2011-09-29T00:16:46 *** dvladim has quit IRC (Read error: Operation timed out) 2011-09-29T00:27:16 contestbot: later tell mleise mind if I changed the duration and speedFastest default settings in the visualizer to 75 and 8 respectively? 2011-09-29T00:27:16 janzert: Job's done. 2011-09-29T00:35:30 *** Chris_0076 has quit IRC (*.net *.split) 2011-09-29T00:35:30 *** Lithosphere has quit IRC (*.net *.split) 2011-09-29T00:44:06 *** Lithosphere has joined #aichallenge 2011-09-29T00:44:06 *** Chris_0076 has joined #aichallenge 2011-09-29T00:47:24 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has quit IRC (Read error: Operation timed out) 2011-09-29T00:51:40 *** xathis has joined #aichallenge 2011-09-29T01:34:43 *** K333ES has joined #aichallenge 2011-09-29T01:42:01 *** Palmik has joined #aichallenge 2011-09-29T01:42:13 *** Kynes38 has joined #aichallenge 2011-09-29T01:54:28 "if all of a player's hills are destroyed they can't spawn any more ants." 2011-09-29T01:54:37 should that be players or player's? my english is terrible 2011-09-29T01:54:45 also, i'm having a go at giving that page a clean 2011-09-29T01:57:25 also, do ant hills have vision? 2011-09-29T02:00:50 *** Kynes38 has quit IRC (Ping timeout: 260 seconds) 2011-09-29T02:30:33 *** okay| is now known as okayzed 2011-09-29T02:36:37 *** eashoka has joined #aichallenge 2011-09-29T02:38:51 antimatroid : found an impregnable ant hill! (multi_maze_10, location (24:103)) :-) http://aichallengebeta.hypertriangle.com/visualizer.php?game=2779&user=9 2011-09-29T02:39:05 old -problem :) 2011-09-29T02:39:17 antimatroid : yea 2011-09-29T02:46:41 *** mcstar has joined #aichallenge 2011-09-29T02:59:17 @later tell janzert Sure, go ahead. 2011-09-29T02:59:17 mleise: Yep. 2011-09-29T03:06:43 mleise: morning 2011-09-29T03:06:58 i hope youre not still mad at me 2011-09-29T03:15:53 mcstar: Morning, I hope that question didn't keep you up all night! No, I just accept you the way you are. 2011-09-29T03:16:16 @later tell janzert ...but update the default values in the JSDoc as well! 2011-09-29T03:16:16 mleise: As you wish. 2011-09-29T03:17:26 technically, it wasnt a question, and i had a wonderful sleep, thanks 2011-09-29T03:18:29 you sound as if i were gay btw 2011-09-29T03:18:48 *** Doddzy39 has joined #aichallenge 2011-09-29T03:19:16 i found a "webinar" where melis explains the prev. contest 2011-09-29T03:20:13 You make it difficult to love you *g* 2011-09-29T03:20:18 who is melis? 2011-09-29T03:20:24 mega1 2011-09-29T03:21:02 so what does he write? anything interesting? 2011-09-29T03:21:30 explains 3 tactics 2011-09-29T03:21:57 i dont think he will go into implementation details, but i only watched half of it 2011-09-29T03:22:06 third 2011-09-29T03:22:20 *** ltriant has quit IRC (Quit: Computer has gone to sleep) 2011-09-29T03:23:00 i think the changes are quite significant, i need to rethink my approach 2011-09-29T03:23:32 the changes from what mega1 posted or the changes from without-hills to with-hills? 2011-09-29T03:23:49 otherwise im starting to think that its not really a bad idea to have ant-hills 2011-09-29T03:23:55 second 2011-09-29T03:25:03 have you decided yet how long the contest gonna be live this time? 2011-09-29T03:25:25 *** aerique has joined #aichallenge 2011-09-29T03:25:33 (i have quite a lot to do in the next couple of weeks) 2011-09-29T03:25:47 I'm not the right person to ask, but I assume it will be 2 months with the option to extend it to 3. 2011-09-29T03:26:05 lets go with 3 2011-09-29T03:26:58 I think antimatroid is on your side with that. You could ask amstan or janzert or jeff about that. 2011-09-29T03:27:09 *** ibdknox has quit IRC (Remote host closed the connection) 2011-09-29T03:27:38 *** ibdknox has joined #aichallenge 2011-09-29T03:27:58 there are only 83 registered users though 2011-09-29T03:28:04 kind of few 2011-09-29T03:28:15 I'm fine with 1 month or so. Usually I don't keep hacking for 3 months anyway. 2011-09-29T03:28:38 mcstar: There was no email about the changes. 2011-09-29T03:29:10 For the official start I hope, all the registred users will be informed. 2011-09-29T03:30:09 oh 2011-09-29T03:30:18 so its still not official? 2011-09-29T03:30:22 i see 2011-09-29T03:30:28 it is still aichallengebeta 2011-09-29T03:30:40 ai-contest.com has the old content 2011-09-29T03:30:47 yeah i see 2011-09-29T03:31:56 191 days ago this is what aerique said: 2011-09-29T03:31:58 *** ibdknox has quit IRC (Ping timeout: 256 seconds) 2011-09-29T03:32:08 "The official start is still a couple of weeks away," 2011-09-29T03:33:06 191/7 = 27 weeks ... 2011-09-29T03:33:59 but *this* time we are really close to the official start, closer than before at least, but that is just a matter of time i guess 2011-09-29T03:34:34 i guess the changes brought about new bugs for mcleo 2011-09-29T03:34:39 to fix 2011-09-29T03:36:23 and for the starter bots 2011-09-29T03:37:23 i am more and more surprised by the score difference that melis had 2011-09-29T03:37:31 pw was a very clean game 2011-09-29T03:38:18 and yet, in ants the top players were very close in score 2011-09-29T03:38:26 melis was god-like to us. we thought he had some crazy uber code in the bot. turned out he did what everyone else does, just a little better 2011-09-29T03:39:03 interesting 2011-09-29T03:39:36 "heurisitcs" was the buzz-word. "Are you still using heuristics only?" 2011-09-29T03:40:03 Heuristics won the game and it took a while until someone wrote a bot that could beat melis 2011-09-29T03:40:06 in ants "heuristics" has much bigger role 2011-09-29T03:41:00 You don't have complete knowledge in ants, you had that in PW 2011-09-29T03:41:33 yes, you could predict the future with more certainty in pw 2011-09-29T03:41:42 but you really cant predict in ants globally 2011-09-29T03:42:06 first thing i will do is to identify what is happening on the map 2011-09-29T03:42:22 you could start by merging groups of ants into larger blobs 2011-09-29T03:42:26 and with the addition of maze maps, the game becomes significantly harder too 2011-09-29T03:42:44 i think i will ned a waypoint graph 2011-09-29T03:42:46 need* 2011-09-29T03:42:58 at least its logical to have one 2011-09-29T03:43:21 There are improvements to A* that divide a map into 'rooms'/'dead ends' 2011-09-29T03:43:46 oh 2011-09-29T03:43:56 whats that called? 2011-09-29T03:43:57 I have my own ideas to pathfinding. Maybe I'll just code that for 2 months and nothing else ^^ 2011-09-29T03:44:38 *** antimatroid1 has joined #aichallenge 2011-09-29T03:44:45 *** antimatroid has quit IRC (Read error: No route to host) 2011-09-29T03:44:51 I fogot, but the PDF had nice example graphics 2011-09-29T03:46:32 unfortunately its a bit annoying to hear him talk english 2011-09-29T03:47:26 does he have a strong accent or bad grammar? 2011-09-29T03:49:48 bad pronunciation, and sometimes has trouble finding words 2011-09-29T03:49:52 so the usual 2011-09-29T03:50:37 plus imagine a noisy microphone with cracking noise 2011-09-29T03:51:38 from my experience hungarians have an annyoing accent when they talk english 2011-09-29T03:51:47 i think i have it too, but i dont talk much 2011-09-29T03:52:29 that nice rhytm and stressing of syllables are missing 2011-09-29T03:52:57 and the inflexions 2011-09-29T03:53:09 hm i cannot find that PDF again, sorry. but I can explain the how the algorithm works 2011-09-29T03:53:28 sure 2011-09-29T03:53:41 fire away captain 2011-09-29T03:54:38 you use your normal A* but you black out rooms that cannot be in the path. 2011-09-29T03:54:56 *** amstan_ has quit IRC (Ping timeout: 248 seconds) 2011-09-29T03:55:09 a room is a flood filled area where you have no concave parts in the area 2011-09-29T03:55:44 *** cichociemno has joined #aichallenge 2011-09-29T03:56:17 so when you reach an opening (a door) to the left and the pixel above you is also free (land) then you have to stop there. 2011-09-29T03:56:59 actually I forgot how they determine what room is to be included in the search. 2011-09-29T03:57:30 it may be a waypoint system 2011-09-29T03:57:46 i guess after a couple of searches these areas will be clear, and you need to store them permanently 2011-09-29T03:57:54 as if they were water for example 2011-09-29T04:00:34 I'm still not sure how optimized algorithms perform with the ants game, where you have no knowledge of the map and may have to move ants in packs 2011-09-29T04:00:58 here's some interesting stuff http://harablog.wordpress.com/ 2011-09-29T04:01:16 http://www.google.com/url?sa=t&rct=j&q=filetype%3Apdf%2Bpath-finding&source=web&cd=21&ved=0CBwQFjAAOBQ&url=http%3A%2F%2Fwww.cs.ucl.ac.uk%2Fstaff%2FD.Silver%2Fweb%2FPublications_files%2Fcoop-path-AIWisdom.pdf&ei=syWETrjBOauP4gSWza2pDw&usg=AFQjCNFNKB6uCUwq9lGaOH6EtC0ZTPKo_Q 2011-09-29T04:02:02 xathis: seems detailed, thx 2011-09-29T04:02:11 mleise: i cannot copy that url 2011-09-29T04:02:17 har har 2011-09-29T04:03:22 damn google 2011-09-29T04:03:33 got it 2011-09-29T04:04:07 *** boegel has joined #aichallenge 2011-09-29T04:04:19 seems you guys got me some extra curricularactivities 2011-09-29T04:05:06 in my last uploads i stopped using a* btw 2011-09-29T04:05:35 i used dijsktra it was better to search from food to the nearest ants 2011-09-29T04:06:34 when you dont know the destination coordinates, only a criterion for the end tile, its better to use dijkstra 2011-09-29T04:06:45 in fact i dont see if a* could be used than 2011-09-29T04:06:52 or there 2011-09-29T04:11:18 *** berak has joined #aichallenge 2011-09-29T04:12:09 if you know where the enemy hill is and want to send your ants there all the way through a maze? 2011-09-29T04:19:51 PW -> initial ants -> ants with hills... It's curious to see the game gradually converging towards a generic RTS 2011-09-29T04:20:38 yeah, its natural though, the complexity increases 2011-09-29T04:20:46 and the hot topic of ai is rts i think 2011-09-29T04:20:52 Next game will have archers I'm sure, and ability to build walls/turrets or other structures. 2011-09-29T04:21:14 i think there will be units that fly and that can swim 2011-09-29T04:21:53 maybe there will be tesla-coils too 2011-09-29T04:22:13 *** patrisk has joined #aichallenge 2011-09-29T04:22:31 I have a bit of a problem with 1 sec / move... I feel that with increasingly complex game you need a bit more for intelligent play.. 2011-09-29T04:22:56 You know, the viewer on the front page is zoomed in on the center area, but there is no way to zoom out or move it, even using the minimap. Is that supposed to happen? 2011-09-29T04:23:02 in beta it was 0.5 sec 2011-09-29T04:25:39 Personally I'd prefer a chess-like time management. Where an engine has N minutes (or seconds) for M moves. 2011-09-29T04:27:40 Which would introduce a time management concept. Like, thinking much on some moves and little on other moves. 2011-09-29T04:28:25 Which would also mean that a time left on the clock is received by a bot at the beginning of each move. Which would allow to change the time control without rewriting all bots. 2011-09-29T04:30:50 mcstar: I found the article I was looking for: www.ru.is/faculty/yngvi/pdf/BjornssonH06.pdf 2011-09-29T04:32:19 mleise: thx, i need to work now though 2011-09-29T04:33:32 patrisk: We should reset the zoom level for non-interactive maps, you are right 2011-09-29T04:43:28 *** mleise has quit IRC (Ping timeout: 248 seconds) 2011-09-29T04:46:20 *** cichociemno has quit IRC (Ping timeout: 252 seconds) 2011-09-29T04:48:32 *** eashoka has quit IRC (Quit: Page closed) 2011-09-29T04:50:38 K333ES: http://ai-contest.com/forum/viewtopic.php?f=21&t=1223 2011-09-29T04:59:50 are you going to participate in the next contest? - well, my wife may watch this presentation on the net later, and im afraid to say anything here, so far i denied any intention to do so 2011-09-29T05:00:07 (quote 'melis) 2011-09-29T05:00:25 damn, i f'd up 2011-09-29T05:00:30 i meant: 2011-09-29T05:00:33 (quote melis) 2011-09-29T05:05:16 aerique: Thanks 2011-09-29T05:11:07 *** K333ES has left #aichallenge 2011-09-29T05:30:58 *** sigh has joined #aichallenge 2011-09-29T05:48:20 *** Accoun has quit IRC (Ping timeout: 260 seconds) 2011-09-29T06:05:54 *** patrisk has quit IRC (Quit: Leaving) 2011-09-29T06:25:09 *** onensora has joined #aichallenge 2011-09-29T06:30:53 aichallenge: janzert epsilon * r9be2a6a / worker/compiler.py : Fix earlier Erlang change - http://git.io/91lUbw 2011-09-29T06:37:47 aichallenge: janzert epsilon * r0fd8e9d / ants/visualizer/js/Config.js : Try a faster default duration and speedFastest setting for the visualizer - http://git.io/ug8MhQ 2011-09-29T06:38:18 *** onensora has quit IRC () 2011-09-29T06:46:40 *** antimatroid1 has quit IRC (Ping timeout: 260 seconds) 2011-09-29T06:55:25 *** onensora has joined #aichallenge 2011-09-29T07:03:43 *** onensora has quit IRC (Ping timeout: 248 seconds) 2011-09-29T07:05:14 *** olexs has joined #aichallenge 2011-09-29T07:29:52 *** delt0r_ has quit IRC (Read error: Operation timed out) 2011-09-29T07:34:18 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has joined #aichallenge 2011-09-29T07:34:29 *** Accoun has joined #aichallenge 2011-09-29T07:41:04 *** Accoun has quit IRC (Ping timeout: 244 seconds) 2011-09-29T07:42:39 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has quit IRC (Ping timeout: 248 seconds) 2011-09-29T07:44:32 *** delt0r_ has joined #aichallenge 2011-09-29T07:48:56 *** olexs1 has joined #aichallenge 2011-09-29T07:49:48 *** eashoka has joined #aichallenge 2011-09-29T07:51:08 *** olexs has quit IRC (Ping timeout: 260 seconds) 2011-09-29T08:11:48 *** Accoun has joined #aichallenge 2011-09-29T08:16:47 *** Accoun has quit IRC (Ping timeout: 276 seconds) 2011-09-29T08:17:07 I'm having trouble logging in to my account on the beta site; is there a password recovery mechanism? 2011-09-29T08:17:55 when was the last time you logged in? there was a reset of all accounts a few days ago 2011-09-29T08:25:01 oh, a few days 2011-09-29T08:25:18 so create my account anew? 2011-09-29T08:25:36 check the rankings if your bot is still there, if not, then yes 2011-09-29T08:25:57 I haven't submitted one, was just thinking about doing so 2011-09-29T08:25:59 cool 2011-09-29T08:26:06 I'll try recreating 2011-09-29T08:26:08 *** Accoun has joined #aichallenge 2011-09-29T08:29:36 that worked, cheers 2011-09-29T08:30:33 *** Accoun has quit IRC (Ping timeout: 255 seconds) 2011-09-29T08:35:42 *** Accoun has joined #aichallenge 2011-09-29T08:42:45 *** Accoun has quit IRC (Ping timeout: 260 seconds) 2011-09-29T08:46:22 *** xathis_ has joined #aichallenge 2011-09-29T08:49:19 *** xathis has quit IRC (Ping timeout: 248 seconds) 2011-09-29T09:01:16 *** olexs1 has quit IRC (Quit: Leaving.) 2011-09-29T09:08:44 *** sigh has quit IRC (*.net *.split) 2011-09-29T09:10:41 *** sigh has joined #aichallenge 2011-09-29T09:22:02 *** Asmoo has joined #aichallenge 2011-09-29T09:27:16 *** Accoun has joined #aichallenge 2011-09-29T09:28:13 *** jsmith has joined #aichallenge 2011-09-29T09:34:40 *** amstan_ has joined #aichallenge 2011-09-29T09:34:40 *** ChanServ sets mode: +o amstan_ 2011-09-29T09:41:11 *** mceier has joined #aichallenge 2011-09-29T09:41:58 *** antimatroid has joined #aichallenge 2011-09-29T10:11:09 *** aerique has quit IRC (Quit: ...) 2011-09-29T10:11:25 *** amstan_ has quit IRC (Ping timeout: 260 seconds) 2011-09-29T10:12:58 *** patrisk has joined #aichallenge 2011-09-29T10:43:23 *** Keth has joined #aichallenge 2011-09-29T10:45:15 *** Keth_ has quit IRC (Ping timeout: 260 seconds) 2011-09-29T10:47:41 *** Keth has quit IRC (Ping timeout: 245 seconds) 2011-09-29T10:51:41 *** onensora has joined #aichallenge 2011-09-29T10:51:58 *** Keth has joined #aichallenge 2011-09-29T11:03:17 *** sigh has quit IRC (Remote host closed the connection) 2011-09-29T11:03:51 *** sigh has joined #aichallenge 2011-09-29T11:08:30 *** sigh has quit IRC (Ping timeout: 258 seconds) 2011-09-29T11:13:37 *** sir_macelon has joined #aichallenge 2011-09-29T11:14:44 I don't know if you have already noticed this, but the maze_4 map is generating almost only draws, because the starting locations are so far away 2011-09-29T11:14:46 http://aichallengebeta.hypertriangle.com/map.php?map=maze/maze_4.map 2011-09-29T11:15:21 there should some mechanism to identify draw-prone maps and disqualify them 2011-09-29T11:15:28 like it was in planet wars 2011-09-29T11:16:16 these draws affect ranks of bots which get to be paired with some low ranked bot 2011-09-29T11:17:14 sir_macelon : very true, when everything else in the competition has been designed to end games quickly. 2011-09-29T11:17:35 the other thing is I've noticed odd matchups, like two consecutive games on same map, or two consecutive 1v1 games with the same opponent 2011-09-29T11:18:11 eashoka: I've read janzert has been removing some maps already 2011-09-29T11:18:27 but they were related to the hills on water problem 2011-09-29T11:18:31 maze_14 is another 2011-09-29T11:23:07 *** boegel has quit IRC (Read error: Operation timed out) 2011-09-29T11:42:27 *** Keth_ has joined #aichallenge 2011-09-29T11:42:27 *** Keth has quit IRC (Read error: Connection reset by peer) 2011-09-29T11:46:18 *** olexs has joined #aichallenge 2011-09-29T11:58:52 *** sir_macelon has quit IRC (Quit: Page closed) 2011-09-29T11:59:54 *** Palmik has quit IRC (Remote host closed the connection) 2011-09-29T12:02:17 *** mleise has joined #aichallenge 2011-09-29T12:03:31 *** UncleVasya has joined #aichallenge 2011-09-29T12:07:13 *** boegel has joined #aichallenge 2011-09-29T12:19:50 *** Palmik has joined #aichallenge 2011-09-29T12:22:10 *** mcstar has quit IRC (Quit: WeeChat 0.3.5) 2011-09-29T12:36:40 *** okayzed is now known as okay| 2011-09-29T12:52:13 *** McLeopold has joined #aichallenge 2011-09-29T12:52:43 hi janzert 2011-09-29T12:54:56 *** okay| is now known as okayzed 2011-09-29T12:55:19 @later tell janzert 2 games that are different, 11 that are the same, I think my probable turn code is bad, so I'm not sure what the savings would be 2011-09-29T12:55:19 McLeopold: Ready to serve. 2011-09-29T12:56:12 @later tell janzert also, the unicode changes you made are funny because that was my original method, then I thought it might be to intrusive and switched it to the type error :) 2011-09-29T12:56:12 McLeopold: As you wish. 2011-09-29T13:02:29 *** okayzed is now known as okay| 2011-09-29T13:08:54 *** dr0b3rts has joined #aichallenge 2011-09-29T13:09:54 *** mcstar has joined #aichallenge 2011-09-29T13:15:28 @later tell janzert both "different" scenarios had to do with tie leaders, so it was the leader that was being waited on, game 2327 looks like it gave 4 points to a bot for no reason? 2011-09-29T13:15:28 McLeopold: Ready to serve, my lord. 2011-09-29T13:16:16 *** amstan has joined #aichallenge 2011-09-29T13:16:16 *** ChanServ sets mode: +o amstan 2011-09-29T13:18:58 *** dvladim has joined #aichallenge 2011-09-29T13:22:55 *** harley has joined #aichallenge 2011-09-29T13:31:05 *** ibdknox has joined #aichallenge 2011-09-29T13:32:15 *** ibdknox has quit IRC (Read error: Connection reset by peer) 2011-09-29T13:32:35 *** ibdknox has joined #aichallenge 2011-09-29T13:33:12 aichallenge: McLeopold epsilon * r5dc6e97 / ants/ants.py : fix bug where probably score was modifying the score list - http://git.io/Zv0Xpg 2011-09-29T13:33:12 aichallenge: McLeopold epsilon * rb66b48f / (6 files in 2 dirs): Merge branch 'epsilon' of github.com:aichallenge/aichallenge into epsilon - http://git.io/37s5cw 2011-09-29T13:39:01 *** UncleVasya has quit IRC () 2011-09-29T13:42:51 *** McLeopold has quit IRC (Read error: Connection reset by peer) 2011-09-29T13:51:48 *** mleise has quit IRC (Ping timeout: 258 seconds) 2011-09-29T13:53:03 *** xathis_ has quit IRC () 2011-09-29T14:09:44 *** okay| is now known as okayzed 2011-09-29T14:09:59 *** okayzed is now known as okay| 2011-09-29T14:21:08 *** mleise has joined #aichallenge 2011-09-29T14:35:30 *** Accoun has quit IRC (Ping timeout: 258 seconds) 2011-09-29T14:39:39 *** Accoun has joined #aichallenge 2011-09-29T14:49:18 *** amstan has quit IRC (Ping timeout: 258 seconds) 2011-09-29T14:57:15 *** dvladim has quit IRC (Ping timeout: 260 seconds) 2011-09-29T14:58:47 *** celio has joined #aichallenge 2011-09-29T15:08:57 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has joined #aichallenge 2011-09-29T15:09:13 *** UncleVasya has joined #aichallenge 2011-09-29T15:13:48 *** Palmik has quit IRC (Remote host closed the connection) 2011-09-29T15:19:32 *** UncleVasya has quit IRC () 2011-09-29T15:32:16 *** delt0r_ has quit IRC (Ping timeout: 245 seconds) 2011-09-29T15:33:48 *** patrisk has quit IRC (Remote host closed the connection) 2011-09-29T15:36:45 *** Accoun has quit IRC () 2011-09-29T15:38:23 RE maps like maze_4 we've already discussed the need to measure and limit the travel time between hills in the maps. It just needs to be done still ;) 2011-09-29T15:42:19 *** amstan has joined #aichallenge 2011-09-29T15:42:19 *** ChanServ sets mode: +o amstan 2011-09-29T15:44:53 *** delt0r_ has joined #aichallenge 2011-09-29T15:49:46 *** celio has quit IRC (Quit: Leaving) 2011-09-29T15:50:05 *** harley has quit IRC (Ping timeout: 252 seconds) 2011-09-29T15:51:00 *** Accoun has joined #aichallenge 2011-09-29T15:57:20 when did the hills change go in btw? 2011-09-29T15:57:27 i haven't been keeping up 2011-09-29T15:57:35 this past Sunday 2011-09-29T15:57:40 oh wow 2011-09-29T15:58:00 when do we launch? 2011-09-29T15:58:15 really changes the dynamics of the game. once the hill is gone you can't spawn any more ants, i take it? 2011-09-29T15:58:25 a1k0n: exactly 2011-09-29T15:58:32 not all the docs have been updated for hills 2011-09-29T15:58:34 you can have multiple hills though 2011-09-29T15:58:39 what? how? 2011-09-29T15:58:50 the map starts you with them 2011-09-29T15:58:53 depending on the map, some maps have 4 hills per player 2011-09-29T15:58:54 oh 2011-09-29T15:59:07 ok, so you can't gain hills; only lose them 2011-09-29T15:59:09 you can not capture enemy hills 2011-09-29T15:59:14 right 2011-09-29T15:59:38 so if you have four hills they round-robin where ants spawn? 2011-09-29T15:59:51 and there's some kind of spawn queue if you eat multiple food per turn? 2011-09-29T15:59:58 yes 2011-09-29T16:00:04 and yes 2011-09-29T16:00:04 contestbot: bot later tell McLeopold can you look at https://github.com/aichallenge/aichallenge/issues/271 I see the issue showing up on the server but I'm not sure the fix is correct 2011-09-29T16:00:05 janzert: User error -- Replace user. 2011-09-29T16:00:26 contestbot: later tell McLeopold can you look at https://github.com/aichallenge/aichallenge/issues/271 I see the issue showing up on the server but I'm not sure the fix is correct 2011-09-29T16:00:26 janzert: Aye, aye, sir 2011-09-29T16:00:44 a1k0n: If you don't move a spawned ant away, you will effectively block that hill and make your ants spawn elsewhere 2011-09-29T16:00:50 well, it's not quite round-robin, it's least recently occupied 2011-09-29T16:01:16 ah, heh. an interesting way to control that 2011-09-29T16:01:30 yeah the internal counter goes up until an ant is on top of the hill 2011-09-29T16:01:57 at which point it resets to 0 (lowest spawn priority) 2011-09-29T16:03:22 *** mleise has quit IRC (Quit: Leaving.) 2011-09-29T16:03:27 *** boegel has joined #aichallenge 2011-09-29T16:04:29 *** harley has joined #aichallenge 2011-09-29T16:13:56 *** harley has quit IRC (Ping timeout: 245 seconds) 2011-09-29T16:25:51 *** Dlayne has joined #aichallenge 2011-09-29T16:26:02 *** Dlayne has left #aichallenge 2011-09-29T16:34:28 *** harley has joined #aichallenge 2011-09-29T16:35:32 wait what? 2011-09-29T16:35:45 if the hill is blocked the ant spawns randomly on the map? 2011-09-29T16:36:28 *** harley has quit IRC (Client Quit) 2011-09-29T16:36:31 a blocked hill is by definition more recently occupied than any unblocked hill ;) 2011-09-29T16:36:58 i mean can ants spawn anywhere else toehr than a hill? 2011-09-29T16:37:04 if there is a tie in how recently occupied hills are it is broken randomly 2011-09-29T16:37:09 no 2011-09-29T16:37:12 *** boegel has quit IRC (Quit: This computer has gone to sleep) 2011-09-29T16:37:24 if all hills are blocked no ant spawns 2011-09-29T16:38:25 can those ants later spawn that were not spawned because of blocking? 2011-09-29T16:38:35 are they buffered so to speak? 2011-09-29T16:38:40 yep 2011-09-29T16:38:56 it's called showed as the hive in the visualizer 2011-09-29T16:39:52 http://antsbeta.janzert.com/visualizer.php?game=3563 see towards the end of that game 2011-09-29T16:40:19 so the ant-hill can be used for an ambush, lure in the opponent, release the buffer voila, they never knew what hit them 2011-09-29T16:40:22 A.Smith builds up a large queue 2011-09-29T16:40:36 :) you could try it anyway 2011-09-29T16:40:41 guess whose bot is that 2011-09-29T16:41:21 a star treck fan ;) 2011-09-29T16:41:50 you are correct 2011-09-29T16:41:54 i am 2011-09-29T16:42:09 my prev. bot was called Lore 2011-09-29T16:42:23 what is -.-.-- though? 2011-09-29T16:42:31 figure it out 2011-09-29T16:42:34 its easy 2011-09-29T16:42:44 is that supposed to be the enterprise? 2011-09-29T16:42:53 *** berak has quit IRC (Quit: ChatZilla 0.9.84 [SeaMonkey 2.0a3/20090223135443]) 2011-09-29T16:42:53 and you will be even more "right" after you read that message 2011-09-29T16:43:06 nnm? 2011-09-29T16:43:17 ahh 2011-09-29T16:43:32 ? 2011-09-29T16:43:49 dash-dot = n dash-dash = m 2011-09-29T16:44:08 well, yeah, but you can't just concatenate morse code unambiguously 2011-09-29T16:44:30 since -.- is k 2011-09-29T16:44:37 right, it's actually supposed to be ascii art 2011-09-29T16:44:52 mcstar: right? 2011-09-29T16:44:59 nah 2011-09-29T16:45:05 a1k0n was spot on 2011-09-29T16:45:13 now pls decode 2011-09-29T16:45:22 the message? 2011-09-29T16:45:25 sure 2011-09-29T16:45:32 you really want me to tell you? 2011-09-29T16:45:35 live long and prosper, of course 2011-09-29T16:45:40 ah ok 2011-09-29T16:45:42 :) 2011-09-29T16:45:50 how did you think I knew it was star trek fan? :) 2011-09-29T16:46:09 whoa, what? 2011-09-29T16:46:26 well, i though it was a coutner-joke to matrix 2011-09-29T16:46:38 i also don't get the a.smith reference 2011-09-29T16:46:45 oh 2011-09-29T16:46:48 at least morse code doesn't mean trekky to me ;) 2011-09-29T16:46:48 right 2011-09-29T16:46:49 you say star trek, but the nick is obviously related to the matrix 2011-09-29T16:48:23 heh, I'm old fashioned I suppose. morse code makes me think the person is probably a ham operator 2011-09-29T16:49:22 i have to re-watch the first movie, the one with V'ger 2011-09-29T16:49:30 didnt they use morese-code in it? 2011-09-29T16:49:35 morse* 2011-09-29T16:49:56 actually I've only ever watched the most recent movie, and maybe a dozen episodes 2011-09-29T16:49:58 im not sure, anyway, i dont know klingon 2011-09-29T16:50:36 and aside from star trek, live long and prosper has a nice ring to it, i liek it 2011-09-29T16:50:59 :) 2011-09-29T16:51:09 i really hope that i can write a bot that lives up to its name 2011-09-29T16:51:24 i might have bitten more than i can chew... 2011-09-29T16:52:12 well you can decide at contest launch whether you want to try or switch to a less demanding name :P 2011-09-29T16:52:39 oh, i forget that its still half-live 2011-09-29T16:52:49 yeah, i think ill stick with it 2011-09-29T16:52:55 it will give me motivation 2011-09-29T16:52:57 http://aichallengebeta.hypertriangle.com/visualizer.php?game=3575&user=39 2011-09-29T16:53:03 can you take a look at this game? 2011-09-29T16:53:13 sort of interesting the final scoring 2011-09-29T16:53:54 what about it? no one captured a hill throughout 2011-09-29T16:54:17 ok 2011-09-29T16:54:27 seems im not familiar with the changed rules 2011-09-29T16:54:50 points are only given for destroying a hill 2011-09-29T16:55:05 and you lose a point for losing one of your own 2011-09-29T16:56:08 2 per hill are given 2011-09-29T16:56:40 instead of losing a point when you lose a hill I'd rather say it as you get a point each hill you keep control of until the end of the game, but McLeopold doesn't like it that way and it works out to the same thing ;) 2011-09-29T16:56:56 yes, 2 points for each hill you destroy 2011-09-29T16:59:36 janzert: have you considered re-opening an ant-orifice after another hive's ant stuck dirt in it? 2011-09-29T16:59:44 (by the same ant it belonged to) 2011-09-29T17:00:04 i mean the possibility of this 2011-09-29T17:00:48 alias somethingVeryComplicatedSounding="./essay --topic somethingVerySimpleAndDumb -vvvvvvvvvvvvvvvvv" 2011-09-29T17:01:18 considered it as in rejected it yes ;) 2011-09-29T17:02:18 I think a hill capture being non-reversible is a fairly important aspect 2011-09-29T17:03:18 i guess you want short games 2011-09-29T17:07:34 contestbot: later tell McLeopold http://antsbeta.janzert.com/visualizer.php?game=3560 that ends on "ants not razing hills" I think the population calculation needs to include ants in hives for players with active hills, since the hill can't be captured until it is emptied 2011-09-29T17:07:34 janzert: Ready to serve. 2011-09-29T17:08:02