2010-12-17T00:02:41 hmm... actually might be able to set it to nonblocking, but definitely blocking by default 2010-12-17T00:10:06 *** boegel has joined #aichallenge 2010-12-17T00:10:22 sigh: yep, but my bot should be sending back that its done, it's weird 2010-12-17T00:10:34 the first bot does, but the second doesn't, even though it's exactly the same file 2010-12-17T00:11:15 *** yasith has joined #aichallenge 2010-12-17T00:11:24 i'm going to have to work out how to know when there is something to collect later, i just want to get the bot running, once i fix these issues up i can just a version of games being debugged to file 2010-12-17T00:11:32 do me a favor... set the fgets buffer size to 1 character and tell me what it does? 2010-12-17T00:11:36 just watch* 2010-12-17T00:12:39 lol, if i do that it works :P 2010-12-17T00:12:45 well, gets further than before anyway 2010-12-17T00:12:55 what caused that? 2010-12-17T00:13:30 well, I'm guessing that fgets blocked waiting for more data because the buffer wasn't full yet 2010-12-17T00:13:40 *** tapwater has quit IRC (Quit: tapwater) 2010-12-17T00:13:41 which is the problem with using non-blocking IO 2010-12-17T00:13:45 *blocking IO 2010-12-17T00:15:20 hmm... are you sure your bot is sending newlines after the done command? 2010-12-17T00:15:37 positive 2010-12-17T00:15:54 does "go \n" do what i think? 2010-12-17T00:16:02 yeah 2010-12-17T00:16:03 to do i need to add '\n' explicitly 2010-12-17T00:16:08 or*, yeah okay 2010-12-17T00:17:47 you said it got further... did it stall later down the track? 2010-12-17T00:19:32 reading the docs for fgets... a buffer size of 1 should be useless, I don't know why it worked o_O 2010-12-17T00:20:43 i have lots of stuff commented out 2010-12-17T00:20:51 it went all the way through the game and crashed at the end 2010-12-17T00:21:03 but it's not doing much past passing start information 2010-12-17T00:21:15 i see 2010-12-17T00:22:15 it seems to pass perfectly fine with buffer of 1? 2010-12-17T00:22:39 can you paste the exact fgets line? 2010-12-17T00:22:50 Process& operator>>(Process &p, stringstream &instream) 2010-12-17T00:22:50 { 2010-12-17T00:22:50 char line[1]; 2010-12-17T00:22:50 fgets(line, 1, p.in); 2010-12-17T00:22:50 instream << line; 2010-12-17T00:22:50 return p; 2010-12-17T00:22:50 }; 2010-12-17T00:23:20 yeah... this makes no sense... the only valid string you can get out of that is "\0" 2010-12-17T00:23:32 or at least, the only valid string you *should* be able to get 2010-12-17T00:23:46 make the buffer size = 2 and tell me if it still works 2010-12-17T00:23:56 well my bots correctly outputed the input parameters passed to them 2010-12-17T00:24:07 including maxTurns of 200 and maxTime of 1.1 2010-12-17T00:24:35 I thought this code was in the engine 2010-12-17T00:24:49 is your engine getting the correct data from the bot? 2010-12-17T00:24:53 it is, but it uses that code to pass the bots the information 2010-12-17T00:24:58 oh err 2010-12-17T00:25:03 hang on 2010-12-17T00:25:36 fgets is the reading part, so got nothing to do with what is written to the agents 2010-12-17T00:26:54 yeah, it's not reading stuff properly 2010-12-17T00:27:07 hmmm... 2010-12-17T00:27:18 ok, set the buffer size to 2 2010-12-17T00:27:58 i got 'g' from them with bsize = 2 2010-12-17T00:29:09 do i see \n when i get information from them? 2010-12-17T00:29:21 if so i can probably keep getting a single character until that? 2010-12-17T00:29:51 ahh... yes you can... I thought you were already doing that in the 50 case 2010-12-17T00:30:22 don't worry about the loop[ 2010-12-17T00:30:30 but the buffer back to 50 (or whatever) 2010-12-17T00:30:44 but just make sure you stop asking the bot for data after it says go 2010-12-17T00:31:19 if line == "go" return p? 2010-12-17T00:31:47 errr... can you pastebin the code that you have atm? 2010-12-17T00:32:08 hmm, there's quite a few files :P 2010-12-17T00:32:21 this is why you should set up git 2010-12-17T00:32:27 or equivelent 2010-12-17T00:32:56 i should be able to do this though i think, i'll try and see 2010-12-17T00:33:49 I suspect the engine was stalling when you had buffer = 50 because you were asking the agent for data when it didn't have any available 2010-12-17T00:33:51 antimatroid: I'll try to implement what you posted earlier. I don't think I can get the ants to move smoothly though, now that movement is only 1/3 of the turn 2010-12-17T00:34:53 I think you should be able to implement what you need just fine with blocking IO (although it wont be robust against bad bots), but don't quote me on that 2010-12-17T00:35:59 Frontier: I was thinking maybe if the visualiser had 3 steps per turn, even when replaying? 2010-12-17T00:36:22 so you can also sort of see what happened each turn if you go back through it 2010-12-17T00:36:45 You either go with non-blocking io, or - what many programmers prefer - multiple threads with blocking io. As threads seem to be easier to handle than non-blocking io. 2010-12-17T00:36:49 sigh: my engine is terrible, but it will do as an example 2010-12-17T00:36:52 I don't know if this will be helpful, but the planet wars canvas visualiser's move number variable was a float, and it used that to interpolate fleet positions 2010-12-17T00:37:31 i don't think that'll help for this 2010-12-17T00:37:36 antimatroid: I understand that I should implement 3 separate states. 2010-12-17T00:38:26 Frontier: can you also return me go once you have processed each round? 2010-12-17T00:38:28 sigh: with three states and only one of them concerning movement, there will be some time inbetween where no ant moves (attack for example) 2010-12-17T00:39:16 antimatroid: The visualizer is a passive JavaScript that just receives some string out of a database ... isn't it Oo? 2010-12-17T00:40:24 Frontier: i have no idea? :P 2010-12-17T00:40:42 do i run the visualiser as a process of its own or not? 2010-12-17T00:40:51 i can just give you a .txt file at the end if you like 2010-12-17T00:41:23 Frontier: the string from the database is part of the generated page. 2010-12-17T00:41:27 anyway I implemented your map generator in JavaScript for testing and found some bugs. When you use %, remember to make the first value positive (x - 1) % dim will probably not work, while (x - 1 + dim) % dim is sure to work. Don't know if Python handles that differently. There was something else, I don't remember. 2010-12-17T00:41:43 (at least in pw) 2010-12-17T00:41:57 Frontier: ah yeah, i'll fix those up later, cheers 2010-12-17T00:42:28 sigh: I know, yet the visualizer can be in a separate .js file to abstract the viszalization engine and the web site a little. 2010-12-17T00:42:39 Frontier: python will work correctly with (x - 1) % dim, most other languages not 2010-12-17T00:43:08 python's result takes the sign from dim, but languages like c++ take it from the (x-1) 2010-12-17T00:43:29 sigh: thx for the information. it would have been odd if it worked at all if python didn't handle that correctly 2010-12-17T00:43:52 python's is mathematically nicer, but c++ is faithful to x86 2010-12-17T00:44:29 ah I'm technically wrong... c++ is implementation dependent... which on x86 boils down to what I just said :P 2010-12-17T00:46:46 i see. thought there was some mathematical consens on how that case has to be handled. 2010-12-17T00:47:47 across programming languages, it's all over the place... no consistency at all :P 2010-12-17T00:48:36 this page has a useful summary: http://en.wikipedia.org/wiki/Modulo_operation#Remainder_calculation_for_the_modulo_operation 2010-12-17T00:48:51 in the right-hand table 2010-12-17T00:48:54 hmmm getting input is tricky now 2010-12-17T00:49:04 i don't like processing input character at a time 2010-12-17T00:49:21 you should be able to set the buffer back up to something reasonable 2010-12-17T00:49:34 wont it always have that stall issue? 2010-12-17T00:49:36 sigh: This detailed information is just a bit too much for me today :p. 2010-12-17T00:49:38 So the three parts of a turn are: 1(attack-harvest-deposit) 2(walk) 3(spawn-collision fight-fusion) 2010-12-17T00:50:08 antimatroid: only if you ask the bot for data when it doesn't have any 2010-12-17T00:50:20 i wan't before 2010-12-17T00:50:28 other than not 50 chars worth or whatever 2010-12-17T00:51:09 put it up to 50, and tell me what it reads before it stalls 2010-12-17T00:51:25 antimatroid: what exactly does 'fusion' mean? What should I draw on the map there? 2010-12-17T00:52:30 can an ant attack and move in one turn? 2010-12-17T00:52:32 fusion and collision fight all happen together, if ants are on the same square, sum hp for both players, do hp1-hp2, then hp < == or > 0 gives which player the ant is, if hp != 0 then antgrains = sumgrains1 + sumgrains2 else square grains += that, but you don't need to process anything like that, i do it all for you 2010-12-17T00:52:54 Frontier: no, only the last "move" issued for an ant will be processed 2010-12-17T00:53:49 that's good to know. so I can abuse the 'attack' time for rotation of ants that move to another square 2010-12-17T00:56:34 antimatroid: So you can have any number of friendly and enemy ants on one square (max 5 I guess) and the weaker stack dies. 2010-12-17T00:56:57 Frontier: yep 2010-12-17T00:57:16 so each player can only ever see one ant per square, but ants can move and land on the same square as any other ant 2010-12-17T00:57:30 and it's really not a good idea to merge your own ants, except for a few cases i think 2010-12-17T00:57:46 particularly when the win condition at the end is the player with the most ants 2010-12-17T00:57:48 *** Mathnerd314 has quit IRC (Quit: ChatZilla 0.9.86-rdmsoft [XULRunner 1.9.2.12/20101026210630]) 2010-12-17T00:57:48 is hp still capped? 2010-12-17T00:57:53 no 2010-12-17T00:57:57 but you start with 5hp 2010-12-17T00:57:59 you can make mega ants 2010-12-17T00:58:17 uh, ok. there goes my health bar :p 2010-12-17T00:58:27 hmmm yeah, sorry :P 2010-12-17T00:58:30 *** Rubicon-|-Cross has quit IRC (Quit: Leaving) 2010-12-17T00:58:31 i think it's better this way though 2010-12-17T00:59:21 I'll try to increase an ant's size according to their hp. That should look nice, too. 2010-12-17T01:00:07 yep, possibly just give it a max size 2010-12-17T01:00:14 Unless someone thinks he's super cool and merges ants up to 100 hp :p 2010-12-17T01:00:23 ^^ :P 2010-12-17T01:00:57 It's like some godzilla ant drawn twice over the infinite board 2010-12-17T01:01:08 scale the size logarithmically... that should stop it getting too out of control 2010-12-17T01:01:10 that's actually maybe be cool to leave 2010-12-17T01:01:24 watch a big ass ant trot around 2010-12-17T01:02:07 be a man, use an exponential scale :P 2010-12-17T01:02:25 *** boegel has quit IRC (Quit: Leaving) 2010-12-17T01:02:35 ok, my exponent will be ... -3 2010-12-17T01:02:42 no wait... 1/3 2010-12-17T01:03:28 no no, the exponent should be i*pi/5 2010-12-17T01:03:31 that makes the volume of the ant scale proportinal to hp 2010-12-17T01:03:35 i have left walls as see through, i think it makes the game cooler 2010-12-17T01:03:35 hp makes the ants rotate :P 2010-12-17T01:03:43 but the way i've written it, it's like a line or two to change 2010-12-17T01:04:13 what's the rational behind see through walls? 2010-12-17T01:04:26 so you can see something attractive you can't actually reach 2010-12-17T01:04:33 should i replace the walls with glass blocks now ^ 2010-12-17T01:04:44 Frontier: they're robot ants, no need 2010-12-17T01:05:20 @ranbow a rainbow map would be cool though 2010-12-17T01:05:21 antimatroid: An error occured while trying to show the previous error. 2010-12-17T01:05:23 fuck 2010-12-17T01:05:28 @rainbow a rainbow map would be cool though 2010-12-17T01:05:28 antimatroid: a rainbow map would be cool though 2010-12-17T01:06:12 antimatroid: join up with the creator of this map please: http://www.youtube.com/watch?v=-9daJWnADjw 2010-12-17T01:06:40 amstan: i should be able to message rainbow bot @rainbow whatever, then have him repeat "antimatroid: rainbow message" in the main chan :) 2010-12-17T01:06:48 It's so funny to see people rage on a map like that. 2010-12-17T01:07:29 "Ignore the rainbow! It's supposed to be fun." lol 2010-12-17T01:08:13 antimatroid: that's what the "say" command is for, but amstan disabled it 2010-12-17T01:08:14 this is not supposed to be fun 2010-12-17T01:08:30 antimatroid: I'm referring to the video 2010-12-17T01:08:46 Frontier: so am i :P 2010-12-17T01:10:37 *** ebrahim has joined #aichallenge 2010-12-17T01:10:49 That guy also created a map called octagons which consosted of hexagons and poeple discussed over and over what kind of geometry it is. 2010-12-17T01:11:03 people will be people 2010-12-17T01:11:07 (idiots) 2010-12-17T01:11:37 It turned out that the geometry might be octahedrons 2010-12-17T01:12:07 but then they started to discuss the grammer of the word. is it written with c or with k? 2010-12-17T01:16:12 oh, one problem i did have with the visualiser was i wasn't translating home coordinated, but i fixed that yesterday 2010-12-17T01:16:21 visauliser == mapgen* aha 2010-12-17T01:16:38 antimatroid: I'm also currently using a JSON syntax for the data, but don't bother about that too much. 2010-12-17T01:17:28 oh well, I fixed that too when I was wondering why the ants spawn on walls ^^ 2010-12-17T01:18:50 :) 2010-12-17T01:19:20 what do you think about the home bases having their own little clearings? 2010-12-17T01:19:25 with no grains in the clearings too 2010-12-17T01:19:38 In the JavaScript version I also moved all global vars into the generate function, because I don't know if there is something like namespaces in JavaScript. Once the server engine is running I'll just use the "real" replays though. 2010-12-17T01:20:15 antimatroid: First of all it makes the starting positions more recognizable and the map more interesting to look at. 2010-12-17T01:20:44 yep, i think it makes it easier to deal with depositing and moving spawned ants without merging etc. 2010-12-17T01:20:48 I don't know if it is good or bad for the game. Depends on wether walls are see-through or not i guess 2010-12-17T01:20:55 i think it's good for the game 2010-12-17T01:21:39 i can make it so that nothing but blank squares are seen on the first turn 2010-12-17T01:21:41 that might be cool 2010-12-17T01:21:47 and home 2010-12-17T01:21:54 You know, when the whole ant idea first came up I thought there would be just a few longer walls (maybe 4 squares) scattered around the map. 2010-12-17T01:22:15 i prefer this map gen, it's more ant trail like 2010-12-17T01:22:34 and with the home hole game ending objective i'm imposing 2010-12-17T01:22:41 it makes it very important to defend your home hole 2010-12-17T01:22:55 and to seek out theirs 2010-12-17T01:23:58 Is it feasable to have several types of maps. Some with fewer walls and few grain locations, but high concentration for example. I would like to see how that turns out. 2010-12-17T01:25:36 (now for the snarky comment out of nowhere) I don't understand why you call them walls though, things I step into only to end up dead on top of, well, I've never heard called walls before :P 2010-12-17T01:26:53 *** CIA-24 has quit IRC (Ping timeout: 272 seconds) 2010-12-17T01:27:02 janzert: i removed that, i just ignore all invalid moves atm 2010-12-17T01:27:16 W.A.L.L.: Wreck an-ant laser line 2010-12-17T01:27:27 ahh, ok. that seems better to me 2010-12-17T01:28:13 a wall is basically what we did as children with the magnifying glass 2010-12-17T01:28:31 i can make the walls non-see through 2010-12-17T01:28:32 hehe, but then they have to wander around in little circles at least 2010-12-17T01:28:39 but i think it's more interesting if they are 2010-12-17T01:29:21 yeah I think, while maybe less realistic, see through probably works out better 2010-12-17T01:29:43 it's robot ants, the story can fit for anything 2010-12-17T01:30:20 the squad tactics game has 'hard' and 'soft' cover, how about two types of walls :D 2010-12-17T01:30:44 i think this game already has enough moves for people to consider 2010-12-17T01:30:57 as well as both combat strategy and a larger whole game strategy 2010-12-17T01:31:01 *** sigh has quit IRC (Quit: Leaving) 2010-12-17T01:31:48 i think the way i've got it set up now, it should be easy for people to get started, yet still alow huge amounts of room for the tp players to play around with different ideas 2010-12-17T01:32:09 which is what i see as the most important criteria 2010-12-17T01:32:46 i can change stuff if you really want though :) 2010-12-17T01:33:44 you should be ignoring me and making a good game ;) 2010-12-17T01:33:50 no, it's fine. the only think i would really like to see is different densities for walls and grains 2010-12-17T01:34:04 i had a possible idea for a different map gen 2010-12-17T01:34:40 randomly create 2 symmetrically placed "tron" snakes, keep them moving symmetrically and randoly until they can't move without cutting off a section of the board, repeat untill finished adding walls 2010-12-17T01:35:21 the biggest trouble with the planet wars map generator was that it had too few variations in the first version 2010-12-17T01:35:41 yep, i think it's better if a good map gen is provided right from the start with the next contest 2010-12-17T01:36:19 certainly at the very least the earlier the better 2010-12-17T01:36:33 try that snake idea. could become boring though if the walls are TOO long :p 2010-12-17T01:36:46 yes, i think a date should be specified right at the start by which time the only future changes need to be for actual problems 2010-12-17T01:39:58 do stringstreams not actually clear with clear()? 2010-12-17T01:42:12 it would appear not, alright, i can deal with that 2010-12-17T01:50:31 *** CIA-36 has joined #aichallenge 2010-12-17T01:57:21 wtf, in another place it seems to work 2010-12-17T01:57:24 i am so confused 2010-12-17T01:59:58 antimatroid: I don't want to disturb you, but what kind of field is 'baseNo'? 2010-12-17T02:00:29 integer 2010-12-17T02:00:32 0 1 or 2 2010-12-17T02:00:49 i can probably not pass that to you 2010-12-17T02:00:54 as i give it to you at the start 2010-12-17T02:01:23 Unless the bases can move or ants create new bases we can really drop that. 2010-12-17T02:01:29 they can't 2010-12-17T02:01:40 yep, assume it does'n't show up 2010-12-17T02:02:37 *** iNtERrUpT has quit IRC (Ping timeout: 240 seconds) 2010-12-17T02:03:25 i'm thinking maybe grain squares should be increased 2010-12-17T02:03:27 a square can 2010-12-17T02:03:33 't start with more than 9 grains though 2010-12-17T02:04:52 you can always extend to chineese numbers for 10 and up ;) 2010-12-17T02:05:28 or i could space row elements out 2010-12-17T02:11:41 antimatroid: one more thing - what about an initial game state? I would like to have something to start from, like a round3 for turn 0, that just places the starting ants. 2010-12-17T02:12:30 so i can use one piece of code for the turn transitions (interpolate last round3 and current round1) and the start of the game 2010-12-17T02:16:01 as an alternative I can just create a round3 object with one ant (5hp,0harvest) on each home base. 2010-12-17T02:22:16 *** boegel has joined #aichallenge 2010-12-17T02:25:07 *** mebrahim has joined #aichallenge 2010-12-17T02:25:08 *** mebrahim has joined #aichallenge 2010-12-17T02:25:19 *** ebrahim has quit IRC (Ping timeout: 264 seconds) 2010-12-17T02:26:35 Frontier: start of the game there are 9 grains on home and 1 ant there 2010-12-17T02:26:50 no other ants exist 2010-12-17T02:30:49 *** mebrahim has quit IRC (Remote host closed the connection) 2010-12-17T02:31:17 That sounds like something that can change in the future. And as we already have a format for round3, we can just use that. It's just 2 additional object rows 2010-12-17T02:32:42 (object in the JSON version, row in the plain text version) 2010-12-17T02:33:51 *** sigh has joined #aichallenge 2010-12-17T02:33:56 antimatroid: will the start grains be in the map data, or will the map data contain 'a' and 'b'? 2010-12-17T02:36:22 Frontier: grain data on map 2010-12-17T02:36:36 i'm feeding you base locations right at the start 2010-12-17T02:37:07 what if i feed you 2 rounds after the map 2010-12-17T02:37:13 giving locations of ants for each player 2010-12-17T02:39:28 That's how I like it. You can add more ants if you want, change hp give more than 9 starting grains etc... 2010-12-17T02:39:39 *** Palmik has joined #aichallenge 2010-12-17T02:43:51 Frontier: how about this... 2010-12-17T02:43:56 http://pastebin.com/p3gbSiRg 2010-12-17T02:44:10 antimatroid: I don't know in which direction the ants attack. The current format looks like it was written for a server engine, not for a visualizer :p 2010-12-17T02:44:27 Frontier: i am not sending you attack moves 2010-12-17T02:44:35 i don't see the point 2010-12-17T02:45:11 same as i don't send you harvest/depoist, i process them, then send you information at squares that changed 2010-12-17T02:45:14 the point is: i can only draw points 2010-12-17T02:45:42 were you planning to animate attack/harvest/deposit? 2010-12-17T02:45:53 i can send you each of those moves, but you'll get a lot more data 2010-12-17T02:46:02 if we want anything that looks logical on screen i need to turn the ants toward what they are attacking 2010-12-17T02:47:10 harvest takes place on the same square, right? I only need the direction for deposit and attack. No matter how much data that is. "N" "S" "E" "W" 2010-12-17T02:47:11 easy, add in a round before round 1 2010-12-17T02:48:39 do you want direction or destination coords for moves/rotations? 2010-12-17T02:49:44 *** fgump has joined #aichallenge 2010-12-17T02:49:51 direction, just because it is less data 2010-12-17T02:50:16 and... you can't do stupid things with it like use coordinates 10 squares away from the ant :p 2010-12-17T02:50:24 what about for walks? 2010-12-17T02:51:09 you can basically omit the destination. walk and attack are exclusive so I would know what to apply the direction to. 2010-12-17T02:51:47 http://pastebin.com/CJpQ3Jgz 2010-12-17T02:51:50 try that format 2010-12-17T02:51:55 (I'll pick the spawn heading for the ants at random) 2010-12-17T02:52:54 strip the direction from round3 (move) 2010-12-17T02:53:02 oh yep, easy 2010-12-17T02:54:14 I will build up a lot of data internally anyway (spawn fade in, death/fusion fade out, offset from origin square during movement, current angle, ...) 2010-12-17T02:54:32 do you want me to pass you spawn move? 2010-12-17T02:55:00 you could add it internally between round 2 and 3 if there is a grain on home base 2010-12-17T02:55:03 you could pass me a spawn direction if you want 2010-12-17T02:55:39 i think it's maybe easier to just have you add that one 2010-12-17T02:58:24 in your current specs you have round1 and round4. If I get a pair of those at start for each initial ant that would be good. 2010-12-17T02:59:11 say what? 2010-12-17T03:00:01 *** aerique has joined #aichallenge 2010-12-17T03:00:02 never mind. i forgot that you send the map data differently now and that it contains the ants positions 2010-12-17T03:06:48 *** Azrathud has joined #aichallenge 2010-12-17T03:07:14 I'll just make it smaller quickly... 1024 lines of "S row column isWall antPlayer antHP antHarvest squareGrains" will be like 18 KB. A PlanetWars map was 1/2 KB in size in the visualizer. 2010-12-17T03:09:08 hmmmm 2010-12-17T03:09:09 at this point in time, the size doesn't really matter too much 2010-12-17T03:09:42 the only reason I suggested a char array for the map the other day was because it also had the advantage of being human readable 2010-12-17T03:11:18 ok i wont change it 2010-12-17T03:12:05 i don't really see the point of it being human readable 2010-12-17T03:12:20 actually, i do want the start map to be human readable 2010-12-17T03:12:25 haha 2010-12-17T03:12:44 so a square cannot start with more than 9 grains 2010-12-17T03:12:55 trust me, human readable is always better (all else being equal) 2010-12-17T03:13:07 antimatroid: there are bases other than 10 :P 2010-12-17T03:13:34 yeah but then it gets messy 2010-12-17T03:14:30 I think for getting the concept up and running a max of 9 is fine 2010-12-17T03:14:33 anyway, i need to go and work something out for dinner 2010-12-17T03:14:34 but do whatever 2010-12-17T03:14:35 i'll be back later 2010-12-17T03:14:50 cu 2010-12-17T03:20:13 http://cable-lynx.net/botIdea.png 2010-12-17T03:21:04 thinking of how to rewrite my bot 2010-12-17T03:46:05 *** perror has joined #aichallenge 2010-12-17T04:28:38 *** javagamer has quit IRC (Ping timeout: 240 seconds) 2010-12-17T04:43:53 *** javagamer has joined #aichallenge 2010-12-17T05:16:31 *** Frontier has left #aichallenge 2010-12-17T05:17:16 *** Frontier has joined #aichallenge 2010-12-17T05:17:18 *** Frontier has left #aichallenge 2010-12-17T05:18:11 *** Frontier has joined #aichallenge 2010-12-17T05:42:06 *** Naktibalda has joined #aichallenge 2010-12-17T05:44:03 *** yasith has quit IRC (Ping timeout: 272 seconds) 2010-12-17T05:56:19 *** yasith has joined #aichallenge 2010-12-17T06:06:55 *** Migi32 has joined #aichallenge 2010-12-17T06:15:24 *** delt0r___ has joined #aichallenge 2010-12-17T06:16:39 *** delt0r_ has quit IRC (Ping timeout: 240 seconds) 2010-12-17T06:27:06 *** sigh has quit IRC (Remote host closed the connection) 2010-12-17T06:32:16 *** medrimonia has joined #aichallenge 2010-12-17T06:40:19 *** mega1 has joined #aichallenge 2010-12-17T07:04:40 *** Migi32 has quit IRC (Read error: Operation timed out) 2010-12-17T07:04:40 i'm slowly getting bugs removed 2010-12-17T07:04:58 the ants are starting to see their surroundings now 2010-12-17T07:06:49 God said, 'There shall be light,' and light came into existence. 2010-12-17T07:19:41 *** Migi32 has joined #aichallenge 2010-12-17T07:22:11 Frontier: what do you think about the ant game anyway? :P 2010-12-17T07:22:16 well, the concept of 2010-12-17T07:30:37 I'm writing a visualizer for it :), so I can't hat it too much. Recently I wasn't quite sure if I quite like it as I did when Mistmanov presented it. Maybe it is because of all the changes to the concept and that the ants are no separate processes or because the project is starting to be work. 2010-12-17T07:32:46 delt0rs specification does a pretty good job at modelling the individual processes 2010-12-17T07:33:12 i really like the home idea thing, but i played a game like that when i was younger 2010-12-17T07:33:16 i've forgotten the name 2010-12-17T07:33:57 it was a 3d game on playstation, where you made tanks and flying thing and claimed turrets and you had to get a tank into their base 2010-12-17T07:34:14 although you didn't get to control the tanks or flying things 2010-12-17T08:08:31 *** medrimonia1 has joined #aichallenge 2010-12-17T08:10:07 *** medrimonia has quit IRC (Ping timeout: 276 seconds) 2010-12-17T08:45:05 *** mceier has joined #aichallenge 2010-12-17T08:53:15 *** fgump has quit IRC (Quit: Leaving) 2010-12-17T09:17:03 *** medrimonia has joined #aichallenge 2010-12-17T09:17:28 *** mega1 has quit IRC (Read error: Operation timed out) 2010-12-17T09:18:33 *** mega1 has joined #aichallenge 2010-12-17T09:18:47 *** medrimonia1 has quit IRC (Ping timeout: 264 seconds) 2010-12-17T09:19:00 i haz a walking ant 2010-12-17T09:24:57 *** Palmik_ has joined #aichallenge 2010-12-17T09:25:17 *** Palmik has quit IRC (Read error: Connection reset by peer) 2010-12-17T09:32:09 *** amriedle has joined #aichallenge 2010-12-17T09:35:54 *** Azrathud has quit IRC (Quit: Leaving) 2010-12-17T09:36:43 *** Sylph2 has quit IRC (Read error: Connection reset by peer) 2010-12-17T09:36:50 *** amriedle has quit IRC (Read error: Operation timed out) 2010-12-17T09:36:55 *** delt0r has quit IRC (Ping timeout: 265 seconds) 2010-12-17T09:39:32 *** amriedle has joined #aichallenge 2010-12-17T09:40:36 *** mceier has quit IRC (Quit: leaving) 2010-12-17T09:43:03 *** delt0r has joined #aichallenge 2010-12-17T09:46:34 *** Sylph has joined #aichallenge 2010-12-17T09:49:00 *** antimatroid1 has joined #aichallenge 2010-12-17T09:50:08 *** antimatroid has quit IRC (Ping timeout: 255 seconds) 2010-12-17T09:53:30 *** antimatroid has joined #aichallenge 2010-12-17T09:53:49 *** antimatroid1 has quit IRC (Ping timeout: 265 seconds) 2010-12-17T10:03:33 *** Migi32 has quit IRC (Read error: Connection reset by peer) 2010-12-17T10:20:56 *** iNtERrUpT has joined #aichallenge 2010-12-17T10:21:37 *** Kingpin13 has quit IRC (Ping timeout: 276 seconds) 2010-12-17T10:23:23 *** Migi32 has joined #aichallenge 2010-12-17T10:24:56 *** JamesMG has joined #aichallenge 2010-12-17T10:50:40 *** aerique has quit IRC (Quit: ...) 2010-12-17T10:51:54 *** amstan has quit IRC (Ping timeout: 240 seconds) 2010-12-17T10:57:03 *** amriedle has quit IRC (Quit: Lost terminal) 2010-12-17T11:16:54 *** Kingpin13 has joined #aichallenge 2010-12-17T11:18:48 *** p4p4p4 has joined #aichallenge 2010-12-17T11:21:48 *** boegel has quit IRC (Ping timeout: 265 seconds) 2010-12-17T11:28:40 *** mceier has joined #aichallenge 2010-12-17T11:36:12 *** Accoun has quit IRC (Ping timeout: 276 seconds) 2010-12-17T11:41:16 *** Migi32 has quit IRC (*.net *.split) 2010-12-17T11:41:16 *** iNtERrUpT has quit IRC (*.net *.split) 2010-12-17T11:41:16 *** mega1 has quit IRC (*.net *.split) 2010-12-17T11:41:17 *** janzert has quit IRC (*.net *.split) 2010-12-17T11:41:17 *** holmar has quit IRC (*.net *.split) 2010-12-17T11:42:34 *** Accoun has joined #aichallenge 2010-12-17T11:43:05 *** Migi32 has joined #aichallenge 2010-12-17T11:43:05 *** iNtERrUpT has joined #aichallenge 2010-12-17T11:43:05 *** mega1 has joined #aichallenge 2010-12-17T11:43:05 *** janzert has joined #aichallenge 2010-12-17T11:43:06 *** holmar has joined #aichallenge 2010-12-17T11:47:42 *** p4p4p4_ has joined #aichallenge 2010-12-17T11:50:13 *** p4p4p4 has quit IRC (Ping timeout: 240 seconds) 2010-12-17T11:53:57 *** Rubicon-|-Cross has joined #aichallenge 2010-12-17T11:54:54 *** p4p4p4_ has quit IRC (Quit: ChatZilla 0.9.84 [SeaMonkey 2.0a3/20090223135443]) 2010-12-17T11:56:37 *** amstan_ has joined #aichallenge 2010-12-17T11:56:38 *** ChanServ sets mode: +o amstan_ 2010-12-17T11:58:42 *** Accoun has quit IRC (Ping timeout: 240 seconds) 2010-12-17T12:04:29 *** amstan_ has quit IRC (Ping timeout: 260 seconds) 2010-12-17T12:04:47 *** Accoun has joined #aichallenge 2010-12-17T12:08:58 *** tapwater has joined #aichallenge 2010-12-17T12:12:07 *** Error323 has joined #aichallenge 2010-12-17T12:13:33 *** Accoun has quit IRC (Ping timeout: 240 seconds) 2010-12-17T12:16:36 *** KP13 has joined #aichallenge 2010-12-17T12:18:50 *** Kingpin13 has quit IRC (Ping timeout: 265 seconds) 2010-12-17T12:20:05 *** Accoun has joined #aichallenge 2010-12-17T12:27:52 *** antimatroid1 has joined #aichallenge 2010-12-17T12:28:59 *** antimatroid has quit IRC (Ping timeout: 265 seconds) 2010-12-17T12:41:24 *** antimatroid has joined #aichallenge 2010-12-17T12:41:33 *** antimatroid1 has quit IRC (Ping timeout: 240 seconds) 2010-12-17T12:51:58 *** iNtERrUpT has quit IRC (Remote host closed the connection) 2010-12-17T12:54:55 *** Rubicon-|-Cross has quit IRC (Quit: Leaving) 2010-12-17T13:10:38 *** bhasker has joined #aichallenge 2010-12-17T13:11:57 *** perror has quit IRC (Quit: Bye all !) 2010-12-17T13:35:28 *** Apophis_ has quit IRC (Read error: Connection reset by peer) 2010-12-17T13:41:25 *** Apophis_ has joined #aichallenge 2010-12-17T13:46:26 *** Rubicon-|-Cross has joined #aichallenge 2010-12-17T13:56:23 *** jaspervdj has quit IRC (Quit: NEVER GONNA GIVE YOU UP NEVER GONNA LET YOU DOWN) 2010-12-17T14:05:23 *** Naktibalda has quit IRC (Quit: ChatZilla 0.9.86 [Firefox 3.6.8/20100722155716]) 2010-12-17T14:11:10 *** yasith has quit IRC (Ping timeout: 255 seconds) 2010-12-17T14:14:26 *** choas has joined #aichallenge 2010-12-17T14:15:30 *** delt0r_ has joined #aichallenge 2010-12-17T14:15:46 *** delt0r___ has quit IRC (Read error: Operation timed out) 2010-12-17T14:22:48 *** Naktibalda has joined #aichallenge 2010-12-17T14:48:23 *** amriedle has joined #aichallenge 2010-12-17T14:55:08 *** KP13 has left #aichallenge ("You're a kitty!") 2010-12-17T15:00:30 *** Mathnerd314 has joined #aichallenge 2010-12-17T15:16:33 *** Accoun has quit IRC () 2010-12-17T15:26:26 *** Accoun has joined #aichallenge 2010-12-17T15:35:19 *** amstan has joined #aichallenge 2010-12-17T15:35:19 *** ChanServ sets mode: +o amstan 2010-12-17T15:55:56 *** Kingpin13 has joined #aichallenge 2010-12-17T16:12:45 working on a new version of cyndrebot - anyone want to see the design so far? 2010-12-17T16:20:15 *** sigh has joined #aichallenge 2010-12-17T16:34:16 *** Naktibalda has quit IRC (Remote host closed the connection) 2010-12-17T17:03:46 *** sigh has quit IRC (Remote host closed the connection) 2010-12-17T17:06:37 *** Kingpin13 has quit IRC (Quit: You're a kitty!) 2010-12-17T17:11:58 *** JamesMG has quit IRC (Quit: home) 2010-12-17T17:20:13 *** Azrathud has joined #aichallenge 2010-12-17T17:29:20 *** choas has quit IRC (Quit: leaving) 2010-12-17T17:35:36 *** mega1 has quit IRC (Ping timeout: 265 seconds) 2010-12-17T17:43:46 *** amstan has quit IRC (Remote host closed the connection) 2010-12-17T17:47:24 *** sigh has joined #aichallenge 2010-12-17T18:12:54 *** mceier has quit IRC (Quit: leaving) 2010-12-17T18:28:35 hey, I know what game we should use for the next AI contest: 2010-12-17T18:28:39 HaxBall! 2010-12-17T18:28:41 http://haxball.appspot.com/ 2010-12-17T18:33:06 oh, it looks like you've already picked a game 2010-12-17T18:37:17 i don't think we have? 2010-12-17T18:44:09 *** Palmik_ has quit IRC (Remote host closed the connection) 2010-12-17T18:50:38 the ant thing? 2010-12-17T18:54:33 *** Naktibalda has joined #aichallenge 2010-12-17T18:59:54 I haven't heard the in-charge-of-things people say anything about a decision 2010-12-17T19:02:04 Migi32: people have gone ahead and begun to implement parts of the ant game, but that was not the result of any official decision 2010-12-17T19:02:15 Migi32: that is a nice game and it could be played by one program per field player 2010-12-17T19:02:48 Frontier, yes I was thinking exactly the same 2010-12-17T19:02:50 delt0r, antimatroid and I were pushing the ant game a lot recently :) 2010-12-17T19:03:25 well, delt0r says that he is still on completely on board with the idea :P 2010-12-17T19:03:50 in particular, no one else organizing has made any comment on my request for suggestions 2010-12-17T19:03:51 well I came to the idea just by discovering the game on the haXe mailing list, playing it a lot (cause it's really fun) and then wondering what the best strategy is 2010-12-17T19:04:45 Zannick: I can't make comments in your fancy members only section :) 2010-12-17T19:05:03 sigh: i did say no one else organizing :) 2010-12-17T19:05:03 oh "organising" 2010-12-17T19:05:07 yeah 2010-12-17T19:05:21 but it is open if you click the right button 2010-12-17T19:05:35 mainly i want hoping to start a discussion 2010-12-17T19:05:57 oh, then you went about it the wrong way 2010-12-17T19:06:07 where is the most up-to-date specification of that ant game btw? 2010-12-17T19:06:19 you need some controversial statements for people to disagree with 2010-12-17T19:06:21 Migi32: You may need to put some work into a prototype for the game. I have a feeling part of the code for next time will come from the community. 2010-12-17T19:06:38 :P 2010-12-17T19:06:44 i did say i didn't want to do the ant game 2010-12-17T19:06:53 :) 2010-12-17T19:07:12 well, whatever, it's still early and i haven't started any backend work yet 2010-12-17T19:07:33 Migi32: delt0r and antimatriod are working on two different specs 2010-12-17T19:07:58 ok, so where can I find one? 2010-12-17T19:08:05 Migi32: just ask, what do you want to know about the ant game? delt0r is experimenting with multiple processes I think while antimatroid is going for a single process solution and I write a vissualizer 2010-12-17T19:08:20 the forum is the best bet, unless one of them decides to post here 2010-12-17T19:08:33 just what it's all about. I know no more than that it's a basic RTS game 2010-12-17T19:08:35 with ants :) 2010-12-17T19:08:42 antimatroid's pastebin'd his latest specs to the channel every so often 2010-12-17T19:08:52 but neither of them have a public repo :( 2010-12-17T19:09:23 Migi32: look at this - http://marco-leise.homedns.org/index.html 2010-12-17T19:09:34 *** smellyhippy has quit IRC (Ping timeout: 255 seconds) 2010-12-17T19:10:21 The ants will try to harvest the numbers on the map and carry them home to base. There the harvest will be used to spawn new ants. (1 ant per 1 point, 1 ant each turn) 2010-12-17T19:11:08 nice visualiser 2010-12-17T19:11:27 the ants have a limited view radius, so they wont know where the enemy base is. But once an ant enters the enemy base the game is over. 2010-12-17T19:11:31 thx 2010-12-17T19:11:42 yea with this amount of work in it, this is 90% official already 2010-12-17T19:12:07 maybe haxball next time 2010-12-17T19:12:32 if you are serious about haxball, then at least post in the forum about it 2010-12-17T19:12:46 Migi32: I could live with it, if another game was chosen. I learnt a lot about the HTML canvas on the way :), but as I said, you may have to put a lot of effort into it yourself. 2010-12-17T19:12:46 ok I will 2010-12-17T19:13:48 Frontier, yes and I'm only trying to propose a new idea. The only reason I would push it would be because of my own interest in what the best strategy would be. But I'm not really prepared to write visualisers for it etc. 2010-12-17T19:14:28 yeah, HTML canvas is nice... I want to have a better look at it sometime. My only experience with it so far is hacking things into the existing planet wars visualiser :P 2010-12-17T19:14:39 *** smellyhippy has joined #aichallenge 2010-12-17T19:15:18 Migi32: start out with a scetch and see if people like the idea. this is what happened with the ant idea. and now there are 3 people (the original poster not included) putting some work into it to flesh it out. 2010-12-17T19:16:22 sigh: You can read out the pixel data of a canvas and modify it. So you are basically free to do anything. Write a software 3D engine if you like :p 2010-12-17T19:16:43 :) 2010-12-17T19:18:54 You can also create new canvases in the 'background' and use them the same way you use regular images. You can paint a canvas into another canvas. So you can use cache some expensive drawings as well. 2010-12-17T19:34:50 Migi32: the game is in no way set, even with the work done, like frontier i've managed to learn a fair bit 2010-12-17T20:00:06 I will try to write a proposal for haxball tomorrow, if not for the next AI contest then for the one after that. 2010-12-17T20:00:30 probably won't be easy because there's some basic physics and collision detection involved 2010-12-17T20:00:33 but I have to go now 2010-12-17T20:00:34 bye 2010-12-17T20:01:14 *** Migi32 has quit IRC (Remote host closed the connection) 2010-12-17T20:16:36 *** Naktibalda has quit IRC (Remote host closed the connection) 2010-12-17T20:20:11 *** Rubicon-|-Cross has quit IRC (Quit: Leaving) 2010-12-17T21:01:15 *** bhasker has quit IRC (Ping timeout: 272 seconds) 2010-12-17T21:11:05 yay 2010-12-17T21:11:17 i have an ant spawning every move for the first ten moves and them walking around 2010-12-17T21:11:40 all the hardest stuff is done i think, now it's just adding in the other moves 2010-12-17T21:30:53 *** Mathnerd314 has quit IRC (Ping timeout: 255 seconds) 2010-12-17T22:16:07 *** delt0r___ has joined #aichallenge 2010-12-17T22:17:53 *** delt0r_ has quit IRC (Ping timeout: 272 seconds) 2010-12-17T22:25:43 http://www.youtube.com/watch?v=5qwi5XY9CZ4 2010-12-17T22:25:48 did any of you ever play this? 2010-12-17T22:26:01 this is what my ant game reminds me of 2010-12-17T22:27:16 although it's the different part of the game where your objective is to get into the enemies home base 2010-12-17T22:29:26 that kid couldn't handle battletoads 2010-12-17T22:29:56 cut to 6:35 for where it explains my ant gameish 2010-12-17T22:31:12 http://www.youtube.com/watch?v=W2rPuRDtL_s 2010-12-17T22:31:16 this is just that clip 2010-12-17T22:34:09 http://www.youtube.com/watch?v=cLM9WTOsVU0 and that's the game being played with a horrible soundtrack 2010-12-17T22:40:41 *** Azrathud has quit IRC (Ping timeout: 272 seconds) 2010-12-17T23:17:23 *** amriedle has quit IRC (Quit: Lost terminal) 2010-12-17T23:24:56 *** Azrathud has joined #aichallenge 2010-12-17T23:42:25 just got my first non-draw game 2010-12-17T23:42:42 meaning ants are going out, harvesting, returning home, depositing and getting new ants 2010-12-17T23:42:51 still need to do attacks, but then i think it's just cleaning