2010-10-21T00:07:44 janzert: nice 2010-10-21T00:08:56 *** synmnky_ has quit IRC (Ping timeout: 255 seconds) 2010-10-21T00:10:52 Top 10 players: smatcher(3754), cfaftw(3673), dmj111(3667), rsergio(3630), deepblue(3518), Polsky(3518), ruilov(3508), savchenko(3493), ofrias(3493), chotting(3493) 2010-10-21T00:13:49 *** davidjliu has quit IRC (Quit: Page closed) 2010-10-21T00:29:28 *** Azrathud has quit IRC (Ping timeout: 265 seconds) 2010-10-21T00:32:32 *** Zepp has joined #aichallenge 2010-10-21T00:32:56 *** JCS^ has quit IRC (Quit: \x00) 2010-10-21T00:33:13 amstan_: so phpbb can't send emails right now? 2010-10-21T00:33:40 Zepp: no 2010-10-21T00:33:47 *** rebelxt_ has joined #aichallenge 2010-10-21T00:33:54 Zepp: error log's full of these: http://pastebin.com/ZdqkckMw 2010-10-21T00:34:06 Zepp: hold on, let me make you admin 2010-10-21T00:34:20 amstan_: alright 2010-10-21T00:34:55 oh shit. it's trying to use google servers still? 2010-10-21T00:34:56 *** narnach has joined #aichallenge 2010-10-21T00:34:59 Zepp: yeah 2010-10-21T00:35:13 Zepp: there should be an admin link at the bottom of each page now 2010-10-21T00:35:17 hehe 2010-10-21T00:36:19 *** hornairs has joined #aichallenge 2010-10-21T00:38:44 Ok I'm going nuts... it would appear my bot keeps crashing on the TCP... but it works perfect on the local bots? 2010-10-21T00:39:11 Add more logging :} 2010-10-21T00:39:19 How can you log? 2010-10-21T00:39:50 to a file or standard error while you're running it locally on the tcp server 2010-10-21T00:40:00 just shut it off when you upload 2010-10-21T00:40:09 k 2010-10-21T00:41:27 *** amstan_ has quit IRC (Remote host closed the connection) 2010-10-21T00:41:42 bah where the hell does phpbb keep the smtp server settings haha 2010-10-21T00:43:10 *** noid05826 has joined #aichallenge 2010-10-21T00:43:29 *** noid05826 has left #aichallenge 2010-10-21T00:43:30 should be an email settings page linked to from the general configuration page 2010-10-21T00:43:34 *** Accoun has joined #aichallenge 2010-10-21T00:44:07 janzert: yeah that is what i was expecting from a quick look at docs but failing to see it hrmm 2010-10-21T00:44:31 *** Utkarsh has quit IRC (Ping timeout: 276 seconds) 2010-10-21T00:44:55 damnit amstan left 2010-10-21T00:45:02 older version of phpbb maybe? 2010-10-21T00:45:11 *** Azrathud has joined #aichallenge 2010-10-21T00:45:14 i think he didn't promote me high enough 2010-10-21T00:45:18 i can't access everything 2010-10-21T00:45:21 ahh, maybe 2010-10-21T00:46:48 http://www.ai-contest.com/profile.php?user_id=8112 look out 2010-10-21T00:47:27 *** amstan has joined #aichallenge 2010-10-21T00:47:27 *** ChanServ sets mode: +o amstan 2010-10-21T00:47:39 Agh, am copying the server code to work on locally for syncing with svn and just saw that someone has left a 1.2GB file lying around in the backend directory 2010-10-21T00:48:27 no wonder it's taking so long 2010-10-21T00:49:12 oops 2010-10-21T00:49:19 yeah, i wouldn't check that it 2010-10-21T00:49:24 antimatroid: at 1 game/hour (with corresponding rise in rank by 1), it will probably be at least 1 more day until it's even possible to get to the coveted 1st place 2010-10-21T00:49:25 in * 2010-10-21T00:49:54 hehe, no. Already knew it was going to take a bit of picking through before checking in. 2010-10-21T00:50:15 antimatroid: seen sin_sun on the tcp a few times 2010-10-21T00:50:28 *** Utkarsh has joined #aichallenge 2010-10-21T00:50:29 amstan: give me more permissions 2010-10-21T00:50:54 Zepp: you should have all 2010-10-21T00:50:56 hmm 2010-10-21T00:51:20 MathNerd314: is that you? 2010-10-21T00:51:34 Zepp: k, incoming 2010-10-21T00:51:36 rebelxt_: yeah i've seen them on there 2010-10-21T00:51:57 Zepp: try now 2010-10-21T00:52:02 antimatroid: no; I currently have some junk bot in 2000th place or something 2010-10-21T00:52:43 my current bot crashes every time i run it 2010-10-21T00:52:49 amstan: hmm i still don't see it. also i can't manage the admin group 2010-10-21T00:52:51 but i'm in the middle of a big overhaul of my back end 2010-10-21T00:53:03 just watched "A Beautiful Mind" for the second time. what a great movie, especially now that Nash equilibrium actually makes sense... anyone remember the movie? 2010-10-21T00:53:19 yeah, movie was kind of lame :P 2010-10-21T00:53:21 Zepp: sigh, hold on, this interface is confusing 2010-10-21T00:53:38 I've seen it a couple times. 2010-10-21T00:53:41 Zepp: now? 2010-10-21T00:53:42 john nash was a mathematician not an economist 2010-10-21T00:53:46 amstan: yeah, i am pretty close to just changing it from the mysql shell :P 2010-10-21T00:53:58 was/is if you will 2010-10-21T00:53:59 antimatroid: i was still surprised how much of it was in sync with Nash's real life story 2010-10-21T00:54:00 amstan: that seems to have worked 2010-10-21T00:54:14 Zepp: for some reason the admin group was not fully admin, idk why 2010-10-21T00:54:26 i'm more of a neumann fanboy anyway :P 2010-10-21T00:54:50 that man was a true freak 2010-10-21T00:55:23 nash equilibrium isn't really all that surprising tbh, the idea of finding a fixed point in a structure is nothing new 2010-10-21T00:55:50 and i'm pretty sure he didn't come up with the fixed point theorem used to prove the existence of a mixed strategy nash equilibrium 2010-10-21T00:56:05 having said that i still think it's cool, just not as astounding as some of the things neumann did 2010-10-21T00:56:52 amstan: fixed i believe 2010-10-21T00:57:10 he was a grad student back then, so that's pretty impressive in my book. neumann did tons that's true 2010-10-21T00:58:25 well, i still had fun watching the movie for the 2nd time. first time i watched it i did not even remember much of game theory since i graduated long long ago 2010-10-21T00:58:44 Zepp: cool 2010-10-21T00:58:49 if i do a phd i expect the thesis to be on game theory :P 2010-10-21T00:59:01 amstan: so people want to reset their contest account too? 2010-10-21T00:59:14 after this contest, you gotta do it! 2010-10-21T00:59:17 Zepp: oh yeah, they want lots of things, check the introductions forums 2010-10-21T00:59:55 hehe i don't really have time to look at everything sadly 2010-10-21T01:00:26 i was sort of trying to avoid this time sink this term :-/ 2010-10-21T01:02:21 amstan: do you watch the mailing list much? 2010-10-21T01:02:40 Zepp: yeah, but i choose to not to much 2010-10-21T01:02:45 i don't have time for a lot of this either 2010-10-21T01:03:12 amstan: yeah, understandable. I am just wondering if you have been following the cloud server talk and such 2010-10-21T01:03:24 not really 2010-10-21T01:03:38 i still do not know why we picked cloud servers and not user sponsored servers 2010-10-21T01:03:51 no idea 2010-10-21T01:04:09 and like... 2010-10-21T01:04:14 contestbot: seen j3camero 2010-10-21T01:04:14 amstan: j3camero was last seen in #aichallenge 2 weeks, 2 days, 2 hours, 44 minutes, and 28 seconds ago: amstan: yes, but no names, haha. 2010-10-21T01:04:21 that ^^ 2010-10-21T01:04:39 ouch 2010-10-21T01:05:43 poor communication from jeff isn't helping yeah :-/. Plus the drama that has been causing, sigh. 2010-10-21T01:06:31 *** hornairs has quit IRC (Quit: hornairs) 2010-10-21T01:06:48 i've been getting countless pms and emails from people ranting about various things 2010-10-21T01:07:01 i can only forward them to jeff 2010-10-21T01:07:06 haters gonna hate 2010-10-21T01:07:25 just make sure things work for the final contest 2010-10-21T01:07:33 jeff was jere less than 2 weeks ago 2010-10-21T01:07:52 constestbot: seen jeffcameron 2010-10-21T01:08:09 and even the profile changes and stuff... i was able to do them, in tron, now we have over 3000 people, mostly everyone registering has been requesting a change 2010-10-21T01:08:15 I usually just ping him on gtalk 2010-10-21T01:08:33 you need to allow people to change their about me and organisation :P 2010-10-21T01:08:33 amstan: yeah. What sort of things do they need changed? 2010-10-21T01:08:40 Zepp: he doesn't go on much there either, when he does, he's in gmail 2010-10-21T01:08:53 Zepp: organizations, change names, change passwords, change bio 2010-10-21T01:09:02 there's a need for an interface that does these things 2010-10-21T01:09:05 *** rebelxt_ has quit IRC (Quit: Page closed) 2010-10-21T01:09:29 *** delt0r has joined #aichallenge 2010-10-21T01:09:52 amstan: hmm, yeah. do you know if all the site code is in the general public repo? 2010-10-21T01:09:55 *** delt0r___ has quit IRC (Ping timeout: 240 seconds) 2010-10-21T01:10:01 it is 2010-10-21T01:10:40 yeah, a bunch of that stuff will need to be reworked before the next contest 2010-10-21T01:10:54 Top 10 players: smatcher(3809), cfaftw(3720), dmj111(3707), rsergio(3669), deepblue(3577), Polsky(3565), savchenko(3564), ruilov(3563), ofrias(3554), Slin-.-(3550) 2010-10-21T01:10:57 I wonder how easy it would be to make some changes... 2010-10-21T01:11:10 password reset shouldn't be too bad 2010-10-21T01:11:32 changing profiles, I don't know how the site auth works at the moment 2010-10-21T01:11:35 Zepp: there is a code review up for quite a while for the password reset now 2010-10-21T01:11:49 wait someone coded it and it just needs approval? 2010-10-21T01:11:59 contestbot: repo 2010-10-21T01:12:00 amstan: repo = http://code.google.com/p/ai-contest/ 2010-10-21T01:12:07 * janzert nods, there are a lot of things like that 2010-10-21T01:12:16 that is depressing 2010-10-21T01:12:43 Zepp: yeah, though, one of the problems is that there were changes in the site that were not commited to svn 2010-10-21T01:12:45 I'm hoping to work through some of that backlog soon 2010-10-21T01:13:00 janzert: are you admin now? 2010-10-21T01:13:18 yes, jeff got me setup this evening 2010-10-21T01:13:29 janzert: oh cool 2010-10-21T01:13:37 working on syncing the code from the server back to svn now 2010-10-21T01:13:55 *** amstan sets mode: +o janzert 2010-10-21T01:14:08 janzert: what's your forums username? 2010-10-21T01:14:12 same 2010-10-21T01:15:07 janzert: k, you're red now 2010-10-21T01:15:43 uh oh have to watch out, now I'm a big fat target :) 2010-10-21T01:15:50 thanks though 2010-10-21T01:16:50 someone needs to respond to this http://ai-contest.com/forum/viewtopic.php?f=18&t=877&start=20 2010-10-21T01:17:11 respond what? 2010-10-21T01:17:30 just let people know some details about the server situation 2010-10-21T01:17:31 starting the servers up might be enough? 2010-10-21T01:17:46 janzert: probably 2010-10-21T01:18:02 so who can do the code reviews? 2010-10-21T01:18:15 someone who's admin 2010-10-21T01:18:29 like 2010-10-21T01:18:32 all 3 of us 2010-10-21T01:18:46 which things need code reviews by admins only? 2010-10-21T01:18:50 *** rommel has quit IRC (Quit: Page closed) 2010-10-21T01:19:13 suppose to be all, before merging to trunk I believe 2010-10-21T01:19:20 amstan: how do these work? is it something on google code? because i only see committers and j3camero as owner? 2010-10-21T01:19:21 oh.. right.. anyone who's a commiter that's not the original guy who wrote the commit 2010-10-21T01:19:33 ah okay 2010-10-21T01:19:37 Zepp: oh yeah, you have to be added to the repo 2010-10-21T01:19:40 any commiters can do a review though 2010-10-21T01:19:50 janzert: yeah, amstan is right regarding that 2010-10-21T01:20:01 jeff changed that before he vanished 2010-10-21T01:20:09 because otherwise we all bottlenecked on him 2010-10-21T01:20:09 damn i don't have committ access hah 2010-10-21T01:20:15 *** Utkarsh has quit IRC (Ping timeout: 240 seconds) 2010-10-21T01:20:41 ahh, someone should update the wiki http://code.google.com/p/ai-contest/wiki/HowToContribute 2010-10-21T01:20:57 since that's all he pointed me at :P 2010-10-21T01:22:22 janzert: update what? 2010-10-21T01:22:54 in what situations people are allowed to merge to trunk I suppose 2010-10-21T01:23:03 or maybe I misunderstood what Zannick meant 2010-10-21T01:23:10 hm, yeah, jeff never updated that 2010-10-21T01:23:26 well.. it's still applicable 2010-10-21T01:23:40 exceptions are urgent stuff 2010-10-21T01:24:08 2010-09-10.log:1984:23:18 < j3camero> Did you guys see the new policy for merging changes into trunk? It's decentralizing. Now your code reviews can be approved by any other developer. Be responsible, though. Try to get negative reviews. Don't avoid critical reviewers just to push your change. 2010-10-21T01:24:24 Zannick: isn't that already in? 2010-10-21T01:24:53 maybe i misunderstood what he meant? 2010-10-21T01:25:10 the first line of For Commiters says in big bold letters "Do not ever commit any changes to the trunk." 2010-10-21T01:25:11 idk.. his message implies that policy is on the wiki already, but i can't see it 2010-10-21T01:25:53 yeah, a little confusing 2010-10-21T01:26:00 *** Utkarsh has joined #aichallenge 2010-10-21T01:26:18 well, i'd go with try not to break anything while fixing things 2010-10-21T01:26:46 too bad the wiki doesn't seem to have a history of the page to see what it was before his last change 2010-10-21T01:26:50 Zepp: do you have commit access then? 2010-10-21T01:27:01 janzert: lol, it doesn't? wow.. 2010-10-21T01:27:10 not that I'm seeing 2010-10-21T01:27:13 amstan: i do not i believe 2010-10-21T01:28:17 Zepp: i think you can use my password, since google code has a separate password for svn stuff 2010-10-21T01:29:03 Zepp: won't work for commenting on stuff and writing reviews 2010-10-21T01:29:04 amstan: meh i dont need to commit anything right now 2010-10-21T01:29:09 ok 2010-10-21T01:29:45 i might ping jeff but at the moment i know pretty well nothing of the code base anyway 2010-10-21T01:30:02 i've only seen parts of it during the summer when it was still being worked on haha 2010-10-21T01:30:37 Zepp: so.. there's 2 important folders in planetwars/: backend and www 2010-10-21T01:30:56 in each one there's a login file for the system 2010-10-21T01:31:07 and lots of other files 2010-10-21T01:31:21 hacked together scripts 2010-10-21T01:31:45 *** narnach has quit IRC (Quit: Leaving.) 2010-10-21T01:31:52 sounds awesome haha 2010-10-21T01:32:03 yeah.. 2010-10-21T01:32:38 this contest needs a rewrite 2010-10-21T01:32:53 then maybe i would be open to coding for it too 2010-10-21T01:33:12 i will be around in december to help with that 2010-10-21T01:33:26 also.. git, or something decentralized 2010-10-21T01:33:26 i have many ideas, hopefully most of them good :) 2010-10-21T01:33:49 right now we have this monstrosity of a svn repo 2010-10-21T01:33:52 with branches.. 2010-10-21T01:34:06 you could use git-svn 2010-10-21T01:34:16 but yeah git/hg would be nicer probably 2010-10-21T01:34:31 Zepp: the problem's mostly with deploying those changes, and updating the server 2010-10-21T01:34:37 it's not the frontend that bothers me 2010-10-21T01:34:47 ah okay 2010-10-21T01:45:57 *** Mathnerd314 has quit IRC (Remote host closed the connection) 2010-10-21T01:46:57 *** hellman has quit IRC (Remote host closed the connection) 2010-10-21T01:47:12 *** d3c1ph3r has joined #aichallenge 2010-10-21T01:47:22 i have a java problem anybody able to help 2010-10-21T01:47:37 stupidest thing is happening to my program and i dont understand why 2010-10-21T01:54:07 *** amstan_ has joined #aichallenge 2010-10-21T01:54:09 *** ChanServ sets mode: +o amstan_ 2010-10-21T01:54:16 *** Utkarsh has quit IRC (Read error: Connection reset by peer) 2010-10-21T01:56:59 *** amstan__ has joined #aichallenge 2010-10-21T01:57:00 *** ChanServ sets mode: +o amstan__ 2010-10-21T01:57:05 d3c1ph3r: it would help if you actually described your problem 2010-10-21T01:57:28 *** amstan has quit IRC (Ping timeout: 252 seconds) 2010-10-21T01:57:32 *** amstan__ is now known as amstan 2010-10-21T01:57:52 i was waiting for a reply before waisting my time 2010-10-21T01:57:55 lol 2010-10-21T01:58:30 you just succeeded in wasting your time, this way people that do not know if they can help will respond 2010-10-21T01:58:31 like me 2010-10-21T01:58:35 or.. won't respond at all 2010-10-21T01:59:18 *** amstan_ has quit IRC (Ping timeout: 245 seconds) 2010-10-21T01:59:20 d3c1ph3r: just go ahead 2010-10-21T01:59:35 the problem is that i have a 2-d array just containing my remaining ships from each planet as i recurse through my tree it takes away from my remaining ships pending on what ship i used to attack 2010-10-21T01:59:55 i made it global because i didn't know if i was passing it wrong (which i wasn't) 2010-10-21T01:59:59 ARGH, there are two or more copies of the code on the server now I have to figure out which of each is actually used 2010-10-21T02:00:03 but anyways one second its one value and the other its a nother 2010-10-21T02:00:18 janzert: where? 2010-10-21T02:00:37 under /home/contest and under /var/www... 2010-10-21T02:00:37 janzert: there's one in ~contest/ai-challenge.com/ and one for www, which i forgot where it was 2010-10-21T02:00:39 so the values in your array are being changed while you are looking at them? 2010-10-21T02:00:50 janzert: yes, one's for the backend, one's for the www 2010-10-21T02:01:04 yeah, that's what I'm coming to think 2010-10-21T02:01:09 in my method its a recursive tree where it takes one planet over it calls that same method loosk for a nother planet to attack with my remaining ships 2010-10-21T02:01:17 why though, why? 2010-10-21T02:01:37 * janzert runs in circles some more :P 2010-10-21T02:01:39 now my global method keeps track of all my ships which at this point its size 1 with just 100 ships 2010-10-21T02:01:44 *** Utkarsh has joined #aichallenge 2010-10-21T02:02:16 when i attack a planet it takes away from the ships then moves on to the next move when it pops back of the system stack it changes it back to its original value but my problem is before i pop off 2010-10-21T02:02:55 janzert: yeah, that seems excessive 2010-10-21T02:03:01 just use symlinks or some such 2010-10-21T02:03:02 amstan: are you pretty certain that the backend code under /var/www isn't actually used? 2010-10-21T02:03:04 one minute i have 61 shits then i do a method to check to see if its full! a safety check for later calls, and it returns true when it should return false, i did a print before the method one second i have 61 ships then i have zero ships hwen it goes into that method 2010-10-21T02:03:13 ships* 2010-10-21T02:03:13 Zannick: yeah 2010-10-21T02:04:03 so before you call this method you have 61 ships, and after you call it you have 0? 2010-10-21T02:04:16 im pretty good at java and at first thought something was fishy with the system stack so i made new copies of the array and passed that now im just doing it globally so it only gets changed one step at a time but its changing somewhere with out me saying to change it 2010-10-21T02:04:21 yeah 2010-10-21T02:04:26 and its a global array 2010-10-21T02:04:28 ? 2010-10-21T02:04:35 anyone remember the svn command to see what revision the current checkout is at? 2010-10-21T02:04:36 this is your check that you have...what? 2010-10-21T02:04:41 *** amstan_ has joined #aichallenge 2010-10-21T02:04:41 *** ChanServ sets mode: +o amstan_ 2010-10-21T02:04:48 damn DSL 2010-10-21T02:04:59 ill show you my check 2010-10-21T02:05:12 *** amstan has quit IRC (Ping timeout: 250 seconds) 2010-10-21T02:05:15 janzert: svn info 2010-10-21T02:05:23 janzert: no, i'm not 2010-10-21T02:05:24 public static boolean noShips(){ int full = 0; for(int i = 0; i < remainShips.length;i++){ test1 = remainShips[i]; if(remainShips[i] <= leaveBehind){ full++; } } test2 = full; if(full == remainShips.length) return true; else return false; } 2010-10-21T02:05:25 and svn status 2010-10-21T02:05:32 oops that did not format nice 2010-10-21T02:05:37 ahh yeah svn info is what I was looking for 2010-10-21T02:05:43 d3c1ph3r: use pastebin if you have lots of code 2010-10-21T02:05:49 d3c1ph3r: try not pasting in here 2010-10-21T02:06:06 yes 2010-10-21T02:06:09 well, anyway 2010-10-21T02:06:21 what are test1 and test2? 2010-10-21T02:06:37 http://pastebin.com/6L2sfeKT 2010-10-21T02:06:38 so the backend code under /var/www is an umodified checkout from revision 250 so definitely not used :/ 2010-10-21T02:06:49 i use that for when i print my file i see what values they receive 2010-10-21T02:07:12 janzert: ok, does it matter though? 2010-10-21T02:07:13 after every added node i create a file showing the new state and the old state of game plus variable outputs 2010-10-21T02:07:16 no 2010-10-21T02:07:31 it doesnt miscellaneous code 2010-10-21T02:07:34 so the results are test1 is 61 and test2 is 0? 2010-10-21T02:07:44 no the file prints 0 and 1 2010-10-21T02:07:55 just that there are misleading bits laying around 2010-10-21T02:08:05 okay, so test1 is 0 and test2 is 1, you're saying? 2010-10-21T02:08:12 yeah 2010-10-21T02:08:13 and then it returns true 2010-10-21T02:08:14 and right before the file 2010-10-21T02:08:16 yeah 2010-10-21T02:08:20 it has zero ships 2010-10-21T02:08:23 but listen to this 2010-10-21T02:08:30 at the start of the method i do a copy into a new array 2010-10-21T02:08:36 guys.. address to people, there's 2 conversations going on, it's hard to differentiate 2010-10-21T02:08:49 amstan_: it's really easy. :P 2010-10-21T02:08:53 and print those values simultaneousness and it has the r ight values 2010-10-21T02:09:11 but no where between i change the valeus of the global array 2010-10-21T02:09:13 values* 2010-10-21T02:09:13 janzert: so what are you trying to do? 2010-10-21T02:09:28 amstan_: get svn back in sync with the server 2010-10-21T02:09:50 makes no sense 2010-10-21T02:10:06 btw my conversation with everyone so i can't private i need multiple inputs 2010-10-21T02:10:09 amstan_: so it appears now that I need to get the backend directory in sync with what is under ~/contest and the www directory with what is under /var/www 2010-10-21T02:10:23 amstan_: getting there :) 2010-10-21T02:10:29 yeah, pretty much 2010-10-21T02:10:36 watch out to not commit login stuff 2010-10-21T02:10:38 d3c1ph3r: well, with test1 being 0, remainships[i] is definitely being set to 0 somewhere. 2010-10-21T02:10:48 amstan_: yep, being careful 2010-10-21T02:10:53 Top 10 players: smatcher(3784), dmj111(3764), rsergio(3762), deepblue(3752), cfaftw(3693), savchenko(3676), chotting(3639), DocBot(3619), Slin-.-(3608), ruilov(3599) 2010-10-21T02:10:57 janzert: svn ignore was not implemented/exist 2010-10-21T02:11:13 yeah your totally right zannick but its no where in my code can i paste my method and show you in pastebin 2010-10-21T02:11:20 sure 2010-10-21T02:11:21 just dont steal it :P 2010-10-21T02:11:25 * janzert just hopes there aren't too many bits being used laying around elsewhere 2010-10-21T02:11:35 janzert: i don't think so 2010-10-21T02:11:51 amstan_: for instance I *know* whatever is being used to clean up old submissions isn't in there 2010-10-21T02:12:06 janzert: that's in backend, under ~contest 2010-10-21T02:12:08 just ignore the testing shit 2010-10-21T02:12:19 janzert: delete_some_old_submissions or something like that 2010-10-21T02:12:29 ahh, I missed it 2010-10-21T02:12:35 d3c1ph3r: sure 2010-10-21T02:12:50 how do you private 2010-10-21T02:12:59 d3c1ph3r: you click on a nick 2010-10-21T02:13:21 /msg Zannick 2010-10-21T02:13:24 or that 2010-10-21T02:16:13 whee, just caught tournament_manager in a crash 2010-10-21T02:17:51 ugh, the worker #21 at least is still running with the pre-compile_anything cleanup code 2010-10-21T02:24:58 *** mega1 has joined #aichallenge 2010-10-21T02:25:08 *** jmpespxoreax has quit IRC (Ping timeout: 250 seconds) 2010-10-21T02:27:55 *** Accoun has quit IRC (Ping timeout: 240 seconds) 2010-10-21T02:28:29 *** smloh1 has joined #aichallenge 2010-10-21T02:33:56 *** smloh1 has quit IRC (Quit: Page closed) 2010-10-21T02:37:11 *** perestrelka has quit IRC (Remote host closed the connection) 2010-10-21T02:41:42 *** wvdschel has joined #aichallenge 2010-10-21T02:41:43 *** wvdschel has joined #aichallenge 2010-10-21T02:42:25 *** narnach has joined #aichallenge 2010-10-21T02:49:01 *** Appleman1234 has joined #aichallenge 2010-10-21T02:50:55 *** perestrelka has joined #aichallenge 2010-10-21T02:51:11 *** noid05826 has joined #aichallenge 2010-10-21T02:51:24 *** noid05826 has left #aichallenge 2010-10-21T02:51:29 *** Accoun has joined #aichallenge 2010-10-21T02:59:30 *** d3c1ph3r has quit IRC (Quit: Page closed) 2010-10-21T02:59:40 i should get paid for this :P 2010-10-21T03:02:15 which this? remote debugging :) 2010-10-21T03:02:25 yes, debugging 2010-10-21T03:03:02 otoh, i do not want to debug 3000 contestants' entries :P 2010-10-21T03:03:25 heh, no kidding 2010-10-21T03:05:34 btw, I got worker 21 updated to latest svn code... worker 25 went away before I got it updated though 2010-10-21T03:07:40 giving rudimentary access to exit code and stderr of bots would go a long way 2010-10-21T03:10:53 Top 10 players: deepblue(3819), rsergio(3804), savchenko(3779), cfaftw(3776), dmj111(3759), smatcher(3759), Slin-.-(3668), chotting(3668), DocBot(3641), ofrias(3622) 2010-10-21T03:20:39 *** perror has joined #aichallenge 2010-10-21T03:27:48 *** nullkuhl_ has joined #aichallenge 2010-10-21T03:28:02 *** nullkuhl has quit IRC (Ping timeout: 265 seconds) 2010-10-21T03:35:09 *** nullkuhl_ has quit IRC (Remote host closed the connection) 2010-10-21T03:44:50 *** aerique has joined #aichallenge 2010-10-21T03:46:51 *** smellyhippy has joined #aichallenge 2010-10-21T03:47:35 *** Olathe has quit IRC (Ping timeout: 240 seconds) 2010-10-21T03:53:19 *** tapwater has quit IRC (Quit: tapwater) 2010-10-21T03:53:41 *** amstan_ has quit IRC (*.net *.split) 2010-10-21T03:53:43 *** wvdschel has quit IRC (*.net *.split) 2010-10-21T03:56:20 *** amstan_ has joined #aichallenge 2010-10-21T03:56:20 *** jordan.freenode.net sets mode: +o amstan_ 2010-10-21T03:58:44 *** Itkovian has quit IRC (Ping timeout: 250 seconds) 2010-10-21T04:00:55 *** boegel has joined #aichallenge 2010-10-21T04:09:09 *** p4p4 has joined #aichallenge 2010-10-21T04:10:08 *** phreeza has joined #aichallenge 2010-10-21T04:10:54 Top 10 players: cfaftw(3849), rsergio(3828), deepblue(3826), smatcher(3813), savchenko(3811), dmj111(3788), Hazard(3763), chotting(3708), ofrias(3688), DocBot(3677) 2010-10-21T04:20:18 *** Utkarsh has quit IRC (Ping timeout: 252 seconds) 2010-10-21T04:25:01 *** Utkarsh has joined #aichallenge 2010-10-21T04:29:03 *** Florian95 has joined #aichallenge 2010-10-21T04:31:39 *** j3camero has quit IRC (Ping timeout: 240 seconds) 2010-10-21T04:32:48 *** j3camero has joined #aichallenge 2010-10-21T04:39:21 *** Utkarsh has quit IRC (Ping timeout: 265 seconds) 2010-10-21T04:45:12 *** Utkarsh has joined #aichallenge 2010-10-21T04:46:44 *** wvdschel has joined #aichallenge 2010-10-21T04:48:53 *** wvdschel has joined #aichallenge 2010-10-21T04:48:53 *** wvdschel has joined #aichallenge 2010-10-21T04:50:27 *** wvdschel has joined #aichallenge 2010-10-21T04:50:28 *** wvdschel has joined #aichallenge 2010-10-21T04:54:20 delt0r: does JBotManager actually read the .jbotmanager file? can I enable to always show lines in it? 2010-10-21T04:54:43 It should always read it yes 2010-10-21T04:55:20 you can just add a line and next time you start it --that cmd line bot should be loaded 2010-10-21T04:55:34 whoohoo moar servers :D 2010-10-21T04:55:35 aerique: you are not using WITH-TIMEOUT yet, are you? 2010-10-21T04:55:44 mega1: nope 2010-10-21T04:56:07 delt0r: what do i need to put in it exactly to enable lines to be drawn from fleets to planets in replays? 2010-10-21T04:56:21 aerique: it's broken in 1.0.11, I had to backport timer.lisp 2010-10-21T04:56:47 oh--so the .jbotmanger file only saves bot cmd lines 2010-10-21T04:56:55 _and_ lower the timeout to 0.8s but I guess that's due to main server being so weak. 2010-10-21T04:56:57 no other config is saved 2010-10-21T04:57:10 mega1: oh dear, could you blog it up for others and make it available? or on the forums? 2010-10-21T04:57:22 aerique: yes, I will 2010-10-21T04:57:30 aerique: so you need to click the drawLines check box at the bottom... 2010-10-21T04:57:32 delt0r: ah, got it! 2010-10-21T04:57:42 *** Naktibalda has joined #aichallenge 2010-10-21T04:57:59 mega1: awesome, i've got a feeling i'll be needing it in the near future 2010-10-21T04:58:10 It draws a line from the fleet to the destination... it does not draw anything to the source planet 2010-10-21T04:58:28 mega1: already had to take out a naive pathfinding algo because of timeouts 2010-10-21T04:59:02 the main server is very slow, it favors fast algos 2010-10-21T04:59:03 delt0r: i know, i always turn it on in replays 2010-10-21T04:59:20 right... 2010-10-21T04:59:31 well saving the gui state is a pita 2010-10-21T04:59:38 so i didn't do it 2010-10-21T04:59:38 delt0r: i'll just unpack the jar and enable it in the source then :) 2010-10-21T04:59:43 yea 2010-10-21T05:00:01 I can change the default... some others didn't like it however. 2010-10-21T05:00:11 but not till next bug fix 2010-10-21T05:00:34 delt0r: nah, otherwise others have to turn it off again :) 2010-10-21T05:00:40 k 2010-10-21T05:00:56 *** phreeza has quit IRC (Quit: kthxbai) 2010-10-21T05:10:55 Top 10 players: rsergio(3887), dmj111(3877), deepblue(3854), cfaftw(3850), savchenko(3841), smatcher(3824), Hazard(3790), chotting(3731), Rexxar(3711), ofrias(3705) 2010-10-21T05:18:30 *** mceier has joined #aichallenge 2010-10-21T05:32:02 *** farbe has quit IRC (Read error: Operation timed out) 2010-10-21T05:33:26 *** farbe has joined #aichallenge 2010-10-21T05:34:01 ahwwww its down again. janzert ur there? 2010-10-21T05:36:01 *** farbe has quit IRC (Read error: Operation timed out) 2010-10-21T05:36:31 *** farbe has joined #aichallenge 2010-10-21T05:52:37 *** HakanD has joined #aichallenge 2010-10-21T06:04:19 *** phreeza has joined #aichallenge 2010-10-21T06:10:55 Top 10 players: rsergio(3912), dmj111(3909), deepblue(3890), cfaftw(3874), savchenko(3866), smatcher(3845), Hazard(3812), chotting(3753), Rexxar(3736), ofrias(3725) 2010-10-21T06:13:54 *** wvdschel has quit IRC (Ping timeout: 240 seconds) 2010-10-21T06:16:24 *** smellyhippy has quit IRC (Read error: Connection reset by peer) 2010-10-21T06:16:40 *** smellyhippy has joined #aichallenge 2010-10-21T06:16:41 *** smellyhippy has joined #aichallenge 2010-10-21T06:22:07 *** mceier has quit IRC (Read error: Operation timed out) 2010-10-21T06:22:59 *** mceier has joined #aichallenge 2010-10-21T06:35:13 *** smellyhippy has quit IRC (Quit: BUT WAIT! What are those penguins up to?!) 2010-10-21T06:47:12 *** smellyhippy has joined #aichallenge 2010-10-21T06:55:06 *** Utkarsh has quit IRC (Ping timeout: 250 seconds) 2010-10-21T07:00:11 *** Utkarsh has joined #aichallenge 2010-10-21T07:02:55 *** jesionaj` has quit IRC (Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com )) 2010-10-21T07:04:15 *** p4p4 has quit IRC (Read error: Connection timed out) 2010-10-21T07:07:15 *** p4p4 has joined #aichallenge 2010-10-21T07:08:14 *** wvdschel has joined #aichallenge 2010-10-21T07:08:14 *** wvdschel has joined #aichallenge 2010-10-21T07:10:55 Top 10 players: rsergio(3888), dmj111(3881), Hazard(3872), cfaftw(3850), savchenko(3831), deepblue(3825), smatcher(3820), chotting(3734), ofrias(3723), Rexxar(3713) 2010-10-21T07:11:16 *** Accoun has quit IRC (Ping timeout: 276 seconds) 2010-10-21T07:23:20 *** noid51468 has joined #aichallenge 2010-10-21T07:23:38 *** noid51468 has left #aichallenge 2010-10-21T07:23:48 *** Accoun has joined #aichallenge 2010-10-21T07:27:59 *** oPless has quit IRC (Quit: This computer has gone to sleep) 2010-10-21T07:44:59 *** Veracious has joined #aichallenge 2010-10-21T07:45:51 *** avoinea has joined #aichallenge 2010-10-21T07:51:16 how big is the benefit of using a compiled language like c++ when compared with an interpreted one in terms of this challenge? 2010-10-21T07:53:23 In the last challenge (tron), compiled languages made a bigger difference. 2010-10-21T07:54:03 This time, so far, interpreted languages can be fast enough. If someone discovers how to do minimax or something like that, intepreted languages will fall behind. 2010-10-21T07:55:30 *** HakanD has quit IRC (Ping timeout: 240 seconds) 2010-10-21T07:56:35 hasn't someone found a way yet? =) 2010-10-21T07:56:47 btw secondary server seems offline again 2010-10-21T07:57:17 someone may have, but it is hard to tell ... bocsimacko 2010-10-21T07:57:33 's bot is awesome, but I doubt it is minimax (not enough time to search the log to see if he says...) 2010-10-21T08:01:09 he's not in the top 10? or does he use a different name 2010-10-21T08:01:43 *** perestrelka has quit IRC (Read error: Connection reset by peer) 2010-10-21T08:02:18 http://72.44.46.68/getratings 2010-10-21T08:02:18 *** perestrelka has joined #aichallenge 2010-10-21T08:02:23 he's on that rating list 2010-10-21T08:02:45 Error323: http://ai-contest.com/profile.php?user_id=8565 2010-10-21T08:03:47 Last submission's from yesterday so he's still climbing the rankings. 2010-10-21T08:06:13 *** p4p4 has quit IRC (Read error: Connection timed out) 2010-10-21T08:06:22 i see 2010-10-21T08:07:09 *** p4p4 has joined #aichallenge 2010-10-21T08:07:12 aerique: good that he got the lisp problems figured out. 2010-10-21T08:07:48 dmj111: yeah, like most other languages we're also stuck with an ancient implementation 2010-10-21T08:07:49 Error323: sorry about the name dropping out of context. I am not sure who is here when. bocsimacko is mega1 here. 2010-10-21T08:08:10 *** yasith has joined #aichallenge 2010-10-21T08:08:32 aerique: yeah... I don't mind the old python, because at the defense contractor I used to work at, we had just got python 2.3 two years ago. 2.5 is like a dream compared to that! 2010-10-21T08:08:32 lol bug? http://ai-contest.com/visualizer.php?game_id=5970112 2010-10-21T08:08:47 (Well... the difference between 2.5 and 2.3 seems much smaller than 2.3 to 2.2) 2010-10-21T08:09:24 maybe next time around, the server will be updated. 2010-10-21T08:10:05 raedwulf: I think it's timing out and returns no orders. 2010-10-21T08:10:43 dmj111: i hope so, this server version was even older than the one for tron 2010-10-21T08:10:47 I had to lower the timeout 0.8s to avoid failing outright but the official server is way too slow. 2010-10-21T08:10:48 ouch. 2010-10-21T08:10:55 Top 10 players: rsergio(3895), dmj111(3887), Hazard(3877), cfaftw(3855), savchenko(3837), deepblue(3830), smatcher(3824), chotting(3740), ofrias(3729), Rexxar(3718) 2010-10-21T08:11:05 I timeout at 0.5 seconds, just to be sure. 2010-10-21T08:11:05 mega1: ah ok 2010-10-21T08:11:07 dmj111: i think we're more than three years behind on bug fixes and optimisations for sbcl 2010-10-21T08:11:35 my bot is still in its infancy hehe 2010-10-21T08:11:41 hopefully get a working version today 2010-10-21T08:11:44 mega1: have you timed out yet on the lasted upload? 2010-10-21T08:11:51 yikes. I am a wanna-be lisper, so I do cheer for you guys. Sorry that you are having technical difficulties. 2010-10-21T08:11:59 aerique: not yet 2010-10-21T08:12:18 dmj111: there is no try, come over to the dark side :) 2010-10-21T08:12:23 I have on loss and that's legitimate 2010-10-21T08:13:22 on a related note, it would certainly be good to know the memory/cpu power that is available 2010-10-21T08:13:34 *** HakanD has joined #aichallenge 2010-10-21T08:13:50 mega1: janzert should be able to answer that 2010-10-21T08:13:57 no point in tree searching if it's 5 times slower than my laptop 2010-10-21T08:14:22 aerique: :) I have read a bunch of the books, and written some toy programs. My excuse is still that I cannot use it at work. 2010-10-21T08:15:33 In fact, tron was the first time I implemented minimax, and I had PAIP on my desk for the whole contest. 2010-10-21T08:15:41 dmj111: ah right, same problem here. java & jboss... *shudder* i sneak it in sometimes instead of shell scripts or to solve a problem at work that doesn't need source code (just output / results) 2010-10-21T08:15:58 * mega1 is thankful to have been programming in Lisp professionally for 6 years 2010-10-21T08:16:23 yeah.... I have been lucky (?) enough to use F# lately. Its much better than using the C variants. 2010-10-21T08:16:51 I did java but it was incredibly boring. 2010-10-21T08:16:54 i'm trying to get out of wage slavery and start for myself, then it will be Lisp all the way 2010-10-21T08:18:14 well. better go. good luck with the server issues guys. 2010-10-21T08:18:43 * dmj111 uses guys as a plural noun for any group of people :/ 2010-10-21T08:27:01 *** Titankiller has joined #aichallenge 2010-10-21T08:30:21 *** p4p4 has quit IRC (Read error: Connection timed out) 2010-10-21T08:30:25 im writing it in C (yikes) 2010-10-21T08:30:26 lol 2010-10-21T08:30:45 i was considering haskell, but i don't know enough haskell yet 2010-10-21T08:31:57 *** p4p4 has joined #aichallenge 2010-10-21T08:34:41 *** Mathnerd314 has joined #aichallenge 2010-10-21T08:37:43 i'm ready to rage kill people 2010-10-21T08:38:05 i've spend a day rewriting the backend and have had to scratch and start over again cause it just got too messed up 2010-10-21T08:44:52 *** Appleman1234 has quit IRC (Ping timeout: 252 seconds) 2010-10-21T08:52:26 *** p4p4 has quit IRC (Read error: Connection timed out) 2010-10-21T08:53:45 lol 2010-10-21T08:53:49 *** p4p4 has joined #aichallenge 2010-10-21T08:58:55 *** Titankiller has quit IRC (Read error: Connection reset by peer) 2010-10-21T08:58:56 *** rommel has joined #aichallenge 2010-10-21T08:59:12 g'day all :) 2010-10-21T09:05:25 *** Tankh has quit IRC (Remote host closed the connection) 2010-10-21T09:08:59 *** delt0r___ has joined #aichallenge 2010-10-21T09:10:29 *** delt0r has quit IRC (Ping timeout: 265 seconds) 2010-10-21T09:10:56 Top 10 players: rsergio(3899), dmj111(3883), Hazard(3883), cfaftw(3858), savchenko(3837), deepblue(3834), smatcher(3775), chotting(3758), Accoun(3753), ofrias(3740) 2010-10-21T09:11:53 *** hellman has joined #aichallenge 2010-10-21T09:26:32 *** yasith has quit IRC (Remote host closed the connection) 2010-10-21T09:28:45 *** Azrathud has quit IRC (Quit: Meh) 2010-10-21T09:31:55 *** yasith has joined #aichallenge 2010-10-21T09:39:43 *** p4p4 has quit IRC (Ping timeout: 245 seconds) 2010-10-21T10:10:57 Top 10 players: rsergio(3899), Hazard(3883), dmj111(3882), cfaftw(3857), savchenko(3834), deepblue(3833), smatcher(3776), chotting(3756), Accoun(3754), ofrias(3740) 2010-10-21T10:13:16 *** bhasker has quit IRC (Ping timeout: 255 seconds) 2010-10-21T10:15:15 *** Itkovian has joined #aichallenge 2010-10-21T10:20:04 *** bhasker has joined #aichallenge 2010-10-21T10:21:23 *** JamesMG has joined #aichallenge 2010-10-21T10:34:18 *** krokkrok has joined #aichallenge 2010-10-21T10:41:36 *** p4p4 has joined #aichallenge 2010-10-21T10:51:37 *** krokkrok has quit IRC (Quit: Page closed) 2010-10-21T10:52:27 yo antimatroid 2010-10-21T10:52:28 Are you there? 2010-10-21T10:55:32 *** aerique has quit IRC (Quit: ...) 2010-10-21T11:02:59 *** Itkovian has quit IRC (Quit: Itkovian) 2010-10-21T11:04:19 my bot does weird stuff now 2010-10-21T11:04:20 http://72.44.46.68/canvas?game_id=609617 2010-10-21T11:05:23 *** Bobng has joined #aichallenge 2010-10-21T11:06:43 *** hornairs has joined #aichallenge 2010-10-21T11:09:43 *** Bobng has quit IRC (Remote host closed the connection) 2010-10-21T11:10:11 *** Bobng has joined #aichallenge 2010-10-21T11:10:57 Top 10 players: Hazard(3919), rsergio(3885), dmj111(3873), cfaftw(3846), ofrias(3819), deepblue(3811), savchenko(3800), Rexxar(3766), smatcher(3764), chotting(3746) 2010-10-21T11:14:39 *** Bobng has quit IRC (Client Quit) 2010-10-21T11:15:56 rommel: i am 2010-10-21T11:16:07 i am hopefully not too far off having a new and amazing bot 2010-10-21T11:16:58 niceee 2010-10-21T11:17:14 I've just finished writing the whole thing from scratch :P 2010-10-21T11:18:29 i have redone my backend quite a lot 2010-10-21T11:18:39 hoping that will give it the extra kick i'm looking for 2010-10-21T11:18:48 strategically it will, if i don't fuck it up 2010-10-21T11:19:20 well, this consists of my main... I did away completely with the planetwars.cc :P http://www.pastie.org/1238296 2010-10-21T11:19:29 antimatroid: is it playing yet? 2010-10-21T11:23:46 phreeza: not even testing yet 2010-10-21T11:23:50 but not far off trying to test 2010-10-21T11:25:14 i rewrote my backend in C :P 2010-10-21T11:26:11 i have pluggable bots, currently i can run either a C port of the rage bot or bully bot 2010-10-21T11:26:22 not sure if thats any use, but i needed something to test my backend 2010-10-21T11:27:55 *** hornairs has quit IRC (Quit: hornairs) 2010-10-21T11:30:36 *** hornairs has joined #aichallenge 2010-10-21T11:34:43 *** boegel has quit IRC (Ping timeout: 255 seconds) 2010-10-21T11:35:40 *** Itkovian has joined #aichallenge 2010-10-21T11:39:13 *** wvdschel has quit IRC (Quit: Leaving) 2010-10-21T11:43:22 *** Sarin has joined #aichallenge 2010-10-21T11:45:22 *** lavalamp_ has joined #aichallenge 2010-10-21T11:54:54 *** avoinea has quit IRC (Quit: Leaving) 2010-10-21T11:55:55 *** Itkovian has quit IRC (Ping timeout: 240 seconds) 2010-10-21T11:58:26 *** Mekanik has joined #aichallenge 2010-10-21T12:03:31 *** krokkrok has joined #aichallenge 2010-10-21T12:05:02 *** Mekanik has joined #aichallenge 2010-10-21T12:05:55 raedwulf 2010-10-21T12:05:57 in C!? 2010-10-21T12:05:59 You crazy :P 2010-10-21T12:06:05 *** Bobng has joined #aichallenge 2010-10-21T12:06:06 Surely the memory management was a hassle? 2010-10-21T12:06:21 rommel: not really, most of mine was alloc alloc alloc 2010-10-21T12:06:25 Its made my first time usage of vectors a positive one :P 2010-10-21T12:06:26 and one big free at the end 2010-10-21T12:06:32 haha 2010-10-21T12:06:35 I'll say. 2010-10-21T12:06:43 i have a simple memory management syste 2010-10-21T12:06:45 system* 2010-10-21T12:06:51 Do you rebuild the entire gamestate each time? Or just update 2010-10-21T12:06:57 it is a slight bit of a hassle, but im familiar with it 2010-10-21T12:07:16 i separate the static gamestate from the dynamic gamestate in my structures 2010-10-21T12:07:28 ie planets from fleets? 2010-10-21T12:07:42 sorry thats wrongly said (static data) and (dynamic state) 2010-10-21T12:07:47 yep, planets are static mostly 2010-10-21T12:07:51 apart from ships/ownership 2010-10-21T12:09:00 This is my parser, which sends the data packets to my instance which handles all the planets/fleets : http://www.pastie.org/private/q2c2gd6trvhfc68uymwagg 2010-10-21T12:09:18 nice 2010-10-21T12:09:42 *** yasith has quit IRC (Ping timeout: 250 seconds) 2010-10-21T12:09:48 mine is a bit longer 2010-10-21T12:09:58 * raedwulf is pastebinning 2010-10-21T12:10:32 errrm wait i have some of my implementation detail in there 2010-10-21T12:10:43 i.e. i have the driver inside it as well 2010-10-21T12:10:58 Top 10 players: Hazard(3910), rsergio(3874), dmj111(3863), cfaftw(3836), ofrias(3811), deepblue(3802), savchenko(3790), Rexxar(3756), smatcher(3755), chotting(3734) 2010-10-21T12:11:30 the driver? 2010-10-21T12:11:51 i call it a driver, i.e. the main loop 2010-10-21T12:11:56 gotcha 2010-10-21T12:12:24 essentially the key bit is sscanf(p+2, "%lf %lf %d %d %d",....) 2010-10-21T12:12:26 for parsing 2010-10-21T12:12:29 haha 2010-10-21T12:12:35 that would have been useful :) 2010-10-21T12:12:52 I actually wrote the entire parser in C 2010-10-21T12:12:58 ^^ 2010-10-21T12:13:00 then decided I'd be better of using C++ and the STL 2010-10-21T12:13:14 i started off with a C++ implementation 2010-10-21T12:13:19 why write a new parser different from the starter bot, whats the point? 2010-10-21T12:13:20 but im so much more comfortable with C 2010-10-21T12:13:57 krokkrok: they seeem to do a lot of appends and its not as clean 2010-10-21T12:14:07 also, reinventing the wheel is cool 2010-10-21T12:14:08 why does it matter ? 2010-10-21T12:14:34 gives me time to keep my hands busy as i think about the algorithm (but i wrote it in C) 2010-10-21T12:14:37 Well krokkrok, I wanted to learn a bit, had never used the STLs before... seemed a good place to start 2010-10-21T12:14:38 there's no C starter pack afaik 2010-10-21T12:15:09 plus it looks pwetty haha 2010-10-21T12:15:59 I also needed to change a few things in how it handled the overall system 2010-10-21T12:16:01 STL is the devil 2010-10-21T12:16:14 So regardless I had to modify it 2010-10-21T12:17:00 Hated how I couldn't store variables to planets and then they'd be deleted on the vector reset 2010-10-21T12:17:37 It only allowed for reaction, no prediction or prior knowledge could be had at all 2010-10-21T12:18:00 *** cfaftw has joined #aichallenge 2010-10-21T12:18:16 And checking to see where incoming fleets were coming from was a big process, cycling through all the planets... every time... now its handled in 1 line ... 2010-10-21T12:18:33 and only once per fleet. 2010-10-21T12:18:53 i had difficulty initially because i did a big fread, and for some reason it didn't work 2010-10-21T12:19:09 fread? 2010-10-21T12:19:18 i think i know what the problem is now, but it made some fo the code lots simpler when i used a fgets and a sscanf 2010-10-21T12:19:40 fread for reading stdin 2010-10-21T12:19:47 i was trying to read an entire state into one huge buffer 2010-10-21T12:19:47 you guys get any rank in the tournament yet? 2010-10-21T12:20:03 im ~1990 with my bot 2010-10-21T12:20:04 I only started yesterday... :) 2010-10-21T12:20:07 lol 2010-10-21T12:20:29 its actually winning though, and testing my base code - with a reimplementation of the rage bot in C 2010-10-21T12:21:05 But I have yet to implement my AAN and get it learning... that'll take a while... THEN, and only then, once its trained... will it hopefully work :D 2010-10-21T12:21:32 neat 2010-10-21T12:21:37 im going for a less complex approachj 2010-10-21T12:21:53 but making my stuff reusable so i can try some other stuff 2010-10-21T12:22:00 my last bot hit rank 31, and its dumb as a stump, written in java which i don't even know, and hacked to high heaven :D 2010-10-21T12:22:07 lmao 2010-10-21T12:22:13 *** tapwater has joined #aichallenge 2010-10-21T12:22:17 Well I plan until I finish certain part of my ANN**, to just use a heuristics approach :P... and lol krokkrok 2010-10-21T12:22:21 the thing is that i believe the search space is immense 2010-10-21T12:23:36 ANN mean neural network right? How you gonna train it? 2010-10-21T12:24:08 play bots i guess 2010-10-21T12:24:12 lol 2010-10-21T12:24:55 i was thinking maybe you can download all the previous games and learn from them, playing bots is really slow these days. it was 80 an hour for the first 3 weeks, now its down to 5 2010-10-21T12:25:30 *80 a minute 2010-10-21T12:25:31 *** HakanD has quit IRC (Ping timeout: 265 seconds) 2010-10-21T12:26:21 krokkrok: it was 40-50 earlier 2010-10-21T12:26:26 but the second server went down again 2010-10-21T12:26:33 it looks like the main server is the slowest 2010-10-21T12:27:05 i think they are not going down so much as not being paid for. I think they have a pay scale that they only play when server time is especially cheap. 2010-10-21T12:27:22 ah ok 2010-10-21T12:27:29 um 2010-10-21T12:27:50 playing locally without the visualiser you can get up to 20 games a second? 2010-10-21T12:27:56 krokkrok: no they're really going down for some unknown reason 2010-10-21T12:27:57 atleast 2010-10-21T12:27:58 krokkrok: do you believe that it's so low that servers are down for 5 days? 2010-10-21T12:28:12 if you haven't seen this, this is useful for testing http://72.44.46.68/ 2010-10-21T12:28:27 errr why do i have an ip in that url 2010-10-21T12:28:29 wait a sec 2010-10-21T12:28:34 yeah, I've done most of my testing on that :) 2010-10-21T12:28:41 oh ok ^^ 2010-10-21T12:28:45 JBotManager ftw :P 2010-10-21T12:28:46 http://www.benzedrine.cx/planetwars/ i meant 2010-10-21T12:28:48 hehe 2010-10-21T12:28:51 i've just been scanning this chat and the forums, and people say that the number of servers is limited by cost...its just a rumor i guess 2010-10-21T12:29:13 maybe someone keeps kicking the server over, and it goes down with a crash 2010-10-21T12:29:14 @seen j3camero 2010-10-21T12:29:14 Bobng: j3camero was last seen in #aichallenge 2 weeks, 2 days, 14 hours, 9 minutes, and 29 seconds ago: amstan: yes, but no names, haha. 2010-10-21T12:29:22 RIP 2010-10-21T12:29:34 *** JensTi has joined #aichallenge 2010-10-21T12:30:09 krokkrok: I think several people have volunteered to fund some cloud server instances 2010-10-21T12:30:14 my reason is by far more plausible ^^ 2010-10-21T12:30:18 *** JCS^ has joined #aichallenge 2010-10-21T12:30:34 yeah i saw that people offered to fund it, but still, we are at 5 games a minute :( 2010-10-21T12:31:22 I am willing to donate a fair few resources in the form of a VM 2010-10-21T12:31:27 in return for some publicity 2010-10-21T12:31:47 I just don't think any organizers with the right access have been around 2010-10-21T12:32:36 it's bad when the truck number of the organization is 0 2010-10-21T12:32:49 krokkrok: how am I going to train it? Well the inputs are easy; but I'm still looking into how I'm going to determine a fitness for it; I might just end up using a mix of Evolution, ACO for path determination and Backprop for the planet and fleet decisions... one of which I've already tested and it works well in this situation... just whether it will actually be worth it in the end is my interest... it probably just won't be ab 2010-10-21T12:33:46 *** smhanov__ has quit IRC (Remote host closed the connection) 2010-10-21T12:34:29 luckily I've written them all in plug and play format in C++ so implementing isn't the difficult part, just the pre and post processing on the inputs and outputs is where the hours will be spent 2010-10-21T12:35:03 i have no idea how to train, if I did, I'd do that and with the competition :D 2010-10-21T12:36:21 *with=win 2010-10-21T12:36:51 rommel, what's ACO ? 2010-10-21T12:37:00 ant colony optimisation 2010-10-21T12:37:19 k. 2010-10-21T12:37:37 *** smellyhippy has quit IRC (Ping timeout: 265 seconds) 2010-10-21T12:38:21 hmm ant colony ..is that what mcleopold's bot uses too? 2010-10-21T12:38:26 lol "ant colony optimisation" made me giggle.. started thinking about optimising those poor little ants 2010-10-21T12:39:36 *** JCS^ has quit IRC (Quit: \x00) 2010-10-21T12:39:48 krokkrok, to train, you need to loop back the output into the input so it can learn from itself and its previous, but the most important part is the fitness of net, basically what your saying is a good outcome and a bad one... how to determine that; since the feedback can only come after a result 2010-10-21T12:40:08 is going to be a fun task 2010-10-21T12:40:25 input is huge, isn't it? 2010-10-21T12:41:01 I was thinking of game theory as a comparison... but then its sortive like... well I'm just going to train the network to do game theory... which could be done heuristically so... 2010-10-21T12:41:13 sort of* 2010-10-21T12:41:13 well be sure to use the tcp server, you can do everything as fast as you want instead of waiting for the official servers 2010-10-21T12:41:22 already have been :) 2010-10-21T12:41:54 so i should stop using it so i dont traing your bot...wait, im retarded, don't learn from me hehe 2010-10-21T12:42:09 hahah 2010-10-21T12:42:14 *** smellyhippy has joined #aichallenge 2010-10-21T12:42:15 *** smellyhippy has joined #aichallenge 2010-10-21T12:42:36 thats just it 2010-10-21T12:42:40 *** Naktibalda has quit IRC (Quit: ChatZilla 0.9.86 [Firefox 3.6.8/20100722155716]) 2010-10-21T12:42:43 to be able to beat the pros... I'd need samples of their bots :P 2010-10-21T12:43:00 any plans in effect to prevent it learning effective strategies for beating say, Ragebot 2010-10-21T12:43:02 Unless I set it up so it could learn from replays... infact that would probably be the best way... that way you can determine a fitness function easily ... :) 2010-10-21T12:43:04 dmj11 is up there, and he's among the best 2010-10-21T12:43:05 and applying them every time? 2010-10-21T12:43:29 James 2010-10-21T12:43:35 Ragebot is the simplest thing to beat haha 2010-10-21T12:43:39 Just don't do anything... it loses 2010-10-21T12:43:44 yeah exactly 2010-10-21T12:43:47 Not hard to determine a ragebot either 2010-10-21T12:44:06 I mean to prevent it from learning that as a valid strategy based on wins against ragebot 2010-10-21T12:44:27 yeah... hence I think selective large samplings of replays will be the best solution 2010-10-21T12:44:42 Real time learning sounds like it may be very difficult... 2010-10-21T12:45:04 Simply because the inputs and outputs are not immediately determined 2010-10-21T12:55:55 *** seisatsu has quit IRC (Quit: Leaving) 2010-10-21T12:59:19 this oldman3 bot is the bane of my existence 2010-10-21T12:59:20 :-\ 2010-10-21T12:59:25 and Accoun* variants 2010-10-21T13:00:05 it's pretty weak on close-start maps at least 2010-10-21T13:00:06 haha 2010-10-21T13:00:31 yea 2010-10-21T13:00:37 but its strong on maps where it spreads out 2010-10-21T13:00:57 haha shit http://72.44.46.68/canvas?game_id=610506 2010-10-21T13:00:57 once that bot starts up 2010-10-21T13:01:00 all i see are losses 2010-10-21T13:01:07 my bot timed out on that one. 0.5.1 fixes that, i might as well kill 0.5 2010-10-21T13:01:22 heh 2010-10-21T13:01:52 *** JCS^ has joined #aichallenge 2010-10-21T13:01:54 not sure why my bot stopped trying to attack 2010-10-21T13:01:59 this bot is interesting: http://72.44.46.68/canvas?game_id=605129 2010-10-21T13:02:02 *** JCS^ has quit IRC (Client Quit) 2010-10-21T13:02:14 i never would have thought to do that but it makes sense. i'll ahve to alter my defense a little. 2010-10-21T13:02:29 and looks like it made a mistake 2010-10-21T13:02:32 it sent 22 ships to capture 2010-10-21T13:02:36 on the second attempt 2010-10-21T13:02:50 which was wrong 2010-10-21T13:02:55 ah bugs 2010-10-21T13:03:38 off by two 2010-10-21T13:03:43 yea 2010-10-21T13:03:54 yeah that was the critical mistake i guess 2010-10-21T13:06:00 *** krokkrok has quit IRC (Quit: Page closed) 2010-10-21T13:07:19 i just got killed by oldman3 too 2010-10-21T13:07:35 heh 2010-10-21T13:07:37 the center two planets just kept blinking blue and red 2010-10-21T13:08:34 heh 2010-10-21T13:10:42 i need to change this parse_game_state_url.py to swap players around when my bot is player 2 on the page 2010-10-21T13:10:59 Top 10 players: Hazard(4004), dmj111(3844), rsergio(3813), cfaftw(3799), savchenko(3779), ofrias(3773), deepblue(3769), Accoun(3766), sin_sun(3744), Rexxar(3728) 2010-10-21T13:11:10 *** hornairs has joined #aichallenge 2010-10-21T13:14:44 k fixed the parse script 2010-10-21T13:14:57 bhasker: http://pastebin.com/XqLYEhzA by appending swap as 2nd argument (after the game id) it gets swapped 2010-10-21T13:15:24 ah too late 2010-10-21T13:15:27 yea i just did that 2010-10-21T13:15:31 i had an earlier version 2010-10-21T13:15:33 which didnt do it 2010-10-21T13:15:36 so i just added it 2010-10-21T13:16:03 *** hellman has quit IRC (Remote host closed the connection) 2010-10-21T13:21:33 @rankings 2010-10-21T13:21:34 amstan_: Top 10 players: Hazard(4004), dmj111(3845), rsergio(3812), cfaftw(3800), savchenko(3777), ofrias(3773), deepblue(3771), Accoun(3767), sin_sun(3746), Rexxar(3726) 2010-10-21T13:21:40 *** gra has joined #aichallenge 2010-10-21T13:21:58 my bot crashes :S 2010-10-21T13:22:15 my bot isn't complete :S 2010-10-21T13:22:59 *** gra has quit IRC (Client Quit) 2010-10-21T13:23:42 lol 2010-10-21T13:23:49 *** wh1teside has joined #aichallenge 2010-10-21T13:34:05 *** RainCT has joined #aichallenge 2010-10-21T13:41:33 *** Florian95 has quit IRC (Remote host closed the connection) 2010-10-21T13:43:49 *** boegel has joined #aichallenge 2010-10-21T13:45:13 *** narnach has quit IRC (Ping timeout: 255 seconds) 2010-10-21T13:50:08 *** boegel has quit IRC (Ping timeout: 245 seconds) 2010-10-21T13:50:30 *** narnach has joined #aichallenge 2010-10-21T13:50:33 *** brunswick has joined #aichallenge 2010-10-21T13:54:03 *** Itkovian has joined #aichallenge 2010-10-21T13:57:11 http://72.44.46.68/canvas?game_id=611635 2010-10-21T13:58:44 * p4p4 "To reduce the numbers of unhappy people, always beat up the same individuals." 2010-10-21T13:59:06 is there a first move that doesn't result in a loss on this that I'm not seeing? 2010-10-21T13:59:31 *** smellyhippy has quit IRC (Ping timeout: 265 seconds) 2010-10-21T14:04:09 *** Itkovian has quit IRC (Quit: Itkovian) 2010-10-21T14:04:19 *** smellyhippy has joined #aichallenge 2010-10-21T14:04:19 *** smellyhippy has joined #aichallenge 2010-10-21T14:06:01 *** Itkovian has joined #aichallenge 2010-10-21T14:07:57 nope, that's one of those maps that best play is to sit and tie 2010-10-21T14:08:38 Yeah, looks like a tie to me 2010-10-21T14:08:43 *** amstan_ is now known as amstan 2010-10-21T14:08:59 Sigh... I'm going nuts over this stupid algo :P 2010-10-21T14:09:02 just noted the symmetry in al those maps, is there a reason for it? 2010-10-21T14:09:13 it's so each game is fair 2010-10-21T14:09:33 Anyone got any ideas why this *doesnt* work? 2010-10-21T14:09:34 http://www.pastie.org/1238696 2010-10-21T14:09:55 It returns the closest sometimes... but then randomly choses planets not close at all 2010-10-21T14:10:12 Yet AFAIK... it should chose n amount iteratively... 2010-10-21T14:10:34 what is it supposed to do? 2010-10-21T14:10:52 *** BtbN has joined #aichallenge 2010-10-21T14:10:56 Find n amount of planets, closest to a planet 2010-10-21T14:11:01 Top 10 players: Hazard(4036), dmj111(3867), rsergio(3842), Accoun(3804), savchenko(3804), deepblue(3791), sin_sun(3789), cfaftw(3781), ofrias(3776), Rexxar(3760) 2010-10-21T14:11:39 you want the n closest planets to planet x? 2010-10-21T14:11:39 rommel: you should get an array of those planets, then sort them by distance 2010-10-21T14:11:56 *** delt0r has joined #aichallenge 2010-10-21T14:11:58 correct lavalamp 2010-10-21T14:12:03 if so, what amstan said 2010-10-21T14:12:04 I didn't want to have to do that amstan :P 2010-10-21T14:12:12 You can cache it 2010-10-21T14:12:13 *** delt0r___ has quit IRC (Ping timeout: 245 seconds) 2010-10-21T14:12:14 this should work... lol 2010-10-21T14:12:24 its like a minesweeping method 2010-10-21T14:12:31 rommel, did you take a class on neural networks? 2010-10-21T14:12:44 yes cfaftw 2010-10-21T14:12:55 rommel, well mostly i'm just interested in how you learned to use them... i've been wanting to learn for a while 2010-10-21T14:12:57 rommel: what does side mean? 2010-10-21T14:12:59 but i'm out of school now 2010-10-21T14:13:28 side of source, ie don't return friendly planets... can remove it for now 2010-10-21T14:13:33 oh 2010-10-21T14:15:22 rommel, make a 2d array with every planet sorted by distance. Then this can be an O(n) function (just check every planet in the list to see if it's friendly) 2010-10-21T14:15:50 yeah that was easy and worked lava... but this *SHOULD* work... I don't get why not lol :S 2010-10-21T14:17:07 dist > (d-2) 2010-10-21T14:17:14 what does that do? 2010-10-21T14:17:58 ok it works now... NFI idea why... 2010-10-21T14:18:08 like I said 2010-10-21T14:18:15 to get the 4 closest 2010-10-21T14:18:36 I just did the following: sweep 2m, if not N found, then increase sweep radius, but, dont sweep old 2010-10-21T14:19:24 I'm thinking now though that a 2D array may just be better... lol 2010-10-21T14:19:58 a 2d array is definitely better 2010-10-21T14:21:16 also it's probably more useful to return indexes (vector) rather than copies of the planets 2010-10-21T14:22:25 I know :) 2010-10-21T14:22:43 :) 2010-10-21T14:22:45 And I do... I was just doing something quick and dirty lol 2010-10-21T14:25:18 *** p4p4 has quit IRC (Read error: Connection timed out) 2010-10-21T14:26:52 http://www.pastie.org/1238741 2010-10-21T14:27:04 That's how I do it (golang) 2010-10-21T14:29:34 now that I paste that there are a bunch of things I want to improve about it but it gets the job done 2010-10-21T14:32:19 *** brunswick has quit IRC (Quit: Page closed) 2010-10-21T14:36:40 any luck getting it on the official server lavalamp_? 2010-10-21T14:38:05 Still haven't tried what we were talking about yesterday 2010-10-21T14:38:20 I spent my time debugging it instead... 2010-10-21T14:38:45 I think pretty much everything is working now except my move generation code 2010-10-21T14:39:47 atm it chooses between doing nothing or simulating ragebot... not exactly useful 2010-10-21T14:40:01 haha 2010-10-21T14:40:07 well... its 5:40am ... 2010-10-21T14:40:11 :| 2010-10-21T14:40:17 well, at least it simulates ragebot intelligently 2010-10-21T14:40:23 So I'm off to sleep haha, ttyl 2010-10-21T14:40:26 although the look-ahead keeps it from choosing ragebot's moves most of the time, so that's good 2010-10-21T14:40:31 see ya rommel 2010-10-21T14:40:52 *** rommel has quit IRC (Quit: Page closed) 2010-10-21T14:41:08 and the evaluation steadily gets more negative as it's losing worse, so that's good too 2010-10-21T14:41:32 that's definitely a big step right there 2010-10-21T14:42:26 I think it will suffer from a horizon effect... :/ 2010-10-21T14:42:55 yeah, when mine runs it has a very greedy outlook on things 2010-10-21T14:43:33 I need to work on it's move generation code to give it only moves that have a chance of paying off 2010-10-21T14:43:39 Also I run the opponent's moves as if mine were already made... I'm still debating if that will help my bot or paralyze it 2010-10-21T14:43:41 and adjustable depth tree, I think 2010-10-21T14:45:32 mine spends 50% of its time predicting the future... I think I can get that down by not predicting all the way through the end until someone asks for it 2010-10-21T14:46:07 how far do you predict/plan? I was thinking I'd have mine add plies until it was nearly out of time 2010-10-21T14:46:14 then pick one and act, each turn 2010-10-21T14:46:49 I do that too, but my move generators will be looking into the future 2010-10-21T14:46:58 *** McLeopold has joined #aichallenge 2010-10-21T14:47:25 I might have to learn enough Go to read your bot after this is all done 2010-10-21T14:47:52 right now it can do ~2000 nodes in a second, but most of them are doing nothing so I expect that to drop :( 2010-10-21T14:48:04 lol, you should learn it anyway :) 2010-10-21T14:48:13 I just hope it's worth reading! 2010-10-21T14:48:41 *** treyryte has joined #aichallenge 2010-10-21T14:48:47 1 2010-10-21T14:48:55 I'm most interested in how you build the tree, since that's where mine is falling down at the moment 2010-10-21T14:49:41 I don't actually keep a tree around, maybe I should... 2010-10-21T14:49:57 mine is moving a lot more data than it needs to also, though 2010-10-21T14:50:25 I just remember the best thing I've found so far and all the moves on all the current stack levels 2010-10-21T14:50:37 Yours is Java? 2010-10-21T14:50:42 C# 2010-10-21T14:50:46 ah ok 2010-10-21T14:51:44 Maybe I could squeeze out a few more nodes if I stored a tree, hm 2010-10-21T14:51:48 I originally had it copying the game board for each node, and then each planet evaluation/ship count prediction would return a planet object as well 2010-10-21T14:51:53 есть русские? 2010-10-21T14:51:59 *** delt0r___ has joined #aichallenge 2010-10-21T14:52:11 ah yeah, no I just have one board and everything uses it 2010-10-21T14:52:21 *** delt0r has quit IRC (Read error: Operation timed out) 2010-10-21T14:52:21 but I was trying to track down a state issue when I did that 2010-10-21T14:52:24 and just never undid it 2010-10-21T14:52:59 I would make a second board if I made it multi-threaded 2010-10-21T14:53:04 *** treyryte has quit IRC (Client Quit) 2010-10-21T14:53:09 Did I miss a source code link? 2010-10-21T14:53:16 nope 2010-10-21T14:53:19 nope 2010-10-21T14:53:37 I'm just hoping lavalamp_ opens his source after the contest :) 2010-10-21T14:53:51 unless you want to see my ugly distance calculations that I showed rommel earlier 2010-10-21T14:54:05 ah 2010-10-21T14:54:21 If I'm not embarrassed about it, I'll open it :) 2010-10-21T14:54:23 Are you guys having any success with minimax then? 2010-10-21T14:54:34 remains to be seen, still 2010-10-21T14:55:05 How do you solve the problem of too many branches? Do you just go 1 or 2 ply? 2010-10-21T14:55:07 mine operates at the sample bot level, +1 2010-10-21T14:55:51 yeah mine is probably going to end up at 1 or 2 ply... it does 3 right now but most of my move generators do nothing so that will change 2010-10-21T14:56:02 I solved that initially by giving it a very limited list of options, mostly individual orders my other bot would make 2010-10-21T14:56:18 And then the same for the opponent I would guess. 2010-10-21T14:56:22 I'm doing something similar 2010-10-21T14:56:38 for the opponent I will have some other generators too, such as ragebot 2010-10-21T14:56:48 I assumed(obviously incorrectly) that the other would run similar tactics 2010-10-21T14:56:52 just for the sake of getting it running 2010-10-21T14:57:48 Do you pair down insignificate boards? 2010-10-21T14:57:49 *** Itkovian has quit IRC (Quit: Itkovian) 2010-10-21T14:58:14 For instance, if every planet is > 2 away, keep going until a planet changes hands? 2010-10-21T14:58:22 Then consider that a node. 2010-10-21T14:58:51 Kinda like forced moves in chess. 2010-10-21T14:58:59 hadn't got that far into it 2010-10-21T14:59:27 I was thinking about doing that but I think it might skip some important decisions (like when you build up enough ships to take a neutral) 2010-10-21T14:59:47 Okay, so let's think about it, on map 7. 2010-10-21T14:59:53 What's the best opening? 2010-10-21T15:00:37 Obviously you don't want to take the center +5 planets with minimal force, you will get sniped. 2010-10-21T15:01:04 my current subset/knapsack code is obviously broken now that I ran it on map 7 2010-10-21T15:01:06 so thanks for that 2010-10-21T15:01:10 So, throw everything to the cheap +2 right there, and react. 2010-10-21T15:01:41 But, if you throw everything to the closer +5, you will get it for sure, and have better position. 2010-10-21T15:01:48 is it the same map 7 as on the tcp server? 2010-10-21T15:01:54 I think so. 2010-10-21T15:02:02 this looks like a case of throwing all 100 to the 12 ship +5 neutral 2010-10-21T15:02:22 ok yeah I would throw all 100 at the 4 cost neutral and see what happens 2010-10-21T15:03:23 so, if you had minimax only concerned about the home planets, +5 center and +2 planets, what would it do? 2010-10-21T15:03:38 I think it must be a diferent one on tcp, can't see what you are talking about 2010-10-21T15:03:47 http://72.44.46.68/canvas?game_id=612185 2010-10-21T15:03:54 phreeza: it's the same 2010-10-21T15:04:13 you could try ignoring all other planets and see how far your minimax goes. 2010-10-21T15:04:19 I'm going to give it more options than that, I think 2010-10-21T15:04:34 clicked the wrong line X( 2010-10-21T15:05:34 I've already got a chess engine written in python. 2010-10-21T15:05:40 I should try an use it. 2010-10-21T15:06:03 haha 2010-10-21T15:06:15 *** boegel has joined #aichallenge 2010-10-21T15:06:23 I think the big difference is that moves happen at the same time 2010-10-21T15:06:40 also fleet ship counts complicate the decisions 2010-10-21T15:06:43 yeah that is a very big difference 2010-10-21T15:07:02 just for kicks I ran my last workable minimax bot on map 7 2010-10-21T15:07:16 what happened? 2010-10-21T15:07:26 it opens with a quickest return on ship investment move to the 12 and 4 ship neutrals 2010-10-21T15:07:35 kinda dumb 2010-10-21T15:07:51 mine would do nothing 2010-10-21T15:07:53 lol 2010-10-21T15:08:08 leaves the 12 snipable, and then went after the 12 closer to the enemy, haha 2010-10-21T15:09:11 how many ply? 2010-10-21T15:09:49 5 right now, I think 2010-10-21T15:10:00 I think it would take at least 20 to see the snipe. 2010-10-21T15:10:33 the way mine is set up it would take 1 ply to see the snipe 2010-10-21T15:10:34 wait, I took it down to 3 2010-10-21T15:10:49 because my generators look into the future 2010-10-21T15:11:00 Top 10 players: Hazard(4014), dmj111(3940), rsergio(3923), sin_sun(3879), Accoun(3835), savchenko(3830), cfaftw(3819), ofrias(3809), deepblue(3807), Rexxar(3807) 2010-10-21T15:11:41 mine is supposed to be seeing the state of the planet as many turns into the future as the map is wide 2010-10-21T15:11:56 so if the furthest planet is 10 turns, it's supposed to know what enemy ships are there 2010-10-21T15:12:00 but it's clearly screwing that up 2010-10-21T15:12:03 P 9.31956732508 21.8088737532 1 100 5 2010-10-21T15:12:03 P 14.2884238258 0.622568806433 2 100 5 2010-10-21T15:12:04 P 8.86481414847 9.73662367883 0 12 5 2010-10-21T15:12:04 P 14.7431770024 12.6948188808 0 12 5 2010-10-21T15:12:04 P 17.0287479269 6.65976901033 0 4 2 2010-10-21T15:12:04 P 6.57924322402 15.7716735493 0 4 2 2010-10-21T15:12:22 That's map 7 with the other planets removed. What kind of ply do you get with just that? 2010-10-21T15:12:43 *** Naktibalda has joined #aichallenge 2010-10-21T15:13:59 I think an good way to prune moves would be only consider neutrals with incoming fleets or "good" planets. 2010-10-21T15:14:23 When the enemy expands to a "bad" planet, you could add it's twin to the list. 2010-10-21T15:14:39 interesting idea 2010-10-21T15:14:45 It would be a good way to get more ply at the start, where you really need it. 2010-10-21T15:15:16 I'm expanding the plies and running it on that map, be a minute though 2010-10-21T15:15:50 So, for moving at the same time, I suppose you would pick the set of moves that raise your average expected num_ships the most? 2010-10-21T15:16:22 Kinda like a prisoners delima matrix, but hugh? 2010-10-21T15:16:23 os i'm a bit late, mine goes for the +5 in the middle with 12 2010-10-21T15:16:25 I actually build a table and play prisoner's dilemma 2010-10-21T15:16:38 and sends 45 consistently, even though its monte carly 2010-10-21T15:16:40 o 2010-10-21T15:16:53 McLeopold: yeah, exactly 2010-10-21T15:17:10 but the again, my bot sucks ;) 2010-10-21T15:17:26 it's better than mine is atm 2010-10-21T15:17:45 mine only runs 4 turns before it times out at 20 plies 2010-10-21T15:17:56 interestingly, it does the same dumb opener 2010-10-21T15:18:12 I guess it isn't using the extra depth correctly 2010-10-21T15:18:23 do you stop-I'm-out-of-time code in there? 2010-10-21T15:18:31 *do you have 2010-10-21T15:18:34 not in this version 2010-10-21T15:18:41 i let mine run for 10s for this 2010-10-21T15:18:53 I'll bump up the time limit and see if it recovers the start 2010-10-21T15:19:12 that was a pain to implement, it makes my minimax function look awful with all the state keeping 2010-10-21T15:19:46 As soon as I perfect RamBot, I will probably move on to minimax 2010-10-21T15:19:47 mine's very recursive, and each iteration takes nearly no time 2010-10-21T15:20:00 what's RamBot do? 2010-10-21T15:20:01 so I should be able to just check the time before each iteration and stop it 2010-10-21T15:20:21 ah, I expect my iterations to be kinda slow 2010-10-21T15:20:39 It's an attempt to put constant pressure on the enemy instead of waiting. 2010-10-21T15:20:51 So, there would be no giant fleet buildups. 2010-10-21T15:21:14 I figure I've got a better bot in a miser map. 2010-10-21T15:21:32 interesting 2010-10-21T15:21:37 Oh, it doing a bad start is me being dumb 2010-10-21T15:21:47 I never gave it the move to throw all ships at a single planet 2010-10-21T15:22:01 how about 99? 2010-10-21T15:22:18 it only has the option to take them with the minimum required right now 2010-10-21T15:22:25 oh 2010-10-21T15:22:54 so 0 or 13 to the center only? 2010-10-21T15:23:04 yup 2010-10-21T15:23:30 okay, that would also prune your branches down quite a lot. 2010-10-21T15:23:46 I'll add a move to the center real quick, be a few 2010-10-21T15:23:55 just to see if it's smart enough to spot the snipe 2010-10-21T15:23:59 the proper amount would be.... 2010-10-21T15:24:08 can it snipe? 2010-10-21T15:24:30 in theory. it hasn't run long enough to do it yet 2010-10-21T15:25:09 it's move list right now is a subset of planets, with ship count estimations when the fleets would arrive 2010-10-21T15:25:14 so when you say only as many as neccesary, that looks ahead to the future? 2010-10-21T15:25:19 *** perror has quit IRC (Quit: Bye all !) 2010-10-21T15:25:23 I have "look for the best time to take the planet" code in mine 2010-10-21T15:25:24 aha 2010-10-21T15:28:08 gotta go...looking forward to max ply you acheive JamesMG 2010-10-21T15:32:41 i need mercurial advice 2010-10-21T15:32:49 shoot 2010-10-21T15:33:19 what's the best way to commit a bug fix to my bot, then revert back to before the change to work on a new fix, then commit that 2010-10-21T15:33:33 two orthogonal bug fixes, basically 2010-10-21T15:33:49 that i want to test independently to make sure they are orthogonal :) 2010-10-21T15:34:14 easiest way might be to clone your directory 2010-10-21T15:34:19 but after hg commit 2010-10-21T15:34:39 hg update -r xxxx will take you back to the prior revision 2010-10-21T15:34:40 branch, or clone a new repo 2010-10-21T15:34:48 so... hg commit, hg clone, hg revert, hg commit? 2010-10-21T15:34:52 where (xxxx is the rev number) 2010-10-21T15:35:03 oh, update 2010-10-21T15:35:04 ok 2010-10-21T15:35:08 thats ound simple enough 2010-10-21T15:35:11 if you clone a new repo, you can work on them both at the same time, but you are taking 2x the space 2010-10-21T15:35:11 sounds* 2010-10-21T15:35:22 hg commit, hg update -r xxx, hg commit, hg merge 2010-10-21T15:35:42 *** perestrelka has quit IRC (Read error: Connection reset by peer) 2010-10-21T15:35:58 there's some way to do branches but I haven't tried it 2010-10-21T15:36:17 *** perestrelka has joined #aichallenge 2010-10-21T15:36:21 ok... so for branching i would make a branch for each fix, then merge them all back once i'm done 2010-10-21T15:36:24 commit, update, commit => this gets you two separate revisions with bugfixes, but only one is in the latest 2010-10-21T15:37:48 it will give you multiple heads, which you'll then merge together 2010-10-21T15:38:26 branching sounds like the correct way to do it... cloning sounds like the easy way, haha 2010-10-21T15:38:38 cloning is the easy way, yes :) 2010-10-21T15:38:52 i have not actually used branching in either git or hg 2010-10-21T15:38:57 *** Mekanik has quit IRC (Quit: WeeChat 0.3.3) 2010-10-21T15:39:33 i'll figure out branching... half the reason i'm using hg is to learn a new skill 2010-10-21T15:39:57 good idea :) 2010-10-21T15:41:03 thanks for the help 2010-10-21T15:41:16 the graph of my personal repo is basically o-o-o-o-o-o-o-o... 2010-10-21T15:41:31 lavalamp_, do you have a bot on TCP? 2010-10-21T15:42:27 cfaftw, I had the pwwax.x series of bots a week or so ago... right now I'm rewriting so no :( 2010-10-21T15:42:48 oh 2010-10-21T15:42:56 my best one got up to 150 Elo or so 2010-10-21T15:42:59 yeah... my bot is still purely heuristics 2010-10-21T15:43:13 started rewriting it a week and a half ago 2010-10-21T15:45:10 is the rewrite heuristics, too? 2010-10-21T15:45:15 *** aerique has joined #aichallenge 2010-10-21T15:45:20 yeah 2010-10-21T15:45:25 i'm going to milk this heuristic bot for all its worth, then in early november i might investigate some more "advanced" techniques 2010-10-21T15:47:15 It was getting hard for me to add stuff to my first bot (heuristics) without breaking stuff 2010-10-21T15:47:33 lavalamp_, that's why i decided to rewrite... it got too complicated 2010-10-21T15:47:38 *** Accoun has quit IRC () 2010-10-21T15:47:43 i took what i learned from my old bot and organized it better 2010-10-21T15:48:00 I tore all the nasty code out of my old bot yesterday also to start fresh 2010-10-21T15:48:11 only kept the framework I'd built 2010-10-21T15:48:19 Yeah, if my current minimax thing doesn't work out at least I will have a pretty decent framework 2010-10-21T15:48:40 i'm really curious about using a neural-net 2010-10-21T15:48:50 but i don't know anything about programming ANNs 2010-10-21T15:49:10 I have a fuzzy text matching neural net laying around from work... 2010-10-21T15:49:23 wonder if I could apply it here 2010-10-21T15:49:46 maybe you could use one to figure out what "good" planets are... I'm not sure how you would get full moves out of one 2010-10-21T15:50:04 yeah that's why I didn't start with it 2010-10-21T15:50:46 I'd bet you could use one to evaluate the value of a planet at various stages though 2010-10-21T15:51:15 do you guys know where to find examples of ANN code? meaning, somebody gives some code, explains what it does and how it learns, and how it can be applied? 2010-10-21T15:51:36 (no, i haven't thoroughly googled yet :) 2010-10-21T15:51:45 *** Palmik has joined #aichallenge 2010-10-21T15:52:22 *** Itkovian has joined #aichallenge 2010-10-21T15:52:53 not off the top of my head, no 2010-10-21T15:53:15 *** codefun has joined #aichallenge 2010-10-21T15:53:17 I have the sample one I used way back when, can send that off to you if you want 2010-10-21T15:53:28 unfortunately i didn't get a chance to take the AI class before graduating 2010-10-21T15:53:32 :[ 2010-10-21T15:53:39 You can issue more than one order per turn, if you have the issueorder() function used in a loop, right? 2010-10-21T15:53:46 neural networks? 2010-10-21T15:54:01 codefun: sure, even without a loop 2010-10-21T15:54:04 JamesMG, yeah that'd be great... my email is my nick @ gmail.com 2010-10-21T15:54:19 Thanks laval 2010-10-21T15:54:21 lavalmp* 2010-10-21T15:54:22 i got linked to a good example some time ago, right here in this channel, let me dig up the link 2010-10-21T15:54:44 sweet 2010-10-21T15:55:06 http://www.ai-junkie.com/ann/evolved/nnt1.html 2010-10-21T15:55:36 Zannick, thanks! i'll be giving this a read tonight after work 2010-10-21T15:55:41 sure 2010-10-21T15:55:55 AFK.... headache making me wanna kill myself. 2010-10-21T15:56:39 lol: http://www.ai-junkie.com/ann/evolved/nnt1_files/image001.gif 2010-10-21T15:56:50 cfaftw: sent, hope it's helpful 2010-10-21T15:57:08 yeah, someone's a little fail with html and images 2010-10-21T15:57:21 "before you start ... make sure you understand how to use genetic algorithms thoroughly ... You can find my short tutorial on genetic algorithms here." 2010-10-21T15:57:46 JamesMG, thanks, i'll check it out 2010-10-21T15:58:09 i wonder if his short tutorial will make me understand genetic algorithms thoroughly 2010-10-21T15:58:27 based on that image I have my doubts 2010-10-21T15:58:42 LOL 2010-10-21T15:58:44 it's not even a link 2010-10-21T15:58:45 it's not terrible 2010-10-21T15:59:06 i will still think of picking up russell and norvig 2010-10-21T15:59:12 *** Accoun has joined #aichallenge 2010-10-21T15:59:21 * Zannick gtg meeting 2010-10-21T15:59:24 c'ya 2010-10-21T15:59:54 *** rebelxt has joined #aichallenge 2010-10-21T16:01:57 *** torgeir has joined #aichallenge 2010-10-21T16:02:35 JamesMG, no email yet... how big is the file? 2010-10-21T16:03:49 580k or so 2010-10-21T16:03:57 does gmail let you receive .tar.gz? 2010-10-21T16:04:34 yeah 2010-10-21T16:05:03 well... not sure how it handles archives that contain executables 2010-10-21T16:05:12 feel like you would have gotten a return to sender or something 2010-10-21T16:05:22 or i'd get the message without the attachment 2010-10-21T16:05:46 *** boegel_ has joined #aichallenge 2010-10-21T16:06:24 *** codefun has quit IRC (Quit: Page closed) 2010-10-21T16:07:49 *** tittan has joined #aichallenge 2010-10-21T16:07:49 whoops, I sent it to "nick@gmail.com" instead your nick @... 2010-10-21T16:07:56 haha, paying attention 101 I guess. 2010-10-21T16:08:23 haha 2010-10-21T16:08:28 * janzert wonders how many emails a day nick@gmail.com gets because of that 2010-10-21T16:08:29 lol 2010-10-21T16:08:39 * lavalamp_ does too 2010-10-21T16:08:53 resent, maybe to the right place this time 2010-10-21T16:09:06 if i type lavalamp, does it alert you? 2010-10-21T16:09:14 or do i need the _ 2010-10-21T16:09:15 apparantly not 2010-10-21T16:09:18 lame 2010-10-21T16:09:22 ...yeah 2010-10-21T16:09:34 JamesMG, got it, thanks 2010-10-21T16:09:47 I've just downloaded the starting pack and tried to run it, but both players time out. Somebody got any idea why? 2010-10-21T16:10:03 the bots aren't running, try to run one standalone 2010-10-21T16:10:06 They took too long? :) 2010-10-21T16:10:09 and type "go" if it goes to a console 2010-10-21T16:10:14 see if it hits you back 2010-10-21T16:11:01 Top 10 players: Hazard(3956), dmj111(3950), sin_sun(3905), rsergio(3899), cfaftw(3875), Accoun(3858), savchenko(3857), deepblue(3851), Rexxar(3816), smatcher(3783) 2010-10-21T16:11:11 How do I run standalone? 2010-10-21T16:11:31 java -jar examplebots/RageBot.jar 2010-10-21T16:11:48 Thank you. 2010-10-21T16:11:55 example_bots, sorry 2010-10-21T16:13:12 *** boegel_ has quit IRC (Quit: Leaving) 2010-10-21T16:13:20 I'm having the same problem as tittan. After running RageBot, what should it do? 2010-10-21T16:13:43 If you type "go", does it respond with another "go"? 2010-10-21T16:13:52 No, silence 2010-10-21T16:14:40 *** Accoun has quit IRC (Ping timeout: 276 seconds) 2010-10-21T16:14:52 whoa, do the example bots not reply without data? interesting 2010-10-21T16:15:02 you press enter? (sorry if that's a dumb question... just gotta be sure) 2010-10-21T16:15:26 oh wait 2010-10-21T16:15:39 if the bot gets no map data on turn one, what would it do if you just tell it "go" 2010-10-21T16:15:51 or on any turn, really 2010-10-21T16:16:39 haha... i wonder what my bot would do with no data... 2010-10-21T16:16:42 mine responds with "go" 2010-10-21T16:16:51 RageBot doesn't seem to 2010-10-21T16:18:35 *** phreeza has quit IRC (Quit: Computer has gone to sleep) 2010-10-21T16:18:48 *** rebelxt has left #aichallenge 2010-10-21T16:19:11 None of the example bots responds. 2010-10-21T16:22:26 Hah, lavalamp_: I fixed my scoring method and my minimax bot throws all 100 to the 12 +5 2010-10-21T16:22:37 nice! 2010-10-21T16:22:37 and then times out... baby steps. 2010-10-21T16:22:41 lol 2010-10-21T16:23:14 also nick@gmail.com isn't registered 2010-10-21T16:23:27 really?? amazing 2010-10-21T16:23:32 I'm shocked 2010-10-21T16:23:39 wow, I get every nick just assumes it's taken 2010-10-21T16:25:53 that's at 5 plies by the way 2010-10-21T16:26:11 but each one estimates until there are no fleets in the air 2010-10-21T16:26:50 *** tittan has quit IRC (Quit: Page closed) 2010-10-21T16:26:52 You should fix the timeouts and set it loose on the tcp server... 2010-10-21T16:26:53 *** boegel has quit IRC (Quit: *poof!*) 2010-10-21T16:27:04 it'll get destroyed for sure 2010-10-21T16:27:13 haha oh ok 2010-10-21T16:27:22 it's options are too limited 2010-10-21T16:27:22 I'm still curious 2010-10-21T16:27:55 i would love to play poker ai challenge 2010-10-21T16:28:30 *** Spike_007 has joined #aichallenge 2010-10-21T16:28:55 too simple 2010-10-21T16:29:09 oh, really? 2010-10-21T16:29:15 calculate chance of win, and if >50%, bet 2010-10-21T16:29:21 lol :D 2010-10-21T16:29:26 hahahaha 2010-10-21T16:29:28 have you ever played poker? 2010-10-21T16:29:46 It would be best if it played multiple rounds against the same bots so it learned their strategies. 2010-10-21T16:29:46 Yeah, but all bots have perfect poker-face builtin 2010-10-21T16:30:03 javagamer: yeah, that would need somekind own database for storing information 2010-10-21T16:30:27 so it's all about pokerface? 2010-10-21T16:30:38 hmm... maybe i should register nick@gmail.com even though my name is not nick 2010-10-21T16:30:44 lol 2010-10-21T16:31:10 it's certainly easier to convey vocally than my current name (cfaftw) 2010-10-21T16:31:33 that's why I use the super simple one I e-mailed you with 2010-10-21T16:31:53 oh... usernames must be at least 6 chars long 2010-10-21T16:31:59 that's why i couldn't get brett@gmail.com 2010-10-21T16:32:00 ah, that explains that 2010-10-21T16:32:01 Just about to say that 2010-10-21T16:32:16 lol 2010-10-21T16:32:23 I was checking if a@gmail.com was registered 2010-10-21T16:32:39 why do account management systems always impose some lower limit on name length? 2010-10-21T16:33:03 beats me 2010-10-21T16:34:05 also... their 6 character minimum does not count periods 2010-10-21T16:35:39 yeah, IIRC you can add periods into any existing username and it will still work (e.g. janzert@gmail.com will work as j.an.zert@gmail.com too) 2010-10-21T16:37:34 *** Palmik has quit IRC (Remote host closed the connection) 2010-10-21T16:38:32 oh 2010-10-21T16:42:49 *** aerique has quit IRC (Quit: ...) 2010-10-21T16:46:15 *** choas has joined #aichallenge 2010-10-21T16:47:38 *** Mathnerd314 has quit IRC (Ping timeout: 245 seconds) 2010-10-21T16:48:34 *** iFire has quit IRC (Remote host closed the connection) 2010-10-21T16:52:13 *** bhasker has quit IRC (Ping timeout: 245 seconds) 2010-10-21T16:53:24 *** cfaftw has quit IRC (Quit: Going home...) 2010-10-21T16:54:37 *** Mathnerd314 has joined #aichallenge 2010-10-21T16:56:54 *** iFire has joined #aichallenge 2010-10-21T17:04:30 *** torgeir has quit IRC (Ping timeout: 265 seconds) 2010-10-21T17:04:43 &quit 2010-10-21T17:04:47 *** nrub_ has quit IRC (Quit: leaving) 2010-10-21T17:10:19 *** delt0r has joined #aichallenge 2010-10-21T17:10:59 *** delt0r___ has quit IRC (Ping timeout: 245 seconds) 2010-10-21T17:11:01 Top 10 players: rsergio(3978), dmj111(3978), cfaftw(3969), savchenko(3951), Hazard(3944), deepblue(3916), sin_sun(3851), Accoun(3845), felixcoto(3829), ofrias(3810) 2010-10-21T17:19:33 *** hvnsr has joined #aichallenge 2010-10-21T17:22:47 *** bhasker has joined #aichallenge 2010-10-21T17:23:52 *** hornairs has quit IRC (Quit: hornairs) 2010-10-21T17:26:44 *** hornairs has joined #aichallenge 2010-10-21T17:30:56 *** bhasker_ has joined #aichallenge 2010-10-21T17:32:06 *** bhasker__ has joined #aichallenge 2010-10-21T17:32:14 *** hornairs has quit IRC (Quit: hornairs) 2010-10-21T17:33:02 *** bhasker has quit IRC (Ping timeout: 252 seconds) 2010-10-21T17:35:34 *** bhasker_ has quit IRC (Ping timeout: 265 seconds) 2010-10-21T17:39:28 *** Nargil has joined #aichallenge 2010-10-21T17:39:30 *** Bobng has quit IRC (Quit: Leaving) 2010-10-21T17:39:41 *** jmpespxoreax has joined #aichallenge 2010-10-21T17:39:53 Buah! So many many games the servers play ! yuppi 2010-10-21T17:41:31 will it be permanent ? 2010-10-21T17:41:39 or are they expected to fail again ? 2010-10-21T17:42:58 they seem to fail in a couple hours at most, but I'm trying to track down why 2010-10-21T17:43:08 in the meantime I'm keeping up what I can 2010-10-21T17:43:29 Btw. http://ai-contest.com/organization_profile.php?org_id=195 <- Gordon Freeman @ CERN ! Nice humour. Wonder if it's him: http://api.ning.com/files/8fil5gmLgTbnvHiB9qe7ZlP*PesdnleY89Wa1XSwTN2VJOF-MRR*27IOl5yulIJ9a4-EFM6xPdeCTe4IvZZRdFoRF1G9upzR/GordonFreemanSpottedAtCERN.jpg 2010-10-21T17:43:40 janzert: thanks for the effort 2010-10-21T17:44:16 to be clear, there's no restrictions on organization on signup 2010-10-21T17:44:38 So the CERN part might be fake ? 2010-10-21T17:45:43 *** Azrathud has joined #aichallenge 2010-10-21T17:47:09 *** hornairs has joined #aichallenge 2010-10-21T17:48:36 *** hornairs has joined #aichallenge 2010-10-21T17:49:29 hey all 2010-10-21T17:50:05 would it be against the rules, if I would add 1 simple method into MyBot.java class ? 2010-10-21T17:50:35 rewrite it from scratch if you want 2010-10-21T17:50:40 thx :) 2010-10-21T17:50:48 my bot is crashing on the main site again i think :( 2010-10-21T17:51:03 *** choas has quit IRC (Quit: leaving) 2010-10-21T17:51:23 *** mega1 has quit IRC (Ping timeout: 245 seconds) 2010-10-21T17:53:16 My bot has never compiled on the main site, but I've yet to have something good enough that it matters. 2010-10-21T17:53:36 *** BtbN is now known as BtbN_ 2010-10-21T17:54:13 *** Bobng has joined #aichallenge 2010-10-21T18:00:02 *** JensTi has quit IRC (Quit: Leaving.) 2010-10-21T18:01:01 *** Appleman1234 has joined #aichallenge 2010-10-21T18:04:14 *** zerd has quit IRC (Ping timeout: 264 seconds) 2010-10-21T18:05:19 *** Sarin has quit IRC (Remote host closed the connection) 2010-10-21T18:05:29 *** gerald512kk has joined #aichallenge 2010-10-21T18:05:35 hello everybody :) 2010-10-21T18:06:50 *** zerd has joined #aichallenge 2010-10-21T18:09:11 *** Bobng has quit IRC (Ping timeout: 265 seconds) 2010-10-21T18:09:55 can i create in java an arraylist of arraylist ? 2010-10-21T18:10:28 yep 2010-10-21T18:10:47 thanks you very much :) 2010-10-21T18:11:03 Top 10 players: dmj111(4004), rsergio(4002), cfaftw(3994), savchenko(3977), Hazard(3966), deepblue(3944), Accoun(3872), sin_sun(3870), felixcoto(3862), hutabhu(3845) 2010-10-21T18:11:28 hm, ELOs are rising again 2010-10-21T18:12:11 *** JensTi has joined #aichallenge 2010-10-21T18:15:28 *** JamesMG has quit IRC (Quit: Leaving.) 2010-10-21T18:16:50 *** Spike_007 has quit IRC () 2010-10-21T18:18:52 *** lavalamp_ has quit IRC (Quit: ttfn) 2010-10-21T18:19:44 *** RainCT has quit IRC (Remote host closed the connection) 2010-10-21T18:19:52 *** Itkovian has quit IRC (Quit: Itkovian) 2010-10-21T18:29:08 *** Itkovian has joined #aichallenge 2010-10-21T18:29:58 *** narnach has quit IRC (Quit: Leaving.) 2010-10-21T18:30:16 *** narnach has joined #aichallenge 2010-10-21T18:30:42 *** Mathnerd314 has quit IRC (Ping timeout: 240 seconds) 2010-10-21T18:36:41 *** gerald512kk has quit IRC () 2010-10-21T18:36:44 *** gerald512kk has joined #aichallenge 2010-10-21T18:37:07 *** davidjliu has joined #aichallenge 2010-10-21T18:39:35 wow my first draw :) map where 1st planet are very very close are good for debuging, but bad for a good match :( 2010-10-21T18:44:49 *** Mathnerd314 has joined #aichallenge 2010-10-21T18:45:23 try map 30 2010-10-21T18:51:36 *** hornairs has quit IRC (Quit: hornairs) 2010-10-21T18:59:34 *** Itkovian has quit IRC (Remote host closed the connection) 2010-10-21T19:02:14 *** Itkovian has joined #aichallenge 2010-10-21T19:11:02 Top 10 players: dmj111(3988), rsergio(3967), savchenko(3944), sin_sun(3916), deepblue(3913), Accoun(3887), cfaftw(3883), Hazard(3856), chotting(3840), davidjliu(3831) 2010-10-21T19:12:59 *** Appleman1234 has quit IRC (Ping timeout: 265 seconds) 2010-10-21T19:13:47 *** BtbN_ has quit IRC (Quit: Verlassend) 2010-10-21T19:14:05 *** Itkovian_ has joined #aichallenge 2010-10-21T19:14:59 *** Appleman1234 has joined #aichallenge 2010-10-21T19:16:17 *** BtbN has joined #aichallenge 2010-10-21T19:16:28 *** Itkovian has quit IRC (Ping timeout: 240 seconds) 2010-10-21T19:16:28 *** Itkovian_ is now known as Itkovian 2010-10-21T19:19:53 *** narnach has quit IRC (Quit: Leaving.) 2010-10-21T19:20:17 *** Appleman1234 has quit IRC (Excess Flood) 2010-10-21T19:20:55 *** Appleman1234 has joined #aichallenge 2010-10-21T19:27:30 *** BtbN has quit IRC (Quit: Verlassend) 2010-10-21T19:27:32 *** BtbN_ has joined #aichallenge 2010-10-21T19:29:09 *** hvnsr has quit IRC (Quit: Page closed) 2010-10-21T19:29:36 *** BtbN_ has quit IRC (Client Quit) 2010-10-21T19:29:42 *** BtbN has joined #aichallenge 2010-10-21T19:46:25 *** mceier has quit IRC (Quit: leaving) 2010-10-21T19:48:17 *** reckoner has joined #aichallenge 2010-10-21T20:01:40 *** McLeopold has quit IRC (Quit: Leaving) 2010-10-21T20:02:38 *** Appleman1234 has quit IRC (Remote host closed the connection) 2010-10-21T20:05:00 *** reckoner has quit IRC (Remote host closed the connection) 2010-10-21T20:05:18 is the owner of Lbot here? 2010-10-21T20:05:25 *** gerald512kk has quit IRC () 2010-10-21T20:05:29 *** gerald512kk has joined #aichallenge 2010-10-21T20:05:39 *** Itkovian has quit IRC (Quit: Itkovian) 2010-10-21T20:05:44 is ower of lbot here? 2010-10-21T20:06:09 gerald512kk: no idea 2010-10-21T20:06:15 gerald512kk: why? 2010-10-21T20:06:34 coz his bot is chain killing mine so i wanted to ask for some peace 2010-10-21T20:07:04 so fun to see our very different strats, and his strat his the main weakness of mine so i have no chance to win 2010-10-21T20:08:54 gerald512kk: kill your bot for 20 seconds :) 2010-10-21T20:09:08 let Lbot to find another opponent 2010-10-21T20:09:30 yes it is wat i did when i realized i lost 12 in a row :) 2010-10-21T20:10:01 add a sleep between games 2010-10-21T20:11:03 Top 10 players: dmj111(3981), rsergio(3981), savchenko(3953), deepblue(3917), sin_sun(3911), Accoun(3894), chotting(3890), cfaftw(3882), Hazard(3865), davidjliu(3839) 2010-10-21T20:15:01 *** Frontier has joined #aichallenge 2010-10-21T20:28:50 *** McLeopold has joined #aichallenge 2010-10-21T20:29:11 This beats my previous record: http://72.44.46.68/canvas?game_id=617702 2010-10-21T20:30:34 yep 2010-10-21T20:30:42 he shouldnt take the 0 middle planet 2010-10-21T20:31:58 *** Mathnerd314 has quit IRC (Ping timeout: 265 seconds) 2010-10-21T20:33:46 *** Olathe has joined #aichallenge 2010-10-21T20:43:13 *** Mathnerd314 has joined #aichallenge 2010-10-21T20:44:38 *** Itkovian has joined #aichallenge 2010-10-21T20:48:44 hooray, cloud servers up for pretty much an entire day! 2010-10-21T20:48:58 *** hornairs has joined #aichallenge 2010-10-21T20:51:12 I see only 1 cloud server 2010-10-21T20:53:10 er... better than zero, I guess 2010-10-21T20:53:31 still, like a 400% increase in games 2010-10-21T20:55:31 *** hornairs has quit IRC (Quit: hornairs) 2010-10-21T20:56:48 they should figure out some way to get Google to host it ;-) 2010-10-21T20:57:17 hmm, maybe Google App Engine 2010-10-21T20:58:14 and the primary language is python, as the contest people seem to prefer... 2010-10-21T20:59:18 oldman3 is good but terrible on close maps 2010-10-21T20:59:19 hey, nobody prefers python! 2010-10-21T20:59:20 ;D 2010-10-21T20:59:24 c++ ftw 2010-10-21T21:03:05 *** gerald512kk has quit IRC () 2010-10-21T21:03:57 wait, how could I forget? nobody except a crazy person (particularly, not Google) would let you run arbitrary executables 2010-10-21T21:10:22 Mathnerd314: this contest isn't related to google in any other way than title 2010-10-21T21:11:04 Top 10 players: rsergio(3961), Accoun(3939), savchenko(3930), deepblue(3914), sin_sun(3903), dmj111(3901), cfaftw(3869), chotting(3866), felixcoto(3815), Hazard(3814) 2010-10-21T21:11:10 no, they did something with the name 2010-10-21T21:17:29 *** amstan sets mode: -q *!*@213.88.2.78 2010-10-21T21:17:41 *** Mathnerd314 has quit IRC (Ping timeout: 265 seconds) 2010-10-21T21:17:58 dudes.. 2010-10-21T21:18:00 accoun is winning 2010-10-21T21:18:24 *** hornairs has joined #aichallenge 2010-10-21T21:20:23 amstan: is that significant? 2010-10-21T21:20:33 McLeopold: it's the guy who i just unbanned 2010-10-21T21:20:50 McLeopold: you know.. the russian guy, that wants to blow everyone up, with nuclear bombs? 2010-10-21T21:21:02 Well, he's got a good bot... 2010-10-21T21:21:23 yeah, but.. he's annoying 2010-10-21T21:21:24 lol 2010-10-21T21:21:31 lol 2010-10-21T21:21:37 I'm reviewing my losses to him right now... 2010-10-21T21:21:41 it would be interesting if he wins.. 2010-10-21T21:21:56 pgpaskar_: ^^ 2010-10-21T21:22:01 @rankings 2010-10-21T21:22:01 Hi 2010-10-21T21:22:02 amstan: Top 10 players: rsergio(3957), Accoun(3935), savchenko(3927), deepblue(3911), sin_sun(3899), dmj111(3898), cfaftw(3865), chotting(3864), Hazard(3811), felixcoto(3808) 2010-10-21T21:22:18 oh thats unfortunate 2010-10-21T21:22:20 pgpaskar_: oh man.. what will we do 2010-10-21T21:22:21 lol 2010-10-21T21:22:26 @rainbow WHAT, WHAT, WHAT TIME IS IT, WHAT, WHAT, WHAT TIME IS IT 2010-10-21T21:22:26 pgpaskar_: WHAT, WHAT, WHAT TIME IS IT, WHAT, WHAT, WHAT TIME IS IT 2010-10-21T21:22:51 i don't know lets hope that rsergio can pull us through! 2010-10-21T21:22:52 lol 2010-10-21T21:23:18 Hey! What about me? 2010-10-21T21:23:43 McLeopold: you haven't been in top 10 in a while, sssh and code 2010-10-21T21:23:44 lol 2010-10-21T21:23:59 *** McLeopold has quit IRC (Quit: Leaving) 2010-10-21T21:24:11 i hope he realized i was kidding 2010-10-21T21:24:21 *** bhasker__ has quit IRC (Quit: bhasker__) 2010-10-21T21:24:28 pgpaskar_: http://ai-contest.com/forum/viewtopic.php?f=18&t=941, is rsergio that guy? 2010-10-21T21:24:51 I don't know I haven't been following but possibly 2010-10-21T21:25:00 we might want to get him that 2010-10-21T21:25:16 I've been kind f sidetracked with the whole living with people experience 2010-10-21T21:25:16 lol 2010-10-21T21:25:49 he doesn't use the same email address 2010-10-21T21:26:12 yeah, not the same language either 2010-10-21T21:28:18 *** Mathnerd314 has joined #aichallenge 2010-10-21T21:31:05 *** Accoun has joined #aichallenge 2010-10-21T21:31:39 1 2010-10-21T21:32:23 @rankings 2010-10-21T21:32:25 amstan: Top 10 players: rsergio(3939), Accoun(3920), savchenko(3909), deepblue(3895), sin_sun(3891), dmj111(3879), cfaftw(3843), chotting(3818), felixcoto(3802), Hazard(3795) 2010-10-21T21:32:27 *** yasith has joined #aichallenge 2010-10-21T21:32:32 Accoun: not yet 2010-10-21T21:32:36 Accoun: though, congrats 2010-10-21T21:32:47 Accoun: we were just talking about you 2010-10-21T21:34:58 *** Azrathud has quit IRC (Read error: Connection reset by peer) 2010-10-21T21:38:33 wait... accoun, the guy who begged people to recompile playgame.jar for him is on place 2? 2010-10-21T21:38:53 Frontier: yes.. 2010-10-21T21:39:07 *** phreeza has joined #aichallenge 2010-10-21T21:39:18 either he's hacking.. or he's really lazy on compiling java 2010-10-21T21:40:16 Lol, that was exactly what I thought. Either way he is a far better coder than I expected. 2010-10-21T21:43:28 *** rebelxt_ has joined #aichallenge 2010-10-21T21:44:33 *** rommel has joined #aichallenge 2010-10-21T21:45:12 g''day all :) 2010-10-21T21:49:44 *** rebelxt_ has quit IRC (Quit: Page closed) 2010-10-21T21:52:04 *** cYmen has quit IRC (Ping timeout: 252 seconds) 2010-10-21T21:52:15 *** Naktibalda has quit IRC (Remote host closed the connection) 2010-10-21T21:56:23 *** krokkrok has joined #aichallenge 2010-10-21T21:57:07 did the main site go down? 2010-10-21T21:59:58 has the apocalypse begun! <<.<< >>.>> 2010-10-21T22:00:14 *** _W_ <_W_!~w@irc.w-wins.com> has quit IRC (Excess Flood) 2010-10-21T22:00:52 *** cYmen has joined #aichallenge 2010-10-21T22:01:37 *** _W_ <_W_!~w@irc.w-wins.com> has joined #aichallenge 2010-10-21T22:01:48 ah its back up now yay 2010-10-21T22:02:47 *** Frontier has quit IRC (Ping timeout: 272 seconds) 2010-10-21T22:04:29 *** amstan has quit IRC (Remote host closed the connection) 2010-10-21T22:05:31 1 2010-10-21T22:05:54 ? 2010-10-21T22:11:05 Top 10 players: rsergio(3953), Accoun(3935), savchenko(3923), deepblue(3906), sin_sun(3903), dmj111(3892), cfaftw(3859), chotting(3830), felixcoto(3811), Hazard(3808) 2010-10-21T22:12:18 :P 2010-10-21T22:18:57 *** Azrathud has joined #aichallenge 2010-10-21T22:20:42 *** JensTi has quit IRC (Quit: Leaving.) 2010-10-21T22:20:43 *** perestrelka has quit IRC (Read error: Connection reset by peer) 2010-10-21T22:21:18 *** perestrelka has joined #aichallenge 2010-10-21T22:30:45 !top 2010-10-21T22:30:46 Accoun: I do not know about 'top', but I do know about these similar topics: 'bottopic' 2010-10-21T22:32:08 *** seisatsu has joined #aichallenge 2010-10-21T22:34:16 *** krokkrok has quit IRC (Quit: Page closed) 2010-10-21T22:34:32 *** krokkrok has joined #aichallenge 2010-10-21T22:44:20 *** krokkrok has quit IRC (Quit: Page closed) 2010-10-21T22:47:30 congrats accoun 2010-10-21T22:47:41 *** airhorns has joined #aichallenge 2010-10-21T22:50:26 *** hornairs has quit IRC (Ping timeout: 264 seconds) 2010-10-21T22:55:33 its version form 13 oct 2010-10-21T22:55:59 not all is posted last versions 2010-10-21T23:00:14 *** Nargil has quit IRC (Ping timeout: 265 seconds) 2010-10-21T23:02:19 Accoun: do you think your latest bot would win the games it lost now? (http://ai-contest.com/visualizer.php?game_id=5990073 for example) 2010-10-21T23:03:00 Mathnerd314> i don know 2010-10-21T23:03:26 is your official one still on TCP? 2010-10-21T23:04:19 Mathnerd314> from TCP Accoun203_4 is posted 2010-10-21T23:05:21 ah, okay 2010-10-21T23:05:39 g++ need any envelopment? 2010-10-21T23:06:58 *** hellman has joined #aichallenge 2010-10-21T23:07:35 form bash worck well 2010-10-21T23:07:52 from cmd.exe print nothing 2010-10-21T23:08:06 styped cygwin 2010-10-21T23:11:06 Top 10 players: Accoun(3946), rsergio(3934), savchenko(3928), sin_sun(3926), deepblue(3912), cfaftw(3903), dmj111(3885), Hazard(3843), chotting(3839), felixcoto(3822) 2010-10-21T23:20:58 *** bhasker has joined #aichallenge 2010-10-21T23:24:05 *** Cyndre has joined #aichallenge 2010-10-21T23:26:52 *** amstan has joined #aichallenge 2010-10-21T23:26:52 *** ChanServ sets mode: +o amstan 2010-10-21T23:33:13 *** rebelxt_ has joined #aichallenge 2010-10-21T23:33:33 *** amstan has quit IRC (Remote host closed the connection) 2010-10-21T23:34:22 *** amstan has joined #aichallenge 2010-10-21T23:34:22 *** ChanServ sets mode: +o amstan 2010-10-21T23:34:47 evening 2010-10-21T23:35:00 hey 2010-10-21T23:36:36 *** amstan has quit IRC (Remote host closed the connection) 2010-10-21T23:36:37 *** amstan_ has joined #aichallenge 2010-10-21T23:36:37 *** ChanServ sets mode: +o amstan_ 2010-10-21T23:45:07 it's very hard to get a tree bot right for this 2010-10-21T23:51:56 heh 2010-10-21T23:52:04 tree bot? 2010-10-21T23:52:04 i am still struggling with getting my staging code right 2010-10-21T23:52:13 accoun your bot is annoying 2010-10-21T23:52:21 all your 2xx versions 2010-10-21T23:52:31 who write CCCF-ex12 ? 2010-10-21T23:53:00 no idea 2010-10-21T23:53:14 I've heard people say it might be felixcoto 2010-10-21T23:53:19 not actualyl sure though 2010-10-21T23:53:31 looks like my staging code might be working better finally 2010-10-21T23:53:32 http://72.44.46.68/canvas?game_id=620523 2010-10-21T23:54:16 *** Azrathud has quit IRC (Ping timeout: 265 seconds) 2010-10-21T23:56:20 Oct 17th 18:59:30 sucessfully entered into contest Java <=felixcoto last Submission 2010-10-21T23:57:13 wait, where does it say CCCF isn't java? 2010-10-21T23:58:48 *** foucist has quit IRC (Read error: Operation timed out) 2010-10-21T23:59:16 *** Olathe has quit IRC (Ping timeout: 276 seconds)