2011-07-18T00:00:06 gives this output 'status eliminated survived' 2011-07-18T00:00:44 Qua: "18 2011-07-18 07:42:07 Test Error: compiled, but failed test cases - C# turn 0 bot 0 timed out ", the part where it says "failed test cases" executes "./test_bot.sh Yourbot.exe" 2011-07-18T00:01:00 you can reproduce that exact test buy running it locally 2011-07-18T00:01:02 *** confab has quit IRC (Ping timeout: 252 seconds) 2011-07-18T00:01:09 if it fails locally it will probably fail when you upload it 2011-07-18T00:02:04 that exact command gives: status eliminated survived 2011-07-18T00:02:31 should be status survived survived 2011-07-18T00:02:37 i think 2011-07-18T00:02:41 let me see with a c++ bot 2011-07-18T00:04:29 wait.. no 2011-07-18T00:04:31 wtf? 2011-07-18T00:04:32 same output 2011-07-18T00:06:07 Yeah 2011-07-18T00:06:15 eliminated survived 2011-07-18T00:06:15 Qua: do you have a visualizer? 2011-07-18T00:06:19 Qua: http://aichallengebeta.hypertriangle.com/cpp-starter.replay 2011-07-18T00:06:19 watched the replay 2011-07-18T00:06:26 the test bot owns 2011-07-18T00:06:27 :D 2011-07-18T00:06:31 Qua: it does eh? 2011-07-18T00:06:51 Qua: so then it must mean that not only the c# package is getting owned 2011-07-18T00:06:53 but the others too 2011-07-18T00:07:25 It's kinda strange having to beat a test bot before your upload is validated 2011-07-18T00:07:33 Qua: it's not 2011-07-18T00:07:37 shouldn't it be (status != 'crashed') 2011-07-18T00:07:38 ? 2011-07-18T00:07:41 Qua: they could crash their first turn 2011-07-18T00:07:53 it's not cool to be let in the tournament like that 2011-07-18T00:07:57 they'll just crash every game 2011-07-18T00:08:00 true 2011-07-18T00:08:06 same with timeouts 2011-07-18T00:08:22 still, a bot that doesn't crash will still not validate 2011-07-18T00:08:31 because the test bot owns it 2011-07-18T00:08:37 Qua: a bot that can't own a simple test shouldn't really be let in 2011-07-18T00:08:46 but i think the problem now is that the test is too smart 2011-07-18T00:08:52 or something like that 2011-07-18T00:08:57 no, but the introduction section on your web site says to upload starter package 2011-07-18T00:08:58 freaking rankings not loading 2011-07-18T00:09:01 which you cant 2011-07-18T00:09:06 Qua: yes, and that's supposed to work 2011-07-18T00:09:10 idd 2011-07-18T00:09:21 Qua: can you find me links to other starter bots? 2011-07-18T00:09:26 the page really doesn't want to load for me 2011-07-18T00:09:44 http://aichallengebeta.hypertriangle.com/starter_packages/scala_starter_package.zip 2011-07-18T00:09:49 http://aichallengebeta.hypertriangle.com/starter_packages/ruby_starter_package.zip 2011-07-18T00:09:51 Qua: not the zips 2011-07-18T00:09:54 the bots in the game 2011-07-18T00:10:01 2 secfs 2011-07-18T00:10:28 but still 2011-07-18T00:10:39 are bots that are already uploaded still being tested ? 2011-07-18T00:10:57 yes 2011-07-18T00:10:59 it seemes strange that all c# bots timeouts at turn 0. However, this could be a totally different problem 2011-07-18T00:11:00 against new bots 2011-07-18T00:11:34 http://aichallengebeta.hypertriangle.com/profile.php?user=132 2011-07-18T00:11:40 try loading that page now 2011-07-18T00:12:01 loading...... 2011-07-18T00:12:07 cpp starter bot: http://aichallengebeta.hypertriangle.com/profile.php?user=132 2011-07-18T00:12:10 ohh lol 2011-07-18T00:12:21 what's the status of the latest few games? 2011-07-18T00:12:34 haven't loaded it yet 2011-07-18T00:12:36 ok 2011-07-18T00:13:10 iit plays 2011-07-18T00:13:10 the submission test hasn't been changed since the beginning of april 2011-07-18T00:13:11 and wins 2011-07-18T00:13:12 https://github.com/aichallenge/aichallenge/tree/epsilon/ants/submission_test 2011-07-18T00:13:55 it has won all matched played today 2011-07-18T00:14:02 janzert: http://aichallengebeta.hypertriangle.com/csharp-starter.replay 2011-07-18T00:14:08 http://aichallengebeta.hypertriangle.com/cpp-starter.replay 2011-07-18T00:14:48 janzert: why do both cpp and c# get eliminated? 2011-07-18T00:16:51 It would make sense if the testbot and MyBot changed positions 2011-07-18T00:17:01 *** jmreardon has quit IRC (Quit: jmreardon) 2011-07-18T00:17:09 http://ants.zeroviz.us/XV2NU/ 2011-07-18T00:17:16 looks to me like the players are wrong 2011-07-18T00:17:29 the testbot should be the second player there 2011-07-18T00:18:29 exactly 2011-07-18T00:19:09 was there an engine commit that is changing the order of players or something? 2011-07-18T00:21:12 test_bot.sh and test_bot.cmd both list the test_bot as the second player btw 2011-07-18T00:21:58 yup 2011-07-18T00:22:22 janzert: you sure it's not the test bot owning ppl? 2011-07-18T00:22:30 we did change the battle stuff recently 2011-07-18T00:22:48 or are those settings hardcoded in the test_bot? 2011-07-18T00:23:00 what settings? 2011-07-18T00:25:06 see notice on http://aichallengebeta.hypertriangle.com/ 2011-07-18T00:25:11 hmm, actually relooking at the test map it is the bot being tested that is put in the enclosed positions 2011-07-18T00:26:00 but what does that have to do with the testbot? 2011-07-18T00:26:23 idk, did the game settings change in the test game too? 2011-07-18T00:26:24 starter packages are being eliminated by the test bot, and thus fail the upload test 2011-07-18T00:26:26 or are they still the same? 2011-07-18T00:26:55 they would have changed 2011-07-18T00:27:35 janzert: yes, so that's maybe why they fail now 2011-07-18T00:27:41 test became too hard 2011-07-18T00:28:07 it doesn't have to win the test game, just not crash or timeout 2011-07-18T00:28:35 That makes sense. 2011-07-18T00:28:41 And they don't crash or timeout 2011-07-18T00:28:53 locally anyway, but they still fail upload test 2011-07-18T00:28:53 see around line 451 of worker.py 2011-07-18T00:29:13 that's the way I'm reading it at least 2011-07-18T00:30:05 the result for the bot being tested just can't be one of ('crashed', 'timeout', 'invalid') 2011-07-18T00:30:59 so, it works locally, and it works when amstan tests it on the server 2011-07-18T00:31:07 doesn't work when we upload it to the server though 2011-07-18T00:31:16 then the tournament manager/sandbox doesn't like it 2011-07-18T00:31:44 amstan: you're testing it on a worker? 2011-07-18T00:31:54 janzert: no, just main server 2011-07-18T00:31:59 no guarantee the main server has the same version of stuff at this point 2011-07-18T00:32:34 the main server install was a long time ago 2011-07-18T00:47:03 Any chance you could test it on one of the worker servers? 2011-07-18T00:47:21 *** antimatroid has quit IRC (Ping timeout: 260 seconds) 2011-07-18T00:47:53 Qua: yeah, but tomorrow 2011-07-18T00:48:28 alright, nothing I can do ay? 2011-07-18T00:50:03 *** ibdknox has quit IRC (Ping timeout: 276 seconds) 2011-07-18T00:50:34 *** ibdknox has joined #aichallenge 2011-07-18T01:04:31 *** Mathnerd314 has quit IRC (Read error: Connection reset by peer) 2011-07-18T01:15:01 *** HumWum has joined #aichallenge 2011-07-18T01:16:08 Is it illegal to move into a field containing a piece of food? :p 2011-07-18T01:23:59 ahh nevermind, it is. that is why starter packages get blocked/ignored moves 2011-07-18T01:44:07 *** antimatroid has joined #aichallenge 2011-07-18T02:00:55 *** fbparis has joined #aichallenge 2011-07-18T02:50:24 Qua: can i ask that how can you do that move? 2011-07-18T02:51:22 *** fbparis has left #aichallenge 2011-07-18T02:51:30 *** janzert1 has joined #aichallenge 2011-07-18T02:52:41 *** janzert has quit IRC (Ping timeout: 240 seconds) 2011-07-18T02:57:34 Yeah 2011-07-18T02:58:02 If you start next to a food, which you do in the test case being run when uploading your code, then moving onto it will cause you to do an invalid move 2011-07-18T03:00:20 ah, ok. very rare case tho. 2011-07-18T03:00:52 Indead, just annoyed me seeing that I made blocked/ignored moved 2011-07-18T03:00:53 moves 2011-07-18T03:06:12 *** amstan has quit IRC (Ping timeout: 255 seconds) 2011-07-18T03:14:24 *** ibdknox has quit IRC (Remote host closed the connection) 2011-07-18T03:20:50 *** foRei has quit IRC (Read error: Connection reset by peer) 2011-07-18T03:24:57 *** Palmik has joined #aichallenge 2011-07-18T03:25:00 *** janzert has joined #aichallenge 2011-07-18T03:27:03 *** fbparis has joined #aichallenge 2011-07-18T03:27:18 *** fbparis has left #aichallenge 2011-07-18T03:28:45 *** janzert1 has quit IRC (Ping timeout: 260 seconds) 2011-07-18T03:39:31 *** _will_sell_5_BTC <_will_sell_5_BTC!~1@static-213-88-39-97.chebnet.ru> has joined #aichallenge 2011-07-18T03:40:01 *** Accoun has joined #aichallenge 2011-07-18T03:40:07 *** ltriant has quit IRC (Quit: Computer has gone to sleep) 2011-07-18T03:56:57 *** bhasker has quit IRC (Ping timeout: 258 seconds) 2011-07-18T03:59:25 *** bhasker has joined #aichallenge 2011-07-18T04:16:19 *** amstan_ has joined #aichallenge 2011-07-18T04:16:19 *** ChanServ sets mode: +o amstan_ 2011-07-18T04:24:57 *** amstan_ has quit IRC (Ping timeout: 255 seconds) 2011-07-18T04:29:35 *** chl has quit IRC (Read error: Operation timed out) 2011-07-18T05:01:38 *** greghaynes has quit IRC (Quit: WeeChat 0.3.4) 2011-07-18T05:03:18 *** greghaynes has joined #aichallenge 2011-07-18T05:18:22 *** sigh has joined #aichallenge 2011-07-18T05:25:17 *** sigh_ has joined #aichallenge 2011-07-18T05:26:01 *** sigh__ has joined #aichallenge 2011-07-18T05:27:39 *** sigh___ has joined #aichallenge 2011-07-18T05:27:49 *** sigh has quit IRC (Ping timeout: 252 seconds) 2011-07-18T05:29:43 *** sigh_ has quit IRC (Ping timeout: 258 seconds) 2011-07-18T05:31:19 *** sigh__ has quit IRC (Ping timeout: 258 seconds) 2011-07-18T05:50:05 *** sigh___ has quit IRC (Remote host closed the connection) 2011-07-18T05:52:13 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has quit IRC (Quit: Lost terminal) 2011-07-18T05:58:40 *** stocha has joined #aichallenge 2011-07-18T05:58:50 hi there; what are the current plans for the contest ? 2011-07-18T06:02:37 is there any sponsor yet ? Any planned date for starting ? Any plan for testing ? 2011-07-18T06:02:49 Any plan ? 2011-07-18T06:03:56 is it officially dead ? 2011-07-18T06:10:38 *** stocha has quit IRC (Ping timeout: 252 seconds) 2011-07-18T06:12:48 *** sigh has joined #aichallenge 2011-07-18T06:19:35 *** FireFly has joined #aichallenge 2011-07-18T06:19:39 *** FireFly has joined #aichallenge 2011-07-18T06:20:32 *** Qua_ has joined #aichallenge 2011-07-18T06:20:40 *** mleise has joined #aichallenge 2011-07-18T06:45:50 *** berak has joined #aichallenge 2011-07-18T07:19:57 *** mcstar has joined #aichallenge 2011-07-18T07:52:19 *** Qua has quit IRC (Quit: Page closed) 2011-07-18T08:00:03 *** onensora has joined #aichallenge 2011-07-18T08:45:27 with my second game i already have to fight #1 and #3 .. :( 2011-07-18T08:45:43 with my first version it was with me third game ... 2011-07-18T08:46:03 *me=the 2011-07-18T09:04:04 *** confab_ has quit IRC (Ping timeout: 252 seconds) 2011-07-18T09:24:09 *** Qua has joined #aichallenge 2011-07-18T09:26:07 *** amstan_ has joined #aichallenge 2011-07-18T09:26:07 *** ChanServ sets mode: +o amstan_ 2011-07-18T09:28:28 *** TrumpeT has joined #aichallenge 2011-07-18T09:28:54 *** TrumpeT is now known as Guest78158 2011-07-18T09:29:06 hi 2011-07-18T09:29:31 i think there is some problem with C++ compilator 2011-07-18T09:30:11 *** kincody has joined #aichallenge 2011-07-18T09:30:27 *** sixthgear has quit IRC (Ping timeout: 258 seconds) 2011-07-18T09:30:33 ... 2011-07-18T09:31:43 ? 2011-07-18T09:32:13 there is always timeout on compilation 2011-07-18T09:32:16 g++ on server 2011-07-18T09:32:59 it's on known issues 2011-07-18T09:33:22 but two days ago it works fine 2011-07-18T09:33:27 what happened? 2011-07-18T09:33:36 i don't know 2011-07-18T09:35:34 *** Guest78158 has quit IRC (Quit: Page closed) 2011-07-18T09:35:59 the load is @3 atm 2011-07-18T09:37:48 *** amstan_ is now known as amstan 2011-07-18T09:38:42 *** confab has joined #aichallenge 2011-07-18T09:44:44 *** asdas has joined #aichallenge 2011-07-18T09:44:52 hi 2011-07-18T09:45:03 I can't find this "g++ timeout issue" on github. 2011-07-18T09:45:19 Can anyone help me? I'm not sure if is should open new issue 2011-07-18T09:46:05 asdas: yeah and open one for me, too for top and bash and every other command ;) 2011-07-18T09:46:33 lol ? 2011-07-18T09:46:54 Maybe only g++ 'talks' about its timeout while the other languages just compile for 3 hours until its done :p 2011-07-18T09:47:05 the server is really slow in general recently 2011-07-18T09:48:17 ok so maybe someone should turn off this g++ timeout. Becouse now nobody can compile their C++ bot. 2011-07-18T09:49:43 ok, that would be a good topic for an issue I think :) 2011-07-18T09:52:28 i mean... IF this can be disabled in gcc 2011-07-18T10:00:20 if not, there should be less game threads... (maybe one free processor core for compillers?) 2011-07-18T10:01:48 asdas: I guess the work on the SQL script for the leader board is ongoing. And after that the issue may go away. amstan, janzert? 2011-07-18T10:02:18 hmm 2011-07-18T10:02:22 how about i change the timeout? 2011-07-18T10:02:34 from 120 seconds 2011-07-18T10:02:37 to something like 10 minutes 2011-07-18T10:03:13 *** jmreardon has joined #aichallenge 2011-07-18T10:03:26 thats good idea for now 2011-07-18T10:03:28 amstan: that should do the trick :) 2011-07-18T10:12:02 What is the problem with leader board script? 2011-07-18T10:15:36 There is one trick, how to do ranking ultra fast. If you tell me what is the problem, I can give you the solution 2011-07-18T10:17:14 how is do you calculate the skill? 2011-07-18T10:32:29 @amstan: 10 minutes is not enough. Not all my bot files were compiled. Can you change it from 10 minutes to 1 hour ? 2011-07-18T10:32:30 asdas: No! 2011-07-18T10:32:41 lol ? 2011-07-18T10:32:56 asdas: stuff with @ in front are bot commands 2011-07-18T10:33:08 the 10 minutes is the time for the rankings to calculate 2011-07-18T10:33:11 ok ;) 2011-07-18T10:33:22 compiling is done on another host 2011-07-18T10:33:44 *** sigh has quit IRC (Remote host closed the connection) 2011-07-18T10:34:08 ok, but so what? 10 minutes is not enough - check logs. 2011-07-18T10:34:56 why do you think it's not long enough? 2011-07-18T10:35:19 becouse my bot is not compiled? "Compilation timed out with command g++ -O3 -funroll-loops -c -o ./State.o ./State.cc" 2011-07-18T10:35:36 again, that 10 minutes has nothing to do with compiling 2011-07-18T10:35:50 the 10 minutes is the interval at which the rankings page is regenerated 2011-07-18T10:35:53 ok, but there is some timeout, right? 2011-07-18T10:36:02 compiling timeout? 2011-07-18T10:36:03 for g++ ? 2011-07-18T10:36:18 i increased it so the rankings stay longer in the cache 2011-07-18T10:36:26 compiling timeout is on a different host 2011-07-18T10:36:39 and i'm not sure how to change that 2011-07-18T10:36:48 how long does it take for your thing to compile anyway? 2011-07-18T10:37:59 State.o was last file. It takes 10 minutes to timeout. So at least 15 minutes is necessary 2011-07-18T10:38:37 hm, maybe you link against openoffice? 2011-07-18T10:38:42 But my bot is not complicated. So another bot will be compiled in 1h maybe with this speed? 2011-07-18T10:39:05 asdas: do you have any huge loops with constant time? 2011-07-18T10:39:09 what? 2011-07-18T10:39:17 huge loops that repeat at a constant number 2011-07-18T10:39:27 something like for(i=0;i<100000;i++) 2011-07-18T10:39:51 no. I have big loops like for (int i=0; i nothing like you wrote 2011-07-18T10:40:11 yeah, but ants.size() is not constant, that's fine, it won't get unrolled 2011-07-18T10:40:23 not sure 2011-07-18T10:40:28 how long does it take to compile locally? 2011-07-18T10:40:40 ~20 seconds 2011-07-18T10:40:48 and the timeout is 10 minutes? 2011-07-18T10:40:59 faster 2011-07-18T10:41:03 10 seconds 2011-07-18T10:41:13 yes, timeout is 10 minuts now 2011-07-18T10:41:17 i think so 2011-07-18T10:41:27 and you compile it with the same options? 2011-07-18T10:41:38 i compile it on windows 2011-07-18T10:41:40 localy 2011-07-18T10:41:59 i mean... 2011-07-18T10:42:02 yes, but are you using gcc with the same options to compile? 2011-07-18T10:42:03 with Visual Studio 2011-07-18T10:42:04 not g++ 2011-07-18T10:42:14 ... so it might be the funrollloops 2011-07-18T10:42:48 but my last version were compiled in ~2 minutes on server 2011-07-18T10:42:58 i do not change almost nothing 2011-07-18T10:43:01 and now is timeout 2011-07-18T10:43:37 i do not add ANY loop 2011-07-18T10:43:41 since last version 2011-07-18T10:43:46 i guess bhasker has similar problems, and i might be wrong, but with less optimization, it compiled ok for him 2011-07-18T10:43:55 he should say something on this... 2011-07-18T10:44:29 *I did not add any loops 2011-07-18T10:44:33 sorry for my english 2011-07-18T10:45:11 def compile_anything(bot_dir, timelimit=600): 2011-07-18T10:45:12 yes.. 2011-07-18T10:45:14 10 minutes 2011-07-18T10:45:18 omg, the same answers at the same time here, and on #lisp: sorry for my english 2011-07-18T10:45:20 lol 2011-07-18T10:45:45 mcstar: how did he change the optimization limit? 2011-07-18T10:45:52 s/limit//g 2011-07-18T10:45:58 he used -O1 i think 2011-07-18T10:46:19 yes, but how? you cannot send gcc options when you submit, they're hardcoded in the workers 2011-07-18T10:46:21 but i dont want to mislead anybody 2011-07-18T10:46:42 i didnt know that 2011-07-18T10:46:50 then i dont know 2011-07-18T10:46:52 asdas: do you have a linux comp near you? 2011-07-18T10:47:37 near me... no. But i have one dedicated server ~100km from here ;) 2011-07-18T10:47:52 asdas: in ssh range i mean, :P 2011-07-18T10:47:57 yes ;) 2011-07-18T10:48:02 try compiling your thing there 2011-07-18T10:48:09 with the same options as us ideally 2011-07-18T10:48:16 they should be in the makefile 2011-07-18T10:49:03 ok 2011-07-18T10:49:13 yeah.. just type make 2011-07-18T10:49:33 I must add new files first 2011-07-18T10:49:39 yeah, probably 2011-07-18T10:49:46 there's a source list inside the makefile 2011-07-18T10:50:08 yes i know how it works ;) 2011-07-18T10:50:13 i'm not stupid 2011-07-18T10:50:28 only my english is not good, not my brain ;) 2011-07-18T10:51:29 i think ur eng is okay so stop blaming it all the time .. it doesnt help to solve the problem at all ^^ 2011-07-18T10:53:12 *** smiley1983 has joined #aichallenge 2011-07-18T10:54:24 i think makefiles arent exactly easy, there a couple of gotcha's 2011-07-18T10:54:40 you can be smart, and still have to look up the documentation 2011-07-18T10:55:32 ok give me 5 mintues 2011-07-18T10:55:37 *minutes 2011-07-18T10:55:57 *** amstan has quit IRC (Ping timeout: 264 seconds) 2011-07-18T10:58:52 *** CycloidMike has joined #aichallenge 2011-07-18T11:00:34 OK 2011-07-18T11:00:38 i have the result 2011-07-18T11:00:52 4.5 seconds 2011-07-18T11:01:11 on linux with g++ 2011-07-18T11:01:44 options "CC=g++ CFLAGS=-O3 -funroll-loops -c LDFLAGS=-O2 -lm" 2011-07-18T11:03:14 so the problem is your server. The easies think to do: change "def compile_anything(bot_dir, timelimit=600):" to "def compile_anything(bot_dir, timelimit=6000):" 2011-07-18T11:04:17 *easiest thing 2011-07-18T11:05:22 you should use cpu time, not wall time 2011-07-18T11:05:26 ah 2011-07-18T11:05:33 amstan isnt here 2011-07-18T11:20:25 *** CycloidMike has quit IRC (Quit: Page closed) 2011-07-18T11:28:44 *** jmreardon has quit IRC (Quit: jmreardon) 2011-07-18T11:30:51 *** amstan has joined #aichallenge 2011-07-18T11:30:51 *** ChanServ sets mode: +o amstan 2011-07-18T11:32:14 amstan: hello again 2011-07-18T11:32:47 did you read logs? 2011-07-18T11:35:34 *** amstan_ has joined #aichallenge 2011-07-18T11:35:34 *** ChanServ sets mode: +o amstan_ 2011-07-18T11:35:54 *** amstan has quit IRC (Ping timeout: 252 seconds) 2011-07-18T11:36:40 *** ibdknox has joined #aichallenge 2011-07-18T11:43:33 amstan? 2011-07-18T11:45:21 *** mleise has quit IRC (Ping timeout: 260 seconds) 2011-07-18T11:50:19 *** jmreardon has joined #aichallenge 2011-07-18T11:57:37 *** ibdknox has quit IRC (Remote host closed the connection) 2011-07-18T12:02:19 asdas: yes? 2011-07-18T12:02:36 i'm currently at school, so i can't help a lot 2011-07-18T12:02:49 did you read channel log? 2011-07-18T12:03:19 I had 4.5s compile time, on linux, g++ 2011-07-18T12:03:56 but on contest server, 10 minutes is not long enough ;/ 2011-07-18T12:07:37 *** foRei has joined #aichallenge 2011-07-18T12:12:20 *** kincody has left #aichallenge 2011-07-18T12:21:12 my theory is when there is just one slave 2011-07-18T12:21:15 timeouts happen a lot more 2011-07-18T12:21:27 my compiles work fine when there are more workers available 2011-07-18T12:53:09 *** amstan_ has quit IRC (Ping timeout: 250 seconds) 2011-07-18T13:20:26 *** rabidus has quit IRC (Ping timeout: 260 seconds) 2011-07-18T13:23:10 *** rabidus has joined #aichallenge 2011-07-18T13:37:41 *** rabidus has quit IRC (Ping timeout: 240 seconds) 2011-07-18T13:39:45 *** rabidus has joined #aichallenge 2011-07-18T13:54:42 *** ibdknox has joined #aichallenge 2011-07-18T13:55:42 *** mleise has joined #aichallenge 2011-07-18T14:18:47 bhasker: are you taking a vacation to do the ants challenge? 2011-07-18T14:18:47 *** bhasker has quit IRC (Quit: bhasker) 2011-07-18T14:19:03 wow, i guess he doesn't want to admit it 2011-07-18T14:19:25 you only gave him .76 seconds 2011-07-18T14:35:36 *** ArchMonkey has joined #aichallenge 2011-07-18T14:41:54 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has joined #aichallenge 2011-07-18T14:48:45 *** ibdknox has quit IRC (Remote host closed the connection) 2011-07-18T14:49:26 The compilation is done on a EC2 instance, right? 2011-07-18T15:00:56 *** ibdknox has joined #aichallenge 2011-07-18T15:06:36 *** Palmik has quit IRC (Remote host closed the connection) 2011-07-18T15:06:53 *** Palmik has joined #aichallenge 2011-07-18T15:29:36 *** mleise has quit IRC (Ping timeout: 255 seconds) 2011-07-18T15:34:20 *** ibdknox has quit IRC (Remote host closed the connection) 2011-07-18T15:36:13 *** amstan has joined #aichallenge 2011-07-18T15:36:13 *** ChanServ sets mode: +o amstan 2011-07-18T15:39:50 *** ibdknox has joined #aichallenge 2011-07-18T15:46:42 *** ibdknox has quit IRC (Remote host closed the connection) 2011-07-18T16:01:24 *** Palmik_ has joined #aichallenge 2011-07-18T16:01:27 *** Palmik has quit IRC (Ping timeout: 258 seconds) 2011-07-18T16:09:33 *** ebcube has joined #aichallenge 2011-07-18T16:17:26 *** Accoun has quit IRC () 2011-07-18T16:17:41 *** Reviou is now known as Amtal 2011-07-18T16:19:48 *** jmreardon has quit IRC (Quit: jmreardon) 2011-07-18T16:22:30 The current game rate is about 0.0 players per minute. 2011-07-18T16:22:38 ? 2011-07-18T16:23:55 looks like it.. 2011-07-18T16:24:06 asdas: yo 2011-07-18T16:24:20 actually.. 2011-07-18T16:24:22 nvm 2011-07-18T16:24:31 wait.. nvm nvm 2011-07-18T16:24:37 i can do this 2011-07-18T16:24:50 amstan: actually I'm just rebooting them 2011-07-18T16:24:55 janzert: oh, ok 2011-07-18T16:25:10 janzert: can i have an ip at one point to test the c# compilation? 2011-07-18T16:25:30 sure, if these come back up 2011-07-18T16:25:39 *** ebcube has quit IRC (Read error: Connection reset by peer) 2011-07-18T16:25:40 *** ebcube1 has joined #aichallenge 2011-07-18T16:25:46 I'm starting to think I might need to just terminate these and start fresh 2011-07-18T16:26:46 why? 2011-07-18T16:26:59 we should probably get a real api key at one point 2011-07-18T16:27:00 not just "" 2011-07-18T16:27:17 not sure, they just don't seem to be coming back after the reboot 2011-07-18T16:27:36 amstan: sure 2011-07-18T16:28:38 janzert: can be any length, right? 2011-07-18T16:29:01 *** Accoun has joined #aichallenge 2011-07-18T16:29:11 yep, it is really only used in a string comparison I believe 2011-07-18T16:31:51 ok 2011-07-18T16:31:53 i put it in 2011-07-18T16:32:29 amstan: https://github.com/aichallenge/aichallenge/issues/226 2011-07-18T16:32:58 asdas: oh.. right, no c# problem, you were with the timeout 2011-07-18T16:33:09 I'll need to kill the current ec2 instance requests and restart them with the new key 2011-07-18T16:34:33 you can get keys manually 2011-07-18T16:34:34 also.. 2011-07-18T16:34:40 i should probably clear the worker table 2011-07-18T16:34:48 why? 2011-07-18T16:35:10 all those api_keys were made with the non existent key 2011-07-18T16:35:24 also, it's not like there's any workers there we reuse, right? 2011-07-18T16:35:28 the api_keys really have no relation to the creation key 2011-07-18T16:35:29 you're going to restart yours 2011-07-18T16:35:36 the main server 2011-07-18T16:35:49 ok, i guess 2011-07-18T16:35:50 and I think McLeopold fires one up to test stuff every so often 2011-07-18T16:35:51 fine 2011-07-18T16:40:12 *** HumWum has quit IRC (Ping timeout: 252 seconds) 2011-07-18T16:41:27 hmm, well at least the one worker didn't come back up because of a filesystem error :( 2011-07-18T16:42:14 anyway starting new instance requests now 2011-07-18T16:44:41 why would it do that? 2011-07-18T16:45:02 *** mceier has joined #aichallenge 2011-07-18T16:46:56 when C++ compilation will be possible? 2011-07-18T16:47:47 not sure, two primary possibilities. One is just general ec2 ubuntu problems, we dealt with those a lot in the last contest. Or possibly more likely I do nasty things with the union filesystem for the sandbox and some race condition develops that causes problems eventually. 2011-07-18T16:55:03 *** mcstar has left #aichallenge ("WeeChat 0.3.5") 2011-07-18T16:56:34 *** ibdknox has joined #aichallenge 2011-07-18T17:21:18 *** smiley1983 has quit IRC (Ping timeout: 276 seconds) 2011-07-18T17:22:00 *** smiley1983 has joined #aichallenge 2011-07-18T17:33:47 *** mleise has joined #aichallenge 2011-07-18T17:34:36 *** berak has quit IRC (Remote host closed the connection) 2011-07-18T17:36:55 *** sigh has joined #aichallenge 2011-07-18T17:37:47 *** bhasker has joined #aichallenge 2011-07-18T17:52:37 *** ltriant has joined #aichallenge 2011-07-18T18:07:06 *** asdas has quit IRC (Ping timeout: 252 seconds) 2011-07-18T18:25:39 *** Palmik_ has quit IRC (Remote host closed the connection) 2011-07-18T18:26:55 *** nux67 has quit IRC (Quit: Page closed) 2011-07-18T18:44:52 *** sigh has quit IRC (Remote host closed the connection) 2011-07-18T19:03:07 *** jbrechtel has joined #aichallenge 2011-07-18T19:11:39 *** jmreardon has joined #aichallenge 2011-07-18T19:13:52 *** ibdknox has quit IRC (Ping timeout: 252 seconds) 2011-07-18T19:26:11 *** Apophis_ has joined #aichallenge 2011-07-18T19:28:33 *** nann has quit IRC (Ping timeout: 255 seconds) 2011-07-18T19:37:43 *** Zannick has quit IRC (*.net *.split) 2011-07-18T19:37:43 *** JamesMG has quit IRC (*.net *.split) 2011-07-18T19:40:06 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has quit IRC (Read error: Connection reset by peer) 2011-07-18T19:43:04 *** Zannick has joined #aichallenge 2011-07-18T19:43:04 *** JamesMG has joined #aichallenge 2011-07-18T19:44:22 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has joined #aichallenge 2011-07-18T19:47:01 *** FireFly has quit IRC (Quit: swatted to death) 2011-07-18T19:47:13 *** onensora has quit IRC () 2011-07-18T19:58:06 *** mceier has quit IRC (Quit: leaving) 2011-07-18T20:02:25 *** ebcube1 has left #aichallenge 2011-07-18T20:04:04 Anyone around that I can bug about merging my pull request? 2011-07-18T20:04:10 :) 2011-07-18T20:06:42 jbrechtel: link? 2011-07-18T20:06:49 https://github.com/aichallenge/aichallenge/pull/225 2011-07-18T20:07:28 jbrechtel: does it work on the server? 2011-07-18T20:07:47 Yes, it works on the server I created locally following the instructions in INSTALL.md 2011-07-18T20:07:56 With jailed workers, etc 2011-07-18T20:08:02 oh, ok 2011-07-18T20:08:15 aichallenge: Alexandru Stan epsilon * reb09139 / (6 files in 3 dirs): 2011-07-18T20:08:15 aichallenge: Merge pull request #225 from jbrechtel/epsilon 2011-07-18T20:08:15 aichallenge: Update Scala starter bot to work with Scala 2.9.0-1 2011-07-18T20:08:15 aichallenge: Thanks (+5 more commits...) - http://bit.ly/qs0eT0 2011-07-18T20:08:35 kick-ass! Thanks 2011-07-18T20:08:56 How frequently is the beta server updated? 2011-07-18T20:09:05 i could go update it now 2011-07-18T20:09:21 I'd really appreciate it...then I could finally submit my bot! 2011-07-18T20:09:43 jbrechtel: deoes that license file have to be in root though? 2011-07-18T20:09:51 *** bhasker has quit IRC (Ping timeout: 258 seconds) 2011-07-18T20:10:12 it would be much cleaner if it was in the scala folder instead 2011-07-18T20:10:12 It doesn't make any mention of its location requirement. 2011-07-18T20:10:16 Did I put it in aichallenge root? 2011-07-18T20:10:20 I didn't intend to, sorry about that. 2011-07-18T20:10:24 np 2011-07-18T20:10:29 can i move it? 2011-07-18T20:10:31 Sure 2011-07-18T20:12:27 aichallenge: Alexandru Stan epsilon * raa35e9e / (ants/dist/starter_bots/scala/LICENSE SCALA_LICENSE.txt): moved scala license - http://bit.ly/qoHNQi 2011-07-18T20:15:08 jbrechtel: http://aichallengebeta.hypertriangle.com/starter_packages/scala_starter_package.zip 2011-07-18T20:17:23 Awesome, thanks again 2011-07-18T20:23:17 *** jbrechtel_ has joined #aichallenge 2011-07-18T20:25:19 amstan: The beta server still says I fail test cases. I double checked and the zip I just sent to the beta server works on my local server. 2011-07-18T20:25:44 Is the worker using a different checked out copy of the git repo? 2011-07-18T20:25:55 should be the same 2011-07-18T20:25:59 what language? scala? 2011-07-18T20:26:02 yes 2011-07-18T20:26:15 you sure your local copy works? 2011-07-18T20:26:26 we had some issues 2011-07-18T20:26:37 apparently all bots fail the test case, they get eliminated 2011-07-18T20:26:48 Well, it passes test cases. The submission status is 'Success: ready to play'. 2011-07-18T20:26:58 I use it against bots locally frequently using playgame.py 2011-07-18T20:27:10 Let me upload another bot to my test server and verify that they'll actually play a match 2011-07-18T20:28:10 But it's definitely getting farther on my local server than on the beta server. 2011-07-18T20:30:21 So now I realize that I don't know how to make the server actually start executing games. Perhaps I need a critical mass of players? start_worker isn't starting a match like I would expect it two now that I have multiple players. 2011-07-18T20:37:48 jbrechtel: yeah, it wants like 4 to begin with 2011-07-18T20:37:50 Oh, looks like I needed maps 2011-07-18T20:37:54 that too 2011-07-18T20:37:54 Ah 2011-07-18T20:37:59 running a game now 2011-07-18T20:38:04 but why do you have a server? 2011-07-18T20:38:18 To figure out what was wrong with scala in compiler.py 2011-07-18T20:38:19 you know you can test the bots by just running ./testbot.sh against them 2011-07-18T20:38:31 That's what I've been doing to actually work on my scala bot since I couldn't upload it 2011-07-18T20:39:04 Since the bot code itself was OK I figured it had to be something wrong with the way it was either compiling it or executing it 2011-07-18T20:39:47 turned out it was executing. the execute command was something like "JAVA_OPTS=-Xmx512m; scala...." and trying to start a process like that gave an error 2011-07-18T20:39:57 that is shell syntax so that makes sense. 2011-07-18T20:44:57 You can see my server here: http://xmyth.ath.cx:3500 and http://xmyth.ath.cx:3500/profile.php?user=2 is my player with a Scala bot that's working 2011-07-18T20:53:08 Hmm...it may look like my bot is timing out a lot...I believe that's because the VM I have that server running on is hammered. 2011-07-18T21:01:25 jbrechtel: so what's the error on the main server that you're getting? 2011-07-18T21:01:28 failed test cases? 2011-07-18T21:01:43 Yes. It reports that the bot crashed. 2011-07-18T21:01:53 Which is the exact same behavior when it was trying to execute the scala process the wrong way. 2011-07-18T21:02:30 The only additional information I had was that when I saw the output of the worker that failed to execute it...i.e. from a terminal where I executed start_worker.sh explicitly....I would see the error 'No such file or directory' 2011-07-18T21:02:30 have you tried running your bot locally with ./testbot.sh? 2011-07-18T21:02:46 i don't really care about the worker, that's just extra overhead 2011-07-18T21:02:52 i'm interested in the basic sanity check 2011-07-18T21:02:55 let me try that...I haven't used testbot.sh in a while...I usually just use play_one_game.sh 2011-07-18T21:02:57 one sec 2011-07-18T21:03:10 I only mention it because I suspect that's where the problem lies. 2011-07-18T21:03:16 trying testbot now 2011-07-18T21:07:21 amstan: it does not crash with test_bot.sh locally 2011-07-18T21:07:24 result is: score 11 90 2011-07-18T21:07:25 status eliminated survived 2011-07-18T21:07:40 eliminated though 2011-07-18T21:07:45 it's bot 0? 2011-07-18T21:07:47 yes 2011-07-18T21:07:57 i think everyone is having the same problem too 2011-07-18T21:08:45 Hmm. It's reporting crashed on the beta site though, for my Scala uploads. http://aichallengebeta.hypertriangle.com/profile.php?user=312 2011-07-18T21:10:24 i don't see any crash 2011-07-18T21:10:28 it just says failed test cases 2011-07-18T21:10:35 and i think it's that eliminated 2011-07-18T21:10:54 That output underneath the failed test cases where it says "turn 0 bot 0 crashed" doesn't indicate a crash? 2011-07-18T21:13:41 all i see: 2011-07-19 04:47:31 Test Error: compiled, but failed test cases 2011-07-18T21:13:42 I'm horribly confused now. Looking at test_bot.sh I can't figure out how it's related to compiler.py at all. Yet, compiler.py is what I fixed to make my bot work on my local server. I verified it failing before I made the change and then verified it working afterwards. 2011-07-18T21:14:13 jbrechtel: so.. again, i have a feeling they fail the test cases because the test case is too hard 2011-07-18T21:14:18 so it makes the bot get eliminated 2011-07-18T21:14:27 now.. there's 2 problems with that(from our end, not your bot) 2011-07-18T21:14:31 1. the test case is too hard 2011-07-18T21:14:46 2. the bot fails if it gets eliminated, it should only fail if it crashes 2011-07-18T21:15:29 Fair enough, I'm still quite confused why we're seeing different things rendered on my profile page though. 2011-07-18T21:16:07 http://imgur.com/tAV4c 2011-07-18T21:17:07 jbrechtel: oh.. that.. 2011-07-18T21:17:14 it's probably because those errors are private 2011-07-18T21:17:21 oh, I see 2011-07-18T21:17:25 god.. that is ugly 2011-07-18T21:17:27 I suppose that makes some sense. 2011-07-18T21:17:35 Heh 2011-07-18T21:17:49 jbrechtel: i'm not sure then 2011-07-18T21:18:15 I was also quite confused why my bot was getting eliminated at all in the test scenario. I see now there is no food...that would explain it. 2011-07-18T21:18:41 jbrechtel: you know there might be a game where there no food 2011-07-18T21:18:46 Yea 2011-07-18T21:18:48 anyway.. 2011-07-18T21:18:56 i need mcleopold to clear some things up 2011-07-18T21:19:01 I'm not crashing in that scenario though...just not playing very intelligently yet 2011-07-18T21:26:28 amstan: Thanks for the help. If I see mcleopold around I'll pester him. I still feel like this is a problem with the worker not using an updated compiler.py. 2011-07-18T21:26:41 I really appreciate the help. 2011-07-18T21:26:44 when was the compiler updated? 2011-07-18T21:26:58 My pull request that you merged in. 2011-07-18T21:27:06 The last commit...one sec 2011-07-18T21:27:21 https://github.com/aichallenge/aichallenge/commit/b4de348195c91ae7cfc4e99ae3476915b4d1c32d 2011-07-18T21:27:42 oh yeah, no 2011-07-18T21:27:47 that won't be in the worker yet 2011-07-18T21:27:51 Oh... :) 2011-07-18T21:27:59 That's certainly the problem then. 2011-07-18T21:28:32 *** Amtal has quit IRC (Quit: Leaving) 2011-07-18T21:28:35 Can you update that? 2011-07-18T21:30:16 jbrechtel: yeah, but in a while 2011-07-18T21:30:28 actually 2011-07-18T21:30:36 janzert: ping? are you still working on the workers? 2011-07-18T21:31:54 oh man.. 2011-07-18T21:31:57 now i get it 2011-07-18T21:32:11 how does one do a decreasing for loop where is unsigned? 2011-07-18T21:32:18 for(i=7;i>=0;i--) 2011-07-18T21:32:22 but this will cause it to overflow 2011-07-18T21:34:10 i guess this works: for(i=7;i<8; i--) { 2011-07-18T21:34:44 that will go on forever right? 2011-07-18T21:35:01 oh 2011-07-18T21:35:04 nevermind 2011-07-18T21:35:09 you said unsigned 2011-07-18T21:35:13 that one won't, since it wraps 2011-07-18T21:35:30 but you can't do i>=0 when you have unsigned since it'll always be true 2011-07-18T21:55:30 *** bhasker has joined #aichallenge 2011-07-18T22:04:22 Good morning 2011-07-18T22:05:03 Amstan is there any news on the C# bot 0 timeout issue? 2011-07-18T22:19:09 Qua: you mean the failed test case? 2011-07-18T22:19:15 Yup 2011-07-18T22:19:24 Nope 2011-07-18T22:20:00 The issue is that it receives a timeout in one of the test cases 2011-07-18T22:20:40 not sure 2011-07-18T22:21:25 It appears that your main server might have different settings than the actual servers processing the uploads and compiling them 2011-07-18T22:21:40 any chance you could try the whole compile/test process on one of the worker servers? 2011-07-18T22:22:05 let me check 2011-07-18T22:34:57 *** phillc has joined #aichallenge 2011-07-18T22:42:52 Hiya, I am having no luck submitting a a haskell bot. I am getting "Did not find a recognized MyBot.* file." Even though my zip has a MyBot.hs in it. 2011-07-18T22:43:28 I'm looking through the source, and I have no reason to think the compile everything script shouldn't be picking it up 2011-07-18T22:45:05 phillc: how did you pack your code? 2011-07-18T22:45:22 there might a difference if you include it from a folder or not 2011-07-18T22:45:56 Are you asking if I have my source in a folder in side the zip? 2011-07-18T22:46:09 I am using my same rakefile from planetwars, assuming things haven't changed. 2011-07-18T22:46:26 Err, I mean I am building the zip the same way as I was during planetwars 2011-07-18T22:46:46 The zip file should contain nothing but pure .hs files 2011-07-18T22:46:50 no folders or make files 2011-07-18T22:47:06 No folders, hmm, I do have folders, maybe that is it. 2011-07-18T22:47:18 Going to give that a try, thanks =) 2011-07-18T22:48:45 *** smiley1983 has quit IRC (Quit: leaving) 2011-07-18T22:48:54 phillc: pretty much.. don't right click the folder and archive 2011-07-18T22:49:00 you have to archive what's inside the folder instead 2011-07-18T22:53:36 That is getting me farther, thanks. 2011-07-18T23:00:27 *** ibdknox has joined #aichallenge 2011-07-18T23:22:52 amstan? :p 2011-07-18T23:23:12 Qua: hmm? 2011-07-18T23:23:31 oh.. right.. worker server 2011-07-18T23:23:46 that page took so long to load i forgot i was loading it 2011-07-18T23:23:58 haha yeah alright 2011-07-18T23:24:12 Qua: the c# starter? 2011-07-18T23:24:16 Qua: can i have a link? 2011-07-18T23:24:31 to the zip? 2011-07-18T23:24:33 http://aichallengebeta.hypertriangle.com/starter_packages/csharp_starter_package.zip 2011-07-18T23:24:34 yes 2011-07-18T23:25:58 ugh.. 2011-07-18T23:26:02 nvm, i can't test it in here 2011-07-18T23:26:07 it has to be in a sandbox 2011-07-18T23:26:13 the packages are not installed outside the sandbox 2011-07-18T23:28:29 how is it testable then? 2011-07-18T23:28:51 not sure, janzert's in charge 2011-07-18T23:34:25 *** ibdknox_ has joined #aichallenge 2011-07-18T23:36:39 *** ibdknox has quit IRC (Ping timeout: 250 seconds) 2011-07-18T23:38:14 janzert? 2011-07-18T23:42:51 *** phillc has quit IRC (Quit: Page closed) 2011-07-18T23:53:41 *** jmreardon has quit IRC (Quit: jmreardon)