2011-04-30T00:01:30 mcstar: you could have given the script :P 2011-04-30T00:02:56 *** Cerealklr has left #aichallenge 2011-04-30T00:03:01 Well I understand perfectly 'how' to do it, but yes I understand what you mean. I'll tell you what though, I'm gonna do it anyway, just because so far I've never really dealt with files from python, only in bash, so I'll learn some things in the process of it 2011-04-30T00:03:19 *** sids_aquarius has joined #aichallenge 2011-04-30T00:03:28 I do understand though that it won't be used by other people and in fact I won't even likely use it 2011-04-30T00:03:38 just for the sake of my education 2011-04-30T00:04:43 dealing with files in c++ is nice 2011-04-30T00:04:51 just use ifstreams 2011-04-30T00:05:07 from what I've seen python has been quite nice 2011-04-30T00:05:27 but yes, ifstreams are nice 2011-04-30T00:05:50 i wish c++ had a type system like haskell 2011-04-30T00:06:15 and treated functions like variables, and had operator precedence overloading (and any operators for that matter) 2011-04-30T00:06:20 then c++ would be almost perfect for me 2011-04-30T00:07:34 The reason I wanted to experiment with this was because of situations like when you want to run a local game and you need a map with a certain number of bots to test or wanted to test on a particular map or with particular bots 2011-04-30T00:07:53 I basically wanted to be able to put in what I cared about and have the rest filled in 2011-04-30T00:08:04 or find the best match for that input 2011-04-30T00:17:25 so we have a whole bunch of 4 or 5 player maps, 2 or 3 for each of 7, 8, and 10, and then none of anything else 2011-04-30T00:17:36 is there a reason for that odd distribution? 2011-04-30T00:17:48 animatroid: this would be directed at you 2011-04-30T00:22:49 jaycobcoleman: i imagine a lower number of players is more likely to arise 2011-04-30T00:23:05 i'm not against that happening, but we can tweak the map generator later if necessary 2011-04-30T00:23:15 look at the map generator too 2011-04-30T00:23:23 i may have capped maxPlayers at 10 2011-04-30T00:23:46 It looks as though you might've :/ I'll take a look 2011-04-30T00:23:57 they are just sample maps 2011-04-30T00:24:06 there will be a LOT generator for the server 2011-04-30T00:24:19 yeah I know, and yes, 4-10 2011-04-30T00:25:17 Will new maps be generated specific for each map or are we just going to have a large database to use? 2011-04-30T00:25:25 *match 2011-04-30T00:28:59 i'm in favour of a new map every game 2011-04-30T00:29:03 but not sure that'll happen 2011-04-30T00:29:25 imho the map gen should have been included in the game engine 2011-04-30T00:29:26 why is that? lack of trust in the generator? 2011-04-30T00:29:38 that way it's easy make the map generator dynamic enough 2011-04-30T00:29:46 it isn't too late 2011-04-30T00:29:52 amstan: the map generator is written like a class 2011-04-30T00:29:56 well, maybe 2011-04-30T00:29:58 you can just include it in the engine 2011-04-30T00:30:03 hoho? 2011-04-30T00:30:04 antimatroid: yes, except food depends on the map type 2011-04-30T00:30:22 yes, we're planning to have the map type as a parameter in maps 2011-04-30T00:30:32 then the engine can use that information to select the food spawning method 2011-04-30T00:31:15 but if you had the map gen built in the engine, you wouldn't have to worry about map formats or parameters for maps for that matter 2011-04-30T00:32:11 @seen mcleopold 2011-04-30T00:32:11 amstan: mcleopold was last seen in #aichallenge 4 hours, 54 minutes, and 37 seconds ago: antimatroid: because 12 > 11. It should be 0pm. 2011-04-30T00:33:28 antimatroid: how completed is the engine? do we have anything left in that regard? 2011-04-30T00:34:42 amstan: i think it's all good except reading the symmetry type 2011-04-30T00:34:57 sigh_ would be better to ask 2011-04-30T00:35:48 engine is fine, unless the specs change :P 2011-04-30T00:36:30 *** sids_aquarius has quit IRC (Ping timeout: 246 seconds) 2011-04-30T00:36:43 i'm still a little fuzzy on what's left till we can release a beta 2011-04-30T00:36:59 i know mcleopold's been working on rankings for a while 2011-04-30T00:37:16 tcp server, picking games and ranking them with the stuff? dunno otherwise 2011-04-30T00:37:24 antimatroid: we don't care about tcp for beta 2011-04-30T00:37:34 @seen janzert 2011-04-30T00:37:34 amstan: janzert was last seen in #aichallenge 3 hours, 30 minutes, and 34 seconds ago: yep :) 2011-04-30T00:37:36 i haven't been following the progress much either 2011-04-30T00:37:52 i want mcstar's script 2011-04-30T00:37:53 janzert: how's the sandbox state? 2011-04-30T00:37:58 antimatroid: script? 2011-04-30T00:38:14 (13:20:34) mcstar: i wrote a bash script so that the visualizer can be used conveniently 2011-04-30T00:38:15 (13:20:42) mcstar: no copy-paste 2011-04-30T00:38:15 (13:21:11) mcstar: u just have to give it the replay filename 2011-04-30T00:38:36 antimatroid: it's not that hard.. the php file already does this 2011-04-30T00:38:45 I have one possible but not likely issue with the engine 2011-04-30T00:38:52 um... just use visualizer/visualize_locally.py 2011-04-30T00:38:55 then why aren't we just giving people that? 2011-04-30T00:39:01 yes.. or what sigh said 2011-04-30T00:39:15 the get_perspective and get_map_output functions are designed in such a way that perspective and map output can be found from the point of view of a specific player 2011-04-30T00:39:37 perhaps the local visualizer could have a "reload" button 2011-04-30T00:39:44 McLeopold and I discussed a change that could cause problems if those functions were used in that manner 2011-04-30T00:39:49 so you don't have to keep refreshing the page to get the new game? 2011-04-30T00:40:04 jaycobcoleman: what change? 2011-04-30T00:40:38 Displaying UNSEEN_FOOD on the map for food the player has seen, has not yet received an 'r' line, and is not currently visible 2011-04-30T00:40:42 sigh_: do you know how to setup the pipeline for the manager part? 2011-04-30T00:40:50 sigh_: the cron scripts i mean 2011-04-30T00:40:55 amstan: no... 2011-04-30T00:41:00 sigh.. 2011-04-30T00:41:10 :( 2011-04-30T00:41:32 what? 2011-04-30T00:41:37 jaycobcoleman: that seem really bad 2011-04-30T00:41:43 what does? 2011-04-30T00:41:52 displaying unseen food 2011-04-30T00:41:54 contestbot: later tell mcleopold Do you have time on sunday to code on the contest? I'm very eager to start the beta test server. 2011-04-30T00:41:54 amstan: Yes master! 2011-04-30T00:42:06 why is that? 2011-04-30T00:42:08 yeah, the idea of a fog of war is that you don't have information except for squares you can see 2011-04-30T00:42:13 that is information that is not guarenteed to be accurate 2011-04-30T00:42:19 * amstan signs off 2011-04-30T00:42:26 yes which is why it's made distinct from food 2011-04-30T00:42:30 actually, i understand why they want to do it 2011-04-30T00:42:41 but then you make it hard for people to ignore not visible food 2011-04-30T00:42:48 we now have 3 food functions 2011-04-30T00:42:49 i'm with sigh_, bad idea 2011-04-30T00:42:50 food() 2011-04-30T00:42:57 why are we adding all these things that should be part of the bot? 2011-04-30T00:43:01 food_unseen() 2011-04-30T00:43:07 and food_visible( 2011-04-30T00:43:14 they are a part of the bot 2011-04-30T00:43:22 ? 2011-04-30T00:43:29 this is just a bad idea full stop 2011-04-30T00:43:31 *** Cerealklr has joined #aichallenge 2011-04-30T00:43:32 but get_perspective is created to render the map from the perspective of a player 2011-04-30T00:43:37 *** Cerealklr has left #aichallenge 2011-04-30T00:43:46 it isn't accurate as it is now 2011-04-30T00:43:47 why don't we send the bots suggested moves while we are at it 2011-04-30T00:44:03 i would be in favour of translating each players information so it was the same at the start 2011-04-30T00:44:05 I don't think you understand/// 2011-04-30T00:44:09 *... 2011-04-30T00:44:41 The only change made was bot specific 2011-04-30T00:44:54 antimatroid: what's the benefit of that 2011-04-30T00:45:13 *** Cerealklr has joined #aichallenge 2011-04-30T00:45:17 *** Cerealklr has left #aichallenge 2011-04-30T00:45:37 sigh_: it means symmetric games stay symmetric with more complicated bots :P 2011-04-30T00:45:59 atm, the order in which to consider locations (from wrapping) can cause differences in the paths ants choose 2011-04-30T00:46:22 meh, that's not a problem 2011-04-30T00:46:32 i'm half kidding, i haven't actually requested that, i'm just stating i would be in favour of the change, i'm not requesting it :P 2011-04-30T00:47:23 *** Mathnerd314 has quit IRC (Read error: Connection reset by peer) 2011-04-30T00:47:54 jaycobcoleman: I really don't understand what you want then 2011-04-30T00:48:14 they want it to be easy to give a vector of food locations yeah? 2011-04-30T00:48:28 so implement that in the bot 2011-04-30T00:48:30 I don't want anything. A change was made that doesn't cause any problems for the engine unless we decide that we for some reason want to display the map from a particular player's perspective, which we don't currently do, but have the framework to do 2011-04-30T00:48:32 it would be more complicated to implement 2011-04-30T00:49:11 what more complicated than what? 2011-04-30T00:49:12 i still argue my 'r' line is justified, information about visible squares "state" should not be implicit 2011-04-30T00:49:17 why can't you display the map from a particular player's perspective? 2011-04-30T00:49:21 nevermind what i said about that 2011-04-30T00:50:08 antimatroid: imo, the entire thing could have been side-stepped by passing the state as a character-array 2011-04-30T00:50:09 sigh_: we'd have to make a bunch of changes to the engine to incorporate UNSEEN_FOOD if we wanted to do that 2011-04-30T00:50:27 but there's absolutely no reason to at the moment 2011-04-30T00:50:35 why do you need unseen food? 2011-04-30T00:50:41 sigh_: i fought for the character array and lost 2011-04-30T00:51:17 It was just something implemented to go along with 'r' lines 2011-04-30T00:51:25 why? 2011-04-30T00:51:37 because if we're going to remember them in lists we might as well remember them on the map 2011-04-30T00:51:49 jaycobcoleman: you are then potentially giving false information to bots 2011-04-30T00:51:55 so why can't the bots do that? 2011-04-30T00:52:08 and yes, now you are passing heuristic information 2011-04-30T00:52:11 sigh_: the bots do 2011-04-30T00:52:21 and false information how? 2011-04-30T00:52:23 so then why does the engine need to tell them> 2011-04-30T00:52:24 ? 2011-04-30T00:52:29 sigh: it isn't 2011-04-30T00:52:35 because the square may not currently contain food 2011-04-30T00:52:37 what isn't? 2011-04-30T00:52:44 that's what I meant when I said I don't think you understand 2011-04-30T00:52:57 atm information is only passed for a visible square 2011-04-30T00:52:58 you are not being very clear 2011-04-30T00:53:05 does this require an engine change or not? 2011-04-30T00:53:09 it also makes explaining stuff in the specs more complicated 2011-04-30T00:53:18 animatroid: UNSEEN_FOOD will not overwrite any new information and should clearly be prioritized beneath seen food 2011-04-30T00:53:39 sigh: no 2011-04-30T00:53:51 i'm as confused as sigh now 2011-04-30T00:53:59 not unless you want to display perspective from a player 2011-04-30T00:54:02 we don't do that 2011-04-30T00:54:14 but the function get_perspective is setup so that we can 2011-04-30T00:54:25 ok, you are being extremely unclear 2011-04-30T00:54:39 start over 2011-04-30T00:54:41 haha, it's hard to address all these things so quickly 2011-04-30T00:55:09 https://github.com/aichallenge/aichallenge/pull/92/files 2011-04-30T00:55:17 that's the change 2011-04-30T00:55:32 I talked about it with McLeopold and we decided it would be sompler 2011-04-30T00:55:36 *simpler 2011-04-30T00:55:54 simpler than what? 2011-04-30T00:56:16 'r' lines mean that a player is now remembering all food, which can lead to them not distinguishing between them 2011-04-30T00:56:24 jaycobcoleman: don't say things that rely on so much implicit context 2011-04-30T00:56:46 *shrug* I don't understand 2011-04-30T00:56:49 food that possibly isn't there will be weighted as heavily as visible food 2011-04-30T00:57:40 simpler from a programming stand-point than having to check each food_square to see if it is visible or not 2011-04-30T00:58:00 and simpler to make sure you're distinguishing between the two 2011-04-30T00:58:59 I give up, after all this I don't know what problem you are trying to solve, what this has to do with the engine, why you mentioned player perspectives 2011-04-30T00:59:08 or anything about it really 2011-04-30T00:59:54 are you aware of the get_perspective function? 2011-04-30T01:00:35 and get_map_output? 2011-04-30T01:01:32 ok... look 2011-04-30T01:02:00 when I read you question, I know what functions you are talking about, since I've worked on the engine 2011-04-30T01:02:15 but, I don't know if they are other functions in the code with the same name 2011-04-30T01:02:45 so I have to search the code and see if there are any other get_perspective functions in the code base 2011-04-30T01:03:15 ants.py 2011-04-30T01:03:21 this is what I'm referring to 2011-04-30T01:03:21 luckily there aren't so I can say that, yes, I know those functions, but there would be trouble if this was not the case 2011-04-30T01:03:31 you see why it's hard to follow what you are saying? 2011-04-30T01:03:40 absolutely 2011-04-30T01:03:45 yes, I know the functions 2011-04-30T01:03:57 here it is in one sentence: 2011-04-30T01:04:14 get_map_output combines all the data from each player perspective. If player_perspective were to have a use outside of this function, it would need to include unseen food, which is the only data type we would want displayed in get_map_output as it is ultimately used for replay data 2011-04-30T01:04:16 or two 2011-04-30T01:05:12 get_perspective has potential for other uses, I'm just asserting that with this change it can't be adapted without giving us bad replay information 2011-04-30T01:05:13 why would it need to include unseen_food 2011-04-30T01:06:26 what other uses? 2011-04-30T01:06:50 I don't know why it would. really this issue has been blown way out of proportion as it's more of a cautionary statement that get_perspective or get_map_output should not be used outside getting information for replays 2011-04-30T01:07:20 say somebody wanted to review what their own bots saw 2011-04-30T01:07:38 altering those functions as you appear to be able to do would give you bad replay information 2011-04-30T01:07:44 see... now I don't know what you are talking about again 2011-04-30T01:07:50 you are leaving sooooo much out 2011-04-30T01:08:02 haha here 2011-04-30T01:08:04 how about this 2011-04-30T01:08:20 remove the player parameter from get_map_output 2011-04-30T01:08:56 that'll end my worries 2011-04-30T01:09:14 how will that help? 2011-04-30T01:09:56 get_map_output is never used with the player parameter 2011-04-30T01:09:59 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has joined #aichallenge 2011-04-30T01:10:05 McLeopold seemed pro implementing UNSEEN_FOOD in the starters 2011-04-30T01:10:29 what's that got to do with the engine? 2011-04-30T01:11:06 so getting map output from a player's perspective won't reflect what they really see because the engine doesn't currently handle UNSEEN_FOOD 2011-04-30T01:11:30 remove the player parameter because it isn't used and won't be accurate if that change is made 2011-04-30T01:11:40 ... 2011-04-30T01:11:56 I don't know if you are talking about the engine or the bot 2011-04-30T01:12:01 engine 2011-04-30T01:12:16 it's basically getting rid of temptation 2011-04-30T01:12:29 removing the player parameter will not make a single bit of difference to the engine 2011-04-30T01:12:58 but how will removing it help the bot? 2011-04-30T01:13:41 whatever bot wants to do should have NOTHING to do with the engine 2011-04-30T01:13:54 the bot has all the information it needs to about the game 2011-04-30T01:14:05 It will have no effect on the engine or the bot, it'll have the effect of ensuring that people working on the engine do not use it to render info from a player's perspective 2011-04-30T01:14:15 it's a useless parameter 2011-04-30T01:14:25 I want to make sure nobody tries to use it 2011-04-30T01:14:29 it would be bad 2011-04-30T01:14:36 that's all 2011-04-30T01:15:29 get_map_output should only be used without parameters for replay data. The structure of the function should reflect this so that people know this is its purpose and only purpose 2011-04-30T01:15:30 wait hang on, don't remove the player parameter before the scores 2011-04-30T01:15:39 if you're not talking about that, carry on 2011-04-30T01:15:51 nope :) 2011-04-30T01:16:34 *shrug* I don't really see the problem, if get_map_output is called with a player, then it correctly renders what the map looks like from their perspective 2011-04-30T01:16:44 yes, for now 2011-04-30T01:16:54 it won't if UNSEEN_FOOD is implemented 2011-04-30T01:17:26 You see? that's the only potential issue :/ 2011-04-30T01:17:30 you said unseed food was implemented in the starter bot 2011-04-30T01:18:18 yup, but if it's implemented in the starter bot and not in the engine then the engine will think the bot sees one thing, when the bot is actually seeing that in addition to UNSEEN_FOOD 2011-04-30T01:18:32 that's not the engine's problem 2011-04-30T01:19:10 agreed! 2011-04-30T01:19:24 I hope you understand that I'm just replying to everything you say 2011-04-30T01:19:31 this was never a big issue 2011-04-30T01:19:35 heh 2011-04-30T01:19:39 or really an issue at all 2011-04-30T01:19:46 which I said from the beginning 2011-04-30T01:20:00 on the other hand, putting unseen food in the starter pack has the potential to confuse people 2011-04-30T01:20:22 but I'll let you guys deal with that 2011-04-30T01:21:20 *** orivej has quit IRC (Ping timeout: 258 seconds) 2011-04-30T01:23:32 That was a fun conversation. Why might unseen food be confusing? 2011-04-30T01:24:25 because it's a heuristic... the food might not actually be there 2011-04-30T01:25:55 if we didn't add unseen food, then most people would likely consider it on the same level as visible food, because they're both a part of the food function 2011-04-30T01:26:06 which would be much worse 2011-04-30T01:26:42 a person can choose to ignore the heuristic altogether by using only food_visible() 2011-04-30T01:27:23 fair enough... the other option is to just not care about unseen food in the starter 2011-04-30T01:28:12 *** davidd has quit IRC (Read error: Operation timed out) 2011-04-30T01:28:16 well that's the negative result of 'r' lines. They encourage people to track both seen and unseen food, and unless we make it part of the starter they most likely won't make a distinction 2011-04-30T01:28:50 unseen food? like, have the engine tell the players how much food is out there that they can't see? 2011-04-30T01:28:52 'r' lines don't force people to track food anymore than 'd' lines force them to track ants 2011-04-30T01:29:15 a1k0n: have them remember food they have seen, but isn't visible 2011-04-30T01:29:17 a1k0n: no, just a heuristic for bots to remember food they've seen 2011-04-30T01:29:29 oh. that's up to the bot then. 2011-04-30T01:29:36 so this is a starter pack thing. 2011-04-30T01:29:40 yes 2011-04-30T01:29:44 i c. 2011-04-30T01:30:00 you can name it something more clear like maybe_food() to imply that it is a heuristic 2011-04-30T01:30:07 i am in favor of not caring about unseen food in the starter 2011-04-30T01:30:12 sigh: I'm all for that 2011-04-30T01:30:31 * sigh_ agrees with a1k0n 2011-04-30T01:30:57 what are you suggesting, just ignoring 'r' lines altogether? 2011-04-30T01:31:19 and resetting food_lists every turn 2011-04-30T01:31:22 ? 2011-04-30T01:31:26 I don't see why that's a problem 2011-04-30T01:31:29 i am clearly not up to date on that. the engine tells you about food you can't see? 2011-04-30T01:31:39 a1k0n: no it doesn't 2011-04-30T01:32:04 so, no, then just use the information on the 'r' lines, and pretend no other food exists. this bot isn't supposed to be *good* is it? 2011-04-30T01:32:35 ah... 'f' is for food... 'r' is more complicated 2011-04-30T01:32:41 oh 2011-04-30T01:32:43 r row col # previously seen food no longer at location 2011-04-30T01:32:44 I don't like it too much, because it is annoying to explain :P 2011-04-30T01:32:53 that means someone took the food 2011-04-30T01:32:56 if we don't deal with 'r' lines in the starter then it'll be a huge pain for people to figure out how themselves 2011-04-30T01:33:19 wait a sec, so you know when someone else eats food you saw previously but can't necessarily see now? 2011-04-30T01:33:19 jaycobcoleman: the point is they don't need to! 2011-04-30T01:33:38 many will want to, and they'll have no example to follow 2011-04-30T01:33:43 jaycobcoleman: 'r' doesn't tell them anything they could not have figured out by remembering state 2011-04-30T01:33:44 plus you get the same 'f' squares each frame if none of your ants move, right? 2011-04-30T01:34:10 *** davidd has joined #aichallenge 2011-04-30T01:34:14 the documentation on 'r' is unclear 2011-04-30T01:34:23 it's new :/ 2011-04-30T01:34:32 *** sigh has joined #aichallenge 2011-04-30T01:34:53 a1k0n: what documentation are you looking at? 2011-04-30T01:35:13 https://github.com/aichallenge/aichallenge/wiki/Ants-Game-Specification and https://github.com/aichallenge/aichallenge/wiki/Ant-bot-input-output 2011-04-30T01:35:15 I believe they're mentioned a couple different places 2011-04-30T01:35:33 okay input-output I think is pretty unclear 2011-04-30T01:35:34 a1k0n: yeah, you can't really explain 'r' consisely 2011-04-30T01:35:56 but it does not tell the bot anything that the bot couldn't figure out 2011-04-30T01:35:59 because the example doesn't really show 'r' lines 2011-04-30T01:36:12 (part of the reason I don't like it too much) 2011-04-30T01:36:15 that depends -- does the engine emit 'r' lines for food you saw previously, then moved away from and can no longer see, and then some other ant ate it? 2011-04-30T01:36:26 cuz that's magical information you couldn't get elsewise 2011-04-30T01:36:32 no 2011-04-30T01:36:34 a1k0n: only when that sqaure is visible again 2011-04-30T01:36:43 a1k0n: i am the reason for the 'r' lines, my reasoning being that otherwise information about the "state" of a visible square was implicit in some cases 2011-04-30T01:36:56 oh. so it's basically the absense of 'f' lines for food you think should exist 2011-04-30T01:36:57 and it was rather messy to calculate the information 2011-04-30T01:37:03 a1k0n: right 2011-04-30T01:37:35 I'm actually in favor or removing it given how annoying it was to explain 2011-04-30T01:37:35 not really. if you track state of a square, it would be invisible|food, and then when you observe it it would be visible|no food 2011-04-30T01:38:01 no need for 'r' to tell you of the transition 2011-04-30T01:38:05 except then you need to track what is old and new food information after processing input, then go through and delete food if it's old information 2011-04-30T01:38:06 *** sigh_ has quit IRC (Ping timeout: 246 seconds) 2011-04-30T01:38:25 a1k0n: right, it is not telling the bot anything new 2011-04-30T01:38:38 does the starter pack compute a visibility mask over the map? 2011-04-30T01:38:45 it can, yes 2011-04-30T01:38:55 no, it's making implicit information about the "state" of a visible square explicit 2011-04-30T01:39:44 *** sigh_ has joined #aichallenge 2011-04-30T01:40:17 yeah. so one approach is to have the visibility tracked by the engine and the client dumb and the server notifies about removal of food, or the client could be smarter and compute a visibility mask and then traverse it to update food states *iff* it really cares about food-removal transitions, and isn't just directing its ants to visible food every frame. 2011-04-30T01:40:47 but how are you going to mark food information as old? 2011-04-30T01:40:54 currently i just have bool isFood 2011-04-30T01:40:59 that's the bot's problem 2011-04-30T01:41:31 it's messy to handle and i don't like having implicit information about the state of a visible square 2011-04-30T01:41:41 that's my argument, if people don't think that's important, then go and remove it :P 2011-04-30T01:41:43 yeah, having the server figure that out is more trouble than it's worth 2011-04-30T01:41:56 it's not messy to handle, it's more messy to handle it in the server 2011-04-30T01:41:56 :P 2011-04-30T01:42:05 well it's all implemented already, and it's quite clean 2011-04-30T01:42:15 that's not the problem 2011-04-30T01:42:17 well, any bot will pretty much have to know what it can see and what it can't, and that's defined implicitly by the position of the ants and the radii in the header 2011-04-30T01:42:22 the problem that it is a pain to explain 2011-04-30T01:42:33 look how long it took to tell a1k0n about it 2011-04-30T01:42:44 you have to explain the implicit information otherwise 2011-04-30T01:42:53 with this you can just say ignore it and don't change it 2011-04-30T01:42:58 *** sigh has quit IRC (Ping timeout: 240 seconds) 2011-04-30T01:42:59 antimatroid: that's easier to explain 2011-04-30T01:43:17 "you can't be sure about any food that you weren't told about this turn" 2011-04-30T01:43:32 anything further is the bot's responsibility 2011-04-30T01:44:06 and if the starter packs implement visibility logic, then it's really easy to explain. 2011-04-30T01:44:07 explaining the purpose of the line itself becomes less important when we implement function as we have that do it for us i.e. food(), food_visible(), food_unseen() 2011-04-30T01:44:24 the purpose of the functions is to allow the bot creator to ignore the lines altogether 2011-04-30T01:44:35 and focus on their effect, which is very easy to understand 2011-04-30T01:45:46 well, one could also consider, as a sort of reductio ad absurdam, having the server track which enemy ants were visible once and when the square the enemy ants were on becomes visible again, 'remove' the ants 2011-04-30T01:45:54 :) 2011-04-30T01:45:59 antimatroid: 'a' = visible ants, 'f' = visible food, 'd' = ants which died last turn on visible squares, 'r' = ? 2011-04-30T01:46:08 antimatroid: some up with a short description for 'r' 2011-04-30T01:46:16 removed food 2011-04-30T01:46:19 remove food 2011-04-30T01:46:33 'r' was picked for "remove" 2011-04-30T01:46:35 a complete explanation 2011-04-30T01:46:40 that doesn't tell me anything 2011-04-30T01:47:10 previously seen food that has since been removed 2011-04-30T01:47:14 'r' = visible square that contained food the last time it was visible 2011-04-30T01:47:20 *which 2011-04-30T01:47:52 animatroid: that doesn't explicitly state that the line is only sent if the food is no longer there 2011-04-30T01:48:01 it's implied ... 2011-04-30T01:48:04 ah yes 2011-04-30T01:48:06 add that in 2011-04-30T01:48:09 As much as I hate leaving a heated discussion I must go :/ goodbye everybody 2011-04-30T01:48:14 cya 2011-04-30T01:48:19 i accidentally important part 2011-04-30T01:48:55 :) 2011-04-30T01:49:00 *** jaycobcoleman has quit IRC (Quit: Page closed) 2011-04-30T01:49:00 the full description for r is much longer than the others :P 2011-04-30T01:49:08 <_flag> Can I expect to see this "beta launch" milestone thingy complete in an hour or has it been postponed? :P 2011-04-30T01:49:32 _flag: you muggle :P 2011-04-30T01:49:46 sigh_: but people can ignore it if they want 2011-04-30T01:50:04 most beginners will learn from the starter bot they choose 2011-04-30T01:50:05 antimatroid: they still need to understand it before they can decide to ignore it 2011-04-30T01:50:06 then read the specs 2011-04-30T01:50:12 <_flag> antimatroid: you wizard? I'm confused. 2011-04-30T01:50:25 no, i called you a non-wizard 2011-04-30T01:50:26 big difference 2011-04-30T01:50:33 <_flag> I see that 2011-04-30T01:50:39 it wasn't easy to explain it to a1k0n is the main concern I'm having 2011-04-30T01:50:40 <_flag> Which would imply you are a wizard 2011-04-30T01:50:54 a1k0n: be quicker damnit! 2011-04-30T01:50:58 :p 2011-04-30T01:51:09 haha, sorry 2011-04-30T01:51:09 that doesn't come close to implying that :P 2011-04-30T01:51:13 i just like saying muggle 2011-04-30T01:51:20 i had to put jammies on my kid 2011-04-30T01:51:25 it's people like you who prevented us from having the power rule :P 2011-04-30T01:51:26 also the fact that i woke up at 6am just hit me like a brick wall 2011-04-30T01:52:02 'a' and 'f' are easy to explain, 'd' is slightly trickier, but it is giving the ants new info... 'r' is even harder to explain and is redundant anyway 2011-04-30T01:52:10 hey, for the record i understood immediately, i was just deliberately misinterpreting it 2011-04-30T01:52:20 :) 2011-04-30T01:52:36 ^^ sigh_: your argument is invalid :P 2011-04-30T01:52:52 .* 2011-04-30T01:53:12 if it is possible to misinterpret it, people will :P 2011-04-30T01:53:15 <_flag> 'r' = square that no longer contains food 2011-04-30T01:53:21 <_flag> ta-da 2011-04-30T01:53:28 that's not complete 2011-04-30T01:53:41 <_flag> How so? 2011-04-30T01:53:50 "square that contained food last time it was visible to you, but is now visible to you again and no longer contains food" 2011-04-30T01:54:06 still not complete 2011-04-30T01:54:12 "also next frame, when it is still visible to you, will still not contain food unless there is an 'f'" 2011-04-30T01:54:21 "visible square which doesn't contain food and did the last time it was visible" 2011-04-30T01:54:36 antimatroid: still not complete :P 2011-04-30T01:54:40 oh waut 2011-04-30T01:54:40 hey what if food disappears and reappears in that same location 2011-04-30T01:54:41 why not? 2011-04-30T01:54:42 yes it is 2011-04-30T01:54:48 :p 2011-04-30T01:54:50 do you get a 'r' and 'f' at the same time? 2011-04-30T01:54:50 a1k0n: you get an 'r' and an 'f' 2011-04-30T01:55:09 does the engine really track that correctly? 2011-04-30T01:55:16 yes 2011-04-30T01:55:21 sigh_ is my magical feature wizard 2011-04-30T01:55:24 we can change it if we need to 2011-04-30T01:55:29 i come up with features and he implements them 2011-04-30T01:55:36 :p 2011-04-30T01:55:38 if it happens multiple times? one 'r' and one 'f'? 2011-04-30T01:55:41 so technically, 'r' does give you a tiny bit of new info 2011-04-30T01:55:51 a1k0n: yes, because you only saw one 'r' 2011-04-30T01:56:26 so on every square on the map you have a flag for "player N thinks there is food here" 2011-04-30T01:56:28 hmmm... that info leak is bad 2011-04-30T01:56:30 it's explicitly telling you the outcome of a visible battle? 2011-04-30T01:56:31 need to fix 2011-04-30T01:56:36 is that the difference you refer to? 2011-04-30T01:56:46 and only if a food square spawned on a "dead" food square 2011-04-30T01:56:48 it's just telling you an enemy ant ate it at some point 2011-04-30T01:57:07 yes, which tells you that there *are* enemy ants in that direction 2011-04-30T01:57:21 but food spawning in that exact same location again is unlikely, no? 2011-04-30T01:57:29 so you'd probably know it anyway from other nearby food squares 2011-04-30T01:57:37 food can be waiting to spawn on a location :P 2011-04-30T01:57:44 a1k0n: yes, also the food locations are rotated 2011-04-30T01:57:51 but if maps aren't flooded i don't expect it to be too much of a problem 2011-04-30T01:58:04 so, yes... overall this is a very minor problem 2011-04-30T01:58:18 I would rather get rid of the 'r' lnes 2011-04-30T01:58:38 i will get over it 2011-04-30T01:58:43 *** ahmedhelal has quit IRC (Read error: No route to host) 2011-04-30T01:58:47 but i don't like the implicit information arising from not giving it 2011-04-30T01:59:04 what was mcleopold's opinion on 'r'? 2011-04-30T01:59:14 what implicit information? the exact space of square you can/can't see? 2011-04-30T01:59:21 he was in favour, otherwise it wouldn't have happened 2011-04-30T01:59:32 *** ahmedhelal has joined #aichallenge 2011-04-30T01:59:34 a1k0n: information that you "know" but have to calculate yourself 2011-04-30T01:59:41 visible squares are implicit information 2011-04-30T01:59:47 well, the engine could draw you a huge map every frame 2011-04-30T01:59:55 like in tron 2011-04-30T02:00:06 we had that originally but mcleopold didn't like the size of input 2011-04-30T02:00:11 that's my perefered option :) 2011-04-30T02:00:22 janzert and i preferred it too 2011-04-30T02:00:37 so the input is compressed, and you have to reconstruct the input. big whoop. 2011-04-30T02:00:48 have the starter pack do it, and then everybody can do it without having to worry too much. 2011-04-30T02:01:17 maybe even include a thing to print the map. 2011-04-30T02:01:46 my c++ bot has << for state 2011-04-30T02:02:06 they can also pass "state" information from an ifstream if they want 2011-04-30T02:27:01 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has quit IRC (Quit: Lost terminal) 2011-04-30T02:31:35 So if an "r" and an "f" are issued for the same tile in one turn, are the guaranteed to be in that order? 2011-04-30T03:21:27 *** orivej has joined #aichallenge 2011-04-30T03:31:43 hi 2011-04-30T03:32:10 antimatroid: do you still want the script? 2011-04-30T03:32:21 i can send u by dcc i guess 2011-04-30T03:35:07 *** tron_ has joined #aichallenge 2011-04-30T03:39:09 mcstar: have you seen visualizer/visualize_locally.py 2011-04-30T03:39:25 its open right now 2011-04-30T03:39:41 its doesnt do anything for me 2011-04-30T03:39:50 if you pipe a replay into it, it will visualise it 2011-04-30T03:40:10 ah 2011-04-30T03:40:14 so something like: play_one_game.sh --stdout replay | python visualizer/visualize_locally.py 2011-04-30T03:40:32 i guess cat works too 2011-04-30T03:40:43 yup 2011-04-30T03:40:46 cat 0.replay | python visualize... 2011-04-30T03:41:04 python visualizer/visualize_locally.py < 0.replay 2011-04-30T03:41:10 *** tron_ has quit IRC (Quit: Page closed) 2011-04-30T03:41:14 :) 2011-04-30T03:41:20 no point running an extra process just to read the file :P 2011-04-30T03:41:22 my script embeds the given replay file in base64 in the html, and the opens the browser, and a js function decodes it 2011-04-30T03:41:58 sure, if that works better for your workflow, then use it 2011-04-30T03:42:06 just mentioning that the python script is already there :) 2011-04-30T03:42:25 i didnt know about thus vis_locally 2011-04-30T03:42:37 i talked to janzert 2011-04-30T03:42:48 and he said only copy-paste is possible 2011-04-30T03:43:09 i didnt want to duplicate effort beleive me :) 2011-04-30T03:43:12 :) 2011-04-30T03:43:29 if an "r" and an "f" are issued for the same tile in one turn, are the guaranteed to be in that order? Remove first, then add? 2011-04-30T03:43:38 I guess it should be documented somewhere 2011-04-30T03:43:44 smiley1983: no order guarentees 2011-04-30T03:44:15 in fact, at the moment r is always last 2011-04-30T03:44:43 ... so the starter packs, if they respond to "r", should check for this case. I think I'll just remove "r" support in my starter pack: it creates too many ambiguities. 2011-04-30T03:44:58 yeah, I don't like 'r' in the first place 2011-04-30T03:45:29 I'll write some comments about how to re-enable it. I think amstan will be annoyed about me issuing another pull request. 2011-04-30T03:45:32 it is easy enough to keep track of the state in the bot itself, if you want to expose the same functionality 2011-04-30T03:46:14 (he merged my last two recently without complaint, but a third so soon would be pushing my luck) 2011-04-30T03:46:49 other people can merge too :) 2011-04-30T03:47:21 I guess. It's not an urgent change so I'll wait and see if anything develops in further conversation. 2011-04-30T03:47:56 yeah, that seems a bit better 2011-04-30T03:50:33 @later tell McLeopold I strongly suggest removing the 'r' output. It is complicated to explain, causes a lot of confusion and the implementation makes it easy to introduce bugs for bots that actually use it. 2011-04-30T03:50:33 sigh_: I come to serve. 2011-04-30T03:53:17 I do agree: it will of course break the bots which now expect it, but there's clearly plenty of time before the contest begins. :) 2011-04-30T03:53:35 it's not even beta yet :P 2011-04-30T03:53:58 exactly, so now is the time for such changes if at all 2011-04-30T03:54:46 *** kaemo has joined #aichallenge 2011-04-30T04:00:57 *** berak has joined #aichallenge 2011-04-30T04:16:40 *** orivej has quit IRC (Ping timeout: 260 seconds) 2011-04-30T04:26:00 *** _0x47 <_0x47!~slash@p5B25FAA6.dip.t-dialin.net> has quit IRC (Ping timeout: 260 seconds) 2011-04-30T04:41:49 *** orivej has joined #aichallenge 2011-04-30T04:49:17 *** orivej has quit IRC (Ping timeout: 276 seconds) 2011-04-30T04:58:15 *** _0x47 <_0x47!~slash@p5B25E644.dip.t-dialin.net> has joined #aichallenge 2011-04-30T05:16:11 *** MatmaRex has joined #aichallenge 2011-04-30T05:16:17 *** MatmaRex has joined #aichallenge 2011-04-30T05:17:58 *** Naktibalda has joined #aichallenge 2011-04-30T05:24:19 *** boegel has joined #aichallenge 2011-04-30T05:47:59 *** amstan has quit IRC (Read error: Operation timed out) 2011-04-30T05:48:41 *** Blkt has joined #aichallenge 2011-04-30T05:49:26 good day everyone 2011-04-30T05:49:38 hello 2011-04-30T05:49:47 hi 2011-04-30T05:50:35 *** Kingpin13 has joined #aichallenge 2011-04-30T05:55:29 *** FireFly has joined #aichallenge 2011-04-30T06:31:10 Is it possible to see the terrain at the beginning? Etc where the water is at, before exploring the entire map? 2011-04-30T06:31:42 nope 2011-04-30T06:31:53 you noly see what your ants can see 2011-04-30T06:31:55 only* 2011-04-30T06:32:23 what if i've been there once, do I then know it if my ants move away? 2011-04-30T06:33:47 water is static for the length of a game 2011-04-30T06:34:33 yep I know it doesnt move, but does the starter packages remember where the water is at, once spotted? 2011-04-30T06:34:42 they should 2011-04-30T06:34:47 antimatroid: hi 2011-04-30T06:35:04 they do, the information is only passed once 2011-04-30T06:35:08 mcstar: hi 2011-04-30T06:35:21 ru still interested? 2011-04-30T06:35:44 or u r set up with visualze_locally.py 2011-04-30T06:35:56 ok thx antimatroid 2011-04-30T06:39:23 i'm all good :) 2011-04-30T06:39:29 thanks anyway 2011-04-30T06:39:35 ok 2011-04-30T06:46:21 *** needsch has joined #aichallenge 2011-04-30T06:51:13 Uhm, I cant seem to get to the "starter packages" - Im following: https://github.com/aichallenge/aichallenge/wiki/Ants-Five-Minute-Quickstart-Guide - But it is not possible to create an account? 2011-04-30T06:51:24 yeah, ignore that for now 2011-04-30T06:51:27 we haven't got that far yet 2011-04-30T06:51:54 ah okay, so how do I get started? 2011-04-30T06:52:05 download everything off github 2011-04-30T06:52:44 run play_one_game.sh/bat then open up visualizer/copy_paste.html and paste the contents from /viewer/0replay.something 2011-04-30T06:52:55 *** Naktibalda has quit IRC (Quit: ChatZilla 0.9.86.1 [Firefox 4.0/20110419192255]) 2011-04-30T06:52:58 then inspect play_one_game 2011-04-30T06:53:01 aight, I'll try. thanks =) 2011-04-30T06:53:19 all this is from inside the ants directory :) 2011-04-30T06:57:23 Do I need anything installed before I can run it? Im on windows 7 and I found the files play_one_game.cmd and play_one_game.sh 2011-04-30T07:03:44 you should be able to run the .cmd file 2011-04-30T07:03:52 in a command window 2011-04-30T07:04:08 When doing so, it pops up and asks which program I would like to run playgame.py with 2011-04-30T07:04:14 So I assume I need some sort of python installation to run it 2011-04-30T07:04:33 ah yeah, you need python :P 2011-04-30T07:04:39 make sure you get a 2.* version 2011-04-30T07:04:41 not 3.* 2011-04-30T07:04:50 alright, on it ;) tyvm 2011-04-30T07:05:04 i'm going out sorry, hopefully you should be alright after that 2011-04-30T07:07:51 yep, worked like a charm =) 2011-04-30T07:09:16 cool 2011-04-30T07:09:18 :) 2011-04-30T07:09:23 almost :P 2011-04-30T07:09:53 it you use a c++ bot for example, get rid of the "python ... " to "..." when giving the bots location 2011-04-30T07:09:58 The game ran, but when copying to copy_paste.html I get error "TypeError" - Object doesn't support this property or method 2011-04-30T07:10:14 strange 2011-04-30T07:10:22 should be able to copy the text, paste it and click load 2011-04-30T07:10:37 It's the data from 0.replay and not 0.stream right? 2011-04-30T07:10:45 should be 2011-04-30T07:11:02 did you let the game finish? 2011-04-30T07:11:11 yep 2011-04-30T07:11:31 not sure sorry 2011-04-30T07:11:35 kk np 2011-04-30T07:11:37 anyone else? :p 2011-04-30T07:11:40 sigh_? :P 2011-04-30T07:11:49 u make u mistake when u copied 2011-04-30T07:11:51 i think 2011-04-30T07:12:09 under linux i had to copy using proper clipboard 2011-04-30T07:12:15 I go to viewer -> 0.replay -> open i notepad -> copy all -> paste -> load 2011-04-30T07:12:17 the middle click magic didnt work 2011-04-30T07:12:34 maybe notepad put in extra newlines 2011-04-30T07:12:41 it supposed to be 1 line 2011-04-30T07:13:04 hmm which editor is useful for this? 2011-04-30T07:13:15 sorry im a linux guy 2011-04-30T07:13:51 theres a conversion problem regarding new lines and carriage returns between unix/windows 2011-04-30T07:14:07 hmm it's one line now for sure, still error 2011-04-30T07:14:10 try to open it in ur code editor 2011-04-30T07:14:11 notepad++ <3 for windows 2011-04-30T07:14:34 ye I got 1 line with notepad++, still error on page tho 2011-04-30T07:14:54 go to the end of the line 2011-04-30T07:15:03 does it math up after pasting it in the browser 2011-04-30T07:15:07 and in the editor? 2011-04-30T07:15:13 math->match 2011-04-30T07:15:16 yep 2011-04-30T07:15:22 "storage"} ending in both 2011-04-30T07:15:34 well, then i have no idea :) 2011-04-30T07:15:51 :(:(:( 2011-04-30T07:15:53 hm 2011-04-30T07:15:57 type error 2011-04-30T07:16:01 interesting 2011-04-30T07:16:15 does it matter which browser I use? 2011-04-30T07:16:27 maybe 2011-04-30T07:16:35 have u got js debnug console? 2011-04-30T07:16:41 there we go 2011-04-30T07:16:44 works in firefox 2011-04-30T07:16:47 wohoooooooo =)=) <3<3 thanks guys 2011-04-30T07:16:55 document.getElementById('txtReplay').value; 2011-04-30T07:17:06 i guess this line offends ur other browser 2011-04-30T07:17:19 which was? 2011-04-30T07:17:22 yep, there was some sort of "read from file" that they didn't permit 2011-04-30T07:17:30 internet explorer, avant browser, chrome, opera 2011-04-30T07:17:32 all failed 2011-04-30T07:17:35 oh 2011-04-30T07:17:58 that will cause some trouble for others :) 2011-04-30T07:18:15 yep, hopefully they'll ask in here :) 2011-04-30T07:19:38 Okay so last question before the coding begins: Where do I select which bots to be run in the simulation? atm I see 4 similar bots running 2011-04-30T07:21:12 play_one_game.cmd 2011-04-30T07:21:26 "python dist/sample_bots/python/HunterBot.py" 2011-04-30T07:21:33 there are 4 arguments like this 2011-04-30T07:21:40 ah yes ofc, I understand it =) 2011-04-30T07:22:35 I see it says "python dist/sample_bots/python/HunterBot.py" - What should I specify for c-sharp and is it even supported yet? 2011-04-30T07:22:55 i guess the path to the binary 2011-04-30T07:23:16 havent tried it yet 2011-04-30T07:23:16 kk thx for all the help, greatly appreciated :) 2011-04-30T07:23:32 ur welcome 2011-04-30T07:23:36 but try it first 2011-04-30T07:23:51 i have to look at playgame first 2011-04-30T07:28:15 *** delt0r___ has joined #aichallenge 2011-04-30T07:30:02 *** delt0r has quit IRC (Ping timeout: 276 seconds) 2011-04-30T07:49:50 *** sigh_ has quit IRC (Ping timeout: 250 seconds) 2011-04-30T07:52:18 *** sigh has joined #aichallenge 2011-04-30T08:22:58 *** _0x47 <_0x47!~slash@p5B25E644.dip.t-dialin.net> has quit IRC (Remote host closed the connection) 2011-04-30T08:24:19 *** mceier has joined #aichallenge 2011-04-30T08:25:30 *** _0x47 <_0x47!~slash@p5B25E644.dip.t-dialin.net> has joined #aichallenge 2011-04-30T08:54:19 *** _0x47 <_0x47!~slash@p5B25E644.dip.t-dialin.net> has quit IRC (Remote host closed the connection) 2011-04-30T08:54:59 *** _0x47 <_0x47!~slash@p5B25E644.dip.t-dialin.net> has joined #aichallenge 2011-04-30T08:55:11 *** mceier has quit IRC (Ping timeout: 240 seconds) 2011-04-30T09:03:20 *** MatmaRex has left #aichallenge 2011-04-30T09:14:28 *** mceier has joined #aichallenge 2011-04-30T09:18:42 any date fixed yet for the next contest? 2011-04-30T09:40:42 wth 2011-04-30T09:40:48 map with 10 bots??? 2011-04-30T09:41:59 are you guys planning to really let 10 bots play at the same time on the same map? 2011-04-30T09:42:09 so its not going to be 1 on 1? 2011-04-30T09:42:56 btw 2011-04-30T09:43:06 the visualization http sever works nicely 2011-04-30T09:43:15 it only needs a time slider 2011-04-30T09:48:02 i changed the number of players in the map file 2011-04-30T09:48:24 *** mceier has quit IRC (Ping timeout: 240 seconds) 2011-04-30T09:48:34 but it still says that the number of players font math up 2011-04-30T09:48:37 c 2011-04-30T09:48:42 wth is going on? 2011-04-30T09:49:38 it seems everyone is out, gardening :( 2011-04-30T09:54:38 *** Blkt has quit IRC (Ping timeout: 240 seconds) 2011-04-30T09:54:44 *** Naktibalda has joined #aichallenge 2011-04-30T09:57:34 *** Mathnerd314 has joined #aichallenge 2011-04-30T10:18:14 morning 2011-04-30T10:18:25 where is the http server? 2011-04-30T10:20:09 ants/viewer 2011-04-30T10:20:24 python -m SimpleHTTPServer 2011-04-30T10:20:43 and the readme says open ur browser at localhost:8000 2011-04-30T10:20:50 ah that one yea 2011-04-30T10:20:50 then u get e filesystemview 2011-04-30T10:21:06 its sort of rudimentary 2011-04-30T10:21:14 needs a slider 2011-04-30T10:21:40 bhasker: can u answer my question regarding the participating bots? 2011-04-30T10:21:50 which question is that? 2011-04-30T10:22:00 btw i am not an organizer:) so i may not know the answer 2011-04-30T10:22:13 i saw that certains maps have up to 10 bots 2011-04-30T10:22:37 yes 2011-04-30T10:22:38 i thought it would b ea 1 on 1 battle 2011-04-30T10:22:44 its not 2011-04-30T10:22:48 that much is certain 2011-04-30T10:23:14 so what, at least 4 bots play 1 map? 2011-04-30T10:23:23 i think most maps generated right now have > 4 2011-04-30T10:23:45 man, i have to rethink my strategy alltogether 2011-04-30T10:23:53 not sure what format will be decided upon, i guess it all depends on how scoring/rankging is done 2011-04-30T10:23:59 i proved the perfect startegy already for 2 players 2011-04-30T10:24:00 and ensuring that bots get to play maps evenly 2011-04-30T10:24:29 ok 2011-04-30T10:24:32 well you can still treat it as a 2 player game 2011-04-30T10:24:34 you vs the world 2011-04-30T10:24:41 btw i was joking 2011-04-30T10:24:51 but i guess thats true 2011-04-30T10:25:07 i just thought there would me more room to explorer 2011-04-30T10:25:08 though i guess intelligent bots might push against certain bots 2011-04-30T10:25:09 -r 2011-04-30T10:25:12 to gain more territory 2011-04-30T10:25:18 than go up against stronger bots 2011-04-30T10:25:30 well i think the maps might be bigger than 120x100 2011-04-30T10:25:39 i think nothing is decided on what size maps will be used etc 2011-04-30T10:25:46 i think it will be decided during the beta etc 2011-04-30T10:25:47 but the question is if theres enough time to differentiate between the bots 2011-04-30T10:25:59 as the contest evolves and there is more feedback 2011-04-30T10:26:20 i think the idea is for games to last 1000 turns 2011-04-30T10:26:25 it should be enough to differentiate 2011-04-30T10:26:35 i just tried to run for 1000 2011-04-30T10:26:44 but the game ended b4 2011-04-30T10:26:51 1 bot timed out 2011-04-30T10:26:57 well that may not be true once bots get complicated 2011-04-30T10:26:59 the replay file seemed good 2011-04-30T10:27:05 but i couldnt visualize it 2011-04-30T10:27:11 right now most bots are probably trivial and not very smart 2011-04-30T10:27:21 i am guessing as bots get smarter combat will be entered into less often 2011-04-30T10:27:27 and there will be a lot of back and forth 2011-04-30T10:27:37 yes 2011-04-30T10:27:41 thats possible 2011-04-30T10:27:49 the numbers may oscillate 2011-04-30T10:27:52 bots won't blindly rush in to die 2011-04-30T10:28:04 and i am guessing complex bots will make use of bottlenecks 2011-04-30T10:28:12 and try to mark their territory out 2011-04-30T10:28:35 so as to maximize food availability 2011-04-30T10:28:36 the territory grows with r^2 2011-04-30T10:28:58 true, maybe hard but you could potentially block off certain parts 2011-04-30T10:29:06 and concentrate on access through few corridors 2011-04-30T10:29:08 so where are u now? 2011-04-30T10:29:19 only ideas or u got some programme din? 2011-04-30T10:29:20 well my first version was a basic bfs searcher 2011-04-30T10:29:30 second one added an ability for potential fields 2011-04-30T10:29:35 for exploring 2011-04-30T10:29:52 and some basic battle logistics 2011-04-30T10:29:58 to be able to make move, check result and unmake 2011-04-30T10:30:01 well i dont know these terms 2011-04-30T10:30:12 well the first one does a simple breadth first search 2011-04-30T10:30:15 it takes each ant 2011-04-30T10:30:17 im not a computer science guy 2011-04-30T10:30:22 ah 2011-04-30T10:30:34 hmm basically a bfs is you start at say point A 2011-04-30T10:30:40 and you explore all adjoining points to A 2011-04-30T10:30:46 before going one step further 2011-04-30T10:30:54 so you expand your search space incrementally 2011-04-30T10:31:01 till you hit your goal 2011-04-30T10:31:12 like swiping? 2011-04-30T10:31:16 so starting at an ant i can expand outwards till i find a food, then extract the path 2011-04-30T10:31:16 *** sigh has quit IRC (Read error: Connection reset by peer) 2011-04-30T10:31:17 to the food 2011-04-30T10:31:29 let me giev you a nice link to pathfinding 2011-04-30T10:31:41 but that path can be very long 2011-04-30T10:31:44 it will help you make your bots a little more intelligent on how they get to food and route around obstacles 2011-04-30T10:31:55 well i restrict my space to view radius around the ant 2011-04-30T10:31:58 i don't search the whole map 2011-04-30T10:31:59 true 2011-04-30T10:32:01 everytime 2011-04-30T10:32:08 so its a tunnel u see 2011-04-30T10:32:14 kind of 2011-04-30T10:32:43 in most cases view radius is sufficient to route to most nearby spots 2011-04-30T10:32:49 i rarely send my ants across the map right now 2011-04-30T10:33:08 is the size of the map known beforehand? 2011-04-30T10:33:11 yes 2011-04-30T10:33:12 on turn 1 2011-04-30T10:33:17 it sends rows/cols 2011-04-30T10:33:24 rather turn 0 2011-04-30T10:33:25 ok 2011-04-30T10:33:28 there is a setup turn 2011-04-30T10:33:30 and a setup time 2011-04-30T10:33:34 you get like 3 seconds on turn 0 2011-04-30T10:33:40 to do any one time initializations 2011-04-30T10:33:42 etc 2011-04-30T10:33:53 but i heard that the map is not sent, so u dont have an owerview of the water 2011-04-30T10:33:58 yes map is not sent 2011-04-30T10:34:18 actually i think it is 2011-04-30T10:34:24 let me look at the code 2011-04-30T10:34:39 *** mceier has joined #aichallenge 2011-04-30T10:34:41 oh maybe not 2011-04-30T10:35:04 yea you just get some meta data about the map 2011-04-30T10:35:08 like rows/columns etc 2011-04-30T10:35:15 and attack radius etc 2011-04-30T10:35:52 you get total number of turns for the game 2011-04-30T10:36:02 load time and turn time 2011-04-30T10:36:46 do you agree that this challenge seems harder than the one before? 2011-04-30T10:37:06 yes it is since there are more than 2 people 2011-04-30T10:37:08 per game 2011-04-30T10:37:16 its more like an RTS now 2011-04-30T10:38:05 whats that potential based thingy? 2011-04-30T10:38:24 http://www.hindawi.com/journals/ijcgt/2009/910819/ 2011-04-30T10:38:27 potential fields 2011-04-30T10:38:35 you can use them as a way to do pathfinding/exploration etc 2011-04-30T10:38:51 you assign values to certain areas that spread out radially 2011-04-30T10:38:54 and fall off 2011-04-30T10:39:07 and you move your ant based on potential of the squares surrounding it 2011-04-30T10:39:31 so you could potentially rate unexplored areas higher and if you move the ant towards higher potential areas 2011-04-30T10:39:40 then ti will automatically tned to explore unseen areas 2011-04-30T10:39:41 etc 2011-04-30T10:39:51 you could put higher potential fields around nearby ants 2011-04-30T10:39:53 to make them cluster 2011-04-30T10:40:03 and -ve potential on enemies 2011-04-30T10:40:08 so it uses differentiability to predict potential gain? 2011-04-30T10:40:09 to make your ants automatically avoid enemies 2011-04-30T10:40:43 kind of, mostly its a way to decide what an ant should do at any point 2011-04-30T10:41:05 rather than explicitly telling the ant what to do 2011-04-30T10:41:16 so its like a dynamics potential, the independent variable is the coordinate 2011-04-30T10:41:24 k i lost you there 2011-04-30T10:41:24 and ur bots move along field lines? 2011-04-30T10:41:29 kind of 2011-04-30T10:41:30 yes 2011-04-30T10:41:42 you could make obstacles repell your bots etc 2011-04-30T10:41:48 so your bot will automatically route around them 2011-04-30T10:41:50 etc 2011-04-30T10:42:04 well, i can see that this has analytical value 2011-04-30T10:42:13 yea it helps identify hotspots 2011-04-30T10:42:17 and areas of interest 2011-04-30T10:42:25 like you could give areas with lot of dead ants 2011-04-30T10:42:29 but is good to algorithmically implement this? 2011-04-30T10:42:32 maybe lower potential so that your ants stay away 2011-04-30T10:42:45 its not too expensive to calculate 2011-04-30T10:43:12 though my results have been very mixed and my current implementation is rather crappy 2011-04-30T10:43:16 do they use it in real softwares? 2011-04-30T10:43:23 hmm mostly in games etc 2011-04-30T10:43:40 bc its a 2008 article thats why im asking 2011-04-30T10:43:46 kind of new 2011-04-30T10:43:50 i think there are a few game AI's that use it 2011-04-30T10:44:00 the other thing that RTS AI's use are what are called influence maps 2011-04-30T10:45:39 no wiki 2011-04-30T10:46:53 http://74.6.117.48/search/srpcache?ei=UTF-8&p=influence+maps+RTS&fr=crmas&u=http://cc.bingj.com/cache.aspx?q=influence+maps+RTS&d=4952990835475890&mkt=en-US&setlang=en-US&w=115799ed,80367371&icp=1&.intl=us&sig=1LGLNqBpUswzTPEBqw7cWQ-- 2011-04-30T10:48:31 so this makes a weighted graph from the map? 2011-04-30T10:48:44 kind of yes 2011-04-30T10:49:07 this sounds better 2011-04-30T10:49:15 search for RTS AI 2011-04-30T10:49:19 you will find numerous articles 2011-04-30T10:49:22 at least this is discrete from the start 2011-04-30T10:50:00 i will try to implement something from my physics background 2011-04-30T10:50:09 if im not too stupid for that 2011-04-30T10:50:28 http://aigamedev.com/ 2011-04-30T10:50:37 this is a nice website which has some free articles that you can read 2011-04-30T10:50:43 some of the other articles require you to pay 2011-04-30T10:51:21 thx for the links 2011-04-30T10:51:48 actually i just got an idea 2011-04-30T10:51:57 this is a good forum too http://www.gamedev.net/forum/9-artificial-intelligence/ 2011-04-30T10:53:10 i dont think that the smartest bot will win 2011-04-30T10:53:22 i mean the one that implements state of the art logic 2011-04-30T10:53:25 or methods 2011-04-30T10:53:47 i think that transparent, robust method will prevail 2011-04-30T10:53:57 in this contest the real ai is the developer 2011-04-30T10:54:11 lets see 2011-04-30T10:54:12 http://digestingduck.blogspot.com/ 2011-04-30T10:54:34 this blog is from one of the authors of a commerical game engine, he maintains an opensource 3d pathfinding library 2011-04-30T10:54:40 called recast and detour 2011-04-30T10:54:55 most of it is not relevant to this game but its an interesting read 2011-04-30T10:55:29 he is living my dream 2011-04-30T10:55:36 Our startup just opened a public beta. 2011-04-30T10:55:40 hehe 2011-04-30T10:55:50 another nice article on certain AI behaviours 2011-04-30T10:55:51 http://red3d.com/cwr/steer/ 2011-04-30T10:55:52 that would be nice to say some day 2011-04-30T10:55:59 how to implement like seek/flee 2011-04-30T10:56:03 pursue and evade 2011-04-30T10:56:04 etc 2011-04-30T10:56:18 flocking/leader following etc 2011-04-30T10:56:30 and this is my all time favourite pathfinding page 2011-04-30T10:56:30 http://www-cs-students.stanford.edu/~amitp/gameprog.html 2011-04-30T10:56:36 wait, my browser dont have enough free bookmarks 2011-04-30T10:56:40 this guy has some of the best pathfinding tutorials 2011-04-30T10:56:45 that you can find on the net 2011-04-30T10:57:12 *** Naktibalda has quit IRC (Remote host closed the connection) 2011-04-30T10:57:31 about potential fields nice tutorial 2011-04-30T10:57:31 http://archive.gamedev.net/reference/programming/features/motionplanning/ 2011-04-30T10:57:38 what do you think 2011-04-30T10:58:00 can somebody who isnt educated in com. science get a job dealing with these kind of ai things? 2011-04-30T10:58:54 *** Naktibalda has joined #aichallenge 2011-04-30T10:59:03 if you are good at math i think yes 2011-04-30T10:59:06 AI is more math 2011-04-30T10:59:09 than comp sci 2011-04-30T10:59:18 comp sci part only comes to how to efficiently implement it 2011-04-30T10:59:31 for that you can take a data structures course 2011-04-30T10:59:42 i am crappy at math but decent and comp sci 2011-04-30T10:59:48 i would make an average ai programmer 2011-04-30T11:00:05 im asking if its not a big problem, that i will have another degree 2011-04-30T11:00:26 i don't think most game companies care about your degree as long as you know a few things 2011-04-30T11:00:33 *** Eruonen has joined #aichallenge 2011-04-30T11:00:34 game theory and basic data strcutures 2011-04-30T11:00:37 i would gess that they like to use their own "breed" 2011-04-30T11:00:49 most game programmers are probably not comp sci grads 2011-04-30T11:00:56 games today have a lot of physics too 2011-04-30T11:01:02 so you could be a physics programmer 2011-04-30T11:01:08 if you wanted 2011-04-30T11:01:17 there are companies dedicated to designing physics middleware 2011-04-30T11:01:19 like havok 2011-04-30T11:01:20 ty i will see my future a little brighter :) 2011-04-30T11:01:30 which are used by games for all physics simulations 2011-04-30T11:01:36 there are open source alternatives too like 2011-04-30T11:01:46 yeah i heard of it 2011-04-30T11:01:47 ODE (Open Dynamics Engine) and BulletPhysics 2011-04-30T11:01:53 and Newton 2011-04-30T11:02:04 a lot of open source game engines today integrate these libraries 2011-04-30T11:02:17 like Ogre/CrystalSpace/Irrlicht/Panda3d 2011-04-30T11:02:25 i guess theyre rewriten them to run on gpu-s 2011-04-30T11:02:42 no most of these still run on cpu 2011-04-30T11:02:53 i guess havok makes use of physx 2011-04-30T11:02:59 and i think some game engines support that too 2011-04-30T11:03:01 yeah but they would benefit from it l 2011-04-30T11:03:19 yes some of them are making use of the nvidia PhysX apis 2011-04-30T11:03:20 im kind of certain that its in progress 2011-04-30T11:03:31 its been in progress fo sometime now 2011-04-30T11:03:47 http://www.nvidia.com/object/physx_new.html 2011-04-30T11:04:02 http://www.nvidia.com/object/physx_faq.html 2011-04-30T11:04:29 *** UncleVasya has joined #aichallenge 2011-04-30T11:04:41 *** UncleVasya has joined #aichallenge 2011-04-30T11:04:54 GeForce GPUs with PhysX provide an exponential increase in physics processing power 2011-04-30T11:05:02 not sure what they mean by that 2011-04-30T11:05:27 well since GPUs are inherently paralle 2011-04-30T11:05:36 you can do a lot of physics calculations in parallel 2011-04-30T11:05:48 that sentence is wrong 2011-04-30T11:06:01 why would you say that 2011-04-30T11:06:09 in that sense everything is exponential 2011-04-30T11:06:39 well its marketing speak so take it with a grain of salt 2011-04-30T11:06:47 thats what i mean 2011-04-30T11:07:40 mostly if you do physics calculations on cpu you are limited to one at a time on a single core or at most 4-8 depending on number of cores on the computer 2011-04-30T11:07:47 but most GPU's today have over 400+ cores 2011-04-30T11:07:56 i do them on gpu 2011-04-30T11:08:00 so you could do a lot more parallel calculations 2011-04-30T11:08:08 thats what they mean by exponential increase 2011-04-30T11:08:15 we have a couple of teslas 2011-04-30T11:08:18 ah 2011-04-30T11:08:34 4 tesla and 8 core xenon 2011-04-30T11:08:39 in a box 2011-04-30T11:08:43 and we have 4 of them 2011-04-30T11:08:52 ah k 2011-04-30T11:09:00 48GB memory 2011-04-30T11:09:11 quite noisy though 2011-04-30T11:09:25 heh i guess so 2011-04-30T11:09:29 with that many gpus 2011-04-30T11:09:58 and have 2 rooms filled with 3 clusters 2011-04-30T11:10:23 but im not sure i want to do computational physics though 2011-04-30T11:10:39 i will figure it out 2011-04-30T11:10:59 but this contest will have some say in my future plans 2011-04-30T11:11:42 are you interfacing with outside 3rd party libraies? 2011-04-30T11:11:54 nope 2011-04-30T11:12:06 i am writing in c++ and i doubt i will use any external library 2011-04-30T11:12:20 i dont intend to either 2011-04-30T11:12:33 the game is quite simple to not really require external libraries 2011-04-30T11:12:45 pathfinding a simple bread first search works in most cases 2011-04-30T11:12:45 in the last couple of days i upped my cl knowledge 2011-04-30T11:12:54 so now i can begin write 2011-04-30T11:12:58 to 2011-04-30T11:13:01 which language? 2011-04-30T11:13:10 common lisp 2011-04-30T11:13:14 ah k 2011-04-30T11:13:26 my knowledge of functional programming is very limited 2011-04-30T11:13:48 i used mathematica a lot 2011-04-30T11:13:52 ah k 2011-04-30T11:14:05 and tried a tiny tiny bit of haskell 2011-04-30T11:14:18 but cl is not just functional 2011-04-30T11:14:21 as i understand now 2011-04-30T11:14:32 basically u can do any programming model in it 2011-04-30T11:14:39 hmm i thought lisp was mostly functional 2011-04-30T11:14:42 even lazy evaluation can be implemented 2011-04-30T11:14:45 or at least most people used it that way 2011-04-30T11:14:46 no 2011-04-30T11:14:52 has a good object system 2011-04-30T11:15:00 oh k 2011-04-30T11:15:04 but u dont have to use it of course 2011-04-30T11:26:17 im afraid that in the end my ai will look like a very big if-else forest 2011-04-30T11:35:29 *** FireFly has quit IRC (Quit: swatted to death) 2011-04-30T11:41:34 *** _0x47 <_0x47!~slash@p5B25E644.dip.t-dialin.net> has quit IRC (Remote host closed the connection) 2011-04-30T11:48:01 *** Chris_0076 has joined #aichallenge 2011-04-30T11:52:52 *** delt0r has joined #aichallenge 2011-04-30T11:53:15 *** delt0r___ has quit IRC (Ping timeout: 248 seconds) 2011-04-30T11:59:28 *** Chris_0076 has quit IRC (Remote host closed the connection) 2011-04-30T12:33:49 *** _flag <_flag!~flag@69-165-173-172.dsl.teksavvy.com> has joined #aichallenge 2011-04-30T12:46:23 *** UncleVasya has quit IRC (Ping timeout: 260 seconds) 2011-04-30T12:51:59 *** Redgis has joined #aichallenge 2011-04-30T12:53:25 *** amriedle has joined #aichallenge 2011-04-30T12:59:23 *** _0x47 <_0x47!~slash@p5B25E644.dip.t-dialin.net> has joined #aichallenge 2011-04-30T13:03:28 mcstar: don't do a big if else forest :P 2011-04-30T13:04:07 i dont intend to 2011-04-30T13:04:52 but when im thinking over situations it resembles that 2011-04-30T13:07:08 did you work out what a bfs is in the end? 2011-04-30T13:07:13 sorry didn't read through it fully 2011-04-30T13:07:17 it's really easy to do with a queue 2011-04-30T13:07:34 what is your background anyway? if not cs? 2011-04-30T13:07:37 bhasker told me everything and more 2011-04-30T13:08:18 physics and an apollo 13 poster 2011-04-30T13:08:42 but i frequently change my background 2011-04-30T13:11:36 i had the one background for years 2011-04-30T13:11:45 but i haven't got it off an old hdd since moving to ubuntu 2011-04-30T13:11:51 it is the best desktop image ever 2011-04-30T13:12:25 i'm not a huge "cats" fan (I like them, but not the whole internet hype thing and prefer dogs), but it's like a fake cat picture on a black background that's just awesome 2011-04-30T13:14:38 im sure it looks better then u described it 2011-04-30T13:14:42 :) 2011-04-30T13:15:36 i have a productivity wallpaper: big white letters on black bck. DONT MESS AROUND DO SOME WORK 2011-04-30T13:15:55 it was nice but didnt help 2011-04-30T13:16:13 *** boegel has quit IRC (Quit: Leaving) 2011-04-30T13:16:28 oh yeah, you were using the wrong visualiser :P 2011-04-30T13:16:30 have you fixed that? 2011-04-30T13:16:38 heh? 2011-04-30T13:16:42 use the visualiser, not viwer 2011-04-30T13:16:44 wrong? 2011-04-30T13:16:53 i wasn using the wrong 2011-04-30T13:16:56 wasnt 2011-04-30T13:17:01 i used both 2011-04-30T13:17:04 :) 2011-04-30T13:17:05 ahhh okay 2011-04-30T13:17:12 i just saw the part about the http server 2011-04-30T13:17:15 and was like nooooo 2011-04-30T13:17:24 i just tried it 2011-04-30T13:17:34 the other is obviously better 2011-04-30T13:18:12 i hope u will find a good wallpaper for ur next linux distribution 2011-04-30T13:19:07 * mcstar thinks that sometime soon u will leave the ubuntu crowd and use a real distro 2011-04-30T13:19:17 antimatroid: are you aware that the engine sends food removal lines last? If food spawns on a tile on the same turn as you get a "r row col" line for the same tile, you get both lines, but the "r" will be last. Many bots (including mine currently) will respond incorrectly in this case. 2011-04-30T13:19:39 smiley1983: i saw that conversation, yes that presents problems :P 2011-04-30T13:20:12 My current plan is to clear all food information as before, and present the "removed food" list to the player verbatim to do with as they will. 2011-04-30T13:20:31 i wouldn't be surprised if the 'r' lines get removed 2011-04-30T13:20:49 Which will break all the bots now using it, but oh well 2011-04-30T13:25:02 There is no break. 2011-04-30T13:25:15 Instead only try to realize the truth. 2011-04-30T13:26:01 this contest is 1/3 software engineering 2011-04-30T13:26:02 *** amstan has joined #aichallenge 2011-04-30T13:26:02 *** ChanServ sets mode: +o amstan 2011-04-30T13:26:18 u have to make ur program felxible enough 2011-04-30T13:26:30 to incorporate last minute changes :) 2011-04-30T13:28:05 ... just don't make assumptions about stuff which isn't clearly stated in the specifications 2011-04-30T13:31:48 Yes. Sometimes it's hard to realise when you're making assumptions until after they present problems, though. :) 2011-04-30T13:59:20 *** Eruonenenen has joined #aichallenge 2011-04-30T14:07:19 Has anyone found a C# starter package or did I overlook it? 2011-04-30T14:12:12 antimatroid are all maps symmetrical? 2011-04-30T14:12:30 bhasker: no 2011-04-30T14:12:38 will it be the plan in the final? 2011-04-30T14:12:41 but i expect assymetric maps to be a roaring failure fairness wise 2011-04-30T14:12:50 hopefully not, but it's not my decision 2011-04-30T14:12:58 cause i am seeing my bot play differently depending on which position i play from 2011-04-30T14:13:06 and lose miserably sometimes 2011-04-30T14:13:10 yes, asymmetric maps suck 2011-04-30T14:13:10 sometimes crush the opposition 2011-04-30T14:13:17 even your symmetric maps are not perfect 2011-04-30T14:13:24 they are not symmetrical for all players 2011-04-30T14:13:31 what do you mean? 2011-04-30T14:13:44 they are symmetry excluding relative starting postions 2011-04-30T14:13:51 that's not possible with more than 3 players 2011-04-30T14:13:54 MuTa-ofd00m: https://github.com/aichallenge/aichallenge/pull/44 2011-04-30T14:14:05 symmetric* 2011-04-30T14:14:40 i am looking at the symmetric map but it doesn't look symmetric to me 2011-04-30T14:14:54 let me show you 2011-04-30T14:16:07 http://pastebin.com/P6tmF8f6 2011-04-30T14:16:12 its translation symmetric 2011-04-30T14:16:17 thats a replay form symmetric_10.map 2011-04-30T14:16:26 what does that mean translation symmetric? 2011-04-30T14:16:27 that one too 2011-04-30T14:16:32 theres a unit cell 2011-04-30T14:16:49 which is translated on the 2 basis vectors 2011-04-30T14:17:35 so bots can be placed in places with the same surroundings 2011-04-30T14:17:41 it doesn't look like it 2011-04-30T14:17:43 in the replay 2011-04-30T14:17:51 in 4 player map 2011-04-30T14:18:03 i will look at it 2011-04-30T14:18:15 but as far as i remember i found 10 to be symmetric 2011-04-30T14:18:26 hmm k 2011-04-30T14:19:10 it is symmetric 2011-04-30T14:19:15 *** jaycobcoleman has joined #aichallenge 2011-04-30T14:19:25 look at the start position for blue 2011-04-30T14:19:26 bhasker: have you looked at the map gen 2011-04-30T14:19:27 it is symmetric 2011-04-30T14:19:34 it looks very different from the others 2011-04-30T14:19:38 it was really fun working out how to do that 2011-04-30T14:20:13 if you translate their positions into a single point, the map will perfectly overlap 2011-04-30T14:20:31 if you have rows and row_translate, the number of translations to get back into the same row is rows/gcd(rows, row_translate) 2011-04-30T14:20:32 hmm k 2011-04-30T14:20:34 you know what 2011-04-30T14:20:42 press the window button 2011-04-30T14:20:49 on the right side 2011-04-30T14:20:50 then to get the orders of row and column translates to match up you get lcm of those two numbers 2011-04-30T14:20:55 above AZ 2011-04-30T14:20:55 ah k i see it 2011-04-30T14:21:27 there's some extra checks in there to make sure no two players start in the same row or column etc. 2011-04-30T14:22:19 in the end it's just tessellations on a torus 2011-04-30T14:22:40 i'm pretty sure you get vector spaces if you take the continuous case 2011-04-30T14:23:20 what do you mean? 2011-04-30T14:23:26 it is a vector space 2011-04-30T14:24:03 u just have a bit of congruance 2011-04-30T14:24:57 antimatroid: how come you did the map generator? 2011-04-30T14:25:18 i mean u will participate in the contest right? 2011-04-30T14:25:27 *** Blkt has joined #aichallenge 2011-04-30T14:26:41 mcstar: it's open source, doesn't matter who writes it 2011-04-30T14:26:46 what's wrong with having fun on both sides ? 2011-04-30T14:26:47 i wrote the planet wars map generator too 2011-04-30T14:26:54 *** jaycobcoleman has quit IRC (Ping timeout: 252 seconds) 2011-04-30T14:27:07 others are welcome to write generators for more maps, the more the merrier imo 2011-04-30T14:27:10 yeah, but u have less time for yourslef then :) 2011-04-30T14:27:29 i dont complain 2011-04-30T14:27:31 i've been trying not to do too much bot related before the contest starts 2011-04-30T14:27:32 far from it 2011-04-30T14:27:41 i already have enough of an advantage 2011-04-30T14:27:55 although i'm not sure how much time i will actually have anyway in the end 2011-04-30T14:28:20 i'm here for fun, and making map generators is one of the funnest problems :P so that's my motivation 2011-04-30T14:28:43 McLeopold has a cool one based off my partitioning of the grid in map_analyzer.py 2011-04-30T14:28:49 ok, well ty 2011-04-30T14:29:23 you can basically construct examples where any number of ants can reach a square in the same number of turns 2011-04-30T14:29:58 the map_analyzer partitions the set of locations into sets of which ants can reach them first 2011-04-30T14:30:30 its like a voronoi diagram but here we call it wigner seitz cell 2011-04-30T14:30:51 i call it a voronoi diagram, but not sure if that's correct? 2011-04-30T14:30:59 correct 2011-04-30T14:31:01 so i reverted to just saying it's a partition 2011-04-30T14:31:12 maths talk i can do 2011-04-30T14:31:40 well, to a certain extent :P 2011-04-30T14:31:48 such a large rabbit hole 2011-04-30T14:32:05 hm 2011-04-30T14:32:10 u hunt rabbits ? 2011-04-30T14:32:10 bhasker: what makes map 10 look different to the other maps to you? 2011-04-30T14:32:13 no 2011-04-30T14:32:26 nm its fine i was mistaken 2011-04-30T14:32:33 i feel bad about all the bugs i must kill on a daily basis :\ 2011-04-30T14:32:38 its randomness i guess :) 2011-04-30T14:33:14 i would love to see a 3d viewer on a toroud 2011-04-30T14:33:17 torus* aha 2011-04-30T14:33:45 why? i dont see any advantage of it 2011-04-30T14:33:49 where you could change the angle, and rotate the torus with the arrow keys to translate it around either circle 2011-04-30T14:33:56 it would be novelty and awesome :P 2011-04-30T14:34:02 u mean eyecandy 2011-04-30T14:34:06 yes 2011-04-30T14:34:22 there was an icpc ai-challenge that had an eyecandy 3d visualizer 2011-04-30T14:34:25 then pls write it and i might consider using it as a screensaver 2011-04-30T14:34:41 i don't know how to do guis :P 2011-04-30T14:34:46 hey theres one already 2011-04-30T14:34:59 some ants crawl on mobius strip 2011-04-30T14:35:11 i used ogl before 2011-04-30T14:35:13 its easy 2011-04-30T14:35:15 we have joked about making the maps a klein bottle :P 2011-04-30T14:35:44 i wrote a mandelbrot set viewer in cpp 2011-04-30T14:35:47 some years ago 2011-04-30T14:35:58 it was pretty awesome 2011-04-30T14:36:02 but i lost it since 2011-04-30T14:36:19 some of the 3d fractals you see online are pretty cool 2011-04-30T14:36:45 i have the most fun with stuff from algebra and analysis, where one can fully wrap their head around what's going on 2011-04-30T14:37:06 like what? 2011-04-30T14:37:39 *** amstan_ has joined #aichallenge 2011-04-30T14:37:39 *** ChanServ sets mode: +o amstan_ 2011-04-30T14:37:40 just the way such topics are treated, it's like microeconomics and game theory, you start from the basic definitions and build the theory from there 2011-04-30T14:37:59 rather than just trying to describe what we already see 2011-04-30T14:38:23 well if u want real progress u have to have real foundations 2011-04-30T14:38:45 ? 2011-04-30T14:39:17 u said u like the way mathematics is structured 2011-04-30T14:39:27 mostly pure maths 2011-04-30T14:39:35 applied maths makes me want to rip my hair out 2011-04-30T14:39:38 there are some "sciences" which arent build on such strong foundation 2011-04-30T14:39:52 s 2011-04-30T14:40:04 yeah, i just ignore them :P 2011-04-30T14:40:19 i sometimes laugh too 2011-04-30T14:40:39 i used voronoi diagrams b4 2011-04-30T14:40:44 theres a super duper software 2011-04-30T14:40:47 qhull 2011-04-30T14:40:54 i used them for my evaluation function with tron 2011-04-30T14:40:54 take a look at it if youre interested 2011-04-30T14:41:14 very fast and very capable 2011-04-30T14:41:33 do you know much about game theory? 2011-04-30T14:41:35 can do 3d too 2011-04-30T14:41:43 well i know it exists 2011-04-30T14:42:36 maybe i will read about it sometimes 2011-04-30T14:42:40 i haz a game theory library that i'm slowly building :P 2011-04-30T14:42:45 but for the contest i dont think i will 2011-04-30T14:42:55 https://github.com/antimatroid/GTL 2011-04-30T14:42:55 nice 2011-04-30T14:43:05 how many ppl use it? 2011-04-30T14:43:35 it's not really ready for use yet :P 2011-04-30T14:44:05 well, what's there works, but there's lot's that hasn't been added to what's on github yet, and nothing in the way of a gui, just a library 2011-04-30T14:44:28 i dont think u should bother with guis 2011-04-30T14:44:50 those classes arent implemented elsewhere? 2011-04-30T14:45:03 tensor vector rational? 2011-04-30T14:45:20 i like people to be able to use the library with just stl 2011-04-30T14:45:27 ah 2011-04-30T14:45:31 *** Blkt has quit IRC (Ping timeout: 252 seconds) 2011-04-30T14:45:56 I've added in permutation too, for checking symmetry and equivalence of games 2011-04-30T14:45:56 why do u do it? 2011-04-30T14:46:04 because i like game theory :P 2011-04-30T14:46:05 was some assignment? 2011-04-30T14:46:15 i started it with a friend as a summer research project 2011-04-30T14:46:27 but we were already planning to do it anyway and applied for the scholarship as a bonus 2011-04-30T14:46:58 so u had to present to this somebody? 2011-04-30T14:47:03 no 2011-04-30T14:47:11 we just got to play around all summer 2011-04-30T14:47:18 :_ 2011-04-30T14:47:20 ) 2011-04-30T14:47:32 i changed my layout to en 2011-04-30T14:47:46 somtimes i miss spell 2011-04-30T14:47:46 i need to work out how i'm going to do simplex 2011-04-30T14:47:59 i need that to do a lot of mixed strategy functions 2011-04-30T14:48:15 i was doing one of them with fourier-motzkin elimination, but that's silly 2011-04-30T14:48:39 mixed strategy nash equilibria for n-player normal form games is a beast of a problem 2011-04-30T14:48:50 the proof is alright, but actually calculating them, Oo 2011-04-30T14:49:01 np-complete? 2011-04-30T14:49:25 i'm not very knowledgable on complexity stuff, it's PNP or whatever 2011-04-30T14:49:36 you are guaranteed there exists at least one solution 2011-04-30T14:49:42 i dont know about transistors 2011-04-30T14:49:46 whereas with NP problems there may not even exist a single solution 2011-04-30T14:50:29 and what is its time complexity? 2011-04-30T14:50:42 calculating nash equilibria? 2011-04-30T14:50:45 yep 2011-04-30T14:50:52 i don't believe anyone has actually managed a working implementation 2011-04-30T14:51:05 oh 2011-04-30T14:51:09 gambit has the best functions, but they are far from perfect 2011-04-30T14:51:15 i mean it's one hell of a beast of a problem :P 2011-04-30T14:51:24 i see 2011-04-30T14:51:37 http://www.gambit-project.org/doc/ideas.html 2011-04-30T14:52:00 look for "Implement Herings-Peeters homotopy algorithm to compute Nash equilibria" 2011-04-30T14:52:29 maybe my math friend will be interested in this 2011-04-30T14:52:42 he deals with graphs and coloring problems 2011-04-30T14:52:54 i told him about the challenge 2011-04-30T14:53:01 he may contribute 2011-04-30T14:53:04 or may not 2011-04-30T14:53:16 he cant really program yet :( 2011-04-30T14:53:32 these contests are a great way to learn 2011-04-30T14:53:46 the starter bots basically let someone just start if/elsing if they really want 2011-04-30T14:53:49 he wont participate alone 2011-04-30T14:54:26 for example ur bot isnt easy 2011-04-30T14:54:35 i mean there are quite some concepts 2011-04-30T14:54:39 u have to be familiar with 2011-04-30T14:54:44 to see it through 2011-04-30T14:54:50 and then theres the syntax 2011-04-30T14:55:04 like? 2011-04-30T14:55:16 he doesnt really know c 2011-04-30T14:55:33 i was telling him about pointers and stuff 2011-04-30T14:55:38 and linked lists 2011-04-30T14:55:44 it was quite new to him 2011-04-30T14:56:04 *** jaycobcoleman has joined #aichallenge 2011-04-30T14:56:05 i know a lot of beginners that choose to use the c++ bot might not be familiar with everything i've done, but I've tried to make it so they can learn to use it like a black box and then go back and work out how it works if they are interested 2011-04-30T14:56:15 ok 2011-04-30T14:56:24 then lets pick another language 2011-04-30T14:56:30 unless you are using a game tree, I can avoid touching pointers and delete with c++ bots 2011-04-30T14:56:34 i dont mean cpp in particular 2011-04-30T14:57:08 oh and the whole environment 2011-04-30T14:57:22 the compilation process make linux editors 2011-04-30T14:57:24 ...... 2011-04-30T14:57:34 when all this comes almost at once 2011-04-30T14:57:34 tell them to download something like codeblocks (c++ specific), create a project and add all the files to the project 2011-04-30T14:57:45 one can be pretty much f_cked 2011-04-30T14:57:48 that's what I do, then i just hit f9 and it all just compiles nicely :) 2011-04-30T14:58:09 i am teaching him linux 2011-04-30T14:58:17 he is on arch like me 2011-04-30T14:58:29 i converted 3 of my friends from windows 2011-04-30T14:58:33 ubuntu is couldn't be much easier to use 2011-04-30T14:58:46 another friend previously suggested debian 2011-04-30T14:59:20 when u pick distro i dont think u should pick the easier 2011-04-30T14:59:28 how do you like arch? 2011-04-30T14:59:51 well i searched for distros for like 3 years 2011-04-30T14:59:53 at least 2011-04-30T15:00:06 i used kubuntu for like a year after 2011-04-30T15:00:14 i got fed up with that 2011-04-30T15:00:20 and then at last 2011-04-30T15:00:23 i found arch 2011-04-30T15:00:33 im using it since 2 years 2011-04-30T15:00:37 for 2011-04-30T15:00:40 not since 2011-04-30T15:01:00 Is compatibility a big issue? 2011-04-30T15:01:01 theres nothing not to like about it 2011-04-30T15:01:06 no 2011-04-30T15:01:27 i didnt find any issues regarding compatibility 2011-04-30T15:01:33 cool :P 2011-04-30T15:01:42 I might try it out after the competition is over 2011-04-30T15:01:46 i reinstalled only once 2011-04-30T15:02:00 becaus ein last year december 2011-04-30T15:02:04 my hdd crashed 2011-04-30T15:02:17 i highly recommend 2011-04-30T15:02:27 how much do u know about linux? 2011-04-30T15:02:37 I'm on ubuntu right now and I have no problems with it, but I like to experiment 2011-04-30T15:02:43 ok 2011-04-30T15:02:48 I know more than enought to make my way around 2011-04-30T15:02:51 -t 2011-04-30T15:02:55 ok 2011-04-30T15:03:00 and it has a very good wiki 2011-04-30T15:03:07 and lots of stuff in the forums 2011-04-30T15:03:19 basically i didnt have to write in forums 2011-04-30T15:03:24 since i use it 2011-04-30T15:03:36 have u heard of AUR? 2011-04-30T15:03:52 nope 2011-04-30T15:04:01 i can't find a screenshots page? 2011-04-30T15:04:04 to see what it looks like 2011-04-30T15:04:14 i'm a fan of simple, but not if it looks awful 2011-04-30T15:04:21 theres no look 2011-04-30T15:04:28 what do u mean how does it look like? 2011-04-30T15:04:35 its not a girl 2011-04-30T15:04:35 oh, it's all command line? 2011-04-30T15:04:37 its lego 2011-04-30T15:04:45 hmm 2011-04-30T15:04:51 screw that :P 2011-04-30T15:05:00 u can have everything or nothing 2011-04-30T15:05:01 sooonot not exactly "user-friendly?" 2011-04-30T15:05:09 the core install is done in a couple of minutes 2011-04-30T15:05:09 -not 2011-04-30T15:05:17 In play_one_game.cmd - What do I need to write, instead of "python" to indicate my bot is "csharp" ? 2011-04-30T15:05:20 not click and install 2011-04-30T15:05:34 thats why i asked how much do u know :) 2011-04-30T15:05:40 you still use python 2011-04-30T15:05:45 python runs the engine 2011-04-30T15:05:54 MuTa_ofd00m: try just removing the python and giving the location of the bot file 2011-04-30T15:06:00 is C# compiled? 2011-04-30T15:06:01 just use the path to the exe 2011-04-30T15:06:04 that's what i do for my c++ bots 2011-04-30T15:06:14 i assume it is 2011-04-30T15:07:34 so to wet your apetite 2011-04-30T15:07:44 there are lots of packages 2011-04-30T15:07:48 whet? 2011-04-30T15:07:55 wet 2011-04-30T15:08:07 no? 2011-04-30T15:08:09 wetting 2011-04-30T15:08:43 so in aur there are god know how many packages 2011-04-30T15:08:52 its a user maintained repositry 2011-04-30T15:08:56 you whet your appetite :P 2011-04-30T15:09:02 So I saw 2011-04-30T15:09:02 whet 2011-04-30T15:09:05 i dont care 2011-04-30T15:09:07 It's quite large 2011-04-30T15:09:08 :) 2011-04-30T15:09:17 and it a PKGBUILD file 2011-04-30T15:09:19 its 2011-04-30T15:09:29 instructions on how to build that package 2011-04-30T15:09:42 it downloads the source code for u 2011-04-30T15:09:46 configures makes 2011-04-30T15:09:50 anything 2011-04-30T15:09:58 can use bzr cvs git 2011-04-30T15:10:01 whateveru want 2011-04-30T15:10:07 its quite nice 2011-04-30T15:10:20 so if some package isnt officially supported 2011-04-30T15:10:36 theres a good chance u can find its pkgbuild in aur 2011-04-30T15:10:50 if you install yaourt beside pacman 2011-04-30T15:11:02 then its as easy as pressing enter 2011-04-30T15:11:21 and the offical packages are built through ABS 2011-04-30T15:11:27 arch build system 2011-04-30T15:11:39 basicall the collection af similar PKGBUILD files 2011-04-30T15:12:01 so if u want to recompile something u download its pkgbuild and off you go 2011-04-30T15:12:11 makepkg -s does everything 2011-04-30T15:12:26 *** Bugg has joined #aichallenge 2011-04-30T15:12:26 and it has bsd style init scripts 2011-04-30T15:12:44 not those 20daemon 40daemon2 like in ubuntu or whatever 2011-04-30T15:13:04 so its a breeze 2011-04-30T15:13:24 and u know about rolling release? 2011-04-30T15:13:42 u always have the latest software if you want 2011-04-30T15:13:49 but its not unstable 2011-04-30T15:13:52 yes 2011-04-30T15:14:04 for example i have kde 4.6.2 2011-04-30T15:14:29 and i messed in kde source several times 2011-04-30T15:14:45 and its just easy when u have a solid build system 2011-04-30T15:15:35 but in the end its just linux 2011-04-30T15:15:52 99% the same code as almost any other distro 2011-04-30T15:16:10 its not like switching to bsd :) 2011-04-30T15:16:34 my god, I did not know... http://upload.wikimedia.org/wikipedia/commons/8/8c/Gldt.svg 2011-04-30T15:17:11 *** Cerealklr has joined #aichallenge 2011-04-30T15:17:14 *** Cerealklr has left #aichallenge 2011-04-30T15:17:33 what? 2011-04-30T15:17:53 nice picture but what should i be looking at? 2011-04-30T15:17:58 lol 2011-04-30T15:17:58 lindows 2011-04-30T15:18:09 yeah thats a joke 2011-04-30T15:18:14 i thnk its not free 2011-04-30T15:18:24 like mandrake 2011-04-30T15:19:28 I didn't know there were so many distributions :/ 2011-04-30T15:19:37 thats what i wanted to ask 2011-04-30T15:19:46 if that what u were surprised t 2011-04-30T15:19:53 at 2011-04-30T15:20:00 so 2011-04-30T15:20:08 obviously my search is not complete 2011-04-30T15:20:15 but it completely stopped :) 2011-04-30T15:22:54 jaycobcoleman: what wm do u use? 2011-04-30T15:23:43 GNOME 2011-04-30T15:24:02 ok 2011-04-30T15:24:06 then nothing 2011-04-30T15:24:12 why's that? 2011-04-30T15:24:33 i use kde and wanted to share some tips 2011-04-30T15:25:06 oh, yeah, my Ubuntu install is only about a month old 2011-04-30T15:25:12 maybe a little more 2011-04-30T15:25:18 so I haven't changed much 2011-04-30T15:25:20 (but trying to migrate to a tiling window manager) 2011-04-30T15:25:44 u mean ur dealing with linux for a month? 2011-04-30T15:26:10 I don 2011-04-30T15:26:17 *'t know what you're asking 2011-04-30T15:26:35 since how long do u use linux? 2011-04-30T15:27:17 I used Ubuntu in a virtual machine on a mac for a couple weeks before that, but that's about it 2011-04-30T15:27:31 ok 2011-04-30T15:27:44 well then i wont recommend arch to u 2011-04-30T15:27:50 i mean try it if you want 2011-04-30T15:27:54 but be prepared 2011-04-30T15:27:57 :) 2011-04-30T15:28:47 :) I've spent a pretty good amount of time in terminal. I think I can manage :P 2011-04-30T15:29:03 alright 2011-04-30T15:31:35 So many issues left :( 2011-04-30T15:32:38 amstan: how are we doing for IE on the website? 2011-04-30T15:32:57 *** _0x47 <_0x47!~slash@p5B25E644.dip.t-dialin.net> has quit IRC (Remote host closed the connection) 2011-04-30T15:35:39 *** _0x47 <_0x47!~slash@p5B25E644.dip.t-dialin.net> has joined #aichallenge 2011-04-30T15:44:07 *** Accoun has quit IRC () 2011-04-30T15:46:12 *** _0x47 <_0x47!~slash@p5B25E644.dip.t-dialin.net> has quit IRC (Read error: Connection reset by peer) 2011-04-30T15:54:59 *** Accoun has joined #aichallenge 2011-04-30T16:00:46 *** Naktibalda has quit IRC (Quit: ChatZilla 0.9.86.1 [Firefox 4.0.1/20110422205559]) 2011-04-30T16:01:05 I'm not sure what versions of IE we're looking to support 2011-04-30T16:05:10 *** berak has quit IRC (Quit: ChatZilla 0.9.84 [SeaMonkey 2.0a3/20090223135443]) 2011-04-30T16:30:22 *** amriedle has quit IRC (Quit: Lost terminal) 2011-04-30T16:30:59 *** kaemo has quit IRC (Ping timeout: 258 seconds) 2011-04-30T16:32:16 *** jaycobcoleman has quit IRC (Ping timeout: 252 seconds) 2011-04-30T16:34:49 *** kaemo has joined #aichallenge 2011-04-30T16:47:37 *** amstan_ has quit IRC (Ping timeout: 250 seconds) 2011-04-30T16:48:00 *** amstan has quit IRC (Ping timeout: 260 seconds) 2011-04-30T16:55:40 *** kaemo has quit IRC (Ping timeout: 248 seconds) 2011-04-30T16:56:25 *** kaemo has joined #aichallenge 2011-04-30T17:02:52 *** kaemo has quit IRC (Quit: *puff*) 2011-04-30T17:34:34 IE users make up a little under 6% of visitors to ai-contest.com. IE <= 8 is ~4.5%, <= 7 is ~1% 2011-04-30T17:35:39 from that I would say IE8+ is probably worth supporting, IE7- may or may not depending on the effort required 2011-04-30T17:36:50 You mean IE >= 8, right? 2011-04-30T17:37:06 yes, 8 plus :) 2011-04-30T17:37:56 also of note is Opera has ~6% so should be as important to us as IE 2011-04-30T17:38:14 and of course much easier to support 2011-04-30T17:45:21 *** jaycobcoleman has joined #aichallenge 2011-04-30T17:46:41 I thought Opera support was already taken care of 2011-04-30T17:47:26 so far as I know the visualizer was tested on it, but I've not heard of anyone checking the new site design on it 2011-04-30T17:47:41 doesn't mean it hasn't been done of course :) 2011-04-30T17:48:19 just thought I'd mention that from number of users it should be at least as large a concern as IE for us 2011-04-30T17:48:20 *** Alex3011 has joined #aichallenge 2011-04-30T17:49:42 *** Alex3011 has quit IRC (Client Quit) 2011-04-30T17:51:57 Yup, opera works perfectly. I'll try to figure out what IE8 has problems with later :P 2011-04-30T17:54:06 *** jaycobcoleman has quit IRC (Quit: Page closed) 2011-04-30T17:54:52 I asked earlier but didnt resolve the problem: In play_one_game.cmd, im trying to run the C# starterpackage, what should I write? I tried linking directly to 'dist\starter_bots\c#\CSStarterPackage\bin\Release\CSStarterPackage.exe' but the bot times out? 2011-04-30T18:14:35 *** sigh has joined #aichallenge 2011-04-30T18:26:23 when i do that it works for me 2011-04-30T18:26:38 "path/to/my/bot/binary" 2011-04-30T18:26:47 What do you write in front of the path? 2011-04-30T18:26:53 nothing 2011-04-30T18:26:53 hmm binary, is that the object? 2011-04-30T18:27:05 in windows its the exe 2011-04-30T18:27:25 ok so it should be right 2011-04-30T18:27:26 when u run it in cmd 2011-04-30T18:27:35 does it accept text? 2011-04-30T18:27:40 yep 2011-04-30T18:27:48 and it says go 2011-04-30T18:27:50 back 2011-04-30T18:28:00 dunno, what should I write for it to say go? 2011-04-30T18:28:08 anything 2011-04-30T18:28:11 a newline 2011-04-30T18:28:22 no reply 2011-04-30T18:28:39 thats bad 2011-04-30T18:28:57 well write something 2011-04-30T18:29:04 not juts a newline sry 2011-04-30T18:29:12 still the same? 2011-04-30T18:29:25 yep 2011-04-30T18:29:38 then its some IO problem 2011-04-30T18:29:54 i cant give advice on that part 2011-04-30T18:30:02 only that dont forget to flush 2011-04-30T18:30:06 if IO id buffered 2011-04-30T18:30:08 is 2011-04-30T18:30:10 the starter doesn't need a 'ready' command to signal the end of the pregame information? 2011-04-30T18:30:13 if I write 'go' it says 'go' 2011-04-30T18:30:49 the hell knows rly 2011-04-30T18:31:00 im not there yet 2011-04-30T18:31:02 :) 2011-04-30T18:31:27 i know that the common lisp starter bot says go after anything i give to him 2011-04-30T18:31:43 and it work when run by player.py 2011-04-30T18:31:49 or whatever thats called 2011-04-30T18:32:09 playgame.py 2011-04-30T18:32:59 I've made it reply 'go' to anything now by adding "game.FinishTurn();" to the else {} 2011-04-30T18:33:04 Lets see what hapepns 2011-04-30T18:33:34 nope still error 2011-04-30T18:33:59 cpp bot works for me too 2011-04-30T18:34:02 just compiled it 2011-04-30T18:34:13 take a look at that 2011-04-30T18:34:43 I'll try cpp 1 sec 2011-04-30T18:35:33 ah cant, only have visual studio C# 2011-04-30T18:35:44 :) 2011-04-30T18:35:46 lol 2011-04-30T18:35:51 i feel sry for u 2011-04-30T18:36:27 is the path correct 2011-04-30T18:36:29 ? 2011-04-30T18:36:41 theres an # in it 2011-04-30T18:36:55 and u dont use double quotes 2011-04-30T18:36:55 I changed it from C# to csharp 2011-04-30T18:37:10 "dist/starter_bots/csharp/CSStarterPackage/CSStarterPackage.exe 2011-04-30T18:37:15 there's an " after it 2011-04-30T18:37:21 aha 2011-04-30T18:37:31 what is the error? 2011-04-30T18:37:35 simple timeout? 2011-04-30T18:37:45 *** amstan has joined #aichallenge 2011-04-30T18:37:45 *** ChanServ sets mode: +o amstan 2011-04-30T18:37:57 omg 2011-04-30T18:37:59 @seen mcleopold 2011-04-30T18:37:59 amstan: mcleopold was last seen in #aichallenge 23 hours and 25 seconds ago: antimatroid: because 12 > 11. It should be 0pm. 2011-04-30T18:37:59 I forgot bin/release/ 2011-04-30T18:38:01 in the dir! 2011-04-30T18:38:09 it works now, jezz my bad >.< 2011-04-30T18:38:13 lol 2011-04-30T18:38:20 alright 2011-04-30T18:41:34 well it works and it runs the code, but the bot does nothing, getting closer at least :P 2011-04-30T18:50:15 i think it a north bot 2011-04-30T18:50:18 isnt it? 2011-04-30T18:50:37 nope it's moving towards closest food, else closest enemy 2011-04-30T18:50:40 the rule is that it goest N if it can otherwise stay put 2011-04-30T18:51:09 it spawns with no water around it, and it does nothing 2011-04-30T18:51:24 well at least there is some room for imporvement 2011-04-30T18:51:48 I cant get it to do anything, the game engine says that the bot times out 2011-04-30T18:51:53 so I think it forgets to reply some info 2011-04-30T18:52:08 compare it with the cpp bot 2011-04-30T18:52:09 "bot 9 timed out" 2011-04-30T18:52:19 yeah good idea 2011-04-30T18:56:11 no luck, they're quite different 2011-04-30T19:00:28 mmmmm imma give up for now, I dont think the starterpackage has been developed enough 2011-04-30T19:00:42 *** Eruonen has quit IRC () 2011-04-30T19:02:50 *** Eruonen has joined #aichallenge 2011-04-30T19:03:26 where is the c# starter? 2011-04-30T19:03:54 https://github.com/jmcgonegal/aichallenge/commit/85ccd3ab669cc47b1fe63317b60870b0e4ded801 2011-04-30T19:04:11 sry here is correct link: https://github.com/jmcgonegal/aichallenge 2011-04-30T19:05:01 thx 2011-04-30T19:05:09 never seen c# code before 2011-04-30T19:06:22 it's not very different from c++/java 2011-04-30T19:07:06 different stl 2011-04-30T19:07:13 or standard library 2011-04-30T19:07:46 csproj? 2011-04-30T19:07:50 yep main file 2011-04-30T19:07:52 whats that? 2011-04-30T19:07:56 package definition? 2011-04-30T19:08:04 project solution 2011-04-30T19:08:07 is it a visual shat file ? 2011-04-30T19:08:55 no idea what u are talking about, but if u open .csproj u open the entire project in visual studio and u get the files/etc in the menu list 2011-04-30T19:09:08 rdy to compile -> f6 2011-04-30T19:09:18 yep that what i said 2011-04-30T19:09:22 MS Visual Shat 2011-04-30T19:09:29 kk 2011-04-30T19:09:29 :) 2011-04-30T19:09:42 what the comma for? private int[,] DIRECTIONS 2011-04-30T19:09:57 oh 2011-04-30T19:10:01 its a double list 2011-04-30T19:10:24 you can index lists like ls[x,y] 2011-04-30T19:10:26 ? 2011-04-30T19:10:42 not sure 2011-04-30T19:12:25 class Ant : Location 2011-04-30T19:12:31 is this inheritance? 2011-04-30T19:12:37 ant inherits location? 2011-04-30T19:14:04 can I fold :D 2011-04-30T19:14:16 oh 2011-04-30T19:14:22 only check 2011-04-30T19:14:44 well ok 2011-04-30T19:14:58 i wont ask more stupid questions then :) 2011-04-30T19:15:13 not stupid, I just dont know the code :D 2011-04-30T19:15:21 ok 2011-04-30T19:29:59 does your C# solution move? 2011-04-30T19:46:41 mine? 2011-04-30T19:46:55 i dont have any 2011-04-30T19:47:25 but for the fun of it i will try to compile it 2011-04-30T19:47:31 im on linux 2011-04-30T19:47:35 oh I thought u were trying the C# startpackage 2011-04-30T19:47:39 lets see if mono work 2011-04-30T19:47:42 oh no 2011-04-30T19:47:52 i said i never saw csharp code before 2011-04-30T19:48:26 ye, but u asked for the c# starter etc :P 2011-04-30T19:48:33 so I figured u'd give it a try 2011-04-30T19:48:35 i was just interested 2011-04-30T19:48:46 i see :) 2011-04-30T19:52:16 *** delt0r___ has joined #aichallenge 2011-04-30T19:52:43 *** needsch has quit IRC (Quit: Leaving.) 2011-04-30T19:52:58 *** needsch has joined #aichallenge 2011-04-30T19:54:05 *** delt0r has quit IRC (Ping timeout: 276 seconds) 2011-04-30T19:56:53 *** amstan has quit IRC (Ping timeout: 276 seconds) 2011-04-30T19:57:35 *** needsch has quit IRC (Ping timeout: 260 seconds) 2011-04-30T20:12:12 *** Kingpin13 has quit IRC (Quit: money can't buy you love, but it can if you have enough) 2011-04-30T20:21:42 *** Chris_0076 has joined #aichallenge 2011-04-30T20:22:21 *** mceier has quit IRC (Quit: leaving) 2011-04-30T20:26:03 *** mcstar has quit IRC (Remote host closed the connection) 2011-04-30T20:40:49 *** Accoun has quit IRC (Ping timeout: 246 seconds) 2011-04-30T20:41:27 *** Accoun has joined #aichallenge 2011-04-30T20:42:22 *** amstan has joined #aichallenge 2011-04-30T20:42:22 *** ChanServ sets mode: +o amstan 2011-04-30T20:42:45 *** Bugg has quit IRC (Quit: Page closed) 2011-04-30T20:43:02 *** chris__0076 has joined #aichallenge 2011-04-30T20:43:09 *** amstan_ has joined #aichallenge 2011-04-30T20:43:09 *** ChanServ sets mode: +o amstan_ 2011-04-30T20:45:31 *** Chris_0076 has quit IRC (Ping timeout: 250 seconds) 2011-04-30T20:50:40 *** drhodes has left #aichallenge 2011-04-30T20:57:21 *** contestbot has joined #aichallenge 2011-04-30T20:57:29 *** amstan_ has joined #aichallenge 2011-04-30T20:57:29 *** ChanServ sets mode: +o amstan_ 2011-04-30T20:57:33 *** amstan has joined #aichallenge 2011-04-30T20:57:33 *** ChanServ sets mode: +o amstan 2011-04-30T21:31:18 *** Cerealklr has joined #aichallenge 2011-04-30T21:31:23 *** Cerealklr has left #aichallenge 2011-04-30T21:35:32 *** jaycobcoleman has joined #aichallenge 2011-04-30T21:41:43 *** sigh has quit IRC (Ping timeout: 246 seconds) 2011-04-30T21:42:59 *** sigh has joined #aichallenge 2011-04-30T22:30:25 *** amstan has quit IRC (Ping timeout: 260 seconds) 2011-04-30T22:33:17 *** chris__0076 has joined #aichallenge 2011-04-30T22:36:39 *** chris___0076 has quit IRC (Read error: Connection reset by peer) 2011-04-30T22:57:16 *** smiley1983 has quit IRC (Ping timeout: 260 seconds) 2011-04-30T22:58:58 *** smiley1983 has joined #aichallenge 2011-04-30T23:25:54 *** jaycobcoleman has quit IRC (Quit: Page closed) 2011-04-30T23:49:55 evening 2011-04-30T23:58:41 evening 2011-04-30T23:58:48 *** amstan_ is now known as amstan