-04-04T17:19:01 yeah, that gives me an error 2011-04-04T17:19:18 Error stack: 2011-04-04T17:19:19 that's the one Frontier just added as game_id 3 2011-04-04T17:19:32 hm true 2011-04-04T17:19:33 http://marco-leise.homedns.org/antgui/work/games/0/3.replay 2011-04-04T17:19:41 I'll tell you the error stack I get in a second 2011-04-04T17:21:38 *** javagamer1 is now known as javagamer 2011-04-04T17:22:08 *** mceier has quit IRC (Quit: leaving) 2011-04-04T17:22:10 Error stack: 2011-04-04T17:22:10 TypeError: a 18 45 0 0 37 1 sssssssswwwwwwssseeeesss-ww-ssssssw-n 2011-04-04T17:22:10 Cannot read property '0' of undefined 2011-04-04T17:22:20 http://pastebin.com/qdEqrdye 2011-04-04T17:22:21 *** antimatroid has quit IRC (Ping timeout: 240 seconds) 2011-04-04T17:22:39 TypeError: a 18 45 0 0 37 1 sssssssswwwwwwssseeeesss-ww-ssssssw-n 2011-04-04T17:22:39 Cannot read property '0' of undefined 2011-04-04T17:23:12 So what's wrong? 2011-04-04T17:23:41 git log -1 give you the last commit as a visualizer regression fix? 2011-04-04T17:23:55 *** Ttech has quit IRC (Quit: Your nick was overruled.) 2011-04-04T17:24:14 specifically "visualizer: fixed regression in replay parser" 2011-04-04T17:24:21 *** sigh has joined #aichallenge 2011-04-04T17:24:25 checking 2011-04-04T17:25:51 works, thanks. 2011-04-04T17:25:58 great 2011-04-04T17:26:11 I'm not sure why it didn't update when i downloaded newest repo 2011-04-04T17:26:27 *** iFire` has quit IRC (Remote host closed the connection) 2011-04-04T17:28:28 *** Ttech has joined #aichallenge 2011-04-04T17:36:33 ok, looks like I can go to bed now. ;) 2011-04-04T17:36:44 *** amstan has quit IRC (Remote host closed the connection) 2011-04-04T17:36:52 Frontier 2011-04-04T17:36:54 one more thing 2011-04-04T17:38:29 ? 2011-04-04T17:38:32 actually never mind 2011-04-04T17:38:32 http://pastebin.com/CyXcgy7j 2011-04-04T17:38:39 you might want to put that as game4 if you want 2011-04-04T17:40:08 done, good night 2011-04-04T17:40:12 see ya 2011-04-04T17:40:54 *** skynetz has joined #aichallenge 2011-04-04T17:45:05 *** antimatroid has joined #aichallenge 2011-04-04T17:45:52 *** mega1 has joined #aichallenge 2011-04-04T17:52:12 will there be any compatibility with the vizualiser for IE? 2011-04-04T17:53:20 well, never mind. not really a priority 2011-04-04T18:06:41 *** sigh has quit IRC (Remote host closed the connection) 2011-04-04T18:07:44 *** ltriant has joined #aichallenge 2011-04-04T18:12:27 *** FireFly has quit IRC (Quit: swatted to death) 2011-04-04T18:30:11 *** computer1iz_222 has joined #aichallenge 2011-04-04T18:33:15 how to generate a stream in latest version? 2011-04-04T18:33:23 *** computerwiz_222 has quit IRC (Ping timeout: 240 seconds) 2011-04-04T18:47:32 *** iFire has joined #aichallenge 2011-04-04T19:13:34 *** needsch1 has joined #aichallenge 2011-04-04T19:14:49 *** needsch has quit IRC (Ping timeout: 264 seconds) 2011-04-04T19:14:50 *** needsch1 has quit IRC (Client Quit) 2011-04-04T19:19:25 *** sigh has joined #aichallenge 2011-04-04T19:19:26 *** antimatroid has quit IRC (Ping timeout: 260 seconds) 2011-04-04T19:30:52 *** McLeopold has left #aichallenge 2011-04-04T19:56:16 *** Naktibalda has quit IRC (Quit: ChatZilla 0.9.86.1 [Firefox 3.6.16/20110323143040]) 2011-04-04T20:01:18 *** delt0r_ has joined #aichallenge 2011-04-04T20:02:31 *** delt0r has quit IRC (Ping timeout: 246 seconds) 2011-04-04T20:24:45 *** Mathnerd314 has joined #aichallenge 2011-04-04T20:45:46 *** Eruonen has quit IRC () 2011-04-04T20:51:26 *** Heulsay has quit IRC () 2011-04-04T21:28:29 *** skynetz has quit IRC (Quit: .) 2011-04-04T21:29:11 *** Keth has quit IRC (Ping timeout: 276 seconds) 2011-04-04T21:32:37 *** Keth has joined #aichallenge 2011-04-04T21:33:43 *** antimatroid has joined #aichallenge 2011-04-04T21:35:29 *** suicidolt has joined #aichallenge 2011-04-04T21:40:49 *** smellyhippy has quit IRC (Ping timeout: 246 seconds) 2011-04-04T21:54:23 *** smellyhippy has joined #aichallenge 2011-04-04T21:54:23 *** smellyhippy has joined #aichallenge 2011-04-04T22:05:35 <_flag> Any consensus on battle resolution yet? 2011-04-04T22:08:44 *** kwketh has joined #aichallenge 2011-04-04T22:08:44 *** Keth has quit IRC (Read error: Connection reset by peer) 2011-04-04T22:12:19 _flag: nope :P 2011-04-04T22:33:36 *** mega1 has quit IRC (Ping timeout: 240 seconds) 2011-04-04T22:39:54 <_flag> antimatroid: business as usual then :P 2011-04-04T22:40:43 i haven't given up the good fight :P 2011-04-04T22:51:57 Frontier: where are the specs for the new replay format? 2011-04-04T22:58:34 *** Andy__ has quit IRC (Ping timeout: 252 seconds) 2011-04-04T23:06:00 *** amstan has joined #aichallenge 2011-04-04T23:06:00 *** ChanServ sets mode: +o amstan 2011-04-04T23:06:09 hello 2011-04-04T23:14:58 *** Kingpin13 has quit IRC (Quit: You're a kitty!) 2011-04-04T23:19:03 *** phire has quit IRC (Ping timeout: 264 seconds) 2011-04-04T23:21:29 hello 2011-04-04T23:22:05 amstan: provided mcleopold and sigh agree, i think we really need to get jeff to make some decisions soon 2011-04-04T23:22:13 antimatroid: about? 2011-04-04T23:22:23 it's been over a week since that post was made to reddit, and we've been saying "a few weeks" for a while now 2011-04-04T23:22:29 battle, scoring etc. 2011-04-04T23:23:00 antimatroid: can you setup some page or something so jeff can get up to speed? 2011-04-04T23:23:10 there is one for battle resolutions 2011-04-04T23:23:11 on what he needs to decide 2011-04-04T23:23:14 i will do one for scoring 2011-04-04T23:23:41 i will talk to mcleopold and sigh 2011-04-04T23:23:44 and work something out 2011-04-04T23:23:56 sigh: ^^ 2011-04-04T23:25:42 k, i summoned him 2011-04-04T23:32:12 *** sigh has quit IRC (Ping timeout: 246 seconds) 2011-04-04T23:32:39 amstan: i said not yet :P 2011-04-04T23:33:01 antimatroid: you said not yet? when? lol 2011-04-04T23:33:19 "provided mcleopold and sigh agree", "i will talk to mcleopold and sigh" 2011-04-04T23:34:08 i don't want to pull jeff in for important decisions like that without their consent 2011-04-04T23:34:13 agree to what? jeff summoning? lol 2011-04-04T23:34:24 w/e.. jeff has to be summoned either way 2011-04-04T23:34:30 jeff can come and get up to speed, but yeah 2011-04-04T23:34:42 i would prefer them to be okay with a decision being made if one is going to be 2011-04-04T23:35:04 McLeopold especially as he's done more than most (excepting maybe Frontier) 2011-04-04T23:37:15 *** sigh has joined #aichallenge 2011-04-04T23:51:11 antimatroid: ? 2011-04-04T23:51:37 sigh: what's the ? about 2011-04-04T23:51:53 [13:17] sigh: ^^ 2011-04-04T23:52:19 oh, do you agree that we should get jeff in to make some (at least initial) decisions about battle/scoring 2011-04-04T23:52:34 we've been saying "a few weeks" for a while 2011-04-04T23:52:49 is there any reason not agree? 2011-04-04T23:53:12 i don't know, i just don't want to "go around" you and mcleopold 2011-04-04T23:53:30 particularly when it's on a topic most people seem interested in 2011-04-04T23:53:51 can I outsource my decision making to you? 2011-04-04T23:54:16 you could, but all i was asking was that you have no objections :P 2011-04-04T23:55:20 I have no objections 2011-04-04T23:56:25 *** ltriant has quit IRC (Read error: Connection reset by peer) 2011-04-04T23:56:50 i would appreciate if ppl keep me in the loop though 2011-04-04T23:57:03 amstan: when have you not been? 2011-04-04T23:57:24 well.. it seems like stuff has been slowed down lately 2011-04-04T23:57:27 i was just checking 2011-04-04T23:57:32 maybe i missed something 2011-04-04T23:58:28 for my part people tell me stuff needs to be added to the engine and I add it... so I don't know much about the hold ups