2011-02-20T00:00:53 i'm trying to work out the rule of what battle/spawn radiuses must be relative to each other 2011-02-20T00:02:16 i think it's just when spawnradius is 1 that an issue arises 2011-02-20T00:02:26 like hebbie said :P 2011-02-20T00:02:38 Is the spawnradius currently 1? 2011-02-20T00:03:19 Doesn't look like it 2011-02-20T00:03:42 no, but my ideal parameters were spawnr = 1, attackr = rt(5) 2011-02-20T00:04:16 Ah 2011-02-20T00:04:26 Is the output row, col or col, row? 2011-02-20T00:04:42 row col 2011-02-20T00:04:50 It was col, row before wasn't it? 2011-02-20T00:04:50 everything has gone to row col finally thankgod 2011-02-20T00:04:54 yes 2011-02-20T00:04:54 Okay 2011-02-20T00:05:06 everything is consistent and never needs transposing if one uses row col 2011-02-20T00:05:15 it amazes me for that reason alone that anyone would ever use x y 2011-02-20T00:08:25 Yay, crappy C bot has successfully been ported to volatile spec 2011-02-20T00:30:16 *** antimatroid1 has joined #aichallenge 2011-02-20T00:32:42 *** antimatroid has quit IRC (Ping timeout: 276 seconds) 2011-02-20T00:33:56 *** chris___0076 has joined #aichallenge 2011-02-20T00:36:40 *** chris__0076 has quit IRC (Ping timeout: 240 seconds) 2011-02-20T00:40:44 *** chris___0076 is now known as chris_0076 2011-02-20T00:45:16 *** chris_0076 has quit IRC (Quit: Leaving) 2011-02-20T00:47:21 *** Chris_0076 has joined #aichallenge 2011-02-20T00:55:20 *** pgpaskar_ has joined #aichallenge 2011-02-20T01:35:46 *** amstan has quit IRC (Remote host closed the connection) 2011-02-20T01:37:07 *** amstan has joined #aichallenge 2011-02-20T01:37:08 *** ChanServ sets mode: +o amstan 2011-02-20T01:48:17 *** fmeyer has joined #aichallenge 2011-02-20T02:54:55 *** antimatroid has joined #aichallenge 2011-02-20T02:58:19 *** antimatroid1 has quit IRC (Ping timeout: 272 seconds) 2011-02-20T03:05:59 *** chris__0076 has joined #aichallenge 2011-02-20T03:08:17 *** Chris_0076 has quit IRC (Ping timeout: 240 seconds) 2011-02-20T03:14:49 *** boegel has joined #aichallenge 2011-02-20T03:15:08 *** boegel has joined #aichallenge 2011-02-20T03:25:29 *** boegel has quit IRC (Quit: Leaving) 2011-02-20T03:35:00 *** fmeyer has quit IRC (Quit: leaving) 2011-02-20T03:37:20 *** flag has quit IRC (Quit: Lost terminal) 2011-02-20T03:42:41 *** amstan has quit IRC (Ping timeout: 240 seconds) 2011-02-20T03:55:03 *** boegel has joined #aichallenge 2011-02-20T04:19:22 *** Palmik has joined #aichallenge 2011-02-20T04:19:25 *** Frontier has joined #aichallenge 2011-02-20T04:27:58 *** boegel has quit IRC (Quit: Leaving) 2011-02-20T04:35:38 *** Naktibalda has joined #aichallenge 2011-02-20T04:58:06 *** chris___0076 has joined #aichallenge 2011-02-20T05:00:42 *** chris__0076 has quit IRC (Ping timeout: 240 seconds) 2011-02-20T05:58:03 *** Palmik has quit IRC (Ping timeout: 250 seconds) 2011-02-20T06:05:40 *** delt0r___ has joined #aichallenge 2011-02-20T06:05:47 *** chris__0076 has joined #aichallenge 2011-02-20T06:06:42 *** delt0r_ has quit IRC (Ping timeout: 240 seconds) 2011-02-20T06:09:35 *** chris___0076 has quit IRC (Ping timeout: 276 seconds) 2011-02-20T06:10:24 *** Apophis_ has joined #aichallenge 2011-02-20T06:14:03 *** Apophis has quit IRC (Ping timeout: 272 seconds) 2011-02-20T06:23:35 *** choas has joined #aichallenge 2011-02-20T06:43:27 *** kaemo has joined #aichallenge 2011-02-20T07:31:10 *** Naktibalda has quit IRC (Remote host closed the connection) 2011-02-20T07:32:20 *** Palmik has joined #aichallenge 2011-02-20T07:36:05 *** mceier has joined #aichallenge 2011-02-20T07:52:45 *** sigh has quit IRC (Remote host closed the connection) 2011-02-20T07:55:58 *** Migi32 has joined #aichallenge 2011-02-20T10:17:03 *** Artir has joined #aichallenge 2011-02-20T10:17:20 *** JackyAlcine has joined #aichallenge 2011-02-20T10:31:52 *** antimatroid1 has joined #aichallenge 2011-02-20T10:33:20 *** antimatroid has quit IRC (Ping timeout: 246 seconds) 2011-02-20T10:43:57 *** JackyAlcine has quit IRC (Ping timeout: 240 seconds) 2011-02-20T10:58:51 *** JackyAlcine has joined #aichallenge 2011-02-20T11:05:48 *** Mathnerd314 has joined #aichallenge 2011-02-20T11:15:44 *** JackyAlcine has quit IRC (Read error: No route to host) 2011-02-20T11:17:21 *** JackyAlcine has joined #aichallenge 2011-02-20T11:21:08 *** choas has quit IRC (Ping timeout: 260 seconds) 2011-02-20T11:26:01 *** Naktibalda has joined #aichallenge 2011-02-20T11:33:24 *** McLeopold has joined #aichallenge 2011-02-20T11:33:51 antimatroid: I haven't touched the food spawn code. I just haven't made it to that section yet. 2011-02-20T11:33:55 *** Prillicy has joined #aichallenge 2011-02-20T11:36:00 McLeopold: i don't think you're telling people what team an ant has died from? 2011-02-20T11:36:36 nope 2011-02-20T11:37:23 *** kaemo has quit IRC (Ping timeout: 246 seconds) 2011-02-20T11:37:37 I could put a list of owners at the end of the line? 2011-02-20T11:38:28 just print a line with the player after the row 2011-02-20T11:38:41 if enemy ants collide, print more than one line 2011-02-20T11:39:20 ok 2011-02-20T11:39:33 i think you should scrap the whole land square information stuff, then just send info for all currently visible ants (alive or just dead) and food squares, then any wall squares the first time you see them. I think the input might actually be less then anyway, be easier for the engine to actually deal with, easier to explain to people and easier to deal with 2011-02-20T11:39:59 a lot less, but then bots need to calc vision themselves 2011-02-20T11:40:07 they already have to 2011-02-20T11:40:21 no they don't 2011-02-20T11:40:26 my sample bots don't 2011-02-20T11:40:31 you only get land the first time you see it 2011-02-20T11:41:12 how do you know for example how much you can see on a turn afer that? 2011-02-20T11:42:37 you get more land lines? 2011-02-20T11:42:40 it's also fairly easy then to write a function to create the char map we previously sent, which then gives you the vision amount 2011-02-20T11:43:02 are you talking about total number of squares seen in the game? or squares you can currently see this turn? 2011-02-20T11:43:10 i'm referring to the later 2011-02-20T11:43:17 *** Prillicy_ has joined #aichallenge 2011-02-20T11:43:47 oh, did you see hebbie's new forum thread? 2011-02-20T11:43:58 he has rendered my ideal parameters invalid 2011-02-20T11:44:21 i would prefer to avoid enemy collisions 2011-02-20T11:44:27 *** Prillicy has quit IRC (Ping timeout: 240 seconds) 2011-02-20T11:44:28 *** Prillicy_ is now known as Prillicy 2011-02-20T11:44:30 I'm talking total squares 2011-02-20T11:45:24 total squares is easy to do with anything, just keep an extra variable with each square whether the square has been seen at all yet 2011-02-20T11:45:34 and a counter of how many you've seen 2011-02-20T11:46:30 anyway, i can live with it as is now, but i think it'd make things easier to just send currently visible ants/food and first time seen water squares 2011-02-20T11:47:00 the explanation of that is a lot easier than wording out when you will see a land square again 2011-02-20T11:50:16 *** kaemo has joined #aichallenge 2011-02-20T11:56:51 In the interest of keeping the bots simple, I think I'll stick with giving out the land you can see for the first time. 2011-02-20T11:57:42 the bots would still be very simple 2011-02-20T11:57:55 at the start of each turn, set everything that isn't a wall to land 2011-02-20T11:58:03 then read the state info in, then make moves 2011-02-20T11:58:23 i think it makes everything easier :P 2011-02-20T11:58:30 That would work. But then they are assuming they can walk everywhere at first 2011-02-20T11:58:40 Actually, I kinda like it 2011-02-20T11:58:50 So, no land ever gets sent 2011-02-20T11:58:56 i was always going to assume I can walk anywhere until known otherwise :P 2011-02-20T11:59:23 i will also be marking squares by "last seen" with 0 being not seen yet 2011-02-20T11:59:30 but that's for people to do themselves 2011-02-20T11:59:46 so, you still want a land line to mark disappearing things? 2011-02-20T11:59:51 no 2011-02-20T12:00:24 basic rule would be: you get sent info for a water square the first time you see it, then you also get info for any ants (live or just dead) and food you can see, that's it 2011-02-20T12:00:48 Yeah, that would be a lot less info, and still simple for the sample bots. 2011-02-20T12:01:08 yep, and easier to explain and for you :) 2011-02-20T12:02:19 then if people really want, they can read that info into a char map of ?'s and paint out the '.' squares from 'a' to get the char map, which is what I plan to do 2011-02-20T12:05:06 McLeopold: did you remove the "end" line from bots input? 2011-02-20T12:05:17 i don't really care for it, just checking for taking it out of the spec 2011-02-20T12:17:26 no 2011-02-20T12:17:31 I didn't mean to 2011-02-20T12:18:13 Oh, the sample bot input was a bot that was eliminated 2011-02-20T12:18:50 So, I either need to send the end line to them and the middle game scores, or keep it running and send them the info at the end. 2011-02-20T12:19:11 Either way, this info is for tcp and local debugging/learning only. It wouldn't matter for the official server. 2011-02-20T12:19:35 So really, whatever people using tcp find the most useful 2011-02-20T12:21:54 McLeopold: I just updated the specs page for bot and visualiser input 2011-02-20T12:21:56 *** Artir has left #aichallenge ("Artir is in da house!") 2011-02-20T12:22:11 it might not be perfect, but it's close to what i think should be done 2011-02-20T12:22:23 without land? 2011-02-20T12:22:27 yep 2011-02-20T12:22:31 ok 2011-02-20T12:22:38 you don't need to send w info to the visualiser either 2011-02-20T12:22:50 or loadtime/turntime etc. but it does need the number of players and the version 2011-02-20T12:23:18 ok 2011-02-20T12:23:28 anyway, i'm off to bed 2011-02-20T12:23:35 classes start for the year in 5 hours :\ 2011-02-20T12:23:46 I was thinking that the replay file could be sent to a bot, but it would need to be scrubbed anyway, so I can remove the W lines. 2011-02-20T12:23:54 what class first? 2011-02-20T12:25:17 functional analysis 2011-02-20T12:25:26 ? 2011-02-20T12:25:38 maths :P 2011-02-20T12:25:41 oh 2011-02-20T12:25:48 I use to know math. 2011-02-20T12:25:59 But I never get to use it in the real world. :( 2011-02-20T12:26:23 yeah, i don't think much of what i'll learn this year will be very useful in the real world 2011-02-20T12:26:27 i get take home exams too :D 2011-02-20T12:26:35 nice 2011-02-20T12:31:56 math isn't meant for the real world anyway 2011-02-20T12:32:24 but i bet even functional analysis has some real world applications 2011-02-20T12:42:41 *** flag has joined #aichallenge 2011-02-20T12:54:46 *** kaemo has quit IRC (Ping timeout: 240 seconds) 2011-02-20T12:57:00 *** Prillicy has quit IRC (Quit: I be bangin') 2011-02-20T12:59:47 *** amriedle has joined #aichallenge 2011-02-20T13:11:46 Is the current engine up to date with the everything-is-land until I find it thing? 2011-02-20T13:11:55 And the dead ant owners? 2011-02-20T13:11:58 not yet 2011-02-20T13:12:04 not yet 2011-02-20T13:12:17 okay 2011-02-20T13:15:23 give me out 20 min 2011-02-20T13:19:20 :) 2011-02-20T13:22:04 *** kaemo has joined #aichallenge 2011-02-20T13:37:02 *** Mathnerd314 has quit IRC (Ping timeout: 250 seconds) 2011-02-20T13:49:18 *** Ttech has joined #aichallenge 2011-02-20T14:04:28 *** sumething has joined #aichallenge 2011-02-20T14:05:43 *** delt0r_ has joined #aichallenge 2011-02-20T14:07:25 *** delt0r___ has quit IRC (Ping timeout: 260 seconds) 2011-02-20T14:09:28 aichallenge: McLeopold epsilon * rc7330b8 / (8 files in 4 dirs): Removed 'L' lines from engine. Changed bots to assume everything is land until a 'W' line is sent. Sending an owner along with a dead ant line. - http://bit.ly/gGGs4L 2011-02-20T14:09:29 aichallenge: McLeopold epsilon * rad4a57e / (ants/.gitignore ants/playback/.gitignore): Merge branch 'epsilon' of github.com:aichallenge/aichallenge into epsilon - http://bit.ly/eSNDf7 2011-02-20T14:09:31 aichallenge: McLeopold epsilon * rc04f44a / ants/playback/.gitignore : Removed playback dir and .gitignore file. - http://bit.ly/fkfL7k 2011-02-20T14:09:51 *** Mathnerd314 has joined #aichallenge 2011-02-20T14:10:05 flag: okay, get the latest... 2011-02-20T14:13:04 McLeopold: okay, thanks 2011-02-20T14:13:22 you're doing a c++ bot? 2011-02-20T14:13:24 or just c? 2011-02-20T14:14:08 *** fawek has joined #aichallenge 2011-02-20T14:17:21 c at the moment, but c is valid c++ :P 2011-02-20T14:17:38 true 2011-02-20T14:19:16 The engine doesn't send me the new position of my own ants after I've moved them does it? 2011-02-20T14:19:25 yes, it does 2011-02-20T14:19:34 Oh 2011-02-20T14:19:41 it sends the position of all ants and food and dead ants that you can see 2011-02-20T14:20:03 Even if there's no new information? 2011-02-20T14:20:06 yes 2011-02-20T14:20:20 okay 2011-02-20T14:20:52 it's up to you to clear the food list if it goes out of vision, or to assume it always disappears 2011-02-20T14:24:59 *** boegel has joined #aichallenge 2011-02-20T14:28:07 *** Frontier has quit IRC (Quit: Ex-Chat) 2011-02-20T14:32:08 *** Frontier has joined #aichallenge 2011-02-20T14:45:07 Frontier: hows the visualizer? 2011-02-20T14:46:21 *** sumething has quit IRC (Ping timeout: 245 seconds) 2011-02-20T14:49:52 McLeopold: it is going slow. can a food item spawn and be converted to an ant in the same turn? 2011-02-20T14:50:26 no 2011-02-20T14:50:33 spawn after birth 2011-02-20T14:51:47 that's what I wanted to hear :) 2011-02-20T15:09:53 *** boegel has quit IRC (Quit: Leaving) 2011-02-20T15:15:30 flag: c isn't valid c++ 2011-02-20T15:15:39 int class; 2011-02-20T15:15:40 not even a little? 2011-02-20T15:15:41 ;) 2011-02-20T15:16:25 a subset of c is valid c++ 2011-02-20T15:19:14 *** amstan has joined #aichallenge 2011-02-20T15:19:15 *** ChanServ sets mode: +o amstan 2011-02-20T15:20:37 jmcarthur: did you get the new bot input samples? 2011-02-20T15:20:45 there was 1 change this morning 2011-02-20T15:23:42 jmcarthur: it's close enough for me 2011-02-20T15:24:19 http://tcp.zeroviz.us/viewer.html 2011-02-20T15:24:19 http://tcp.zeroviz.us/0.replay 2011-02-20T15:24:19 http://tcp.zeroviz.us/0.bot0.input 2011-02-20T15:24:19 http://tcp.zeroviz.us/0.bot0.output 2011-02-20T15:24:34 bot1, bot2 and bot3 are there too 2011-02-20T15:24:54 McLeopold: ah thanks, i was going to ask if there were updated versions 2011-02-20T15:25:16 the big change is no 'L' lines, and 'D' lines have an owner 2011-02-20T15:25:28 cool 2011-02-20T15:25:45 glad to see the L lines go, personally :) 2011-02-20T15:25:59 and i figured the D line enhancement would be coming anyway 2011-02-20T15:26:58 *** Accoun has quit IRC () 2011-02-20T15:27:33 the visualizer keeps dead ants around now? is that intentional? 2011-02-20T15:27:41 yeah, it was a hack 2011-02-20T15:27:54 but I'd like it to show area's of high conflict 2011-02-20T15:28:17 cool 2011-02-20T15:28:46 sucks that they disappear when ants run over them, but no biggie i guess 2011-02-20T15:28:58 ah and they don't reset when the game starts over 2011-02-20T15:29:11 or reappear when you go wrap backward to the last turn 2011-02-20T15:30:15 yeah, hack job 2011-02-20T15:30:50 might also be cool to make a square get gradually darker if more than one ant dies on it over time 2011-02-20T15:31:05 like a heat map, i guess 2011-02-20T15:32:37 yes, I thought of that! it would be cool 2011-02-20T15:32:41 dfghbnm, 2011-02-20T15:32:41 j bnm 2011-02-20T15:33:13 just spilled something on the keyboard :/ 2011-02-20T15:34:55 the heat map would allow me to clear the spots properly as well 2011-02-20T15:39:28 do the "turn" lines serve any purpose other than convenience? 2011-02-20T15:41:24 *** Accoun has joined #aichallenge 2011-02-20T15:44:47 No 2011-02-20T15:44:59 For debugging / local learning. 2011-02-20T15:46:42 mkay 2011-02-20T15:47:23 is it guaranteed to always come before any other information on each turn? 2011-02-20T15:48:39 for bot input, yes 2011-02-20T15:48:46 cool 2011-02-20T15:53:06 *** sigh has joined #aichallenge 2011-02-20T15:53:50 i assume that i should probably try to make all this case insensitive 2011-02-20T15:54:03 although i don't see it being an issue in practice 2011-02-20T15:55:03 yes, but I'm trying to keep everything lower 2011-02-20T15:55:40 lol, right as you said that, i saw this line in the repo: sys.stdout.write('O %s %s %s\n' % (order[0], order[1], order[2])) 2011-02-20T16:06:47 *** sigh has quit IRC (Remote host closed the connection) 2011-02-20T16:11:04 *** Palmik has quit IRC (Remote host closed the connection) 2011-02-20T16:13:20 Question: what is the reasoning behind not spawning food symmetrically? 2011-02-20T16:17:54 laziness 2011-02-20T16:18:44 Will it be in the final version? 2011-02-20T16:19:25 *** Frontier has quit IRC (Quit: Ex-Chat) 2011-02-20T16:20:05 *** Frontier1 has joined #aichallenge 2011-02-20T16:27:09 *** Frontier1 has quit IRC (Quit: Leaving.) 2011-02-20T16:27:27 *** Frontier has joined #aichallenge 2011-02-20T16:31:42 flag: maybe not, we are going to try non-symmetrical maps 2011-02-20T16:32:07 The starting food near starting ants should be placed very symmetrical though. 2011-02-20T16:33:10 yeah, early game is more critical than later game 2011-02-20T16:36:05 So I suppose fairness is to be achieved by playing enough games so the good/bad spwans average out? 2011-02-20T16:42:35 *** choas has joined #aichallenge 2011-02-20T16:49:44 *** SheV0 has joined #aichallenge 2011-02-20T16:50:14 *** Prillicy has joined #aichallenge 2011-02-20T16:56:38 *** Prillicy_ has joined #aichallenge 2011-02-20T16:56:48 *** SheV0 has quit IRC (Ping timeout: 240 seconds) 2011-02-20T16:56:52 *** SheV0_ has joined #aichallenge 2011-02-20T16:56:52 *** SheV0_ is now known as SheV0 2011-02-20T16:57:12 *** Prillicy has quit IRC (Ping timeout: 240 seconds) 2011-02-20T16:57:18 *** Prillicy_ is now known as Prillicy 2011-02-20T17:00:20 *** amstan has quit IRC (Remote host closed the connection) 2011-02-20T17:00:33 McLeopold: are you telling us how old a dead ant is? I want to know hold long ago they died if you leave them there like that 2011-02-20T17:00:50 *** amstan has joined #aichallenge 2011-02-20T17:00:50 *** ChanServ sets mode: +o amstan 2011-02-20T17:00:55 I think he only leaves them there on the visualizer 2011-02-20T17:01:08 But I could be wrong 2011-02-20T17:05:23 *** amstan has quit IRC (Remote host closed the connection) 2011-02-20T17:05:50 *** amstan has joined #aichallenge 2011-02-20T17:05:51 *** ChanServ sets mode: +o amstan 2011-02-20T17:07:19 antimatroid: Yeah, visualizer only. Dead ants only appear if you see them die. 2011-02-20T17:07:30 So, always 1 turn old. 2011-02-20T17:11:49 *** SheV0 has quit IRC (Ping timeout: 250 seconds) 2011-02-20T17:12:15 *** SheV0 has joined #aichallenge 2011-02-20T17:12:38 *** sigh has joined #aichallenge 2011-02-20T17:12:47 *** Prillicy_ has joined #aichallenge 2011-02-20T17:13:31 *** Prillicy has quit IRC (Ping timeout: 260 seconds) 2011-02-20T17:13:43 *** Prillicy_ is now known as Prillicy 2011-02-20T17:17:26 *** Naktibalda has quit IRC (Remote host closed the connection) 2011-02-20T17:25:01 *** mceier has quit IRC (Quit: leaving) 2011-02-20T17:31:37 So what's the next step in getting this contest going? 2011-02-20T17:38:43 *** choas has quit IRC (Quit: leaving) 2011-02-20T18:02:57 *** Prillicy has quit IRC (Quit: I be bangin') 2011-02-20T18:03:02 *** SheV0 has quit IRC (Quit: ChatZilla 0.9.86-rdmsoft [XULRunner 1.9.2.10/20100914130356]) 2011-02-20T18:06:33 *** amstan has quit IRC (Remote host closed the connection) 2011-02-20T18:07:01 *** amstan has joined #aichallenge 2011-02-20T18:07:01 *** ChanServ sets mode: +o amstan 2011-02-20T18:08:56 *** amstan has quit IRC (Remote host closed the connection) 2011-02-20T18:09:33 *** amstan has joined #aichallenge 2011-02-20T18:09:33 *** ChanServ sets mode: +o amstan 2011-02-20T18:15:39 amstan: is there a test side of the ai-contest.com? 2011-02-20T18:15:51 McLeopold: no 2011-02-20T19:14:43 *** Migi32 has quit IRC (Quit: boinc) 2011-02-20T19:20:25 *** fawek has quit IRC (Ping timeout: 240 seconds) 2011-02-20T19:31:33 *** McLeopold has quit IRC (Quit: Leaving.) 2011-02-20T19:48:06 *** kaemo has quit IRC (Read error: Operation timed out) 2011-02-20T19:56:28 *** amriedle has quit IRC (Quit: Lost terminal) 2011-02-20T20:40:43 *** chris___0076 has joined #aichallenge 2011-02-20T20:44:08 *** chris__0076 has quit IRC (Ping timeout: 260 seconds) 2011-02-20T22:06:19 *** delt0r___ has joined #aichallenge 2011-02-20T22:07:59 *** delt0r_ has quit IRC (Ping timeout: 272 seconds) 2011-02-20T22:16:11 *** JackyAlcine has quit IRC (Quit: Leaving.)