2012-03-16T00:21:34 *** antonh has quit IRC (Ping timeout: 260 seconds) 2012-03-16T00:24:11 *** antonh has joined #aichallenge 2012-03-16T00:35:21 *** thestinger has quit IRC (Quit: WeeChat 0.3.7) 2012-03-16T01:06:52 *** antimatroid has joined #aichallenge 2012-03-16T01:16:06 *** Areks|2 has joined #aichallenge 2012-03-16T01:35:52 *** mikewintermute has joined #aichallenge 2012-03-16T01:49:35 *** contestbot has joined #aichallenge 2012-03-16T01:49:36 *** ChanServ sets mode: +o contestbot 2012-03-16T02:05:15 *** superflit has joined #aichallenge 2012-03-16T02:05:41 *** Garf has joined #aichallenge 2012-03-16T02:10:01 *** superflit has quit IRC (Remote host closed the connection) 2012-03-16T02:10:18 *** superflit has joined #aichallenge 2012-03-16T02:16:38 *** kurnevsky has joined #aichallenge 2012-03-16T03:06:37 *** epicmonkey has joined #aichallenge 2012-03-16T03:07:12 *** Cyndre has joined #aichallenge 2012-03-16T03:10:28 *** sigh has joined #aichallenge 2012-03-16T03:22:20 *** epicmonkey has quit IRC (Read error: Operation timed out) 2012-03-16T03:39:55 *** X-Scale` has quit IRC (Ping timeout: 272 seconds) 2012-03-16T03:54:52 *** Jak_o_Shadows has joined #aichallenge 2012-03-16T04:35:01 *** epicmonkey has joined #aichallenge 2012-03-16T04:35:15 *** Palmik has joined #aichallenge 2012-03-16T05:15:27 *** KP13 has joined #aichallenge 2012-03-16T05:17:36 *** Kingpin13 has quit IRC (Ping timeout: 245 seconds) 2012-03-16T05:31:17 *** mceier has joined #aichallenge 2012-03-16T05:43:07 *** mikewintermute has joined #aichallenge 2012-03-16T06:03:20 *** Garf has quit IRC (Read error: Connection reset by peer) 2012-03-16T06:21:16 *** Garf has joined #aichallenge 2012-03-16T07:43:59 *** itzkow has joined #aichallenge 2012-03-16T08:02:01 *** smiley1983 has joined #aichallenge 2012-03-16T08:19:43 *** superflit_ has joined #aichallenge 2012-03-16T08:19:54 *** Scooper has joined #aichallenge 2012-03-16T08:19:59 *** Scooper is now known as xScooper 2012-03-16T08:22:39 *** Accoun has quit IRC () 2012-03-16T08:23:03 *** superflit has quit IRC (Ping timeout: 272 seconds) 2012-03-16T08:23:03 *** superflit_ is now known as superflit 2012-03-16T08:25:07 *** welterde has quit IRC (Ping timeout: 260 seconds) 2012-03-16T08:27:54 *** itzkow has quit IRC (Ping timeout: 245 seconds) 2012-03-16T08:28:48 *** Accoun has joined #aichallenge 2012-03-16T08:29:13 *** KP13 is now known as Kingpin13 2012-03-16T08:31:41 *** mikewintermute has quit IRC (Quit: mikewintermute) 2012-03-16T08:46:07 *** mviel_ has joined #aichallenge 2012-03-16T08:49:06 *** mikewintermute has joined #aichallenge 2012-03-16T08:50:17 *** mviel__ has quit IRC (Ping timeout: 272 seconds) 2012-03-16T08:50:24 *** welterde has joined #aichallenge 2012-03-16T09:00:50 *** Jak_o_Shadows has quit IRC (Remote host closed the connection) 2012-03-16T09:18:39 *** foRei has joined #aichallenge 2012-03-16T09:34:06 *** itzkow has joined #aichallenge 2012-03-16T09:50:33 *** ikaros has joined #aichallenge 2012-03-16T09:56:14 *** ikaros has quit IRC (Quit: Ex-Chat) 2012-03-16T10:03:02 *** smiley1983 has quit IRC (Ping timeout: 245 seconds) 2012-03-16T10:04:10 *** smiley1983 has joined #aichallenge 2012-03-16T10:21:29 *** Areks|2 has quit IRC (Ping timeout: 272 seconds) 2012-03-16T10:40:47 *** retre has joined #aichallenge 2012-03-16T10:41:50 *** retre has quit IRC (Client Quit) 2012-03-16T10:46:32 *** amstan has joined #aichallenge 2012-03-16T10:46:32 *** ChanServ sets mode: +o amstan 2012-03-16T10:51:19 *** Chris_0076 has joined #aichallenge 2012-03-16T10:51:34 *** sigh has quit IRC (Remote host closed the connection) 2012-03-16T10:52:58 *** amstan has quit IRC (Quit: Konversation terminated!) 2012-03-16T11:23:47 *** smiley1983 has quit IRC (Ping timeout: 255 seconds) 2012-03-16T11:24:40 *** smiley1983 has joined #aichallenge 2012-03-16T11:30:09 *** mviel__ has joined #aichallenge 2012-03-16T11:30:12 smiley1983: did you see the mspacman contest? 2012-03-16T11:33:59 *** mviel_ has quit IRC (Ping timeout: 260 seconds) 2012-03-16T11:45:06 antimatroid: no, I didn't. Was it recent? 2012-03-16T11:45:19 yep, hasn't even really started yet 2012-03-16T11:45:26 http://www.pacman-vs-ghosts.net/ 2012-03-16T11:45:31 antimatroid: congratulations on your final position in Ants :) 2012-03-16T11:45:39 cheers 2012-03-16T11:45:50 i assume you got the email asking for cv's? 2012-03-16T11:46:02 congrats to you too of course :P 2012-03-16T11:46:42 thankyou :) and yes, I sent them a resume but haven't heard back and don't really expect anything to come of it 2012-03-16T11:47:52 I've just uploaded an untested prototype for a Risk-like game: https://github.com/smiley1983/aichallenge/tree/war 2012-03-16T11:48:01 No visualizer yet 2012-03-16T11:48:33 *** thestinger has joined #aichallenge 2012-03-16T11:50:13 yeah i feel the same about you with the cv thing 2012-03-16T11:50:21 plus i just started phd in january and have no intentions of leaving that 2012-03-16T11:50:48 there needs to be a rewrite of the backend before we hold another contest 2012-03-16T11:51:11 i don't think anyone wants to finalise anything game wise until that is at least well under way 2012-03-16T11:51:18 and as far as i know no one is working on a new backend 2012-03-16T11:52:17 /s/about/as :P 2012-03-16T11:52:28 haha, yes, that sounds about right. I was thinking of putting up a TCP server with pw, ants, asteroids and risk 2012-03-16T11:52:38 AND TRON 2012-03-16T11:52:43 HOW COULD YOU FORGET TRON? 2012-03-16T11:52:46 :D 2012-03-16T11:52:48 yes, and tron. 2012-03-16T11:53:06 i am more keen to play around with tron again than anything else 2012-03-16T11:53:08 I wasn't there for tron, so it's actually the one I have the most interest in. 2012-03-16T11:53:12 although write a new game engine for it 2012-03-16T11:53:26 the old one passed the entire game board each turn i think 2012-03-16T11:53:26 smiley1983: would be a good idea to make a shiny new tcp server 2012-03-16T11:53:32 can just pass the players moves to people 2012-03-16T11:53:40 not a buggy one with resource leaks like the last 2012-03-16T11:53:47 yes, I thought a new engine might be needed 2012-03-16T11:53:48 i looked at doing that a few months ago, current one is so gross :p 2012-03-16T11:53:58 players'* 2012-03-16T11:54:00 it's late :P 2012-03-16T11:54:07 antimatroid: s/late/early/ 2012-03-16T11:54:11 you live in the wrong time zone 2012-03-16T11:54:19 I now know that writing a new game engine is non-trivial but also quite possible 2012-03-16T11:54:41 smiley1983: i did an ants prototype engine in c++ last year, that was fun 2012-03-16T11:54:48 although my processor class was way too slow 2012-03-16T11:55:18 thestinger: I think people would use it, it's just a matter of working out the technical stuff so that once the server is paid for, it can be put to work immediately 2012-03-16T11:55:28 i think a tcp server with past contests up would certainly get people interested 2012-03-16T11:55:47 making a new tcp server/client would be fun 2012-03-16T11:55:51 This is all just hypothetical planning, and may or may not happen. 2012-03-16T11:56:22 smiley1983: it would be nice to actually have it on the official site next time imo 2012-03-16T11:56:30 so starting work on it now would be great 2012-03-16T11:56:34 thestinger: i've been saying that since tron :P 2012-03-16T11:56:52 python with actual threads for each connection is a bad fit for it 2012-03-16T11:56:57 i think the pacman problem is quite interesting 2012-03-16T11:56:59 it won't scale up to all the players 2012-03-16T11:57:20 c++ all the things! 2012-03-16T11:57:23 no 2012-03-16T11:57:25 :p 2012-03-16T11:57:34 C++ is terrible for this :P 2012-03-16T11:57:47 that pretty much forces you to use threads or async with boost::asio just like python does 2012-03-16T11:58:01 erlang or go :) 2012-03-16T11:58:24 https://github.com/antimatroid/GTL/blob/master/tools/Process.h 2012-03-16T11:58:42 that's my really slow process struct which can be used to io with processes like a stream 2012-03-16T11:59:15 i dunno what makes it so slow, maybe the string stream in the input function at the bottom 2012-03-16T12:00:02 you can make a really nice multiple producer and multiple consumer channel in a few lines 2012-03-16T12:00:17 but threads just don't work out well for networking 2012-03-16T12:00:31 *** smiley1983 has quit IRC (Ping timeout: 246 seconds) 2012-03-16T12:00:54 what about what i tried to do with that process struct? but so it isn't slow as hell 2012-03-16T12:01:18 sec, I have code somewhere 2012-03-16T12:02:28 I was messing around with the "C++11" threading stuff (actually boost::thread which is the most compliant implementation of that atm) 2012-03-16T12:02:36 *** smiley1983 has joined #aichallenge 2012-03-16T12:03:22 i no longer use windows on any machines which is handy 2012-03-16T12:03:30 well i do in my office at my uni, but i don't program there 2012-03-16T12:03:46 yeah, the hacky implementations of fork() on windows are crap 2012-03-16T12:03:49 no copy-on-write 2012-03-16T12:04:05 the linux scheduler sees threads/processes as the same thing iirc, so there's only ~50% overhead 2012-03-16T12:04:23 (compared to a posix thread, which is actually not so lightweight) 2012-03-16T12:04:46 what do you mean by overhead? 2012-03-16T12:04:50 and why is there 50%? 2012-03-16T12:05:25 each posix thread has a stack 2012-03-16T12:05:40 so the overhead is whatever the stack size is 2012-03-16T12:05:57 Are the channel logs still available? The link to contestbot.aichallenge.org no longer seems to contain them 2012-03-16T12:06:08 -s: stack size (kbytes) 8192 2012-03-16T12:06:11 default linux stack size 2012-03-16T12:06:17 ulimit -s to check 2012-03-16T12:06:22 so 8MB overhead per thread 2012-03-16T12:06:26 you can't just make 1 per client :P 2012-03-16T12:06:45 antimatroid: threads share the same address space, processes each have their own virtual memory 2012-03-16T12:06:56 linux (and every other *nix) cheats and does copy-on-write 2012-03-16T12:07:32 hmm okay 2012-03-16T12:07:37 that vaguely made sense :P 2012-03-16T12:07:41 windows doesn't, so multiprocessing on windows doesn't really work 2012-03-16T12:07:44 i have a very poor understanding of things at that level 2012-03-16T12:08:12 Go, Erlang, Stackless Python, etc. give you lightweight threads 2012-03-16T12:08:22 they don't use the C stack, so threads have basically zero overhead 2012-03-16T12:08:40 you can make 200k of them and the memory usage will be a few kilobytes for each until they actually do something 2012-03-16T12:09:20 and they map them to posix threads 2012-03-16T12:09:33 that would make life easier 2012-03-16T12:09:39 so you might have 4 real POSIX threads (1 or 2 for each core) and as many threads as you want 2012-03-16T12:09:42 *** mikewintermute has quit IRC (Quit: mikewintermute) 2012-03-16T12:10:15 *** mikewintermute has joined #aichallenge 2012-03-16T12:10:46 they use their own scheduler to deal with their own lightweight threads 2012-03-16T12:10:56 if I/O blocks in one, it does a context switch 2012-03-16T12:11:21 so you get all the benefits of painful asynchronous programming (twisted, node.js, etc.) 2012-03-16T12:12:18 yeah i'm fairly lost :) 2012-03-16T12:12:24 but thanks for the explanation 2012-03-16T12:12:28 :P 2012-03-16T12:12:57 google released this recently https://code.google.com/p/vitess/ 2012-03-16T12:13:02 have any of you every implemented the word problem? 2012-03-16T12:13:37 or more specifically, to generate a dictionary from an alphabet with given relations on words? 2012-03-16T12:13:59 * thestinger doesn't quite understand 2012-03-16T12:14:01 like markov chains? 2012-03-16T12:14:22 ie. has to be valid relations on the words to get a finite dictionary (invalid ones are possible) 2012-03-16T12:14:57 *** mikewintermute has quit IRC (Ping timeout: 260 seconds) 2012-03-16T12:15:24 not like markov chains 2012-03-16T12:15:31 it's basically generating a semigroup from a presentation 2012-03-16T12:15:56 *** smiley1983 has quit IRC (Ping timeout: 244 seconds) 2012-03-16T12:16:03 i think i have a way to do it, but i'm not entirely sure 2012-03-16T12:16:52 https://en.wikipedia.org/wiki/Word_problem_for_groups this? 2012-03-16T12:17:02 *** smiley1983 has joined #aichallenge 2012-03-16T12:17:28 yeah 2012-03-16T12:17:44 although only for semigroups 2012-03-16T12:24:45 thestinger: http://pastebin.com/EFYJvD2u 2012-03-16T12:24:54 that describes how i'm trying to get at the problem 2012-03-16T12:25:00 but i'm a little lost at the end 2012-03-16T12:25:21 1. will i need* to iteratively do this *** 2012-03-16T12:31:32 *** aerique has quit IRC (Remote host closed the connection) 2012-03-16T12:32:11 it's surprisingly hard 2012-03-16T12:32:38 I got distracted by making a package for something 2012-03-16T12:32:44 * thestinger fails at doing 20 things at once 2012-03-16T12:35:08 http://pastebin.com/cES5DZcu 2012-03-16T12:35:11 that's by no means finished 2012-03-16T12:35:20 but that's what i started doing the other day in the middle of the night 2012-03-16T12:36:30 i wish there was a combo break call 2012-03-16T12:36:44 for when i want to break out of more than one loop 2012-03-16T12:37:33 that == operator on words could probably be removed too, you'd want to use something else since i just typedefed words as strings 2012-03-16T12:55:19 I think I'm done packaging that thing 2012-03-16T12:55:36 now to fix up this multiple producer-consumer channel thing 2012-03-16T13:00:16 lol, I forget how to do initializer lists 2012-03-16T13:00:32 the C++11 thing 2012-03-16T13:06:05 antimatroid: a quick example of why clang is awesome: 2012-03-16T13:06:09 http://sprunge.us/XbQj g++ 2012-03-16T13:06:15 http://sprunge.us/ALCg clang++ 2012-03-16T13:07:11 *** kilae has joined #aichallenge 2012-03-16T13:18:28 *** epicmonkey has quit IRC (Ping timeout: 245 seconds) 2012-03-16T13:19:11 *** mviel__ has quit IRC (Quit: Leaving) 2012-03-16T13:23:25 antimatroid: now... exception safety 2012-03-16T13:23:31 I always fail at this 2012-03-16T13:25:47 *** iglo has joined #aichallenge 2012-03-16T13:26:53 heh, boost has a Maybe monad 2012-03-16T13:42:26 antimatroid: there, it works 2012-03-16T13:42:34 no race conditions and I think minimal locking 2012-03-16T13:42:49 and if I understand C++ correctly (I don't) it's exception safe 2012-03-16T13:54:38 antimatroid: https://gist.github.com/cfea7020b44ba4a12e9f 2012-03-16T14:02:09 *** g0llum has joined #aichallenge 2012-03-16T14:05:02 *** smiley1983 has quit IRC (Ping timeout: 245 seconds) 2012-03-16T14:05:57 *** smiley1983 has joined #aichallenge 2012-03-16T14:20:59 *** iglo has quit IRC (Quit: .) 2012-03-16T14:21:31 *** smiley1983 has quit IRC (Quit: leaving) 2012-03-16T14:35:11 *** mikewintermute has joined #aichallenge 2012-03-16T14:37:45 *** epicmonkey has joined #aichallenge 2012-03-16T14:59:35 *** kilae has quit IRC (Quit: ChatZilla 0.9.88.1 [Firefox 10.0.2/20120215223356]) 2012-03-16T15:28:22 *** jcdny has joined #aichallenge 2012-03-16T15:30:43 *** Accoun has quit IRC () 2012-03-16T15:30:53 *** jcdjcd has quit IRC (Ping timeout: 245 seconds) 2012-03-16T15:38:49 *** iglo has joined #aichallenge 2012-03-16T15:40:02 *** rwest_ has joined #aichallenge 2012-03-16T15:44:59 *** amstan has joined #aichallenge 2012-03-16T15:45:00 *** ChanServ sets mode: +o amstan 2012-03-16T15:46:52 *** rwest_ has left #aichallenge 2012-03-16T15:47:17 *** rwest has quit IRC (Disconnected by services) 2012-03-16T15:48:27 *** rwest has joined #aichallenge 2012-03-16T15:52:16 *** mikewintermute has quit IRC (Quit: mikewintermute) 2012-03-16T15:53:22 *** mikewintermute has joined #aichallenge 2012-03-16T15:54:30 *** mikewintermute has quit IRC (Client Quit) 2012-03-16T16:00:21 *** Accoun has joined #aichallenge 2012-03-16T16:05:15 hmm, guess you can't make a gist public once it's pasted 2012-03-16T16:21:52 *** epicmonkey has quit IRC (Ping timeout: 252 seconds) 2012-03-16T16:58:52 *** kurnevsky has quit IRC (Quit: Leaving.) 2012-03-16T17:17:22 *** g0llum has quit IRC (Read error: Connection reset by peer) 2012-03-16T17:28:09 *** sevikkk has quit IRC (Ping timeout: 246 seconds) 2012-03-16T17:38:12 *** Apophis has quit IRC (Read error: Connection reset by peer) 2012-03-16T17:44:22 *** sevikkk has joined #aichallenge 2012-03-16T17:55:37 *** GeorgeH has joined #aichallenge 2012-03-16T18:06:31 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has joined #aichallenge 2012-03-16T18:06:55 <_flag> @seen j3camero 2012-03-16T18:06:55 _flag: j3camero was last seen in #aichallenge 12 weeks, 0 days, 20 hours, 4 minutes, and 31 seconds ago: Some of the game ideas on the forums are awesome. 2012-03-16T18:12:30 *** amstan has quit IRC (Quit: Konversation terminated!) 2012-03-16T18:17:30 *** Palmik has quit IRC () 2012-03-16T18:18:48 i definitely like the idea that contestants can choose which side to program for (or both) 2012-03-16T18:18:54 from the pacman contest 2012-03-16T18:21:02 that's neat 2012-03-16T18:21:08 how does ranking work though? 2012-03-16T18:23:45 separate rankings 2012-03-16T18:24:05 *** X-Scale` has joined #aichallenge 2012-03-16T18:25:04 hm, so it could actually be set up to let people compete in both 2012-03-16T18:28:30 yes 2012-03-16T18:32:46 <_flag> Might be strange then to have "two winners" or something like that 2012-03-16T18:33:33 well, you have a PacMan winner and a Ghosts winner, in their case 2012-03-16T19:20:27 *** itzkow has quit IRC (Ping timeout: 245 seconds) 2012-03-16T19:21:36 *** amstan has joined #aichallenge 2012-03-16T19:21:37 *** ChanServ sets mode: +o amstan 2012-03-16T19:28:39 *** Chris_0076 has quit IRC (Quit: Leaving) 2012-03-16T20:06:14 *** antimatroid has quit IRC (*.net *.split) 2012-03-16T20:06:14 *** janzert has quit IRC (*.net *.split) 2012-03-16T20:06:14 *** _dr <_dr!~johannes@faui32o.informatik.uni-erlangen.de> has quit IRC (*.net *.split) 2012-03-16T20:07:38 *** antimatroid has joined #aichallenge 2012-03-16T20:07:38 *** janzert has joined #aichallenge 2012-03-16T20:07:38 *** _dr <_dr!~johannes@faui32o.informatik.uni-erlangen.de> has joined #aichallenge 2012-03-16T20:20:37 *** foRei has quit IRC (Read error: Connection reset by peer) 2012-03-16T20:29:32 *** treeform_ has quit IRC (Remote host closed the connection) 2012-03-16T20:44:36 *** mceier has quit IRC (Quit: leaving) 2012-03-16T20:51:03 *** amstan has quit IRC (Read error: No route to host) 2012-03-16T20:51:41 *** amstan has joined #aichallenge 2012-03-16T20:51:42 *** ChanServ sets mode: +o amstan 2012-03-16T21:12:42 *** mleise has joined #aichallenge 2012-03-16T21:44:08 *** sigh has joined #aichallenge 2012-03-16T21:44:57 *** xScooper has quit IRC (Quit: Leaving) 2012-03-16T22:00:42 *** iglo_ has joined #aichallenge 2012-03-16T22:03:57 *** iglo has quit IRC (Ping timeout: 244 seconds) 2012-03-16T22:12:58 *** ronchaine has joined #aichallenge 2012-03-16T23:57:30 *** amstan has quit IRC (Quit: Konversation terminated!)