2010-10-09T00:00:07 Top 10 players: rsergio(3825), felixcoto(3814), Hazard(3808), ofrias(3732), cfaftw(3703), Killer(3687), sequoh(3661), davidjliu(3654), adumlah(3619), _iouri_(3615) 2010-10-09T00:02:53 *** rebelxt_ has joined #aichallenge 2010-10-09T00:03:40 crowbot is basically an advanced ragebot, according to mcleopold 2010-10-09T00:04:47 contestbot: later tell McLeopold Use "creeper" as your next codename for your bot. It's the most dreaded mob in minecraft. 2010-10-09T00:04:47 amstan: I think that worked... 2010-10-09T00:06:22 great 2010-10-09T00:06:30 *** Titankiller has joined #aichallenge 2010-10-09T00:07:54 crowbot is doing much better than ragebot.. and i've just lost a game to it ;-) 2010-10-09T00:10:50 this is a bit depressing. another 200 lines of code for a 0.00% improvement over previous version.. lol 2010-10-09T00:11:21 *** delt0r___ has joined #aichallenge 2010-10-09T00:12:38 *** rwa has quit IRC (Quit: rwa) 2010-10-09T00:13:21 *** delt0r_ has quit IRC (Ping timeout: 272 seconds) 2010-10-09T00:13:47 my project continues to grow 2010-10-09T00:14:05 just... slowly 2010-10-09T00:14:11 time to play minecraft 2010-10-09T00:14:21 *** rwa has joined #aichallenge 2010-10-09T00:17:08 is minecraft fun? 2010-10-09T00:17:27 rebelxt_: very 2010-10-09T00:17:49 rebelxt_: trailer: http://www.youtube.com/watch?v=FaMTedT6P0I 2010-10-09T00:21:15 *** davidjliu has quit IRC (Quit: Page closed) 2010-10-09T00:31:55 amstan: looks addictive - gonna try to stay away.. bot taking up free time ;-) 2010-10-09T00:41:21 *** rebelxt_ has quit IRC (Ping timeout: 265 seconds) 2010-10-09T00:51:18 *** Tobu has joined #aichallenge 2010-10-09T00:53:23 I just made my bot win faster. That said it doesn't win all that often, but when it does my opponent doesnt have to wait long on the tcp server ^^. 2010-10-09T01:00:07 Top 10 players: rsergio(3833), felixcoto(3829), Hazard(3817), ofrias(3736), cfaftw(3718), sequoh(3672), davidjliu(3660), adumlah(3640), EBraun(3631), Killer(3624) 2010-10-09T01:19:58 I wish my bot was only taking up my free time :0 2010-10-09T01:32:05 having a few beers while working on my bot is not helping much...lol 2010-10-09T01:35:17 *** yasith_ has quit IRC (Ping timeout: 265 seconds) 2010-10-09T01:37:38 Frontier: are you turtle.frontier? 2010-10-09T01:40:21 *** hellman has quit IRC (Remote host closed the connection) 2010-10-09T01:45:22 Yes, I am 2010-10-09T01:47:52 Frontier: you naming your bots after animals like McLeopold? :P 2010-10-09T01:48:10 *** yasith_ has joined #aichallenge 2010-10-09T01:51:32 *** Naktibalda has quit IRC (Ping timeout: 272 seconds) 2010-10-09T01:51:54 sigh: When I started working on my bot I thought it would be especially slow moving and defensive. So I named it turtle. It is far more aggressive now and likes to attack enemy planets with lots of ships- 2010-10-09T01:53:25 :) 2010-10-09T01:53:29 Next I code a replay-player for my server so I can rerun games from the TCP server and check why my bot is doing some wrong decisions. (Despite the lack of good heuristics like in McLeopolds case) 2010-10-09T01:53:47 yeah, that's useful 2010-10-09T01:53:54 btw, what is your name in the main contest/ 2010-10-09T01:53:55 ? 2010-10-09T01:54:05 <- Frontier 2010-10-09T01:54:12 makes sense :P 2010-10-09T01:54:31 http://www.ai-contest.com/profile.php?user_id=9727 2010-10-09T01:55:49 cool, not doing too bad 2010-10-09T01:56:03 I think your bot would be a lot better if it moved units to the front 2010-10-09T01:56:11 I once had a match where I had considerably more ships, but the enemy had a few 1-growth planets more than me. My bot kept attacking his well defended larger planets instead of just taking two of the 1-growth planets that were only 2 and 3 turns away from me, but a huge leap from the enemy 2010-10-09T01:56:36 yes, planet selection is HARD 2010-10-09T01:57:23 if I could figure out a good way just for selecting planets my bot would be considerably stronger 2010-10-09T01:59:38 *** flag has quit IRC (Ping timeout: 260 seconds) 2010-10-09T01:59:41 I kinda hacked in that it moves ships from the backyard to the front if it is 100% save. Just so matches where I already win end faster. It took me lots of time to get my bot to make good choices in the beginning. It still needs to understand that if it is a neutral just a bit closer to the enemy, then it is useless to attack, because I will just waste ships on it and lose it afterwards. 2010-10-09T02:00:07 Top 10 players: rsergio(3844), felixcoto(3834), Hazard(3810), ofrias(3728), cfaftw(3720), sequoh(3691), davidjliu(3664), adumlah(3652), EBraun(3643), bix0r4ever(3640) 2010-10-09T02:00:32 ah, that neutral thing I just explicitly check for 2010-10-09T02:00:47 my bot is too stupid to figure that out itself 2010-10-09T02:01:29 sigh: I tried to come up with 4 to 5 more or less distinct scores for each planet. I multiply these to get the final score. Then I tweak my bot by doing "pow(score1, 0.5)" or something so certain scores get more weight than others. 2010-10-09T02:02:45 hmm... that sounds complicated :S 2010-10-09T02:02:47 My bot on rank 133 is still the one I hacked together in 2 days. Sometimes 'explicit checks' aren't that bad. They are less likely to fail in special situations. 2010-10-09T02:02:49 *** Utkarsh has quit IRC (Ping timeout: 240 seconds) 2010-10-09T02:03:28 I also score planets independently, but it is just a simple function of growth_rate, distance and cost 2010-10-09T02:03:36 sigh: I learned that in my job as a browser game developer 2010-10-09T02:03:52 (where cost is how many ships are need to take the planet) 2010-10-09T02:04:09 cool, what sort of games do you make? 2010-10-09T02:04:59 Something very peaceful: http://wewaii.de 2010-10-09T02:06:54 can't understand :D 2010-10-09T02:07:04 There are tourists considering several activities, each of which has attributes like its category (sport, party, ...), the quality of the location, the renovation state, the distance, etc... and it is pretty much like in PlanetWars where you have ships that look for planets to have a good time. 2010-10-09T02:07:40 haha, i see 2010-10-09T02:08:36 *** Utkarsh has joined #aichallenge 2010-10-09T02:08:43 The server is written in Java, but I decided to use C++ for faster responses. 2010-10-09T02:09:06 ...I mean for my bot. How is your bot doing sigh? 2010-10-09T02:10:24 hmmm... it's down to 25 now 2010-10-09T02:10:42 i've been trying a lot of stuff, but nothing made it significantly better 2010-10-09T02:11:16 so the previous week I've been re-writing a lot of the framework to make more advanced ideas easier 2010-10-09T02:13:03 *** TomyMMX has joined #aichallenge 2010-10-09T02:13:20 it's been a lot harder than I though to make my bot better :D 2010-10-09T02:16:20 sigh: I hear that 2010-10-09T02:17:21 *** dufflebunk has joined #aichallenge 2010-10-09T02:17:42 I find the hardest thing is to figure out if your idea sucks or if your implementation is just not right yet 2010-10-09T02:18:05 I really need a more systematic testing procedure 2010-10-09T02:19:31 yeah same, atm I just run my bot against several previous versions across like 10 maps 2010-10-09T02:20:50 dhartmei's local 'tournament' stuff is great for that btw 2010-10-09T02:20:56 *** Palmik has joined #aichallenge 2010-10-09T02:22:55 *** flag has joined #aichallenge 2010-10-09T02:31:54 If I tested only 10 maps (1-10) I would have gone wrong. Yesterday I tested a new version and it won mostly on those maps, but lost often on the other 90. 2010-10-09T02:33:03 yeah, I do more thorough test every so often, the 10 maps is just after every change 2010-10-09T02:35:52 sigh: Right now testing agains my old version takes a little longer, becuase I don't have the sources no more so I have to spawn a sub process :O. As soon as I have everything in one executalbe matches on 100 maps against older versions will be back to a few seconds :) But I need a break now. 2010-10-09T02:36:06 :) 2010-10-09T02:37:52 the rate of tech advancement is a little frustrating. I can beat bots now that were in the top 3 not that long ago. But this isn't good enough anymore. 2010-10-09T02:39:21 I fear if I ever update my main server bot it will fall back behind it current rank. 2010-10-09T02:44:23 whatever hexist is doing, it's kicking my ass 2010-10-09T02:48:06 rwa: link? 2010-10-09T02:48:51 http://72.44.46.68/canvas?game_id=385073 2010-10-09T02:53:09 I hate it when my bot starts out winning 8 or 9 games in a row and I think I have a champion bot... then it loses 12 straight 2010-10-09T02:54:33 haha 2010-10-09T03:00:08 Top 10 players: rsergio(3850), felixcoto(3826), Hazard(3800), cfaftw(3729), ofrias(3721), sequoh(3690), Killer(3669), davidjliu(3669), adumlah(3652), EBraun(3641) 2010-10-09T03:03:12 rwa: so you are the timeouter ^^ 2010-10-09T03:03:42 sometimes I don't pare my logging down enough, yeah 2010-10-09T03:04:58 and often when my laptop starts getting low on battery... 2010-10-09T03:06:41 *** TomyMMX has quit IRC (Quit: Page closed) 2010-10-09T03:09:56 *** hellman has joined #aichallenge 2010-10-09T03:25:36 *** Frontier has left #aichallenge 2010-10-09T03:28:28 *** Frontier has joined #aichallenge 2010-10-09T03:30:51 *** HakanD has joined #aichallenge 2010-10-09T03:55:50 *** hellman has quit IRC (Ping timeout: 240 seconds) 2010-10-09T03:56:15 *** hellman has joined #aichallenge 2010-10-09T04:00:08 Top 10 players: rsergio(3829), felixcoto(3816), Hazard(3793), ofrias(3721), cfaftw(3716), sequoh(3685), davidjliu(3668), Killer(3666), adumlah(3648), EBraun(3636) 2010-10-09T04:10:50 *** bhasker has joined #aichallenge 2010-10-09T04:12:17 *** yasith_ has quit IRC (Ping timeout: 264 seconds) 2010-10-09T04:17:41 *** JCS^ has quit IRC (Quit: \x00) 2010-10-09T04:25:09 *** yasith has joined #aichallenge 2010-10-09T04:27:59 *** Migi32 has joined #aichallenge 2010-10-09T04:36:11 *** criis has quit IRC (Ping timeout: 276 seconds) 2010-10-09T04:38:42 *** jaspervdj has joined #aichallenge 2010-10-09T04:40:50 *** Utkarsh has quit IRC (Read error: Connection reset by peer) 2010-10-09T04:46:13 *** Accoun_ has quit IRC (Ping timeout: 265 seconds) 2010-10-09T04:47:52 wats the logic the sample bots use for finding the stronger and weaker bots ? 2010-10-09T04:48:43 *** Utkarsh has joined #aichallenge 2010-10-09T05:00:08 Top 10 players: rsergio(3830), felixcoto(3818), Hazard(3784), cfaftw(3726), ofrias(3721), sequoh(3689), davidjliu(3667), Killer(3665), EBraun(3641), adumlah(3636) 2010-10-09T05:00:23 *** jesionaj` has joined #aichallenge 2010-10-09T05:01:47 *** iFire has quit IRC (Read error: Connection reset by peer) 2010-10-09T05:03:37 *** jesionaj has quit IRC (Ping timeout: 265 seconds) 2010-10-09T05:10:10 *** iFire has joined #aichallenge 2010-10-09T05:12:34 hi 2010-10-09T05:13:47 *** optimum has quit IRC (Ping timeout: 245 seconds) 2010-10-09T05:14:41 *** optimum has joined #aichallenge 2010-10-09T05:20:56 *** Appleman1234 has quit IRC (Ping timeout: 240 seconds) 2010-10-09T05:21:27 *** Appleman1234 has joined #aichallenge 2010-10-09T05:27:23 *** Meatkat has joined #aichallenge 2010-10-09T05:28:39 *** hellman has quit IRC (Read error: Operation timed out) 2010-10-09T05:30:30 *** fawek has joined #aichallenge 2010-10-09T05:31:28 Titankiller, I have no idea what you mean 2010-10-09T05:31:51 also, ruilov's resubmission seems to have a hard time climbing back up 2010-10-09T05:34:06 wel there's this formula in all sample bots saying 2010-10-09T05:34:09 *** Blkt has joined #aichallenge 2010-10-09T05:34:34 code=size/1+growth rate 2010-10-09T05:36:15 Titankiller, well, not in all bots 2010-10-09T05:36:25 at least not in BullyBot and RageBot 2010-10-09T05:36:36 which bot are you looking at? 2010-10-09T05:37:12 ah yes I see it now in ProspectorBot 2010-10-09T05:37:19 so Titankiller what about it? 2010-10-09T05:38:51 shouldn't 1+growth rate be in numerator if u're alculating strongest ? 2010-10-09T05:40:01 Titankiller: this way higher growth rate planets are seen as more "valuable" since the start pack bots attack the weakest enemies 2010-10-09T05:40:21 but definitely play around with the logic, and what difference it makes :) 2010-10-09T05:40:21 Is it ok, to have C++ source code in subdirectories for the submissions? 2010-10-09T05:40:44 Frontier: my initial submission had that, and it wouldn't compile 2010-10-09T05:40:49 Titankiller, this is a very basic formula. What it really calculates is the number of ships ("cost") per growth-rate unit it gets 2010-10-09T05:40:54 ...and why did I upload before asking? 2010-10-09T05:40:59 I don't think anything's changed since then 2010-10-09T05:41:15 *** Appleman1234 has quit IRC (Ping timeout: 240 seconds) 2010-10-09T05:41:56 but if uput growth rate in denmnator ur strongestplane has thelowestscoe am i rigt ? 2010-10-09T05:42:15 so it will try to capture a 5-growth planet with 10 ships before it will try to capture a 1-growth planet with 9 ships 2010-10-09T05:42:29 Titankiller: don't think of it as strongest then, all it is saying is that we want to attack planets with high growth and low ship count 2010-10-09T05:42:46 and we want to attack from planets with low growth but high ship count 2010-10-09T05:42:59 which makes sense, right? 2010-10-09T05:43:26 good day everyone 2010-10-09T05:43:38 hello 2010-10-09T05:44:16 but the comment says its finding MY strongest planet 2010-10-09T05:44:51 Titankiller, the comment is misleading 2010-10-09T05:45:52 it's finding your planet with the highest score, which is the planet with the most ships (so the attack is strong) / growth-rate (it's no big deal to lose a planet with low growth-rate) 2010-10-09T05:46:12 get it? 2010-10-09T05:46:49 yeah 2010-10-09T05:47:24 but as I said, it's by no means a good scoring function, you'll have to come up with better stuff to win this contest ;) 2010-10-09T05:47:24 i guess i shuddelet all the comments nd get my head workng ;-) 2010-10-09T05:47:45 oh yesi'm workng on it 2010-10-09T05:47:53 *** Appleman1234 has joined #aichallenge 2010-10-09T05:48:12 *** hellman has joined #aichallenge 2010-10-09T05:48:38 one more should my entire logic fit into public static void DoTurn(PlanetWars pw) > 2010-10-09T05:48:41 ? 2010-10-09T05:49:25 no 2010-10-09T05:49:34 hi 2010-10-09T05:49:36 you can add as many files and functions as you want 2010-10-09T05:50:14 the only restrictions are: you have to have MyBot.cc (or .java or w/e) with a main() function, and all files have to be in the same programming language 2010-10-09T05:51:35 many thanks 2010-10-09T05:54:07 no problems 2010-10-09T06:00:09 Top 10 players: rsergio(3824), felixcoto(3811), Hazard(3778), cfaftw(3720), ofrias(3707), sequoh(3684), davidjliu(3660), Killer(3659), EBraun(3633), adumlah(3632) 2010-10-09T06:03:49 ehhh lol? 2010-10-09T06:03:49 COMPILER ERRORS 2010-10-09T06:03:49 2010-10-09T06:03:49 GameStateData.cpp:219:2: warning: no newline at end of file 2010-10-09T06:03:49 Orders.cpp:42:2: warning: no newline at end of file 2010-10-09T06:05:13 did it accept it anyway or not? 2010-10-09T06:06:09 No, it considered it a serious problem that there were no blank lines at the ends of my files - I hate blank lines at end of files :/ 2010-10-09T06:06:28 It's like a habit to remove them whenever I find some 2010-10-09T06:07:28 Frontier, in cpp files you definitely shouldn't do that 2010-10-09T06:07:36 #include includes the file exactly as it is 2010-10-09T06:07:37 Migi32: User error, it's not my fault. 2010-10-09T06:07:45 shut up contestbot :P 2010-10-09T06:08:01 if the last line was a comment without a newline, the next line could end up getting commented as well 2010-10-09T06:08:12 contestbot, I understand that it is my fault -.- 2010-10-09T06:08:13 Frontier: Run as fast as you can and don't look back. 2010-10-09T06:08:45 Migi32, thanks for the info. I forgot about that. 2010-10-09T06:08:54 Frontier, no problem 2010-10-09T06:09:35 Now it seems like my bot isn't playing on the server. I checked it on my computer (using the same compiler arguments) before. Does that happen sometimes? 2010-10-09T06:09:59 your first game should happen seconds after your submission 2010-10-09T06:10:38 but there'll be some time before the second and further games 2010-10-09T06:10:47 Sure, what I meant was "It isn't making a move" 2010-10-09T06:11:25 you're not the only one with that problem. I'm not sure but I think it could be the server's fault 2010-10-09T06:12:02 I'll upload again and see if the problem continues. 2010-10-09T06:12:59 Maybe the compiler hadn't finished when the game started. ^^ 2010-10-09T06:13:19 contestbot, what do you say? 2010-10-09T06:13:19 Frontier: Error: No factoid matches that key. 2010-10-09T06:15:31 Oh that doesn't look good. 2010-10-09T06:21:03 Ah ok, I forgot my logger config file. 2010-10-09T06:21:42 *** hellman has quit IRC (Ping timeout: 245 seconds) 2010-10-09T06:35:27 contestbot, what is love? 2010-10-09T06:35:27 wh1teside: Error: No factoid matches that key. 2010-10-09T06:35:59 contestbot, baby don't hurt me 2010-10-09T06:36:00 wh1teside: I have no idea what you mean. 2010-10-09T06:36:31 contestbot, are you in fact a human trying to mimic a turing machine? 2010-10-09T06:36:32 Frontier: No! 2010-10-09T06:36:45 * wh1teside knew contestbot was afraid of commitment 2010-10-09T06:37:57 !commands 2010-10-09T06:37:57 Frontier: action, add, alert, alias, allyourspace, announce, announce add, announce list, announce remove, any, aol, apply, apropos, at, author, azn, ban add, ban list, ban remove, base, binary, bold, bye, cache, calc, call, capabilities, capability add, capability list, capability remove, capability set, capability setdefault, capability unset, capitalize, change, changename, channel, channels, (5 more messages) 2010-10-09T06:38:51 !more 2010-10-09T06:38:51 HakanD: Error: You haven't asked me a command; perhaps you want to see someone else's more. To do so, call this command with that person's nick. 2010-10-09T06:38:58 !more 2010-10-09T06:38:58 Frontier: channelstats, chr, cmd, coin, color, colorize, command, commands, concat, config, connect, contributors, convert, countargs, cpu, crypt, ctime, cut, cycle, decode, default, defaultcapability, defaultplugin, dehalfop, deop, devoice, dice, dict, dictionaries, disable, disconnect, dns, doctype, driver, echo, eightball, elapsed, enable, encode, errno, export, fetch, field, fight, fit, flush, (4 more messages) 2010-10-09T06:39:08 !moar 2010-10-09T06:39:08 wh1teside: Error: You haven't asked me a command; perhaps you want to see someone else's more. To do so, call this command with that person's nick. 2010-10-09T06:39:16 !cpu 2010-10-09T06:39:16 Frontier: I have taken 1988.84 seconds of user time and 63.84 seconds of system time, for a total of 2052.68 seconds of CPU time. My children have taken 0.00 seconds of user time and 0.00 seconds of system time for a total of 0.00 seconds of CPU time. I'm taking up 16116 kB of memory. 2010-10-09T06:39:51 !echo $randomNick 2010-10-09T06:39:51 Janzert 2010-10-09T06:39:59 @predict 2010-10-09T06:39:59 I predict that dmj111 will win! 2010-10-09T06:40:25 !eightball 2010-10-09T06:40:25 Frontier: Yes! 2010-10-09T06:40:50 !fight 2010-10-09T06:40:50 Frontier: Error: I tried to send you an empty message. 2010-10-09T06:41:04 !flush 2010-10-09T06:41:04 Frontier: Error: You don't have the owner capability. If you think that you should have this capability, be sure that you are identified before trying again. The 'whoami' command can tell you if you're identified. 2010-10-09T06:41:29 !more 2010-10-09T06:41:29 Frontier: forget, format, fortune, game, get, ghost, gnu, google, grab, halfop, headers, hebrew, help, hexip, hexlify, hi, hostmask, hostmask add, hostmask list, hostmask remove, icalc, identify, ignore, ignore add, ignore list, ignore remove, info, insert, invite, ircquote, jeffk, join, kban, key, kick, last, latency, learn, leet, len, levenshtein, limit, list, lithp, ln, load, lobotomy add, (3 more messages) 2010-10-09T06:41:29 @whoami 2010-10-09T06:41:30 HakanD: I don't recognize you. 2010-10-09T06:41:52 err... 2010-10-09T06:41:56 !lobotomy 2010-10-09T06:41:57 wh1teside: User error, it's not my fault. 2010-10-09T06:42:06 !remove lobotomy 2010-10-09T06:42:06 sigh: Error: The command "remove" is available in the Alias, BadWords, Dunno, Herald, Later, News, Quote, RSS, Scheduler, Success, Todo, and Topic plugins. Please specify the plugin whose command you wish to call by using its name as a command before "remove". 2010-10-09T06:42:08 !levenshtein 2010-10-09T06:42:08 Frontier: (levenshtein ) -- Returns the levenshtein distance (also known as the "edit distance" between and ) 2010-10-09T06:42:17 @fortune 2010-10-09T06:42:17 HakanD: Error: The fortune command is not configured. If fortune is installed on this system, reconfigure the supybot.plugins.Unix.fortune.command configuration variable appropriately. 2010-10-09T06:42:35 !headers www.google.com 2010-10-09T06:42:36 sigh: x-xss-protection: 1; mode=block, set-cookie: PREF=ID=1fd4df03209a55c2:TM=1286620485:LM=1286620485:S=UUIuI7Kx3J9W7ra-; expires=Mon, 08-Oct-2012 10:34:45 GMT; path=/; domain=.google.ca, NID=39=k7BV7UatQj5RTi9e-32rfKY8kWREa7eI9_78d4j4KhnNL89TJ7MNwLhgNuChKRO2XCU8bexewwZXh18HREsprLiu2bYhkEU0448xZbRsbvpEXB0iBjd1REylPyLhdjWY; expires=Sun, 10-Apr-2011 10:34:45 GMT; path=/; domain=.google.ca; HttpOnly, (1 more message) 2010-10-09T06:42:38 !lobotomy add $randomNick 2010-10-09T06:42:38 Frontier: Error: You don't have the #aichallenge,op capability. If you think that you should have this capability, be sure that you are identified before trying again. The 'whoami' command can tell you if you're identified. 2010-10-09T06:42:58 !levenshtein smile snuke 2010-10-09T06:42:58 wh1teside: 3 2010-10-09T06:43:09 !raninbow $randomNick 2010-10-09T06:43:10 sigh: I'm sorry Dave, I cannot raninbow. 2010-10-09T06:43:11 !headers ai-challenge.com 2010-10-09T06:43:12 Frontier: HTTP Error 401: Authorization Required 2010-10-09T06:43:18 !rainbow $randomNick 2010-10-09T06:43:18 sigh: $randomNick 2010-10-09T06:43:26 !headers ai-contest.com 2010-10-09T06:43:27 Frontier: content-length: 5206, x-powered-by: PHP/5.2.4-2ubuntu5.12, set-cookie: PHPSESSID=242dae88be9eb65fc6eedf6df9b5cb9c; path=/, expires: Thu, 19 Nov 1981 08:52:00 GMT, vary: Accept-Encoding, server: Apache/2.2.8 (Ubuntu) PHP/5.2.4-2ubuntu5.12 with Suhosin-Patch mod_ssl/2.2.8 OpenSSL/0.9.8g mod_perl/2.0.3 Perl/v5.8.8, connection: close, pragma: no-cache, cache-control: no-store, no-cache, must- (1 more message) 2010-10-09T06:43:32 !raninbow [echo $randomNick] 2010-10-09T06:43:33 sigh: User error, it's not my fault. 2010-10-09T06:43:39 !rainbow [echo $randomNick] 2010-10-09T06:43:39 Titankiller 2010-10-09T06:43:54 rainin' bow? 2010-10-09T06:44:24 !rainbow ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 2010-10-09T06:44:24 Frontier: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 2010-10-09T06:45:12 !echo [rankings], $randomNick(3612) 2010-10-09T06:45:13 Top 10 players: rsergio(3831), felixcoto(3818), Hazard(3785), cfaftw(3727), ofrias(3714), sequoh(3690), davidjliu(3671), Killer(3665), EBraun(3639), adumlah(3638) , antimatroid1(3612) 2010-10-09T06:45:20 ^^ still my favorite command 2010-10-09T06:45:36 contestbt: sigh is awesome 2010-10-09T06:45:43 contestbot: sigh is awesome 2010-10-09T06:45:43 i know... 2010-10-09T06:46:11 !rainbow [echo ☀☠☘☁☢☃★♫♲☣☀☠☘☁☢☃★♫♲☣☀☠☘☁☢☃★♫♲☣☀☠☘☁☢☃★♫♲☣☀☠☘☁☢☃★♫♲☣☀☠☘☁☢☃★♫♲☣] 2010-10-09T06:46:11 ☀☠☘☁☢☃★♫♲☣☀☠☘☁☢☃★♫♲☣☀☠☘☁☢☃★♫♲☣☀☠☘☁☢☃★♫♲☣☀ (1 more message) 2010-10-09T06:46:11 there's that one too :P 2010-10-09T06:46:16 o_O 2010-10-09T06:46:19 lol 2010-10-09T06:46:33 was supposed to be some unicode symbols 2010-10-09T06:46:42 !more 2010-10-09T06:46:42 Frontier: ☘☁☢☃★♫♲☣☀☠☘☁☢☃★♫♲☣ 2010-10-09T06:47:45 rankings is cool. you can place yourself first 2010-10-09T06:48:07 !echo Frontier(9876), [rankings] 2010-10-09T06:48:08 Frontier(9876), Top 10 players: rsergio(3831), felixcoto(3818), Hazard(3785), cfaftw(3727), ofrias(3714), sequoh(3690), davidjliu(3671), Killer(3665), EBraun(3639), adumlah(3638) 2010-10-09T06:48:15 haha 2010-10-09T06:48:28 oh.. it is a little obvious 2010-10-09T06:48:44 haha 2010-10-09T06:48:45 fail 2010-10-09T06:49:07 !echo [echo [echo [echo [echo what?] ] ] ] 2010-10-09T06:49:07 what? 2010-10-09T06:49:20 lets cause a stack overflow 2010-10-09T06:49:27 ahh, cycles well wasted 2010-10-09T06:50:20 !echo [echo [echo [echo [echo test 2010-10-09T06:50:20 Frontier: Error: Missing "]". You may want to quote your arguments with double quotes in order to prevent extra brackets from being evaluated as nested commands. 2010-10-09T06:52:02 !echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [ 2010-10-09T06:52:02 Frontier: Error: Missing "]". You may want to quote your arguments with double quotes in order to prevent extra brackets from being evaluated as nested commands. 2010-10-09T06:52:03 [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [ 2010-10-09T06:52:03 [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [ 2010-10-09T06:52:08 [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo [echo Does the stack overflow? ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] 2010-10-09T06:52:13 ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] ] 2010-10-09T06:52:32 damn IRC client, cant you even do haX0rs? 2010-10-09T06:52:55 !cleanupmymess 2010-10-09T06:52:56 Frontier: No! 2010-10-09T06:52:56 heh 2010-10-09T06:53:03 !please 2010-10-09T06:53:04 Frontier: I have no idea what you mean. 2010-10-09T06:53:04 baahah 2010-10-09T06:53:09 :P 2010-10-09T06:55:12 !leet Frontier, thou IRC client hast not the skills 2010-10-09T06:55:12 wh1teside: Fr0n+!3r, +h0u 1RC c1!3n+ h45+ n0+ +h3 5k!11z 2010-10-09T06:57:13 !colorize wheeeee 2010-10-09T06:57:13 wh1teside: wheeeee 2010-10-09T07:00:09 Top 10 players: rsergio(3827), felixcoto(3813), Hazard(3781), cfaftw(3722), ofrias(3709), sequoh(3686), davidjliu(3667), adumlah(3640), EBraun(3636), savchenko(3611) 2010-10-09T07:00:52 !colorize [roulette] 2010-10-09T07:00:52 wh1teside: *click* 2010-10-09T07:01:20 !colorize [roulette] 2010-10-09T07:01:20 delt0r___: *click* 2010-10-09T07:01:24 aww 2010-10-09T07:01:30 its always click? 2010-10-09T07:01:40 only if you "lose" 2010-10-09T07:01:46 lol 2010-10-09T07:01:51 !colorize [roulette] 2010-10-09T07:01:51 delt0r___: *click* 2010-10-09T07:01:53 !colorize [roulette] 2010-10-09T07:01:53 *** delt0r___ was kicked by contestbot (BANG!) 2010-10-09T07:01:54 * contestbot reloads and spins the chambers. 2010-10-09T07:02:08 heh 2010-10-09T07:02:11 * wh1teside mourns 2010-10-09T07:02:12 *** delt0r___ has joined #aichallenge 2010-10-09T07:02:15 !roulette 2010-10-09T07:02:15 rabidus: *click* 2010-10-09T07:02:16 now that was funny 2010-10-09T07:02:23 @roulette 2010-10-09T07:02:23 HakanD: *click* 2010-10-09T07:02:32 oh dear, this is becoming quite an epidemic 2010-10-09T07:02:57 !colorize [roulette] 2010-10-09T07:02:57 wh1teside: *click* 2010-10-09T07:03:04 @roulette 2010-10-09T07:03:04 HakanD: *click* 2010-10-09T07:03:09 !roulette 2010-10-09T07:03:09 *** rabidus was kicked by contestbot (BANG!) 2010-10-09T07:03:10 * contestbot reloads and spins the chambers. 2010-10-09T07:03:17 well i there are not so many channels where a little OT is ok.. and doesn't ruin the channel... and no trolls to f it up... 2010-10-09T07:03:21 its a good channel 2010-10-09T07:03:46 @roulette 2010-10-09T07:03:46 HakanD: *click* 2010-10-09T07:03:58 *** rabidus has joined #aichallenge 2010-10-09T07:04:11 This is the new way to settle flame wars! 2010-10-09T07:04:25 (: 2010-10-09T07:04:27 only with a ban instead of a kick 2010-10-09T07:04:31 ouch 2010-10-09T07:04:55 !nickometer 2010-10-09T07:04:56 wh1teside: No! 2010-10-09T07:05:11 * wh1teside fights tears 2010-10-09T07:05:31 * wh1teside reaches for the pistol 2010-10-09T07:05:37 !colorize [roulette] 2010-10-09T07:05:37 *** wh1teside was kicked by contestbot (BANG!) 2010-10-09T07:05:38 * contestbot reloads and spins the chambers. 2010-10-09T07:06:07 haha 2010-10-09T07:06:23 *** wh1teside has joined #aichallenge 2010-10-09T07:06:52 !colourise british spelling 2010-10-09T07:06:52 delt0r___: I'm sorry Dave, I cannot colourise. 2010-10-09T07:07:09 Americans ! 2010-10-09T07:07:25 !do that 2010-10-09T07:07:26 delt0r___: I'm sorry Dave, I cannot do. 2010-10-09T07:07:32 mm 2010-10-09T07:07:37 *** jesionaj has joined #aichallenge 2010-10-09T07:07:40 should do work instead 2010-10-09T07:09:02 let's say... saving throw for slacking off is 7 2010-10-09T07:09:09 !dice 2d6 2010-10-09T07:09:09 wh1teside: 4 and 2 2010-10-09T07:09:25 that sucks 2010-10-09T07:09:37 did I say seven? I meant six 2010-10-09T07:10:45 *** jesionaj` has quit IRC (Ping timeout: 272 seconds) 2010-10-09T07:12:28 I see a new opportunity for h4x! 2010-10-09T07:12:44 !dice 23d2 2010-10-09T07:12:44 Frontier: Error: You can't roll more than 6 dice. 2010-10-09T07:13:00 no more than 6 ok :( 2010-10-09T07:13:12 I wonder if it can roll lower dimensional dice 2010-10-09T07:13:23 !dice 6d2 2010-10-09T07:13:23 Frontier: Error: Dice can't have fewer than 3 sides. 2010-10-09T07:13:24 !dice 1d3 2010-10-09T07:13:25 wh1teside: 2 2010-10-09T07:14:12 !coin 2010-10-09T07:14:12 sigh: tails 2010-10-09T07:14:38 I guess a cylinder would be a three-sided die 2010-10-09T07:14:59 hard to make it fair 2010-10-09T07:15:15 better to use a 6 sided die, appropriatly marked 2010-10-09T07:16:35 !eightball that thing that sigh just said 2010-10-09T07:16:35 wh1teside: One would be wise to think so. 2010-10-09T07:17:43 *** ichti has joined #aichallenge 2010-10-09T07:18:27 they look like that: http://www.gamescience.com/3-sideddice(d3) 2010-10-09T07:19:41 *** HakanD has quit IRC (Quit: Leaving) 2010-10-09T07:22:20 *** ichti has quit IRC (Ping timeout: 265 seconds) 2010-10-09T07:24:12 ahhh... fair enough 2010-10-09T07:25:17 would have been a bit cooler if the pointy edges weren't reduced to smooth nubs, but cool 2010-10-09T07:26:35 *** ichti has joined #aichallenge 2010-10-09T07:26:50 Heve you ever accidently player a game with a 6-sided dice 3? 2010-10-09T07:27:09 *Have, *played 2010-10-09T07:27:23 *** Meatkat has quit IRC (Ping timeout: 265 seconds) 2010-10-09T07:27:51 !echo [coin][coin] 2010-10-09T07:27:51 heads tails 2010-10-09T07:28:51 !echo $randomNick and $randomNick scratch their [coin]. 2010-10-09T07:28:51 shades and iris1 scratch their heads . 2010-10-09T07:29:05 haha 2010-10-09T07:30:10 way to go 2010-10-09T07:30:15 *** leo2007 has joined #aichallenge 2010-10-09T07:31:02 *** ichti has quit IRC (Ping timeout: 265 seconds) 2010-10-09T07:34:01 *** Blkt has quit IRC (Quit: quit) 2010-10-09T07:35:21 *** ichti has joined #aichallenge 2010-10-09T07:39:03 *** hellman has joined #aichallenge 2010-10-09T07:39:58 *** ichti has quit IRC (Ping timeout: 265 seconds) 2010-10-09T07:40:39 !fight "ai challenge" "leave the house" 2010-10-09T07:40:40 wh1teside: leave the house: 110000000, ai challenge: 2990000 2010-10-09T07:40:56 :( 2010-10-09T07:42:49 *** leo2007 has left #aichallenge ("rcirc on GNU Emacs 23.2.20") 2010-10-09T07:43:42 *** JensT1 has joined #aichallenge 2010-10-09T07:43:47 *** ichti has joined #aichallenge 2010-10-09T07:48:39 *** ichti has quit IRC (Ping timeout: 264 seconds) 2010-10-09T08:00:09 Top 10 players: rsergio(3824), felixcoto(3811), Hazard(3778), cfaftw(3720), ofrias(3703), sequoh(3681), davidjliu(3665), adumlah(3639), EBraun(3635), savchenko(3610) 2010-10-09T08:06:03 *** JensT1 has quit IRC (Quit: Leaving.) 2010-10-09T08:09:22 *** HakanD has joined #aichallenge 2010-10-09T08:11:49 *** delt0r_ has joined #aichallenge 2010-10-09T08:13:22 *** delt0r___ has quit IRC (Ping timeout: 245 seconds) 2010-10-09T08:16:27 *** Naktibalda has joined #aichallenge 2010-10-09T08:22:22 *** Obeleh has joined #aichallenge 2010-10-09T08:23:05 I just switch from Java to C# and I got some weird compile warning: http://pastebin.com/jLHcXchn 2010-10-09T08:23:43 Does anyone have an idea what to do? Looks to me I should contact an admin 2010-10-09T08:25:05 yep, 'No space left on device.' is probably not your fault 2010-10-09T08:26:32 *** Itkovian has joined #aichallenge 2010-10-09T08:27:28 *** Tom_K has joined #aichallenge 2010-10-09T08:28:19 *** Titankiller has quit IRC (Quit: Nettalk6 - www.ntalk.de) 2010-10-09T08:28:30 *** Tom_K has left #aichallenge 2010-10-09T08:29:51 What is a NearestItr? 2010-10-09T08:31:30 *** Bobng has joined #aichallenge 2010-10-09T08:32:24 Obeleh: C# is not so well supported on Linux, especially the contest server is using an old version of Mono. Under these circumstances I would have sticked with Java. 2010-10-09T08:33:20 NearestItr is a class I wrote to iterate through planets starting with the nearest and going further 2010-10-09T08:33:50 Yes I knew mono isn't that stable but I started with Java because there was no C# in the beginning 2010-10-09T08:34:28 And I know C# better than java 2010-10-09T08:37:14 Frontier: Aparently this could have been fixed with a patch: https://bugzilla.novell.com/show_bug.cgi?id=588689 2010-10-09T08:39:13 I hope they update mono for you C# guys. It is a disadvantage. If the language wasn't able to produce fast enough code for the 1 second time limit that's one thing, but the code should at least compile ^^ 2010-10-09T08:41:15 On another note I think about running automated test cases on my bot. I save a few game states in which my bot behaves odd, and when I think I fixed the issue I run it again against that game state and others I previously saved so I don't run into regressions. 2010-10-09T08:44:10 have you implemented any? 2010-10-09T08:45:05 starting conditions - planets, fleets, future orders, etc are huge, maintaining them can take a lot of time 2010-10-09T08:46:35 I have no future orders to save. I thought I could use the format that the server sends to the client on round start. 2010-10-09T08:46:40 well, results are smaller, just a few orders, but small changes to code can affect a lot of testcases 2010-10-09T08:47:33 From there on I could add a note "planet 4 or 6 should be attacked here" and let the bot decide if it wants to do that. 2010-10-09T08:48:07 later you can improve your bot to think that planet 8 is the best target 2010-10-09T08:48:23 or that he has to wait for 2 turns before attack 2010-10-09T08:48:52 hmm... so you think I would spend a lot of time fixing up or deleting obsolete test cases? 2010-10-09T08:49:14 * Naktibalda unit-tested XML generation this week, what a waste of time 2010-10-09T08:50:09 Then again maybe I wouldn't use a test case for decisions that aren't obvious. 2010-10-09T08:52:17 It's the situations where every human would ask why the bot isn't doing xy that give me head aches and I want to go sure I don't repeat such errors when I tune the bot. 2010-10-09T08:55:57 Something like "[don't] send x ships [from/to] planet y in the next z turn" should suffice for a test case. 2010-10-09T09:00:09 Top 10 players: rsergio(3823), felixcoto(3804), Hazard(3775), cfaftw(3718), ofrias(3700), sequoh(3680), davidjliu(3662), adumlah(3641), EBraun(3638), Mistmanovx(3614) 2010-10-09T09:01:20 Or another great idea: Everyone has this 'evaluate planet' function with a few constants to tune. How about predicting what planet the enemy will attack using this function and if the prediction was wrong, tune these constants until you can make a more reliable prediction of the enemy's moves. 2010-10-09T09:03:54 Frontier: i do this 2010-10-09T09:04:07 But i have too many parameters to tune 2010-10-09T09:04:32 and i still have a bug in my prediction code... so it does stupid things at the end game 2010-10-09T09:04:36 So you use the constant function you use for yourself on the enemy? 2010-10-09T09:04:41 yea 2010-10-09T09:04:48 min/max kinda thing 2010-10-09T09:05:14 but then i rank my descisions on possible counters 2010-10-09T09:05:36 still without other hurisitcs it doesn't work well 2010-10-09T09:05:56 What language do you use? 2010-10-09T09:06:01 so things like working out a front line and sending fleets there... 2010-10-09T09:06:07 Java 2010-10-09T09:06:59 interesting, tuning the parameters might work for me, I'm using C++. After all, you can abort the tuning once you run out of time. 2010-10-09T09:06:59 Its just what i have been using lately 2010-10-09T09:07:08 yea 2010-10-09T09:07:13 I do that 2010-10-09T09:07:25 but i abort at 0.5 sec at this stage 2010-10-09T09:08:05 you use that long? 2010-10-09T09:08:49 when is this competition's deadline? 2010-10-09T09:08:54 It is an evolutionary algorith. It has no limit 2010-10-09T09:09:01 *** kire has joined #aichallenge 2010-10-09T09:09:10 27. november 2010-10-09T09:09:23 oh, that's still pretty long 2010-10-09T09:09:57 mind the timezone. for me it is 7 o'clock in the morning of the next day 2010-10-09T09:10:30 what timezone are you in? 2010-10-09T09:10:46 central europe 2010-10-09T09:11:42 same 2010-10-09T09:11:54 Frontier, you're using GA to direct your bot? 2010-10-09T09:12:30 ok same here too 2010-10-09T09:12:42 Well i don't need to use 0.5 --I come up with a plan --then run through permutations to find something better.... There are a lot of permertations and it doesn't improve the greedy aproch much 2010-10-09T09:12:59 you shouldn't listen to me anyways... I am getting my arse kicked 2010-10-09T09:13:32 delt0r, evolutionary algorithms usually require hours or days, not 0.5 second :P 2010-10-09T09:13:53 oPless, no I don't. I'm not sure it will help, but I might try it for auto-tuning the parameters of my bot. 2010-10-09T09:14:07 Oh--yea I am not doing that... I am doing the "predict enemy based on same rules i use" 2010-10-09T09:15:00 I'm cutting some code at the moment that's taking a GA style approach, not sure how good it will be though :) 2010-10-09T09:16:23 I'm always spending at least 2 hours changing some stupid values in my bot and test it against its precedessor. That's work a script can do for me :) 2010-10-09T09:17:09 yeah ... compile in monodevelop, type make ... then check run 2010-10-09T09:17:23 *** BtbN has joined #aichallenge 2010-10-09T09:17:29 *** felixcoto has joined #aichallenge 2010-10-09T09:18:43 when get organizations updated? I thought this would go a bit faster now there is a script? 2010-10-09T09:19:34 *** mceier has joined #aichallenge 2010-10-09T09:20:16 *** felixcoto has quit IRC (Client Quit) 2010-10-09T09:21:58 *** jesionaj` has joined #aichallenge 2010-10-09T09:22:33 *** iFire has quit IRC (Read error: Connection reset by peer) 2010-10-09T09:22:47 *** iFire has joined #aichallenge 2010-10-09T09:25:16 *** jesionaj` has quit IRC (Read error: Connection reset by peer) 2010-10-09T09:25:28 *** jesionaj has quit IRC (Ping timeout: 276 seconds) 2010-10-09T09:25:40 *** jesionaj has joined #aichallenge 2010-10-09T09:27:33 *** jesionaj` has joined #aichallenge 2010-10-09T09:30:08 *** jesionaj has quit IRC (Ping timeout: 240 seconds) 2010-10-09T09:33:09 *** davidjliu has joined #aichallenge 2010-10-09T09:34:24 interesting game 2010-10-09T09:34:25 http://72.44.46.68/canvas?game_id=389123 2010-10-09T09:37:03 *** JensT1 has joined #aichallenge 2010-10-09T09:44:31 stackoverflow maintenance -_- 2010-10-09T09:45:50 *** jesionaj has joined #aichallenge 2010-10-09T09:48:30 *** jesionaj` has quit IRC (Ping timeout: 240 seconds) 2010-10-09T09:57:11 suggestiosn for a good profiler for java ? 2010-10-09T09:59:33 bhasker, how do you get a user/password on 72.44.46.68 ? 2010-10-09T09:59:42 i don't use one 2010-10-09T09:59:52 i think you just pass in a passowrd the first time 2010-10-09T09:59:55 with -p 2010-10-09T09:59:58 or w/e option 2010-10-09T10:00:03 and future attempts it has to match 2010-10-09T10:00:07 if you don't give one ever 2010-10-09T10:00:11 Top 10 players: felixcoto(3794), rsergio(3781), Hazard(3758), cfaftw(3708), ofrias(3687), sequoh(3658), davidjliu(3636), EBraun(3609), adumlah(3609), Mistmanovx(3580) 2010-10-09T10:00:12 it never asks you for it 2010-10-09T10:00:52 I see. 2010-10-09T10:06:06 Hmm, is j3camero on holiday or something@? 2010-10-09T10:09:45 @last seen j3camero 2010-10-09T10:09:45 HakanD: (last [--{from,in,on,with,without,regexp} ] [--nolimit]) -- Returns the last message matching the given criteria. --from requires a nick from whom the message came; --in requires a channel the message was sent to; --on requires a network the message was sent on; --with requires some string that had to be in the message; --regexp requires a regular expression the message must match; (1 more message) 2010-10-09T10:10:24 hmm cant remember the command (: 2010-10-09T10:11:50 !seen j3camero 2010-10-09T10:11:50 rabidus: j3camero was last seen in #aichallenge 4 days, 11 hours, 52 minutes, and 5 seconds ago: amstan: yes, but no names, haha. 2010-10-09T10:12:13 *** noid63464 has joined #aichallenge 2010-10-09T10:24:45 *** Obeleh has quit IRC (Quit: Page closed) 2010-10-09T10:29:00 *** flavor has quit IRC (Ping timeout: 272 seconds) 2010-10-09T10:29:24 *** Appleman1234 has quit IRC (Ping timeout: 265 seconds) 2010-10-09T10:53:26 hmmm... I submitted a new bot and it's been over 15 minutes and it's still not acknowledged on my recent submissions list :( 2010-10-09T10:55:36 *** mega1 has joined #aichallenge 2010-10-09T10:58:14 *** mceier has quit IRC (Quit: leaving) 2010-10-09T10:58:40 *** amstan has quit IRC (Ping timeout: 260 seconds) 2010-10-09T10:59:31 *** Bobng has quit IRC (Quit: Leaving) 2010-10-09T11:00:12 Top 10 players: felixcoto(3798), rsergio(3784), Hazard(3760), cfaftw(3709), ofrias(3695), sequoh(3656), davidjliu(3641), EBraun(3632), adumlah(3628), kcm1700(3595) 2010-10-09T11:06:53 ok, figured out the problem, bot is running now 2010-10-09T11:08:12 *** rwa has quit IRC (Quit: rwa) 2010-10-09T11:08:23 sigh: what was the problem? 2010-10-09T11:08:58 for some reason I had changed MyBot.cc to MyBot.cpp and the compiler didn't like that :P 2010-10-09T11:09:32 it didn't recognise the language so it didn't even so up on the submissions page, but I did get an email 2010-10-09T11:09:39 *** rwa has joined #aichallenge 2010-10-09T11:10:34 *** rwa has joined #aichallenge 2010-10-09T11:10:55 now I've got a bot with the awesome rank of 1226 2010-10-09T11:11:37 ok, time to sleep... I'll see how my bot has faired in the morning 2010-10-09T11:11:52 The build envoiurment is pretty sensative 2010-10-09T11:12:18 *** HakanD has quit IRC (Ping timeout: 264 seconds) 2010-10-09T11:13:22 *** sigh has quit IRC (Quit: Leaving...) 2010-10-09T11:15:06 *** thearchangel has joined #aichallenge 2010-10-09T11:15:29 are matches taking place? 2010-10-09T11:15:50 its been one hr since my last upload,no match yet.. 2010-10-09T11:15:51 :( 2010-10-09T11:19:35 *** thearchangel has quit IRC (Client Quit) 2010-10-09T11:24:07 bartwe: A good profiler is JProfiler. It is not free though. But I have experience with it and have to say that it is very comprehensive. 2010-10-09T11:26:29 thearchangel: I had my second mactch after 2 hrs 30 min :) 2010-10-09T11:26:47 *** hellman has quit IRC (Quit: buffer overflow) 2010-10-09T11:28:55 Hey, I just beat a guy from Micro$oft with my awesome Linux bot creation environment! 2010-10-09T11:29:14 (And he is using.. no C#, but Java ^^) 2010-10-09T11:30:44 *** Titankiller has joined #aichallenge 2010-10-09T11:31:54 *** Bobng has joined #aichallenge 2010-10-09T11:34:43 grats! 2010-10-09T11:34:53 matches do seem to be slow on the servers 2010-10-09T11:35:08 http://www.ai-contest.com/game_stats.php 2010-10-09T11:35:13 5.8 games/minute 2010-10-09T11:35:23 only one server instance is up 2010-10-09T11:36:10 it can change every hour or so 2010-10-09T11:36:19 there is some automation in place to balance price and CPU 2010-10-09T11:36:41 i think its been like that last few hours or more 2010-10-09T11:37:17 jmc what is your bots name? 2010-10-09T11:39:31 *** Itkovian has quit IRC (Quit: Itkovian) 2010-10-09T11:39:47 crow, worm... something like that usually ;) 2010-10-09T11:39:57 oh no wait that was leopold 2010-10-09T11:41:20 *** perror has joined #aichallenge 2010-10-09T11:43:04 I would quite like to donate a VPS to ai-contest 2010-10-09T11:43:13 crowbot is leopold 2010-10-09T11:43:25 who should I talk to? j3camero seems to be permanently away 2010-10-09T11:45:42 prolly danielvf, who is on vacation 2010-10-09T11:47:59 Let's do the math: We have 3000 players, they all have bots that play 200 rounds in each game and think about their turn for one second. So that would be 3000*200 seconds or 167 hours for one iteration of the player list :D 2010-10-09T11:49:51 *** fluffluff1 has joined #aichallenge 2010-10-09T11:50:49 *** davidjliu has quit IRC (Ping timeout: 265 seconds) 2010-10-09T11:50:55 We get 50,000 x more games than that per hour, so lets be thankful ! 2010-10-09T11:51:14 hi, does anyone know what the memory limit for a bot is? 2010-10-09T11:51:21 700 MB 2010-10-09T11:51:25 ty 2010-10-09T11:51:38 Just guessing. Try to stay below ^^ 2010-10-09T11:51:51 There is nothing official I think. 2010-10-09T11:52:04 isn't javas default something like 64Mb ? 2010-10-09T11:52:14 or am I old? 2010-10-09T11:52:17 oh... odd. I'd think many people could use a whole lot 2010-10-09T11:52:22 there was a limit on the sadbox, but i think it was removed when all the java entries started crashing 2010-10-09T11:52:26 opless: you're old :) 2010-10-09T11:52:32 depends on a machine 2010-10-09T11:52:38 I just heard about a Java bot that reached the Java VM limit and they gave it 700 MB 2010-10-09T11:52:50 oic :) 2010-10-09T11:53:12 thinking about writing some actual test code 2010-10-09T11:53:46 *** yellowseed has quit IRC (Remote host closed the connection) 2010-10-09T11:53:55 jmcarthur: tell me more. Do you read out the memory limit and then attack the last planet with as many ships? ^^ 2010-10-09T11:54:08 "read out the memory limit"? 2010-10-09T11:54:30 I think in Java you can query the VM memory limit. 2010-10-09T11:54:36 i'm not using java 2010-10-09T11:54:53 So what will your test code do? 2010-10-09T11:54:57 by test code i meant unit tests 2010-10-09T11:55:13 unrelated to what we were talking about 2010-10-09T11:55:36 was going to ask if anybody else is doing unit tests and what kinds of things they are testing if so 2010-10-09T11:55:39 Like what I said 3 hours ago?: Frontier On another note I think about running automated test cases on my bot. I save a few game states in which my bot behaves odd, and when I think I fixed the issue I run it again against that game state and others I previously saved so I don't run into regressions. 2010-10-09T11:55:55 ah that won't work for my bot anyway 2010-10-09T11:56:10 it's nondeterministic and builds up state over the course of many turns 2010-10-09T11:56:44 oh.. i feel like my bot is very rudimentary now ^^ 2010-10-09T11:57:49 heh 2010-10-09T11:57:53 but at least it's running :P 2010-10-09T11:58:11 What does this state account for? Couldn't you do without for a test case? 2010-10-09T11:58:45 i could test some specific functions, but i don't know how to test the nondeterministic behavior of my bot 2010-10-09T11:59:28 just.. make it deterministic. Put in some default state. 2010-10-09T11:59:43 Do as if it was the first round of the game. 2010-10-09T12:00:01 actually the first round i might make deterministic 2010-10-09T12:00:06 with special case code 2010-10-09T12:00:12 Top 10 players: felixcoto(3802), rsergio(3772), Hazard(3757), cfaftw(3700), ofrias(3699), sequoh(3648), davidjliu(3648), EBraun(3610), adumlah(3610), kcm1700(3603) 2010-10-09T12:00:33 but the other turns are nondeterministic even with a known initial state 2010-10-09T12:01:12 I think good idea is to start testing individual methods, regarding unit testing 2010-10-09T12:01:25 I mean, not testing for a specific response for a given game state 2010-10-09T12:01:26 i mean, there is a way i could make it deterministic (using a known seed, running my algorithm for a known number of iterations per turn, etc.) 2010-10-09T12:01:36 wtf? how does this work under the hood? you cannot know anything in the first round so why is there a chance it will behave differently 2010-10-09T12:01:39 krokokrusa: right, that's the kind of thing i was planning to do 2010-10-09T12:01:52 Frontier: what do you mean not know anything? 2010-10-09T12:03:19 jmcarthur: It doesn't know about the previous rounds, where its defense was weak, what the enemy is doing. You just put in some random numbers to make it non-deterministic and give others trouble to understand how to beat your bot ^^ 2010-10-09T12:03:34 uh 2010-10-09T12:03:57 i think you are making some broad assumptions about where the nondeterminism from my bot is coming from 2010-10-09T12:04:22 ;) 2010-10-09T12:04:34 i'm not just "putting in some random numbers" 2010-10-09T12:04:49 it's possible to iteratively search for a solution for 1 second, and then execute the best move 2010-10-09T12:04:57 right 2010-10-09T12:05:04 then even on he 1st turn there is a place for some non-determinism 2010-10-09T12:05:08 that's essentially what i'm doing 2010-10-09T12:05:17 on top of that, the search is somewhat stochastic 2010-10-09T12:05:37 ok, that explains why the server is so slow now ^^ 2010-10-09T12:05:45 i'm not even running on the server yet 2010-10-09T12:05:52 cool, I'm yet to implement the searching :) 2010-10-09T12:05:53 and no, that doesn't explain it 2010-10-09T12:08:18 can i edit planetwars.java in the starter pack ? 2010-10-09T12:08:53 Well if the top 100 players always end in a draw and think about the best move for 1 second that means 20,000 CPU seconds for their games ;-) And that is about 5,6 hours divided by number of CPUs. But it happened to suddenly now, I don't believe myself that it is the problem, but it could become one. 2010-10-09T12:09:07 Titankiller: sure, why not 2010-10-09T12:09:25 Titankiller: and you can add more files 2010-10-09T12:09:30 any external libraries ? 2010-10-09T12:09:58 hmm... try to stick with .java files and no directory structure. 2010-10-09T12:10:27 Frontier: that's a big "if" 2010-10-09T12:10:42 Frontier: the problem with the server is that there is only one server running games right now 2010-10-09T12:10:53 Frontier: we have been getting our speed by running the games on many servers 2010-10-09T12:11:01 it's pretty easily checkable that the reason games are being run slow is because none of the cloud servers are up 2010-10-09T12:11:19 How many are there usually running? 2010-10-09T12:11:23 http://ai-contest.com/game_stats.php 2010-10-09T12:11:32 2-4 servers 2010-10-09T12:11:33 Frontier: it's variable, but it's usually around 3 i think 2010-10-09T12:11:47 Frontier: the one server running right now is much slower than the others 2010-10-09T12:18:48 As I said, I am willing to donate a VPS 2010-10-09T12:19:04 to help increase the computing power, but nobody seems to be around. 2010-10-09T12:20:55 *** JensT1 has quit IRC (Quit: Leaving.) 2010-10-09T12:23:54 Bobng: be aware unless you can give up all ability to access it (e.g. through the host OS) that will forfeit your ability to participate 2010-10-09T12:24:38 yes, I have been told that several times 2010-10-09T12:24:42 I don't participate anyway 2010-10-09T12:24:53 plus I won't have access. 2010-10-09T12:24:59 I have a feeling the reason for lack of servers running right now is a technical failure and lack of people to fix it not lack of resources 2010-10-09T12:31:35 *** criis has joined #aichallenge 2010-10-09T12:39:35 Janzert: i think it's cost. the cloud servers have variable pricing, iirc 2010-10-09T12:39:52 Janzert: so daniel put a mechanism in place to only use them when their prices are down, i think 2010-10-09T12:43:23 *** BtbN has quit IRC (Quit: Verlassend) 2010-10-09T12:45:50 what's wrong with TCP? 2010-10-09T12:58:16 *** yasith has quit IRC (Ping timeout: 265 seconds) 2010-10-09T13:00:12 Top 10 players: felixcoto(3794), rsergio(3770), Hazard(3754), ofrias(3697), cfaftw(3696), davidjliu(3652), sequoh(3652), EBraun(3619), adumlah(3599), kcm1700(3595) 2010-10-09T13:04:40 *** kire has quit IRC (Quit: Lost terminal) 2010-10-09T13:04:41 TCP offers only matches against people who are currently online and testing their bot. The official servers shows long term results against all participants. 2010-10-09T13:05:59 *** Titankiller has quit IRC (Quit: Nettalk6 - www.ntalk.de) 2010-10-09T13:10:10 *** yasith has joined #aichallenge 2010-10-09T13:12:02 *** CIA-78 has quit IRC (Ping timeout: 265 seconds) 2010-10-09T13:24:10 *** choas has joined #aichallenge 2010-10-09T13:34:30 *** wh1teside has quit IRC (Quit: Lost terminal) 2010-10-09T13:35:09 *** CIA-101 has joined #aichallenge 2010-10-09T13:50:52 *** ichti has joined #aichallenge 2010-10-09T14:00:13 Top 10 players: felixcoto(3785), rsergio(3761), Hazard(3746), ofrias(3689), cfaftw(3689), sequoh(3647), davidjliu(3641), EBraun(3608), kcm1700(3592), adumlah(3583) 2010-10-09T14:04:37 *** Hs has joined #aichallenge 2010-10-09T14:04:46 Hi everyone 2010-10-09T14:06:16 *** wh1teside has joined #aichallenge 2010-10-09T14:12:09 *** jesionaj has quit IRC (Read error: Connection reset by peer) 2010-10-09T14:12:53 *** Hs has quit IRC (Quit: Page closed) 2010-10-09T14:17:15 *** mceier has joined #aichallenge 2010-10-09T14:21:27 *** Bobng has quit IRC (Quit: Leaving) 2010-10-09T14:22:18 *** paddlesteamer has joined #aichallenge 2010-10-09T14:25:01 *** RainCT has joined #aichallenge 2010-10-09T14:29:16 *** Bobng has joined #aichallenge 2010-10-09T14:34:29 *** Bobng has quit IRC (Ping timeout: 255 seconds) 2010-10-09T14:35:19 *** perror has quit IRC (Quit: Bye all !) 2010-10-09T14:36:07 *** Bobng has joined #aichallenge 2010-10-09T14:39:58 *** Bobng has joined #aichallenge 2010-10-09T14:49:47 *** Bobng has quit IRC (Ping timeout: 255 seconds) 2010-10-09T14:50:04 *** optimum has quit IRC (Ping timeout: 245 seconds) 2010-10-09T14:54:22 *** optimum has joined #aichallenge 2010-10-09T14:59:53 *** Bobng has joined #aichallenge 2010-10-09T15:00:14 Top 10 players: felixcoto(3778), rsergio(3759), Hazard(3735), ofrias(3694), cfaftw(3685), sequoh(3648), davidjliu(3642), EBraun(3605), kcm1700(3580), adumlah(3579) 2010-10-09T15:01:41 *** Cyndre has quit IRC (Quit: Leaving) 2010-10-09T15:03:06 *** d3c1ph3r has joined #aichallenge 2010-10-09T15:05:24 Hey question? Whats the difference between your bot being player in the first position or second position for where it calls the jar, it wins under 2 position, it totally screws up at 1, since I've been debugging as 2 I didn't notice. What variables change? is it the owner id? why would second position be player 1 2010-10-09T15:07:20 the id you see is always 1, but your starting planet is not always id 1 2010-10-09T15:07:37 when you are "player 2" to the engine, your starting planet is id 2 2010-10-09T15:08:04 hmm 2010-10-09T15:08:06 could that be it 2010-10-09T15:08:21 i wouldn't imagine i would automatically assume starting position usually i traverse 2010-10-09T15:08:37 that's the only thing i can think of 2010-10-09T15:09:18 *** Cyndre has joined #aichallenge 2010-10-09T15:10:28 also what other things could cause an illegal move, like sending more troops then ur planet has, sending less then zero? is there anything else that illegal besides attacking when enemyplanets == 0; 2010-10-09T15:11:48 attacking when enemyplanets == 0 isn't an illegal move 2010-10-09T15:12:02 sending more ships than you have, sending less than zero, sending ships from a planet to itself 2010-10-09T15:12:06 sending from planet you don't own is illegal 2010-10-09T15:12:25 sending to planets that don't exist 2010-10-09T15:13:18 also keep in mind that there is a subtle point about sending more ships from a planet than are currently there. because you can send more than one fleet in a turn, you have to consider the sum of all fleets leaving a planet, not just each one separately 2010-10-09T15:14:40 sending a fractional number of ships is illegal too 2010-10-09T15:14:49 a comma-separated value is also an illegal move 2010-10-09T15:14:55 perfect this is all really helpfull 2010-10-09T15:15:33 my program fails at me issuing my orders so i'm trying to debug it to see what would cause the illegal move 2010-10-09T15:15:52 right before i win it fails sometimes 2010-10-09T15:16:12 right before you win? 2010-10-09T15:16:21 yay i have some unit testing! 2010-10-09T15:16:32 perhaps you are attempting to do something with an enemy planet when they have none? 2010-10-09T15:16:35 yeah ill be attacking my last planet, and right before the ships get there boom game over my ships dissapear 2010-10-09T15:17:05 *** Bobng has quit IRC (Read error: Connection reset by peer) 2010-10-09T15:17:20 *** Bobng has joined #aichallenge 2010-10-09T15:17:39 you're using playgame.jar? what file are you logging to? 2010-10-09T15:17:43 * jmcarthur tests his parsing over a couple million maps while he works on other stuff... just to be sure, of course 2010-10-09T15:17:44 look there to see what you said 2010-10-09T15:18:35 this is my command in console 2010-10-09T15:18:37 C:\Program Files (x86)\Java\jre6\bin>java -jar tools/PlayGame.jar maps/map7.txt 1000 1000 log.txt "java -jar example_bots/bullyBot.jar" "java -jar example_bots/ google.jar" | java -jar tools/ShowGame.jar 2010-10-09T15:19:21 check log.txt to see what your bot said to do 2010-10-09T15:19:40 change log.txt to log2.txt and run it again, then check it 2010-10-09T15:19:44 *** Bobng has quit IRC (Client Quit) 2010-10-09T15:19:59 "C:\Program Files (x86)\Java\jre6\bin" .... o_O 2010-10-09T15:20:13 you'll want to look for lines that say "player2 > engine" 2010-10-09T15:21:00 where is log kept 2010-10-09T15:21:08 *** Bobng has joined #aichallenge 2010-10-09T15:21:35 your current directory 2010-10-09T15:23:33 log no where to be found :( i've been writing to files after every game state to debug but i can never find the engine log 2010-10-09T15:24:18 *** noid63464 has quit IRC () 2010-10-09T15:25:11 is their a max amount of fleets? 2010-10-09T15:25:43 nah 2010-10-09T15:26:33 woops 2010-10-09T15:26:48 i can't overflow it? 2010-10-09T15:27:23 only in your bot code :) 2010-10-09T15:27:55 lol 2010-10-09T15:28:03 was testing my parsing code. my random map generator also generates comments just to make sure my parser handles it properly. it generates progressively larger data sets as it goes. since i set the number of cases to test to a million is was generating... very large comments. i hit swap and found myself unable to do anything for a bit there :\ 2010-10-09T15:28:57 actually it's not the map generator, it's just the tester that adds the comments, but whatever 2010-10-09T15:29:08 *** Bobng has quit IRC (Quit: Leaving) 2010-10-09T15:30:23 i'm so frustrated i have a pretty good bot here, and before i can add more plys to my tree or fix my evaluation function i need to find all my bugs, and theres one thing here crashing it driving me nuts seems like i gotta get out the paper and pen and manual run through it 2010-10-09T15:30:57 make sure you do it for an endgame situation since that's where it seems to be crashing 2010-10-09T15:31:04 don't do an entire game :o 2010-10-09T15:31:26 yay for bugs 2010-10-09T15:33:57 haha yay! 2010-10-09T15:34:06 whats the winner() method do 2010-10-09T15:34:25 *** Bobng has joined #aichallenge 2010-10-09T15:35:20 *** noid63464 has joined #aichallenge 2010-10-09T15:35:23 *** Bobng has quit IRC (Client Quit) 2010-10-09T15:36:16 *** noid63464 is now known as Accoun 2010-10-09T15:40:59 return true; :) 2010-10-09T15:47:13 *** dmj111 has quit IRC (Remote host closed the connection) 2010-10-09T15:48:53 *** rwa has quit IRC (Remote host closed the connection) 2010-10-09T15:49:31 *** Migi32 has quit IRC (Remote host closed the connection) 2010-10-09T15:49:56 Winner() in C# never works, alas 2010-10-09T15:50:13 I'm test different versions of my bot locally and top shows that PlayGame.jar takes 100% while my bots 10% cpu each. 2010-10-09T15:50:25 is there a faster version of PlayGame.jar? 2010-10-09T15:50:33 (I'm using a1k0n's v1.2) 2010-10-09T15:50:41 *testing 2010-10-09T15:53:10 If they run this on the servers they may be paying hard cash for a few lines of unoptimized IO. 2010-10-09T15:54:15 on they are running a python version 2010-10-09T15:54:17 mega1: you might find this interesting: http://ai-contest.com/forum/viewtopic.php?f=19&t=782&p=5343#p5343 2010-10-09T15:54:23 *no they 2010-10-09T15:54:39 ? 2010-10-09T15:56:18 ? 2010-10-09T15:58:59 *** krokokrusa has quit IRC (Quit: going to sleep) 2010-10-09T16:00:15 Top 10 players: felixcoto(3778), rsergio(3772), Hazard(3733), cfaftw(3694), ofrias(3689), davidjliu(3637), sequoh(3621), EBraun(3614), fglider(3596), adumlah(3577) 2010-10-09T16:03:21 *** Itkovian has joined #aichallenge 2010-10-09T16:03:33 jmcarthur: got you, just missed that your previous comment first 2010-10-09T16:04:08 *** Itkovian has quit IRC (Client Quit) 2010-10-09T16:05:55 *** Itkovian has joined #aichallenge 2010-10-09T16:05:56 *** Palmik has quit IRC (Read error: Connection reset by peer) 2010-10-09T16:07:02 ah 2010-10-09T16:07:57 *** d3c1ph3r has quit IRC (Ping timeout: 265 seconds) 2010-10-09T16:09:22 *** Mathnerd314 has joined #aichallenge 2010-10-09T16:11:42 *** Bobng has joined #aichallenge 2010-10-09T16:12:22 *** delt0r___ has joined #aichallenge 2010-10-09T16:13:56 *** delt0r_ has quit IRC (Ping timeout: 255 seconds) 2010-10-09T16:16:32 *** Bobng has quit IRC (Client Quit) 2010-10-09T16:16:52 *** Bobng has joined #aichallenge 2010-10-09T16:17:31 who hollard 2010-10-09T16:20:59 *** nigg_ has joined #aichallenge 2010-10-09T16:24:14 *** Bobng has quit IRC (Ping timeout: 276 seconds) 2010-10-09T16:28:22 *** nigg_ has quit IRC (Read error: Connection reset by peer) 2010-10-09T16:28:45 *** nrub has joined #aichallenge 2010-10-09T16:28:50 *** nigg_ has joined #aichallenge 2010-10-09T16:31:30 *** nrub_ has quit IRC (Ping timeout: 240 seconds) 2010-10-09T16:33:24 *** optimum has quit IRC (Ping timeout: 245 seconds) 2010-10-09T16:34:27 *** optimum has joined #aichallenge 2010-10-09T16:40:53 *** yasith has quit IRC (Ping timeout: 264 seconds) 2010-10-09T16:41:31 *** Braithy85 has joined #aichallenge 2010-10-09T16:41:32 man, not having scala support sucks. 2010-10-09T16:42:58 *** Blkt has joined #aichallenge 2010-10-09T16:44:16 hey all, does anyone know if theres any fix for the c# "error while compiling submission" bug? 2010-10-09T16:45:57 *** Migi32 has joined #aichallenge 2010-10-09T16:54:32 *** yasith has joined #aichallenge 2010-10-09T17:00:14 Top 10 players: felixcoto(3795), rsergio(3770), Hazard(3752), ofrias(3705), cfaftw(3698), davidjliu(3649), fglider(3633), sequoh(3628), EBraun(3623), Rexxar(3604) 2010-10-09T17:02:41 *** yasith has quit IRC (Ping timeout: 252 seconds) 2010-10-09T17:03:59 *** RainCT has quit IRC (Remote host closed the connection) 2010-10-09T17:10:36 *** Itkovian has quit IRC (Quit: Itkovian) 2010-10-09T17:19:06 *** yasith has joined #aichallenge 2010-10-09T17:19:53 *** choas has quit IRC (Quit: leaving) 2010-10-09T17:20:38 *** Accoun has quit IRC (Ping timeout: 240 seconds) 2010-10-09T17:22:48 *** Accoun has joined #aichallenge 2010-10-09T17:24:15 hmm 2010-10-09T17:28:26 *** Janzert has left #aichallenge 2010-10-09T17:33:19 *** Zeiris_ has quit IRC (Ping timeout: 265 seconds) 2010-10-09T17:37:11 *** Accoun has quit IRC (Ping timeout: 265 seconds) 2010-10-09T17:39:02 *** davidjliu has joined #aichallenge 2010-10-09T17:42:14 *** yasith has quit IRC (Ping timeout: 250 seconds) 2010-10-09T17:45:01 *** noid48962 has joined #aichallenge 2010-10-09T17:45:56 *** noid48962 is now known as Acount 2010-10-09T17:54:24 *** Janzert has joined #aichallenge 2010-10-09T17:55:10 *** ichti has quit IRC (Ping timeout: 240 seconds) 2010-10-09T18:00:15 Top 10 players: felixcoto(3793), rsergio(3777), Hazard(3759), ofrias(3700), cfaftw(3699), davidjliu(3658), sequoh(3637), EBraun(3635), Rexxar(3630), Killer(3624) 2010-10-09T18:00:34 *** yasith has joined #aichallenge 2010-10-09T18:07:44 *** epona has joined #aichallenge 2010-10-09T18:12:11 *** ichti has joined #aichallenge 2010-10-09T18:13:11 *** Braithy85 has quit IRC (Quit: Page closed) 2010-10-09T18:16:11 *** ichti has quit IRC (Ping timeout: 240 seconds) 2010-10-09T18:18:36 *** millertime has left #aichallenge ("WeeChat 0.3.2") 2010-10-09T18:20:56 *** ichti has joined #aichallenge 2010-10-09T18:23:06 *** fawek has quit IRC (Ping timeout: 265 seconds) 2010-10-09T18:23:49 *** sigh has joined #aichallenge 2010-10-09T18:25:44 *** ichti has quit IRC (Ping timeout: 265 seconds) 2010-10-09T18:29:42 *** ichti has joined #aichallenge 2010-10-09T18:30:35 hmm? 2010-10-09T18:30:48 ? 2010-10-09T18:33:28 *** mega1 has quit IRC (Ping timeout: 265 seconds) 2010-10-09T18:33:51 *** ichti has quit IRC (Ping timeout: 240 seconds) 2010-10-09T18:38:31 *** ichti has joined #aichallenge 2010-10-09T18:42:54 *** ichti has quit IRC (Ping timeout: 250 seconds) 2010-10-09T18:46:55 *** ichti has joined #aichallenge 2010-10-09T18:48:07 hmm? 2010-10-09T18:48:37 hmm? 2010-10-09T18:48:47 hmm? 2010-10-09T18:48:51 hmm... 2010-10-09T18:50:42 mmhmm 2010-10-09T18:51:21 *** ichti has quit IRC (Ping timeout: 265 seconds) 2010-10-09T18:56:07 *** ichti has joined #aichallenge 2010-10-09T18:57:01 hrm 2010-10-09T19:00:16 Top 10 players: felixcoto(3791), rsergio(3780), Hazard(3750), ofrias(3724), cfaftw(3720), davidjliu(3676), sequoh(3653), EBraun(3648), dmj111(3639), adumlah(3619) 2010-10-09T19:00:27 *** ichti has quit IRC (Ping timeout: 255 seconds) 2010-10-09T19:00:35 wow, thank goodness for unit tests. i keep finding ridiculous bugs that i know would have taken me forever to track down had i been trying to find it after seeing weird bot behavors 2010-10-09T19:01:24 intel plan to make 32nm processors for socet775 ? 2010-10-09T19:02:29 *** jesusaurus has left #aichallenge ("WeeChat 0.3.2") 2010-10-09T19:04:03 *** yasith has quit IRC (Ping timeout: 255 seconds) 2010-10-09T19:04:17 my core dou to small amout of cires for thid contest 2010-10-09T19:04:23 my core dou to small amout of cores for thid contest 2010-10-09T19:04:26 my core dou to small amout of cores for this contest 2010-10-09T19:05:23 *** Janzert has left #aichallenge 2010-10-09T19:07:00 *** janzert has joined #aichallenge 2010-10-09T19:12:04 please check what you typed before sending it the first time, or if you still have to make a correction please don't resend the entire text 2010-10-09T19:12:35 it's a little annoying for the channel to be flooded with what looks like essentially the same thing over and over 2010-10-09T19:15:04 @later 2010-10-09T19:15:05 nigg_: Run as fast as you can and don't look back. 2010-10-09T19:15:22 @later msg j3camero Please contact me when you get back 2010-10-09T19:15:23 nigg_: I have no idea what you mean. 2010-10-09T19:15:30 or maybe not. 2010-10-09T19:15:31 :< 2010-10-09T19:15:43 try without the msg? 2010-10-09T19:16:03 @later j3camero Please contact me when you get back 2010-10-09T19:16:04 nigg_: User error -- Replace user. 2010-10-09T19:16:06 "later tell" 2010-10-09T19:16:18 @later tell j3camero Please contact me when you get back 2010-10-09T19:16:18 nigg_: Yes master! 2010-10-09T19:16:55 damn... my bot lost before it was even in the top 50 :( 2010-10-09T19:17:22 @seen j3camero 2010-10-09T19:17:22 janzert: j3camero was last seen in #aichallenge 4 days, 20 hours, 57 minutes, and 37 seconds ago: amstan: yes, but no names, haha. 2010-10-09T19:20:07 bleh, spot prices on the rise, servers dropping :( 2010-10-09T19:20:15 spot prices? 2010-10-09T19:20:20 it costs money? 2010-10-09T19:20:22 WTF 2010-10-09T19:20:37 *** yasith has joined #aichallenge 2010-10-09T19:20:38 amazon EC2 spot prices 2010-10-09T19:21:22 whats that 2010-10-09T19:22:01 the extra servers used to run games are amazon ec2 cloud server spot instances and basically the price is set by auction 2010-10-09T19:22:18 the contest is set up to use more servers when the price is low and less when the price goes up 2010-10-09T19:25:24 is the growth rate range of values fixed from 1 to 5 or is it different? 2010-10-09T19:26:10 the current generator will only produce planets from 0 through 5, but it's not specified 2010-10-09T19:26:18 thanks 2010-10-09T19:31:53 i think it's specified to not be below 0, but my memory could be wrong 2010-10-09T19:34:50 yeah 2010-10-09T19:34:59 i think jeff said the visualizer wouldn't be able to handle it or some such 2010-10-09T19:35:21 negative growth_rate? 2010-10-09T19:40:50 fail! 2010-10-09T19:41:27 negative growth would be quite silly 2010-10-09T19:41:50 planets you have to full support in order to have tactical advantage 2010-10-09T19:42:19 mmm, I doubt a planet with negative gf may be of any use, not matter the place 2010-10-09T19:42:35 I agree 2010-10-09T19:42:50 1 or 0 gf may still be of some use 2010-10-09T19:43:29 the only use for negative growth planets that I see would be to catch unwary bots which seems completely unnecessary 2010-10-09T19:43:42 janzert: yeah I agree 2010-10-09T19:45:19 0-growth can be useful as a staging planet 2010-10-09T19:45:30 0-growth can exist 2010-10-09T19:45:39 only as the center planet, at the moment 2010-10-09T19:45:41 yes, in center only atm 2010-10-09T19:45:43 :P 2010-10-09T19:46:37 it depends, if you have a planet further from center that has 5 gf may be better, because it could let you attack earlier on turns instead of this 2010-10-09T19:46:46 *** Meatkat has joined #aichallenge 2010-10-09T19:48:32 djm is back in the top10! 2010-10-09T19:48:35 @rankings 2010-10-09T19:48:37 sigh: Top 10 players: felixcoto(3814), rsergio(3795), Hazard(3766), ofrias(3737), cfaftw(3736), davidjliu(3694), EBraun(3669), sequoh(3661), dmj111(3643), adumlah(3641) 2010-10-09T19:49:25 even after losing his third game to someone ranked 1288 :P 2010-10-09T19:50:51 ahahahaha... is that person still 1288? 2010-10-09T19:51:19 oh, that is their rank now... they've resubmitted 2010-10-09T19:51:29 but either way, they were fairly low ranked :P 2010-10-09T19:51:33 they were ~520 or so when he lost to them 2010-10-09T19:51:40 yeah, something like that 2010-10-09T19:52:02 he had a bug with neutral stealing which messed up the game pretty good 2010-10-09T19:52:18 I told him that the bug wouldn't matter at higher levels 2010-10-09T19:52:22 looks like that was right :P 2010-10-09T19:53:02 Why wouldn't it matter at higher levels? 2010-10-09T19:53:28 because the stronger bots don't take neutrals when the enemy could easily take them 2010-10-09T19:54:12 Right 2010-10-09T19:55:08 it's this game: http://www.ai-contest.com/visualizer.php?game_id=5596456 2010-10-09T19:55:15 look at the 44-ship neutral at the top 2010-10-09T19:55:16 I've got a question for you sigh 2010-10-09T19:55:22 fire away 2010-10-09T19:55:38 *** Blkt has quit IRC (Ping timeout: 265 seconds) 2010-10-09T19:56:54 Say you just concentrate on your side of the board for a moment. You can calculate the optimal order to take planets with the goal of minimizing the amount of turns it takes to take all the planets. Giving you some sort of order to take them. Now add the enemy back in the picture. Does the order have any relevance or use anymore? 2010-10-09T19:58:58 yes I think so... it is obviously an approximation once the enemy enters the picture though 2010-10-09T19:59:01 *** yasith has quit IRC (Ping timeout: 265 seconds) 2010-10-09T19:59:22 I think the most important thing to get out of that is the *paths* that your ships would take 2010-10-09T19:59:56 What do you mean? 2010-10-09T20:00:07 for example, if hopping from planet 1 to 4 to 5 is more efficient then sending units directly from 1 to 4 and 1 to 5 2010-10-09T20:00:16 Top 10 players: felixcoto(3812), rsergio(3795), Hazard(3766), ofrias(3737), cfaftw(3729), davidjliu(3695), EBraun(3672), sequoh(3662), dmj111(3656), adumlah(3649) 2010-10-09T20:00:47 what would happen if I won? 2010-10-09T20:00:48 I don't think any bots handle that case well, because it requires a very annoying search 2010-10-09T20:01:40 (I'm talking about capturing neutrals, rather than just hopping across planets to redistribute ships which many bots do) 2010-10-09T20:02:32 but still... what I'm saying is a guess... I've had plenty of ideas that should work but didn't make much difference ;) 2010-10-09T20:02:36 Hmmm... so you'd be talking about sending extra ships to cap planet 4, planning to cap planet 5 much sooner than if you had to resend from planet 1? 2010-10-09T20:03:18 yeah, for example if planets are in a cluster, it might be best to send units to one planet in the cluster and have the growth from that planet capture the rest 2010-10-09T20:03:48 leaving ships free to take over other planets 2010-10-09T20:04:57 most people at the moment are thinking about attacking only from *current planets* and getting an optimal solution for that is tricky enough, although the knapsak problem helps 2010-10-09T20:05:04 So by saying it's 'more efficient' you're talking about minimizing flight times? 2010-10-09T20:06:20 I suppose, minimising the time to get the highest growth rate... or something similar to that 2010-10-09T20:06:42 figuring out what you should be optimising over is also hard :) 2010-10-09T20:07:02 indeed 2010-10-09T20:08:34 for example, in this post http://www.ai-contest.com/forum/viewtopic.php?f=17&t=854 mcmillen talks about using the knapsack problem to determine the best way to send out ships on the first turn to maximise growth rate 2010-10-09T20:08:36 Cool, well thanks 2010-10-09T20:08:37 How's your bot doing these days? 2010-10-09T20:09:16 I uploaded a recently, it's at 25 right now and still climbing I think 2010-10-09T20:09:19 *** yasith has joined #aichallenge 2010-10-09T20:09:24 so not sure where it will end up 2010-10-09T20:09:36 how is your bot doing? 2010-10-09T20:10:17 Meh, I'm doing okay, my most recent moved from 100 to 60, I just hope my next keeps improving like that :) 2010-10-09T20:10:36 That's an excellent post, thanks 2010-10-09T20:11:45 that's not too bad, a lot of people around your level have been in the top 10 at some point 2010-10-09T20:12:07 the level of the bots is just improving quickly 2010-10-09T20:12:35 Indeed, I wonder if that'll slow down at all... 2010-10-09T20:12:39 my bot is stable in range 70-80 for a week 2010-10-09T20:13:06 maybe even more 2010-10-09T20:13:22 hard to say, it still seems like all the improvements are incremental, no big upsets yet 2010-10-09T20:13:39 and I think that will happen eventually, so still on the lookup for that 2010-10-09T20:13:46 not many people can develop their bots for months 2010-10-09T20:13:55 many gave up 2010-10-09T20:14:35 * Naktibalda did a 9 day break :) 2010-10-09T20:14:56 sometimes breaks are a good thing 2010-10-09T20:15:20 *** yasith has quit IRC (Ping timeout: 264 seconds) 2010-10-09T20:15:22 you can come back with fresh ideas 2010-10-09T20:15:41 let your subconscious work on the problem :P 2010-10-09T20:18:35 *** Meatkat_ has joined #aichallenge 2010-10-09T20:18:49 *** Meatkat has quit IRC () 2010-10-09T20:23:02 *** Migi32 has quit IRC (Quit: http://boinc.berkeley.edu) 2010-10-09T20:27:02 *** mceier has quit IRC (Quit: leaving) 2010-10-09T20:29:28 *** yasith has joined #aichallenge 2010-10-09T20:35:05 *** blbrown_win3 has joined #aichallenge 2010-10-09T20:44:47 *** Accoun has joined #aichallenge 2010-10-09T20:46:31 *** Acount has quit IRC (Ping timeout: 240 seconds) 2010-10-09T20:47:40 Developers Developers Developers 2010-10-09T20:48:34 Developers, Developers ewerewere... 2010-10-09T20:52:27 *** Appleman1234 has joined #aichallenge 2010-10-09T20:58:30 *** jimmerton has joined #aichallenge 2010-10-09T20:59:40 *** Meatkat_ has quit IRC (Ping timeout: 245 seconds) 2010-10-09T21:00:16 Top 10 players: felixcoto(3796), rsergio(3756), Hazard(3747), ofrias(3746), MariusCG(3710), EBraun(3694), cfaftw(3682), adumlah(3680), dmj111(3660), sequoh(3643) 2010-10-09T21:03:08 *** Meatkat_ has joined #aichallenge 2010-10-09T21:03:51 *** Meatkat_ has quit IRC (Client Quit) 2010-10-09T21:04:35 *** nigg_ has quit IRC (Quit: Leaving) 2010-10-09T21:05:42 *** Meatkat has joined #aichallenge 2010-10-09T21:28:02 *** jimmerton has quit IRC (Quit: jimmerton) 2010-10-09T21:36:20 *** Meatkat has quit IRC (Ping timeout: 245 seconds) 2010-10-09T21:48:57 *** Naktibalda has quit IRC (Quit: No Ping reply in 180 seconds.) 2010-10-09T21:49:24 *** Naktibalda has joined #aichallenge 2010-10-09T21:57:11 *** blbrown_win3 has quit IRC (Ping timeout: 245 seconds) 2010-10-09T22:00:17 Top 10 players: felixcoto(3770), rsergio(3741), ofrias(3739), Hazard(3736), MariusCG(3723), EBraun(3682), cfaftw(3674), dmj111(3672), adumlah(3662), sequoh(3631) 2010-10-09T22:04:34 @rankings 2010-10-09T22:04:35 sigh: Top 10 players: felixcoto(3777), ofrias(3755), rsergio(3749), Hazard(3743), MariusCG(3730), EBraun(3695), cfaftw(3684), dmj111(3675), adumlah(3665), sighbot(3632) 2010-10-09T22:04:55 in the top 10 again! :D 2010-10-09T22:05:01 let's see how long it will last 2010-10-09T22:11:01 *** blbrown_win3_ has joined #aichallenge 2010-10-09T22:12:52 *** Frontier has left #aichallenge 2010-10-09T22:16:08 oops, out again... was fun while it lasted 2010-10-09T22:17:24 *** Frontier has joined #aichallenge 2010-10-09T22:40:53 *** agweber has joined #aichallenge 2010-10-09T22:41:29 Soooooooo I need java help. 2010-10-09T22:42:27 http://pastebin.com/qCp4cBTj This section of code causes my bot to not even try. I can comment it out and everything else works as normal. 2010-10-09T22:42:52 There are no code errors, so it must be a logical fault 2010-10-09T22:43:22 Last time I did this one guy pointed out about 5 things that I completely failed at, and I think I fixed all of those. 2010-10-09T22:46:17 Okay, so apparently && != &,.. I changed the & to a && but that didn't change anything. 2010-10-09T22:50:29 what do you mean when you say your bot doesn't work? 2010-10-09T22:50:43 does it timeout, issue illegal moves, just play badly? 2010-10-09T22:52:16 Well, using the playgame thing, it just goes Turn 1; Player 2 Wins 2010-10-09T22:52:51 ok, sounds like an illegal move or a timeout 2010-10-09T22:53:11 hmmm.. you don't seem to check if your source planet has enough ships when issuing an order 2010-10-09T22:53:12 I've tried 80 different ways to make sure that while loop wasn't infinite 2010-10-09T22:53:19 I did 2010-10-09T22:53:27 line 33 2010-10-09T22:54:03 that's not enough, you need to make sure that mine.NumShips() <= p.NumShips() + 1 2010-10-09T22:54:13 since that is how many ships you are sending 2010-10-09T22:54:28 Ahhh 2010-10-09T22:55:30 Tried that, still isn't working. Tried +2 as well 2010-10-09T22:56:39 after sending your first fleet, the number of ships left on the planet is out of date 2010-10-09T22:56:41 *** yasith has quit IRC (Ping timeout: 240 seconds) 2010-10-09T22:56:41 But good catch! 2010-10-09T22:56:53 at least if you are using the stock framework 2010-10-09T22:56:59 It doesn't update? So I need to keep track manually 2010-10-09T22:57:06 Yeah, stock framework 2010-10-09T22:57:17 *** tapwater has joined #aichallenge 2010-10-09T22:57:54 you can keep track manually, or you can update the framework so it keeps track for you 2010-10-09T22:58:22 either way, I would suggest looking at a log of the exact orders that your bot is sending 2010-10-09T22:58:33 log.txt should show you that 2010-10-09T22:59:04 In the base folder? 2010-10-09T22:59:09 says initializing 2010-10-09T22:59:20 and that's it 2010-10-09T22:59:56 ah :( 2010-10-09T23:00:18 Top 10 players: MariusCG(3788), ofrias(3787), felixcoto(3787), Hazard(3764), rsergio(3756), EBraun(3720), cfaftw(3717), adumlah(3688), dmj111(3685), sequoh(3656) 2010-10-09T23:00:33 Lemme relook over my manually track thing again 2010-10-09T23:01:33 just use the RemoveShips() function.... mine.RemoveShips(p.NumShips()+1) 2010-10-09T23:02:08 that should work for now :) 2010-10-09T23:03:31 Hot shit, that was it 2010-10-09T23:03:40 I would've never figured out that it wasn't updating. 2010-10-09T23:04:27 Read the Source, Luke 2010-10-09T23:33:05 *** abd3lnuban has joined #aichallenge 2010-10-09T23:34:29 *** amstan has joined #aichallenge 2010-10-09T23:34:29 *** ChanServ sets mode: +o amstan 2010-10-09T23:34:31 *** abd3lnuban has quit IRC (Client Quit) 2010-10-09T23:40:52 * pgpaskar_ looks over at contestbot 2010-10-09T23:40:53 ◕‿◕ 2010-10-09T23:44:00 *** davidjliu has quit IRC (Quit: Page closed)