2011-06-07T00:03:18 contestbot: later tell McLeopold I've been looking at JSkills but it doesn't seem like it is even compilable, at least with the java 1.6.0 2011-06-07T00:03:18 janzert: I think that worked... 2011-06-07T00:07:37 contestbot: later tell McLeopold ahh, just found a pre-built jar of a slightly older version of JSkills. I'll try working from that. 2011-06-07T00:07:37 janzert: As you wish. 2011-06-07T00:27:56 *** moogiig has joined #aichallenge 2011-06-07T00:29:04 *** moogiig has quit IRC (Quit: Leaving) 2011-06-07T00:41:21 *** nann has quit IRC (Read error: Connection reset by peer) 2011-06-07T00:41:49 *** nann has joined #aichallenge 2011-06-07T00:42:03 *** Zaphus has joined #aichallenge 2011-06-07T00:42:07 *** okayzed is now known as okay 2011-06-07T00:42:39 Testing irc from my phone... 2011-06-07T00:43:25 *** Zaphus has quit IRC (Client Quit) 2011-06-07T00:44:22 *** delt0r_ has joined #aichallenge 2011-06-07T00:44:22 *** komeedipoeg has quit IRC (Ping timeout: 252 seconds) 2011-06-07T00:45:57 *** delt0r___ has quit IRC (Ping timeout: 244 seconds) 2011-06-07T01:20:04 contestbot: later tell McLeopold ok, I got JSkills working and compiling. Also have a simple command line script written that will do what we need and it matches the online calculator. 2011-06-07T01:20:04 janzert: I come to serve. 2011-06-07T01:30:42 *** Palmik has joined #aichallenge 2011-06-07T01:35:15 aichallenge: janzert epsilon * rae14c92 / (5 files): Add JSkills based trueskill updater - http://bit.ly/lvNwma 2011-06-07T01:35:16 aichallenge: janzert epsilon * rd81fd23 / (5 files): Merge branch 'epsilon' of github.com:aichallenge/aichallenge into epsilon - http://bit.ly/iZsuPw 2011-06-07T01:36:06 contestbot: later tell McLeopold see manager/jskills for the jskills based utility 2011-06-07T01:36:06 janzert: OK 2011-06-07T01:50:48 *** amstan has quit IRC (Remote host closed the connection) 2011-06-07T02:10:09 *** Eruonen has joined #aichallenge 2011-06-07T02:23:37 *** Eruonen has quit IRC () 2011-06-07T02:30:37 *** berak has joined #aichallenge 2011-06-07T02:39:21 *** berak has quit IRC (Quit: berak) 2011-06-07T02:54:48 *** UncleVasya has joined #aichallenge 2011-06-07T02:55:21 *** needsch has joined #aichallenge 2011-06-07T03:08:37 *** sir_macelon has joined #aichallenge 2011-06-07T03:09:49 hi 2011-06-07T03:10:38 hi 2011-06-07T03:12:32 *** aerique has joined #aichallenge 2011-06-07T03:18:27 *** dr- has joined #aichallenge 2011-06-07T03:18:33 Hm, if I try uploading the CL starter bot I get "bad submission key found" and no further error messages in my profile 2011-06-07T03:21:41 *** locutus2 has joined #aichallenge 2011-06-07T03:22:01 *** xQuasar has joined #aichallenge 2011-06-07T03:24:05 *** ltriant has quit IRC (Quit: Computer has gone to sleep) 2011-06-07T03:29:18 *** Quasar has joined #aichallenge 2011-06-07T03:29:45 *** Quasar is now known as Guest22227 2011-06-07T03:31:11 *** xQuasar has quit IRC (Ping timeout: 240 seconds) 2011-06-07T03:40:11 *** Guest22227 has quit IRC (Ping timeout: 252 seconds) 2011-06-07T03:44:52 *** qacek has quit IRC (Quit: WeeChat 0.3.5) 2011-06-07T03:45:29 *** xQuasar has joined #aichallenge 2011-06-07T03:49:25 aerique: to me that sounds like a worker error not compilation error. maybe try a different language bot to see if it comes back with the same error 2011-06-07T03:50:09 *** UncleVasya has quit IRC (Ping timeout: 260 seconds) 2011-06-07T03:51:59 *** needsch has quit IRC (Quit: Leaving.) 2011-06-07T03:54:31 janzert: ocaml works right away 2011-06-07T03:54:54 hmm, ok I guess not then 2011-06-07T03:56:25 If I look at the code in website/check_submit.php I don't see why the programming language should matter. 2011-06-07T03:57:35 I'll go kill my session cookie and see what happens. 2011-06-07T03:58:53 hmm, I also don't see how the error in check_submit.php would show up in a submission result 2011-06-07T03:59:57 deleting the cookie didn't work and submitting a JS bot does work 2011-06-07T04:01:09 what profile has the error? 2011-06-07T04:01:47 *** xQuasar has quit IRC (Ping timeout: 240 seconds) 2011-06-07T04:01:52 both "aerique-test" and "lispstarterbot", but i've just been using "aerique-test" for the above results 2011-06-07T04:02:19 *** xQuasar has joined #aichallenge 2011-06-07T04:03:05 I think I misunderstood, the error is displayed immediately on the resulting page when you try and upload or it shows up in the submission list on the profile page? 2011-06-07T04:04:17 it shows after pressing upload (but it takes 15 / 30 seconds, while a succesful upload is pretty much immediate) 2011-06-07T04:04:46 yeah, that's really weird then 2011-06-07T04:05:31 save the current one, but could you try recreating the starter bot archive at say a different compression level 2011-06-07T04:05:49 sure 2011-06-07T04:05:50 I'm wondering if it's some specific byte sequence in the file 2011-06-07T04:06:27 or maybe better try it with the same file but a different browser 2011-06-07T04:07:24 i uploaded a tar.gz instead of a .zip and it worked 2011-06-07T04:07:35 because actually I don't see how the upload file could effect a post field unless either browser or server are doing an encoding or decoding wrong 2011-06-07T04:07:40 i'll try the malfunctioning zip in chrome now 2011-06-07T04:07:48 thanks 2011-06-07T04:11:46 janzert: in chrome with the "lispstaerbot" account the malfunctioning zip didn't work either but the .tar.gz one did 2011-06-07T04:12:06 so it doesn't work with either chrome or ? 2011-06-07T04:12:14 i've ran out of browsers on this machine now :) 2011-06-07T04:12:28 nope, chrome does the same thing as ff 2011-06-07T04:12:36 but the workaround does work 2011-06-07T04:13:02 ok FF, so a completely different vintage than Chrome ;) 2011-06-07T04:13:12 yeah, seems like it must be server side 2011-06-07T04:13:29 so thanks for that :) shall i edit the error message a little with your workaround? 2011-06-07T04:13:37 is there somewhere publicly accessible you could upload the zip? 2011-06-07T04:13:58 it's just the zip from http://aichallengebeta.hypertriangle.com/starter_packages.php 2011-06-07T04:14:01 no, this is something that really shouldn't be able to happen 2011-06-07T04:14:02 for common lisp 2011-06-07T04:14:17 now that you mention it, i'll go check the zip for errors 2011-06-07T04:16:33 no errors locally, but perhaps it does give errors on the server? 2011-06-07T04:17:03 hmm, not an error but it really shouldn't contain the bot binary 2011-06-07T04:17:12 does it? 2011-06-07T04:17:20 and just confirmed that it causes the same error when I try and upload it 2011-06-07T04:17:29 you're right 2011-06-07T04:17:33 i missed that :) 2011-06-07T04:17:46 so almost certainly something server side and not in the network path either 2011-06-07T04:18:13 are there any size limits when unpacking? 2011-06-07T04:18:31 the binary's pretty big and it shouldn't be there 2011-06-07T04:18:34 not sure, but the error is way before that 2011-06-07T04:19:34 for kicks and giggles going to try uploading with IE once :P 2011-06-07T04:19:48 *** Naktibalda has joined #aichallenge 2011-06-07T04:19:54 i haven't got IE here (linux) 2011-06-07T04:20:19 * janzert has both running :/ 2011-06-07T04:22:14 why would the binary be in the starter package? running make in ants/dist/starter_bots doesn't put it there 2011-06-07T04:23:13 actually, i should get back to work, sorry :-( thanks for the workaround 2011-06-07T04:23:32 probably a quirk of the beta server, my guess is someone happened to build the bot before building the starter packages 2011-06-07T04:23:45 thanks for the testing and finding a problem 2011-06-07T04:24:25 and IE causes the same error, so pretty positively a server side error 2011-06-07T04:24:57 my bet's on the binary and the size of the starter package 2011-06-07T04:25:42 not the size but somehow the actual binary, but really the way php works its not suppose to allow this to happen 2011-06-07T04:27:26 *** UncleVasya has joined #aichallenge 2011-06-07T04:33:54 *** UncleVasya has quit IRC (Ping timeout: 252 seconds) 2011-06-07T04:34:55 *** Cyndre has quit IRC (Ping timeout: 260 seconds) 2011-06-07T04:36:14 *** locutus2 has quit IRC (Ping timeout: 260 seconds) 2011-06-07T04:36:43 What are the bounds for the number of players within any single game? 2011-06-07T04:37:31 *** UncleVasya has joined #aichallenge 2011-06-07T04:37:54 well, the map format has a hard limit at 26 2011-06-07T04:38:10 I don't expect more than 15 at the very most though 2011-06-07T04:38:50 and probably somewhere between 9 and 12 is the most likely maximum 2011-06-07T04:38:55 minimum of course is 2 2011-06-07T04:39:35 I have an idea for visualizer. 2011-06-07T04:40:01 To show home many enemy ants bot killed vitoud his own losses. 2011-06-07T04:40:12 *without 2011-06-07T04:43:45 I need to repeat it to the Frontier? 2011-06-07T04:44:47 yes, Frontier handles pretty much all the visualizer development 2011-06-07T04:46:15 I'll try to use contestbot. 2011-06-07T04:46:53 contestbot: later tell Frontier I have an idea for visualizer. To show home many enemy ants bot killed vitoud his own losses. 2011-06-07T04:46:53 UncleVasya: I think that worked... 2011-06-07T04:47:12 wow, success :) 2011-06-07T04:48:35 yay :) 2011-06-07T04:54:09 *** olexs has joined #aichallenge 2011-06-07T04:54:54 Also I think that it will be kindly to insert pictures in sections Battle resolutions and Spawn resolutions of the Game Specifications. 2011-06-07T04:58:35 For example show scheme with 7 ants of 3 different bots and describe Battle resolutions with lines and numbers beetwen these ants. "Ant 3 damage = 0.5 from Ant 2 and 1 from Ant 3 = 1.5 -> he died" 2011-06-07T04:59:14 yeah, they both need a bit of filling out and polishing before the contest opens 2011-06-07T05:01:09 *** Cyndre has joined #aichallenge 2011-06-07T05:11:27 *** olexs has quit IRC (Quit: Leaving.) 2011-06-07T05:30:11 A question concerning the Fog Of War: 2011-06-07T05:30:17 "At the moment this is set at 93, giving a view radius of approximately 9.6.” 2011-06-07T05:30:22 aichallenge: janzert epsilon * r798a74f / manager/manager.py : Change manager to use JSkills based trueskill updates - http://bit.ly/kpRVGf 2011-06-07T05:30:31 that is old sorry 2011-06-07T05:30:38 Ok what’s the new value? 2011-06-07T05:30:45 can anyone remember the current value? 2011-06-07T05:30:59 i know we have spawn2 = 1, attack2 = 4 2011-06-07T05:31:08 I think it should be part of the initial values sent to the clients 2011-06-07T05:31:10 53 rings a bell, but i could be wrong 2011-06-07T05:31:12 it is 2011-06-07T05:31:14 contestbot: later tell McLeopold I changed the manager to use the JSkills based updater. feel free to revert it if you want to track down the python trueskill problem 2011-06-07T05:31:14 janzert: Ready to serve, my lord. 2011-06-07T05:31:29 OH. WHOOPS. 2011-06-07T05:31:36 ok 2011-06-07T05:31:44 contestbot: later tell McLeopold I did not update the beta server in anyway though 2011-06-07T05:31:44 janzert: OK 2011-06-07T05:31:50 Also, say that the view radius is something like 5.3 2011-06-07T05:31:51 *** Frontier has joined #aichallenge 2011-06-07T05:32:02 with parameters? 2011-06-07T05:32:03 what about the .3? 2011-06-07T05:32:12 the reason we sent it squared is so that we don't have rounding errors 2011-06-07T05:32:16 send* 2011-06-07T05:32:23 or have people having to read doubles/etc. 2011-06-07T05:33:52 antimatroid: Ok so what about the extra decimal values then? Is it rounded? 2011-06-07T05:34:01 can you see a square if you can see 0.3 into it? 2011-06-07T05:34:15 look at the distance metric 2011-06-07T05:34:34 all the squared radius' we use will be integral values 2011-06-07T05:34:39 hmm... why contestbot said nothing to the Frontier? Or he did it in private? 2011-06-07T05:34:49 if you have an ant at aloc,then it can see any loc where dist(aloc, loc) <= sqrt(vision2) 2011-06-07T05:35:16 it sends the "later tell" message by private message, yes 2011-06-07T05:36:20 xQuasar: You don't actually need to use the sqrt() of a radius² value. You'll only need comparisons à la "is it within range", so you can do: row²+col² <= viewradius² 2011-06-07T05:36:42 yeah, you can square the distance metric locally if you want 2011-06-07T05:36:58 but the distances aren't intuitive then, so it's easier to officially use the metric we are ;) 2011-06-07T05:38:07 no fog of war is equivalent to having fog of war with the discrete metric and vision2 >=1 2011-06-07T05:40:25 UncleVasya: contestbot always opens a private chat. I don't think what you want can easily be done. If 3 bots attack each other one bot may be killed by multiple bots of multiple players 2011-06-07T05:40:56 ok 2011-06-07T05:41:13 Probably 'damage absorbed' would work somehow. 2011-06-07T05:42:12 *** janzert has left #aichallenge 2011-06-07T05:42:23 *** janzert has joined #aichallenge 2011-06-07T05:42:24 But I have other things on my todo list, like viewing a local game that is in progress 2011-06-07T05:45:06 *** locutus2 has joined #aichallenge 2011-06-07T05:45:51 I don't think it's hard to wait until game will end ant after that watch a replay. 2011-06-07T05:46:36 *and after that... 2011-06-07T05:48:26 What is the point to watch games in progress? I see only an adrenaline because you do not know who will win the fight. 2011-06-07T05:50:35 UncleVasya: 1. We had that feature in the last contest and 2. you can stop the engine when you see that your bot doesn't do what you want 2011-06-07T05:51:41 Especially when the bots run slow and there are more than 3 players and lots of turns it is good to see what is going on. 2011-06-07T05:54:56 I've got it. 2011-06-07T05:56:44 Interesting bot to watch when he have enough ants: http://aichallengebeta.hypertriangle.com/visualizer.php?game=16262 2011-06-07T05:56:57 Reminds me some kind of tetris. 2011-06-07T06:11:29 contestbot: later tel Smiley1983 hah, nice one! http://aichallengebeta.hypertriangle.com/visualizer.php?game=16319 2011-06-07T06:11:30 UncleVasya: Run as fast as you can and don't look back. 2011-06-07T06:11:39 contestbot: later tell Smiley1983 hah, nice one! http://aichallengebeta.hypertriangle.com/visualizer.php?game=16319 2011-06-07T06:11:39 UncleVasya: Aye, aye, sir 2011-06-07T06:16:12 *** UncleVasya has quit IRC (Ping timeout: 252 seconds) 2011-06-07T06:30:20 Frontier: Mcleopold started a bug thing because of colliding ants not being obvious to him 2011-06-07T06:30:36 that might say a lot about what others are going to see, is it possible to make it more obvious when ants die from colliding? 2011-06-07T06:33:36 Any ideas on how it could be visualized? 2011-06-07T06:34:31 a cross on the square? 2011-06-07T06:38:34 Has already someone tested the stability of the trueskill rating by bringing two identical bots on the server? 2011-06-07T06:39:33 If not and its allowed i create a second account and load my bot on this too 2011-06-07T06:40:08 *** UncleVasya has joined #aichallenge 2011-06-07T06:44:17 locutus2: a second account should not be a problem on the test server 2011-06-07T06:44:53 ok i make it so! 2011-06-07T06:45:53 it would be interresting clone bots of different skill levels par example top, middle, bottom 2011-06-07T06:47:08 because i think in the middle of the field the instability highest 2011-06-07T06:47:32 or the opposit :-) 2011-06-07T06:50:40 *** boegel has joined #aichallenge 2011-06-07T06:54:31 contestbot: later tell McLeopold i think for testing stability of the true skill rating its wise upload two identical bots. if allowed i create a second acount with my bot cloned 2011-06-07T06:54:31 locutus2: Ready to serve, my lord. 2011-06-07T07:22:05 *** xQuasar has quit IRC (Ping timeout: 252 seconds) 2011-06-07T07:26:25 *** xQuasar has joined #aichallenge 2011-06-07T07:29:22 *** Quasar has joined #aichallenge 2011-06-07T07:29:48 *** Quasar is now known as Guest92639 2011-06-07T07:33:03 *** xQuasar has quit IRC (Ping timeout: 255 seconds) 2011-06-07T07:47:46 *** olexs has joined #aichallenge 2011-06-07T07:52:21 *** olexs has quit IRC (Client Quit) 2011-06-07T08:02:21 *** guyverthree has joined #aichallenge 2011-06-07T08:02:23 alright guys 2011-06-07T08:02:44 *** ccccccccccc has joined #aichallenge 2011-06-07T08:07:16 *** ccccccccccc is now known as carlop 2011-06-07T08:07:22 *** olexs has joined #aichallenge 2011-06-07T08:13:11 *** guyverthree has quit IRC (Quit: Page closed) 2011-06-07T08:19:57 *** smiley1983 has joined #aichallenge 2011-06-07T08:20:53 does anyone know how to increase the food starting amount / spawn rate in local games? Couldn't find it in playgame options 2011-06-07T08:26:37 smiley1983: If I am correct, I can see it the Ants.py (lines 41-50). 2011-06-07T08:26:42 - 2011-06-07T08:33:31 aichallenge: aerique epsilon * re9a603c / ants/visualizer/js/Replay.js : More elaborate error message for setReplayDuration. - http://bit.ly/lkX9jD 2011-06-07T08:34:18 *** smiley1983 has quit IRC (Ping timeout: 248 seconds) 2011-06-07T08:35:24 *** UncleVasya has quit IRC (Ping timeout: 260 seconds) 2011-06-07T08:35:25 Is the removal of ‘r’ updated on http://aichallengebeta.hypertriangle.com/? 2011-06-07T08:35:27 *** smiley1983 has joined #aichallenge 2011-06-07T08:36:47 *** olexs has quit IRC (Quit: Leaving.) 2011-06-07T08:37:00 *** locutus2 has quit IRC (Ping timeout: 260 seconds) 2011-06-07T08:37:13 *** UncleVasya has joined #aichallenge 2011-06-07T08:37:50 smiley1983: Was I correct? 2011-06-07T08:40:33 *** smiley1983 has quit IRC (Ping timeout: 240 seconds) 2011-06-07T08:41:50 *** smiley1983 has joined #aichallenge 2011-06-07T08:44:42 *** Guest92639 has quit IRC () 2011-06-07T08:44:56 *** delt0r- has joined #aichallenge 2011-06-07T08:45:00 *** Zaphus has joined #aichallenge 2011-06-07T08:46:47 *** delt0r_ has quit IRC (Ping timeout: 276 seconds) 2011-06-07T08:47:46 *** Zaphus has quit IRC (Client Quit) 2011-06-07T08:52:33 *** locutus2 has joined #aichallenge 2011-06-07T08:57:14 *** smiley1983 has quit IRC (Ping timeout: 248 seconds) 2011-06-07T08:57:32 *** sigh has joined #aichallenge 2011-06-07T08:57:54 *** sigh has quit IRC (Remote host closed the connection) 2011-06-07T08:58:21 *** smiley1983 has joined #aichallenge 2011-06-07T08:59:44 *** sigh has joined #aichallenge 2011-06-07T09:04:42 *** smiley1983 has quit IRC (Ping timeout: 252 seconds) 2011-06-07T09:04:56 *** olexs has joined #aichallenge 2011-06-07T09:05:43 *** smiley1983 has joined #aichallenge 2011-06-07T09:13:59 *** amstan_ has joined #aichallenge 2011-06-07T09:13:59 *** ChanServ sets mode: +o amstan_ 2011-06-07T09:21:13 *** olexs has quit IRC (Quit: Leaving.) 2011-06-07T09:32:57 *** locutus2 has quit IRC (Ping timeout: 240 seconds) 2011-06-07T09:37:31 *** amstan_ has quit IRC (Read error: Operation timed out) 2011-06-07T09:40:17 *** smiley1983 has quit IRC (Ping timeout: 276 seconds) 2011-06-07T09:40:33 *** smiley1983 has joined #aichallenge 2011-06-07T09:48:47 *** amstan_ has joined #aichallenge 2011-06-07T09:48:47 *** ChanServ sets mode: +o amstan_ 2011-06-07T09:58:59 *** mcstar has joined #aichallenge 2011-06-07T10:01:14 *** sir_macelon has quit IRC (Quit: Page closed) 2011-06-07T10:03:43 *** berak has joined #aichallenge 2011-06-07T10:11:36 *** sigh has quit IRC (Remote host closed the connection) 2011-06-07T10:12:49 *** UncleVasya has quit IRC (Ping timeout: 260 seconds) 2011-06-07T10:21:14 *** smiley1983 has quit IRC (Ping timeout: 276 seconds) 2011-06-07T10:21:29 *** smiley1983 has joined #aichallenge 2011-06-07T10:29:40 *** needsch has joined #aichallenge 2011-06-07T10:30:19 *** amstan_ has quit IRC (Ping timeout: 260 seconds) 2011-06-07T10:31:45 *** smiley1983 has quit IRC (Ping timeout: 240 seconds) 2011-06-07T10:37:08 *** locutus2 has joined #aichallenge 2011-06-07T10:38:13 *** smiley1983 has joined #aichallenge 2011-06-07T10:47:44 *** amstan_ has joined #aichallenge 2011-06-07T10:47:44 *** ChanServ sets mode: +o amstan_ 2011-06-07T10:48:43 *** needsch has quit IRC (Read error: Connection reset by peer) 2011-06-07T10:49:30 *** UncleVasya has joined #aichallenge 2011-06-07T10:56:52 *** FireFly has joined #aichallenge 2011-06-07T10:57:14 *** needsch has joined #aichallenge 2011-06-07T11:04:55 *** amstan_ has quit IRC (Read error: Operation timed out) 2011-06-07T11:06:24 *** olexs has joined #aichallenge 2011-06-07T11:08:22 Frontier: Instead of Player1, Player2... in the local visualizer I want the name of the bot file without extension (for example MyBot.exe -> MyBot). I think I have to seek somewhere else, not in the visualizer, for this thing to work and I have to talk to someone else, not to you. But also I think you know where I have to seek or whom I have to talk:) 2011-06-07T11:09:17 It's in the replay file, so figure out where that is generated (ants.py?). 2011-06-07T11:09:26 UncleVasya: That's right: Talk to McLeopold or sigh about this :) 2011-06-07T11:09:55 *** olexs has quit IRC (Client Quit) 2011-06-07T11:12:55 Frontier: I thought about bot's kills without losses. It can be change to "how many times your ants where involved in the battle and survived". It seems not very difficult to implement. But again, it's a logic part, not your's? 2011-06-07T11:14:16 *** McLeopold has joined #aichallenge 2011-06-07T11:14:51 *** aerique has quit IRC (Quit: ...) 2011-06-07T11:16:06 Frontier: "player_info":[{},{},{},{},{}] - Does players names must be stored here in the replay? 2011-06-07T11:16:20 *** locutus2 has quit IRC (Ping timeout: 240 seconds) 2011-06-07T11:18:59 Frontier: I can't talk to McLeopold directly because my English makes him really unhappy. :( 2011-06-07T11:19:09 hi 2011-06-07T11:19:56 hi 2011-06-07T11:20:35 *** locutus2 has joined #aichallenge 2011-06-07T11:23:23 McLeopold: I'm sorry, but I will :) Instead of "Player1" in the local visualizer I want the name of the bot file without extension (for example MyBot.exe -> MyBot). What I have to change? 2011-06-07T11:24:35 And that was first. Second one: 2011-06-07T11:24:49 *** needsch has quit IRC (Read error: Connection reset by peer) 2011-06-07T11:26:09 McLeopold: What do you think about idea to show people not only their scores in the visualizer but also "how many times bot's ants where involved in battle and survived." 2011-06-07T11:28:25 When people will end up with food gathering algoritms they will focuse on the fight. And instead of watching your bot's every move you can look at this value and compare to the value of ypur opponents. 2011-06-07T11:29:28 UncleVasya: https://github.com/aichallenge/aichallenge/issues/151 this issue tracks extra stats that we want, add your ideas to this issue 2011-06-07T11:29:35 you can name it "Tactical score" 2011-06-07T11:29:48 I called it kill ratio 2011-06-07T11:30:08 also, I agree that for local testing, the visualizer should have better names 2011-06-07T11:30:25 McLeopold: is this idea good enough to be added there? 2011-06-07T11:30:37 This means playgame.py needs to be modified to capture the replay data and interject the extra bot program name into the json 2011-06-07T11:30:56 UncleVasya: yes, I already added it :) 2011-06-07T11:31:25 *** boegel has quit IRC (Ping timeout: 260 seconds) 2011-06-07T11:31:34 *** needsch has joined #aichallenge 2011-06-07T11:31:38 also, you english doesn't make my unhappy, it just makes me laugh sometimes 2011-06-07T11:31:44 *** needsch has quit IRC (Client Quit) 2011-06-07T11:34:35 McLeopold: Can I discuss ideas here and only if it seems to be good add it to the github? Or this is not proper place and I have to do it only on github? 2011-06-07T11:35:56 Both are fine 2011-06-07T11:36:27 Conversation here is usually faster and gets more done, but if we don't put it on github, we may forget 2011-06-07T11:39:48 Playgame.py: "--rounds" "Number of rounds to play" Is this a number of games on the same map with the same bots? 2011-06-07T11:39:56 Yes 2011-06-07T11:40:20 It was the start of getting a round robin tournament to work, but I didn't finish it. 2011-06-07T11:40:38 *** amstan_ has joined #aichallenge 2011-06-07T11:40:38 *** ChanServ sets mode: +o amstan_ 2011-06-07T11:40:57 Perhaps a separate tournament file should be created for this 2011-06-07T11:41:32 *** locutus2 has quit IRC (Ping timeout: 240 seconds) 2011-06-07T11:46:13 *** needsch has joined #aichallenge 2011-06-07T11:47:03 *** amstan_ has quit IRC (Quit: Konversation terminated!) 2011-06-07T11:47:26 * McLeopold thinks amstan maximized his terminal window... 2011-06-07T11:52:28 *** Eruonen has joined #aichallenge 2011-06-07T11:53:16 *** aeter has joined #aichallenge 2011-06-07T11:54:16 *** FireFly has quit IRC (Read error: Connection reset by peer) 2011-06-07T11:56:18 *** ohcrivens has joined #aichallenge 2011-06-07T11:59:07 *** Kingpin13 has joined #aichallenge 2011-06-07T12:01:37 *** stocha has joined #aichallenge 2011-06-07T12:01:44 (crashed) Could not create the Java virtual machine. 2011-06-07T12:01:52 sometime there is no message at all. 2011-06-07T12:02:01 what to make out of it ? 2011-06-07T12:02:10 *** locutus2 has joined #aichallenge 2011-06-07T12:05:39 it seems all java bot run into the trouble. 2011-06-07T12:08:14 stocha: it's because of betaserver's low RAM. On the main server it will be ok. 2011-06-07T12:08:59 McLeopold: what think you about testing trueskill stability by running a second copy of a bot? 2011-06-07T12:09:13 if it for the server admins ok i would try it 2011-06-07T12:09:43 why not try with less players then ? 2011-06-07T12:09:46 like 3 2011-06-07T12:09:58 so all java bot will not crash all the time. 2011-06-07T12:10:09 stocha: If you wanna test your bot against someone's one I can give you an .exe of my bot (top 10 nowbut still very weak). 2011-06-07T12:10:46 i don't want to test it against another one :) I know my bot behavior is probably in the best right now. 2011-06-07T12:11:16 also the bot is so old, i don't remember anything about the code :p. I just wanted to see how things go. 2011-06-07T12:11:57 if it would have enough time and such. apparently it simply doesn't boot most of the time. 2011-06-07T12:12:17 Heheh. So I have to be gateful to these problems with Java because only with their help I'm in top 10 now :) 2011-06-07T12:12:51 so fare it has 3 crash, and one first position. 2011-06-07T12:19:34 *** stocha has quit IRC (Ping timeout: 252 seconds) 2011-06-07T12:19:34 *** needsch has quit IRC (Read error: Connection reset by peer) 2011-06-07T12:19:59 locutus2: go ahead and create multiple accounts on beta 2011-06-07T12:21:16 *** McLeopold has left #aichallenge 2011-06-07T12:22:17 *** needsch has joined #aichallenge 2011-06-07T12:25:30 hmm, shouldn't the compile errors be private ?, looking at stocha's errors, i can tell, that it's the bot he posted on the forum weeks ago! 2011-06-07T12:26:13 (still beats anything i can come up with, so far;) 2011-06-07T12:27:27 ok, i have created a second account locutus2_copy and uploaded the same version wich is running under locutus2 2011-06-07T12:27:28 *** FireFly has joined #aichallenge 2011-06-07T12:28:05 i'am very interrested in the results 2011-06-07T12:30:13 *** needsch has quit IRC (Quit: Leaving.) 2011-06-07T12:30:25 *** needsch has joined #aichallenge 2011-06-07T12:31:49 insofar i known each new version will start with trueskill stats mu=50 and sigma=50/3 2011-06-07T12:32:36 what if we start a new version of bot with mu from last version but with the sigma of a complete new entry? 2011-06-07T12:33:16 *** needsch has quit IRC (Client Quit) 2011-06-07T12:34:51 it should faster converge and the opponents skill updates are not biased by a more or less 'random' skill of the new version 2011-06-07T12:39:05 McLeopold: if i upload the copy_version i get an empty site and under myprofile i find no submission. has the server problems or are there checks which prevent me from uploading, like same file or Email-Address? 2011-06-07T12:40:17 Help! People are discussing stuff that produces work for me :p 2011-06-07T12:40:51 I need pizza and a coffee. 2011-06-07T12:43:03 crazy mixture! better is pizza and beer 2011-06-07T12:43:58 *** _0x47 <_0x47!~slash@p5B12EF8E.dip.t-dialin.net> has quit IRC (Read error: Operation timed out) 2011-06-07T12:44:10 *** _0x47 <_0x47!~slash@p5B12E016.dip.t-dialin.net> has joined #aichallenge 2011-06-07T12:45:49 wow, McLeopold posted a question on stack overflow 2 hours ago and it is already the top result in google for the related query "php detect max_post_size exceeded" 2011-06-07T12:46:40 after upload the last url is http://aichallengebeta.hypertriangle.com/check_submit.php and a blank site 2011-06-07T12:47:28 locutus2: it shouldn't be doing any duplicate detection 2011-06-07T12:47:53 locutus2: can you try uploading say a starter bot package and see if that works? 2011-06-07T12:48:07 locutus2: it sounds like check_submit.php got broken though 2011-06-07T12:51:07 i try a starter bot kit 2011-06-07T12:52:29 no, the same problem, no new submission and a black site 2011-06-07T12:54:38 i try to resubmit my bot on my normal account 2011-06-07T12:55:53 ok, my guess is McLeopold is doing some work on trying to fix a bug in that file 2011-06-07T12:56:28 at least hopefully he's still actively working and didn't accidentally leave it in a completely broken state 2011-06-07T12:57:17 i have resubmit my own bot, there i get the error my file is too big. A Maximum of 1 KB is allowed! 2011-06-07T12:57:42 lol 2011-06-07T12:57:57 yeah, he's actively making changes 2011-06-07T13:00:58 *** McLeopold has joined #aichallenge 2011-06-07T13:01:36 janzert: I find it hilarious that you answered my stackoverflow questions on a bug that you opened... 2011-06-07T13:01:49 :) 2011-06-07T13:02:11 *** Naktibalda has quit IRC (Quit: ChatZilla 0.9.86.1 [Firefox 4.0.1/20110413222027]) 2011-06-07T13:02:19 just had to find the write google fu to get my IQ boost :P 2011-06-07T13:02:52 I swear I searched stackoverflow for "post_max_size" and didn't get anything, now it's the first result 2011-06-07T13:02:53 and obviously it didn't extend to discerning homonyms 2011-06-07T13:03:25 boost my rep while your at it :) 2011-06-07T13:03:40 I can't I don't have enough rep to upvote yet :( 2011-06-07T13:04:01 yes, my answer was an attempt at karma whoring :P 2011-06-07T13:04:37 geeze, how did I miss 15 duplicate questions? 2011-06-07T13:05:35 hehe 2011-06-07T13:06:30 *** ohcrivens has quit IRC (Ping timeout: 252 seconds) 2011-06-07T13:08:46 since we only have to handle a single upload at a time, I wonder if the simplest/best fix isn't to just make sure max_post_size is larger than upload_max_filesize? 2011-06-07T13:09:46 i.e. set the former to 11MB and the latter to 10MB 2011-06-07T13:20:10 janzert: that's a fun idea 2011-06-07T13:21:12 janzert: they can always exceed post_size, and then the super globals are blank 2011-06-07T13:28:14 also set a MAX_FILE_SIZE hidden input in the upload form that precedes the file input and apparently php will cut the upload off at that size 2011-06-07T13:29:22 doesn't stop a malicious user from going larger but will stop any normal user from seeing the nasty error case 2011-06-07T13:29:31 http://ca2.php.net/manual/en/features.file-upload.post-method.php 2011-06-07T13:32:15 although the above page implies the browser will prevent the upload altogether if it exceeds MAX_FILE_SIZE but from reading elsewhere it is rather confusing whether browsers actually do or not 2011-06-07T13:41:44 *** boegel has joined #aichallenge 2011-06-07T13:43:09 McLeopold: btw, did you get a chance to look at any of the JSkills based stuff yet and maybe try some tests against the online calc? 2011-06-07T13:46:18 *** locutus2 has quit IRC (Ping timeout: 248 seconds) 2011-06-07T13:47:07 janzert: not yet 2011-06-07T13:47:55 If the test_trueskill.py test case now matches the online calculator, then let's switch, even if it isn't perfect. 2011-06-07T13:49:11 yeah, it does 2011-06-07T13:49:38 don't have the time to update the beta server right now but I'll try and get it later 2011-06-07T13:52:44 *** |UncleVasya| <|UncleVasya|!kvirc@178-133-242-98.dialup.umc.net.ua> has joined #aichallenge 2011-06-07T13:56:49 *** UncleVasya has quit IRC (Ping timeout: 260 seconds) 2011-06-07T14:06:11 McLeopold: When the replay is streamed and the Java window is closed, should I keep the visualizer running until the engine has finished or should I close the pipe and exit? 2011-06-07T14:07:00 If I close the pipe the engine would have to handle the closed stdout. 2011-06-07T14:07:41 *** carlop has quit IRC (Ping timeout: 240 seconds) 2011-06-07T14:09:53 *** locutus2 has joined #aichallenge 2011-06-07T14:10:03 On another note, the visualizer is so slow that I had to make it signal to the stdin reader when it is ready for another turn. Before the screen would never update as the new turns came in faster than the JavaScript processing them would execute :p 2011-06-07T14:11:10 Frontier: I haven't actually used the java visualizer with streaming 2011-06-07T14:12:48 I would probably close the pipe and make playgame handle it. Playgame might also be doing other logging, so we wouldn't want to kill the game. 2011-06-07T14:12:49 eh.. how could you. the streaming format isn't implemented yet 2011-06-07T14:13:22 so, yeah, I couldn't :) 2011-06-07T14:14:11 Close the pipe, ok. I think that's best. Just right now the Python script will error out at this point. 2011-06-07T14:24:02 aichallenge: McLeopold epsilon * rfe7d608 / (website/check_submit.php website/submit.php): Fixed large upload issue. Closes #166 - http://bit.ly/iV0I6H 2011-06-07T14:25:27 *** Naktibalda has joined #aichallenge 2011-06-07T14:25:34 *** smiley1983 has quit IRC (Quit: leaving) 2011-06-07T14:31:43 aichallenge: McLeopold epsilon * rd74c6e9 / website/check_submit.php : Better file upload error message - http://bit.ly/jnX1GD 2011-06-07T14:33:53 *** jmpespxoreax has quit IRC (Ping timeout: 276 seconds) 2011-06-07T14:35:56 aichallenge: McLeopold epsilon * r5cefefd / (website/check_submit.php website/submit.php): Better file upload error message - http://bit.ly/ipKFr0 2011-06-07T14:39:29 *** McLeopold has left #aichallenge 2011-06-07T14:40:55 contestbot: later tell amstan Unless you have a protest I'm going to delete the ants-epsilon label from the issue tracker since every open issue either has or should have that label and it is simply lowering the utility of using the label system at all 2011-06-07T14:40:56 janzert: I come to serve. 2011-06-07T14:53:23 something is wrong. since the last few games the skill is everytime zero 2011-06-07T14:55:24 *** locutus2 has left #aichallenge 2011-06-07T14:55:34 *** locutus2 has joined #aichallenge 2011-06-07T14:57:18 locutus2: should be fixed now 2011-06-07T14:57:43 *** okay is now known as okayzed 2011-06-07T14:57:48 new updater code got pulled in but didn't get built 2011-06-07T14:59:21 thanks 2011-06-07T15:04:26 *** locutus2 has quit IRC (Ping timeout: 276 seconds) 2011-06-07T15:10:31 *** locutus2 has joined #aichallenge 2011-06-07T15:11:51 *** computerwiz_222 has joined #aichallenge 2011-06-07T15:15:29 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has joined #aichallenge 2011-06-07T15:15:35 *** amstan has joined #aichallenge 2011-06-07T15:15:35 *** ChanServ sets mode: +o amstan 2011-06-07T15:45:09 *** locutus2 has quit IRC (Ping timeout: 252 seconds) 2011-06-07T15:57:42 *** Cyndre_ has joined #aichallenge 2011-06-07T15:59:47 *** aerique has joined #aichallenge 2011-06-07T16:01:08 *** Accoun has quit IRC () 2011-06-07T16:01:11 *** Cyndre has quit IRC (Ping timeout: 258 seconds) 2011-06-07T16:07:11 *** boegel has quit IRC (Quit: Leaving) 2011-06-07T16:13:50 *** Accoun has joined #aichallenge 2011-06-07T16:14:20 *** stocha has joined #aichallenge 2011-06-07T16:20:33 *** Cyndre_ has quit IRC (Quit: Leaving) 2011-06-07T16:21:05 *** okayzed is now known as okay 2011-06-07T16:22:19 *** locutus2 has joined #aichallenge 2011-06-07T16:22:32 my bot either is Winner ( First rank) or looser (Last rank = java vm could not load) 2011-06-07T16:22:45 that can screw the evaluation of the ranking a bit. 2011-06-07T16:23:36 well every java bot have the problem. and i noticed that other langages were crashing too at load time. 2011-06-07T16:27:24 parsley bot seems to be strong too. that would be nice to see the two bot (stocha and parsley) playing together rather than both crashing :p 2011-06-07T16:30:11 *** davidjliu has joined #aichallenge 2011-06-07T16:30:15 it seems that only one java bot can compete. 2011-06-07T16:30:27 the others will simply crash at jvm load time. 2011-06-07T16:31:32 *** stocha has quit IRC (Quit: Page closed) 2011-06-07T16:31:55 your bot is a winner. it's just retarded by the current beta server running in a vm with low ram 2011-06-07T16:36:56 *** UncleVasya has joined #aichallenge 2011-06-07T16:40:44 *** |UncleVasya| <|UncleVasya|!kvirc@178-133-242-98.dialup.umc.net.ua> has quit IRC (Ping timeout: 260 seconds) 2011-06-07T16:41:44 *** Naktibalda has quit IRC (Quit: ChatZilla 0.9.87 [Firefox 4.0.1/20110518052719]) 2011-06-07T16:44:56 *** delt0r_ has joined #aichallenge 2011-06-07T16:45:54 contestbot: bring McLeopold here. 2011-06-07T16:45:55 UncleVasya: An error occured while trying to show the previous error. 2011-06-07T16:46:25 *** delt0r- has quit IRC (Ping timeout: 260 seconds) 2011-06-07T16:47:34 should strictly one language be used when submitting? Is it allowed e.g. to use Python + ctypes + an optimised .so library in C? 2011-06-07T16:48:03 aeter: if it works, sure 2011-06-07T16:48:06 aeter: it's not against the rules, but it may not be supported 2011-06-07T16:48:12 yes, that ^ 2011-06-07T16:48:24 aeter: it's allowed, but note that the .so library won't be built for you on the server 2011-06-07T16:48:35 aeter: there would be no way to compile your so, so you'll have to compile before upload 2011-06-07T16:48:39 so you'll need to provide the correctly built binary with your upload 2011-06-07T16:48:45 hax 2011-06-07T16:48:48 I'll have to do some guessing when building it then ;] 2011-06-07T16:48:55 thanks :) 2011-06-07T16:49:02 ubuntu natty amd64 2011-06-07T16:49:08 :) 2011-06-07T16:59:46 In my testing application I need to catch a moment when launched local game is already ended (to start another one, I'm working on multimap testing now). Can you recommend me an easy way to obtain that info? 2011-06-07T17:00:12 when the process quits? 2011-06-07T17:03:58 The moment when I already can read the replay file. 2011-06-07T17:05:26 process is python.exe and I'm not sure I can easy determine that it is a process I really need. 2011-06-07T17:06:52 For example, I determine that I already can read game result data from 0.replay file (game_logs\) by checking replay.html existense. 2011-06-07T17:07:15 well.. you're launching the python process from inside your app 2011-06-07T17:07:22 you should be able to monitor the same process 2011-06-07T17:07:28 If replay.html exists than game ended and I can deal 0.replay 2011-06-07T17:08:24 Do not laugh. I am creating a cmd-file like play_one_game.cmd but with user's settings :) 2011-06-07T17:08:44 And then launch it. 2011-06-07T17:08:55 so check when the cmd file quits 2011-06-07T17:09:04 also.. i really can't not laugh 2011-06-07T17:09:29 I'll try to check. 2011-06-07T17:10:19 I've tried to launch playgame.py directly but I found the way with cmd-file easier. 2011-06-07T17:10:47 Need to rejoin. 2011-06-07T17:14:53 bah, 7 15am 2011-06-07T17:14:56 who needs sleep anyway? 2011-06-07T17:15:09 *** UncleVasya has quit IRC (Ping timeout: 260 seconds) 2011-06-07T17:18:08 *** aerique has quit IRC (Quit: ...) 2011-06-07T17:19:31 *** berak has quit IRC (Remote host closed the connection) 2011-06-07T17:21:14 *** computerwiz_222 has quit IRC (Ping timeout: 240 seconds) 2011-06-07T17:22:35 *** amstan has quit IRC (Ping timeout: 260 seconds) 2011-06-07T17:24:48 antimatroid: you pulled an all nighter? 2011-06-07T17:27:07 *** UncleVasya has joined #aichallenge 2011-06-07T17:31:11 *** aeter has left #aichallenge 2011-06-07T17:35:44 *** olexs has joined #aichallenge 2011-06-07T17:38:15 Trying to support the streaming format with what I already have turns out to be harder than I thought. 2011-06-07T17:39:32 *** xQuasar has joined #aichallenge 2011-06-07T17:39:43 How long does it take the test server to compile and run tests on submitted bot source code? 2011-06-07T17:39:57 Mine's been "Compiling: compiling and running tests" for over 20 minutes 2011-06-07T17:40:32 0.3 seconds, for an android, that is eternity 2011-06-07T17:41:42 *** McLeopold has joined #aichallenge 2011-06-07T17:41:51 Do you use a language noone else uses xQuasar? 2011-06-07T17:41:57 Java 2011-06-07T17:42:10 Oh nevermind, it finally finished 2011-06-07T17:42:37 ah sry, its 0.68 seconds, my memory doesnt serve me well 2011-06-07T17:43:32 *** jmpespxoreax has joined #aichallenge 2011-06-07T17:44:49 mcstar: yeah :P 2011-06-07T17:44:58 * mcstar starts to think that not everyone with a computer loves start trek :( 2011-06-07T17:44:58 exam is done though 2011-06-07T17:45:03 not sure how right it is, but it's done 2011-06-07T17:45:11 antimatroid: good news 2011-06-07T17:45:14 * antimatroid hates star trek 2011-06-07T17:45:19 now go get some sleep 2011-06-07T17:45:31 nah, i have to take the exam to uni in a couple of hours 2011-06-07T17:45:40 oh, thats baaad 2011-06-07T17:45:48 meh 2011-06-07T17:45:54 i hate doing exams if im sleepy 2011-06-07T17:45:55 worst part is it's pouring outside 2011-06-07T17:45:55 *** smellyhippy has quit IRC (Ping timeout: 260 seconds) 2011-06-07T17:46:02 so i'll have to drive instead of ride 2011-06-07T17:46:07 the exam is done, it was a take home exam 2011-06-07T17:46:13 i just have to physically take it to uni :P 2011-06-07T17:46:14 McLeopold: I want to have replay.html but do not want launch it after game ends. What I need to change? 2011-06-07T17:46:18 i see 2011-06-07T17:46:43 *** ltriant has joined #aichallenge 2011-06-07T17:46:53 add --nolaunch to playgame command 2011-06-07T17:47:10 wait...you won't get the replay.html file... 2011-06-07T17:47:16 *** smellyhippy has joined #aichallenge 2011-06-07T17:47:16 *** smellyhippy has joined #aichallenge 2011-06-07T17:47:21 yes 2011-06-07T17:47:32 only 0.replay with --nolaunch 2011-06-07T17:47:36 antimatroid: do you mean all star trek? or maybe just tos? 2011-06-07T17:47:49 You could create it yourself 2011-06-07T17:48:15 i've never really watched it tbh 2011-06-07T17:48:20 don't really get down to sci fi 2011-06-07T17:48:42 No, I need it to be created. Even if I'm not gonna use it. 2011-06-07T17:48:47 ah, ok. i breathe it 2011-06-07T17:48:51 *** xQuasar has quit IRC (Ping timeout: 240 seconds) 2011-06-07T17:48:54 :P 2011-06-07T17:48:55 UncleVasya: do you even actually need replay.html or are you still just trying find a way to know when playgame finishes? 2011-06-07T17:48:57 anyway, shower time 2011-06-07T17:50:02 I do have a slightly hard time thinking that there is a language that allows you to launch other processes without being able to check if they've exited later 2011-06-07T17:50:32 UncleVasya: there are blocking function calls, return values, temporary files, process viewver whatever 2011-06-07T17:50:42 janzert: I will copy every replay.html and after multimap testing user can click "Watch replay" on every game. But yes, I'm still hope to use it as a detector. 2011-06-07T17:50:48 there are many solutions to your problem 2011-06-07T17:51:09 UncleVasya: the visualizer can load a replay file, which is a better solution 2011-06-07T17:51:58 McLeopold: ok 2011-06-07T17:53:28 mcstar: As I am a bad programmer (maybe even not a programmer at all) I have pain with dealing with all this staff. 2011-06-07T17:53:40 stuff 2011-06-07T17:53:49 we are the staff 2011-06-07T17:53:56 (well im not) 2011-06-07T17:54:04 thank you 2011-06-07T17:54:13 *** xQuasar has joined #aichallenge 2011-06-07T17:55:41 I need to find some other file-detector :) 2011-06-07T17:56:18 *** olexs has quit IRC (Quit: Leaving.) 2011-06-07T17:56:31 maybe this sounds hard for you, but take a look at bash 2011-06-07T17:56:38 try using cygwin 2011-06-07T17:57:19 i think all that you want can be done in a couple of lines in it, and later you could use it for other purposes as well 2011-06-07T17:57:24 *** xQuasar has quit IRC (Client Quit) 2011-06-07T17:57:46 i dont know cmd scripting so i wont recommend that 2011-06-07T17:58:42 I have another crazy idea: 2011-06-07T17:59:39 1. Create a backup of visualize_locally.py 2011-06-07T18:00:20 Delete these lines: if not nolaunch: webbrowser.open('file://'+os.path.realpath(generated_path)) 2011-06-07T18:00:54 just set nolauch to true 2011-06-07T18:01:12 After testing is over replace visualize_locally.py from backup. 2011-06-07T18:02:19 With --nolaunch script will not create an html file. With my solution it will create it but will not open it in the browser. 2011-06-07T18:03:00 write pass after : and put a # after that 2011-06-07T18:04:21 Just to edit TextFilele much easier to me than deal with these scary words like blocking function calls, return values, temporary files, process viewver, etc :) 2011-06-07T18:04:38 be aware that in python blocks of codes are separated with indentation 2011-06-07T18:04:57 you can easily mess up a program if you cant see the code structure 2011-06-07T18:05:13 thats why instead of deleting code use pass and partial commenting 2011-06-07T18:05:23 valuable advice 2011-06-07T18:07:24 what i would do is: create an array that holds parameters to the script i want to run, id call that script with it in a loop of some kind, then if theres an output file ill be later needing, id rename it, and basically thats all 2011-06-07T18:08:29 if you will be dealing with computers after the contest, too, then its the perfect time to learn basic scripting 2011-06-07T18:09:37 if you can understand us, i dont think ull have any trouble understanding tutorials, and i think theres literature in russian, too 2011-06-07T18:11:45 aichallenge: McLeopold epsilon * r2b8607c / (ants/playgame.py ants/visualizer/visualize_locally.py): update to playgame to control html filename - http://bit.ly/ipsY2B 2011-06-07T18:11:58 UncleVasya: add a --html to control the filename 2011-06-07T18:13:26 aichallenge: McLeopold epsilon * rff1b7bd / ants/playgame.py : update to playgame to control html filename - http://bit.ly/lnEnFh 2011-06-07T18:13:36 you can combine --nolaunch and --html to just produce files 2011-06-07T18:14:49 wow, you've made an update because of me? 2011-06-07T18:14:53 yes 2011-06-07T18:15:23 UncleVasya: everyone is equally important, didnt you know that? 2011-06-07T18:17:54 Thank you but I can't figure out why you decided to do it because my problem is only from my un-formededucation (hi, vocabulary :)). 2011-06-07T18:18:13 I'll let you know if your ideas suck :) 2011-06-07T18:19:04 *** george1914 has joined #aichallenge 2011-06-07T18:19:32 george1914: did you open the new issue on bot positions? 2011-06-07T18:19:33 McLeopold might think that other people will need the switch as well, if a somebody(you) wanted it 2011-06-07T18:20:06 yes, i did 2011-06-07T18:20:29 have you seen it before? 2011-06-07T18:21:03 something different 2011-06-07T18:21:24 we had it where the replay only had 3 name in a 4 player game, but that has been fixed 2011-06-07T18:21:41 *** Palmik has quit IRC (Remote host closed the connection) 2011-06-07T18:23:47 i have seen another game like it since #16846 - as always with my coding, there might be some weird error on my part, but my code is pretty basic atm, and i would put good money on it being some mixup ... 2011-06-07T18:24:05 how much money $) 2011-06-07T18:24:14 McLeopold: Do I need to open an issue on github about lowering skill after playing with frail bots (even if you won) ? 2011-06-07T18:24:52 UncleVasya: not yet, we are going to switch the trueskill implementation to see if it helps any 2011-06-07T18:24:59 :) i generally only risk a few hundred at a time ... 2011-06-07T18:25:35 george1914: I'm going to implement the seed numbers so we can recreate the games if needed 2011-06-07T18:26:19 sounds good - i guess if i knew the seed, i could recreate it myself also? 2011-06-07T18:26:35 if you had the other bots, and if they are deterministic 2011-06-07T18:27:01 we should be able to at least see if your bot has the same opening moves until influenced by the other bots 2011-06-07T18:27:25 well, it looks totally wrong from move 1 or 2, so other bots dont come into it 2011-06-07T18:27:48 george1914: for a quicker test, output something unique on stderr at the start of the game and upload that 2011-06-07T18:28:08 I can check the database if the special string exists 2011-06-07T18:29:50 std::cerr << "hey this is george's bot" << std::endl; for example? 2011-06-07T18:30:02 sure 2011-06-07T18:30:31 ok, i put it in now, and upload code. thanks. 2011-06-07T18:30:38 george1914: is this you? http://ai-contest.com/profile.php?user_id=11173 2011-06-07T18:39:11 yes 2011-06-07T18:39:28 this is my favorite :D http://ai-contest.com/visualizer.php?game_id=8957222 2011-06-07T18:39:45 i tried the first one also, i seem to remember analysing many of your games ... 2011-06-07T18:41:21 aye, happy memories - though i still curse my bot for it's errors ... looks like i was outsmarted there ... :) 2011-06-07T18:41:34 *** olexs has joined #aichallenge 2011-06-07T18:45:59 george1914: paste the game_id of the next game issue you spot to the github issue 2011-06-07T18:47:41 McLeopold: the new updater is running on the beta server now (when you pulled in the latest git changes earlier it got pulled in, although it still had to be built) 2011-06-07T18:47:56 trueskill? 2011-06-07T18:47:59 yes 2011-06-07T18:48:06 is it working? 2011-06-07T18:48:17 seems to be, most of the time ;) 2011-06-07T18:48:32 *** olexs has quit IRC (Quit: Leaving.) 2011-06-07T18:48:33 okay, we'll let that sit and see if we have skill decay 2011-06-07T18:48:33 on minor problem it is having is the same jvm won't start issue the java bots have 2011-06-07T18:48:42 *** okay is now known as okayzed 2011-06-07T18:48:45 ouch 2011-06-07T18:48:57 maybe I'll start a PySkills... 2011-06-07T18:49:01 hehe 2011-06-07T18:49:30 btw, it's amazing the crappy answers you can get on stackoverflow, and how they get upvoted 2011-06-07T18:49:31 it would be fine if weren't trying to run a worker in the same vm with not enough ram available 2011-06-07T18:49:44 hehe 2011-06-07T18:49:56 * janzert goes to look at the question from earlier 2011-06-07T18:50:19 I was comparing the answers on my question to the older questions. 2011-06-07T18:50:45 The first answer I got was within 5 minutes, was dead wrong, and had 5 votes. 2011-06-07T18:51:02 After about 30 mintues, the question is a tumbleweed. 2011-06-07T18:51:09 McLeopold: would it be too expensive to use the ec2 for the beta purposes? 2011-06-07T18:51:23 mcstar: not if you paid for it :) 2011-06-07T18:51:23 McLeopold: ok, i'll look at overnight games tomorrow morning (8-9 hrs from now), and shout when/if i spot one. thanks. 2011-06-07T18:51:54 sry, i dont have money right now, otherwise id donate 2011-06-07T18:52:08 mcstar: I realize the java guys are frustrated, but it's not really an issue we need to solve for beta 2011-06-07T18:52:20 there's lots of other work to do 2011-06-07T18:52:39 Right not the goal is not for participants to get games 2011-06-07T18:52:56 yep 2011-06-07T18:55:19 McLeopold: one good sign is that locutus2 submitted a copy of his bot and they've been hovering within a rank or two of each other after about 10-15 games 2011-06-07T18:55:35 that is good 2011-06-07T18:55:43 probably more accurate than a starter bot test 2011-06-07T18:55:45 I actually find it almost worrisomely good 2011-06-07T18:56:03 ;) 2011-06-07T18:56:14 too good? 2011-06-07T18:56:20 McLeopold: I added --nolaunch --html "game_logs\replay.html" but it still trying to open it in browser. 2011-06-07T18:56:53 UncleVasya: --html will assume the --log_dir directory, so make that "--html replay.html" 2011-06-07T18:57:03 McLeopold: right, almost 2011-06-07T18:58:56 UncleVasya: also, if you are doing a round robin, you may want to change the html file names to mean something 2011-06-07T18:59:53 yes, I will name them with game number. 2011-06-07T19:00:25 *** Kingpin13 has quit IRC (Quit: the coast is clear) 2011-06-07T19:00:31 Butit is still trying to launch. 2011-06-07T19:00:48 UncleVasya: did you know you can pass a game_id? 2011-06-07T19:02:05 No, I didn't. But I'll ask you about this tomorrow. Now it's time to sleep. Bye. 2011-06-07T19:02:13 *** UncleVasya has quit IRC () 2011-06-07T19:09:43 *** locutus2 has quit IRC (Ping timeout: 252 seconds) 2011-06-07T19:25:05 *** okayzed is now known as okay 2011-06-07T19:34:08 *** FireFly has quit IRC (Remote host closed the connection) 2011-06-07T19:36:32 fatal error encountered in SBCL pid 13810(tid 140737354061600): 2011-06-07T19:36:34 Heap exhausted, game over. 2011-06-07T19:36:49 my bot didnt compile :( 2011-06-07T19:38:32 *** Eruonen has quit IRC () 2011-06-07T19:38:57 *** amstan has joined #aichallenge 2011-06-07T19:38:57 *** ChanServ sets mode: +o amstan 2011-06-07T19:53:21 *** computerwiz_222 has joined #aichallenge 2011-06-07T20:14:06 *** mcstar has quit IRC (Ping timeout: 246 seconds) 2011-06-07T20:25:10 *** McLeopold has quit IRC (Read error: Connection reset by peer) 2011-06-07T20:28:54 *** computerwiz_222 has quit IRC (Remote host closed the connection) 2011-06-07T21:23:05 *** Zaphus has joined #aichallenge 2011-06-07T21:46:26 *** Zaphus has quit IRC (Quit: Bye) 2011-06-07T21:59:27 *** qacek has joined #aichallenge 2011-06-07T22:11:14 *** jugde has quit IRC (*.net *.split) 2011-06-07T22:11:14 *** CIA-26 has quit IRC (*.net *.split) 2011-06-07T22:11:15 *** MuTa-ofd00m has quit IRC (*.net *.split) 2011-06-07T22:11:17 *** acieroid has quit IRC (*.net *.split) 2011-06-07T22:11:31 *** jugde has joined #aichallenge 2011-06-07T22:11:32 *** acieroid has joined #aichallenge 2011-06-07T22:12:18 *** MuTa-ofd00m has joined #aichallenge 2011-06-07T22:12:34 *** CIA-45 has joined #aichallenge 2011-06-07T22:41:53 *** Cyndre has joined #aichallenge 2011-06-07T22:50:40 woah, apparently my bot is a little less loltastic than I thought it was