2012-01-03T00:05:24 *** Fandekasp has joined #aichallenge 2012-01-03T00:30:37 *** dorisabayon has joined #aichallenge 2012-01-03T00:34:42 *** Fandekasp has quit IRC (Ping timeout: 268 seconds) 2012-01-03T00:38:06 *** replore has joined #aichallenge 2012-01-03T00:49:48 *** treeform has quit IRC (Remote host closed the connection) 2012-01-03T01:03:49 *** cyphase has quit IRC (Ping timeout: 240 seconds) 2012-01-03T01:14:55 *** yy is now known as Guest48473 2012-01-03T01:26:19 *** ltriant has quit IRC (Quit: Computer has gone to sleep) 2012-01-03T01:29:49 *** Cyndre has quit IRC (Ping timeout: 240 seconds) 2012-01-03T01:33:37 *** treeform has joined #aichallenge 2012-01-03T01:38:30 *** replore has quit IRC (Ping timeout: 252 seconds) 2012-01-03T02:01:18 *** jstrong has quit IRC (Quit: Leaving.) 2012-01-03T02:03:35 *** dorisabayon has quit IRC (Ping timeout: 252 seconds) 2012-01-03T02:10:47 *** treeform has quit IRC (Remote host closed the connection) 2012-01-03T02:11:12 *** treeform has joined #aichallenge 2012-01-03T02:11:48 contestbot: later tell McLeopold https://github.com/aichallenge/aichallenge/tree/amstan-asteroids 2012-01-03T02:11:48 amstan: I think that worked... 2012-01-03T02:14:47 *** TheLinker has quit IRC (Quit: Bye) 2012-01-03T02:49:50 *** Fandekasp has joined #aichallenge 2012-01-03T02:50:35 *** Kurnevsky has joined #aichallenge 2012-01-03T02:55:00 *** treeform_ has joined #aichallenge 2012-01-03T02:58:16 *** treeform has quit IRC (Ping timeout: 240 seconds) 2012-01-03T03:05:11 *** ikaros has joined #aichallenge 2012-01-03T03:08:11 *** Jak_o_Shadows has quit IRC (Ping timeout: 255 seconds) 2012-01-03T03:17:51 *** Palmik has joined #aichallenge 2012-01-03T03:24:14 *** epicmonkey has joined #aichallenge 2012-01-03T03:31:05 *** dorisabayon has joined #aichallenge 2012-01-03T03:33:00 *** Jak_o_Shadows has joined #aichallenge 2012-01-03T03:34:08 *** Fandekasp has quit IRC (Ping timeout: 252 seconds) 2012-01-03T03:41:09 amstan: thanks, i'll go and check yours out as well. hadn't thought of bullets with different speeds 2012-01-03T04:04:00 *** Varan has joined #aichallenge 2012-01-03T04:04:47 Is the asteroids thing going to be the game for the next contest? 2012-01-03T04:08:54 *** grc48 has quit IRC (Quit: This computer has gone to sleep) 2012-01-03T04:09:43 *** mayanks43 has joined #aichallenge 2012-01-03T04:12:51 Varan: it'll be one of the candidates 2012-01-03T04:13:55 my preference: it would become a smaller (4 or 6 weeks) challenge in the spring or early summer 2012-01-03T04:15:31 *** Garf has joined #aichallenge 2012-01-03T04:16:21 i don't think it'll be as accessible to beginners 2012-01-03T04:16:26 and i'm not sure how much variety maps can really have? 2012-01-03T04:16:33 otherwise it's neat 2012-01-03T04:17:32 Hmm yeah i like the idea 2012-01-03T04:17:39 I also like the risk idea 2012-01-03T04:17:53 *** Garf has quit IRC (Client Quit) 2012-01-03T04:19:24 I like asteroids because it's different from other types of games ... more arcade like instead of turn based 2012-01-03T04:22:05 it'll still be turn based 2012-01-03T04:22:32 yup, but it looks more "fluid" perhaps 2012-01-03T04:22:51 i dunno, pw and ants had fairly "fluid" visualisers 2012-01-03T04:22:55 maybe not so much ants 2012-01-03T04:23:21 antimatroid: true, it'll be more planet wars like in the visualization 2012-01-03T04:23:48 i guess the "grid" will be closer to being continuous 2012-01-03T04:24:04 antimatroid: i'm not so sure about the variety either, but it looks like fun nevertheless. that's why i think it's a good candidate for a shorter challenge 2012-01-03T04:24:18 antimatroid: why would it be harder for beginners? 2012-01-03T04:24:30 you could make obstacles in the map 2012-01-03T04:25:04 obstacle might work, you'd need to break hitting asteroids and work out what trajectory they split at or whatever 2012-01-03T04:25:05 Varan: yeah, or indestructible asteroids 2012-01-03T04:25:21 *** kaemo has quit IRC (Ping timeout: 255 seconds) 2012-01-03T04:25:22 aerique: need to understand how motion works 2012-01-03T04:25:28 it's just harder than previous 2012-01-03T04:25:34 it's not like it's that complicated 2012-01-03T04:25:58 ah, right. at least the battle resolution will be easier :-) 2012-01-03T04:26:18 how do you determine if a ship is hit? 2012-01-03T04:26:23 like is the ship really a circle? 2012-01-03T04:27:02 might be easier to visualise stuff as actual circles? then you could have a rotation line indicating the "forward" direction where bullets would shoot out 2012-01-03T04:27:25 currently ships and asteroids are circles 2012-01-03T04:28:02 it's just for the looks that i've drawn them differently but i'm also showing the hit circle 2012-01-03T04:28:04 *** Garf has joined #aichallenge 2012-01-03T04:28:38 i'm down 2012-01-03T04:28:50 i'm interested in playing around with finite optimal control type methods 2012-01-03T04:28:52 How does the ai get the data about the other ships? 2012-01-03T04:28:56 and doing it numerically 2012-01-03T04:29:00 *** amstan_ has joined #aichallenge 2012-01-03T04:29:00 *** ChanServ sets mode: +o amstan_ 2012-01-03T04:29:02 Just the location at all times? 2012-01-03T04:29:11 *** amstan has quit IRC (Quit: Konversation terminated!) 2012-01-03T04:29:23 Varan: i imagine you get a position, radius, direction and velocity 2012-01-03T04:29:45 Varan: yup, just the location. i'm not sure whether the velocity vector (or number) should be send to other players 2012-01-03T04:29:52 (currently it is) 2012-01-03T04:30:05 Because you could maybe also get only a limited view of the field depending on how you are turned 2012-01-03T04:30:27 i assume asteroids split such that the sum of the new asteroids area's is equal to the original asteroids area? 2012-01-03T04:30:52 aerique: location? what about direction? 2012-01-03T04:30:59 and what info do you get about asteroids and bullets? 2012-01-03T04:31:24 and out of curiousity, how do you determine if two objects have collided? 2012-01-03T04:31:51 i'll link a movie and bot input and output in a couple of minutes 2012-01-03T04:32:00 *** sigh has joined #aichallenge 2012-01-03T04:32:19 *** treeform_ has quit IRC (Remote host closed the connection) 2012-01-03T04:32:30 have any of you looked at evader/pursuer problems before? 2012-01-03T04:32:49 1v1 with no asteroids would be solvable i imagine 2012-01-03T04:33:08 antimatroid: currently the players is send another player's id, x and y coordinates, direction (in radians) and speed (a number which has to be combined with the heading for the x and y velocities) 2012-01-03T04:33:16 3d would be more complicated, but visualisation would be harder 2012-01-03T04:33:57 antimatroid: collision is stupid at the moment, just a check at the end points. needs to be improved for the actual challenge but it seems to suffice for my prototype 2012-01-03T04:33:59 antimatroid: 3d and 2d dont differ that much in concepts i think ... only in visualization ... and harder math :P 2012-01-03T04:34:15 *** mviel__ has joined #aichallenge 2012-01-03T04:34:16 Varan: harder to solve stuff in 3d 2012-01-03T04:34:30 http://en.wikipedia.org/wiki/Differential_game 2012-01-03T04:34:38 the game would be one of those if there weren't turns 2012-01-03T04:35:45 hmm 2012-01-03T04:35:47 perhaps it'd be a difference game or something 2012-01-03T04:37:24 Hmm maybe it would be cool to have a game like Osmos .. it's kinda similar to asteroids 2012-01-03T04:37:26 http://www.hemispheregames.com/osmos/ 2012-01-03T04:37:51 Lots of different types of games 2012-01-03T04:37:53 aerique: if such approaches do prove fruitful that's why I expect it to be far less approachable to beginners 2012-01-03T04:38:04 *** mviel_ has quit IRC (Ping timeout: 248 seconds) 2012-01-03T04:38:16 you really need to have at least up to multivariable calculus before you really start trying to learn that stuff 2012-01-03T04:38:51 and more realistically some differential equations 2012-01-03T04:39:08 and /calculus of variations/optimal control :P 2012-01-03T04:42:41 aerique: is it random how asteroids split? in both size? and direction? or is that known before it's hit? 2012-01-03T04:43:57 sorry, i'm on a really shitty connection currently 2012-01-03T04:44:05 http://dl.dropbox.com/u/1908796/asteroids/test_p04_02.mpg 2012-01-03T04:44:05 http://dl.dropbox.com/u/1908796/asteroids/0.bot0.input 2012-01-03T04:44:05 http://dl.dropbox.com/u/1908796/asteroids/0.bot1.output 2012-01-03T04:44:50 antimatroid: asteroids are currently split randomly, but that's what the beta is for 2012-01-03T04:45:55 'm also toying with the idea of being able to config your ship's rate-of-fire, turn speed, hit points, bullet damage, etc to have some hidden variables 2012-01-03T04:47:14 * aerique goes back to work now, will check later 2012-01-03T04:47:28 (real work that is, not the asteroids protoype) 2012-01-03T04:47:42 *** Blkt has joined #aichallenge 2012-01-03T04:51:54 here's the map: http://dl.dropbox.com/u/1908796/asteroids/test_p04_02.map 2012-01-03T05:43:29 *** epicmonkey has quit IRC (Ping timeout: 240 seconds) 2012-01-03T05:45:13 *** UncleVasya has joined #aichallenge 2012-01-03T05:55:55 do people avoid "git commit -a -m "yay no add" "> 2012-01-03T05:55:57 ?* 2012-01-03T06:23:24 jmcarthur, hi. Is this a nice enough solution for your and mcstar yesterdays problem? http://codepad.org/yuD18B1n 2012-01-03T06:27:19 *** sashaSochka has joined #aichallenge 2012-01-03T06:28:24 sashaSochka: Good afternoon. Have you done your exams? 2012-01-03T06:28:34 *** nha_ has joined #aichallenge 2012-01-03T06:29:35 *** grc48 has joined #aichallenge 2012-01-03T06:30:10 of course no. They are in April -Ma 2012-01-03T06:30:16 may* 2012-01-03T06:30:42 right... 2012-01-03T06:31:00 But you were busy with something last week of the contest. 2012-01-03T06:31:01 *** Fandekasp has joined #aichallenge 2012-01-03T06:31:12 finally you beat me!) I'm not even in top 100.. 2012-01-03T06:32:25 I risked and included few new fetures in the last day ;) 2012-01-03T06:32:27 I'm always busy this year, because I pretend on medal and my teachers want I receive it.. They want it more than me 2012-01-03T06:32:46 what features? 2012-01-03T06:33:00 Seems like I am the only one who didn't broke his bot with new last-day features ;D 2012-01-03T06:33:43 I read something on forum that developers(admins) have already decided what is the next game. Is it true? 2012-01-03T06:33:45 1. Correct definition of battle ants. 2012-01-03T06:34:26 I was afraid of breaking of my bot. That was one of the reasons I stopped making him one week before the end of contest 2012-01-03T06:35:19 *** dorisabayon has quit IRC (Ping timeout: 276 seconds) 2012-01-03T06:35:21 2. Crasy mode. When all ants a blocked and cannot reach their destination (happens on maze maps) attack without counting losses. 2012-01-03T06:36:07 Did you read xathis's explanation? 2012-01-03T06:36:24 and many small changes of cost function and such. 2012-01-03T06:36:39 Not carefully. 2012-01-03T06:37:10 I was surprised I didn't find there a lot of cool-winning ideas. 2012-01-03T06:37:19 yes, I have many half-random constants in my code too, it would much better if I tested them and make them more correct 2012-01-03T06:37:57 Yes, but I don't understand even now how he resolves battles with 12+ ants 2012-01-03T06:38:05 I don't know about next game. 2012-01-03T06:38:41 Yesterday am.stan said he coding asteroids on pyQT but I doubt it is for contest. 2012-01-03T06:38:54 http://ants.aichallenge.org/visualizer.php?game=346138&user=4513 for example 100 before end xathis and greentea has very many ants in battle 2012-01-03T06:39:24 There are a lot of talks about communication protocol between bots. Everybody wants it. 2012-01-03T06:39:40 It's the same i read on forum )about asteroids) 2012-01-03T06:40:18 between bots of different players? what for? 2012-01-03T06:40:47 Also I know am.stan was impressed by some aeriques demo on forums. I don't know what it is. Some video on youtube I guess. 2012-01-03T06:41:41 To add diplomacy element to the game. Like 'I will do moves A, B and C' if you do 'H, K'. 2012-01-03T06:41:54 Or let's attack player 4'. 2012-01-03T06:42:02 Just after finishing the contest I read about visualizations with special line commands etc. Did you use them? 2012-01-03T06:42:15 no. 2012-01-03T06:42:41 The only thing I used for testing is AntsTesting (my gui to playgame.cmd). 2012-01-03T06:42:52 I think it can cause mahinations 2012-01-03T06:43:56 So you were creator of that tool? Is it the same I read on habrahabr? 2012-01-03T06:45:12 sorry, I have a dinner, will return soon 2012-01-03T06:45:26 In the channel logs I see some talks about asteroids game mechanic. 2012-01-03T06:49:45 *** mcstar has joined #aichallenge 2012-01-03T06:49:56 *** Fandekasp has quit IRC (Ping timeout: 252 seconds) 2012-01-03T06:50:35 *** Fandekasp has joined #aichallenge 2012-01-03T06:51:56 mcstar: hi. I tried to decrese the size of my code. Not much succes :) http://codepad.org/yuD18B1n 2012-01-03T06:52:07 *decrease 2012-01-03T06:53:27 I'm here 2012-01-03T06:53:47 sashaSochka: I don't know if there is a post about it on habr. Wait few mins, I'll make an upload. 2012-01-03T06:54:49 sashaSochka: Do you have an account on Timus online judge? 2012-01-03T06:55:44 No 2012-01-03T06:57:24 sashaSochka: http://ompldr.org/vYzE4Zw/AntsTesting_0_5_�.7z 2012-01-03T06:58:39 very close to that what I've already seen. Have you already shared it somewhere? 2012-01-03T07:00:21 On forum but it was very long ago (in June, v0.3). 2012-01-03T07:00:33 Also sometimes I put link here. 2012-01-03T07:00:42 wow, it's amazing 2012-01-03T07:00:58 how much time did you spend making it? 2012-01-03T07:01:07 As far as I know there are only 2 users of it except me :) 2012-01-03T07:02:29 During beta were 3 but one guy left irc and now lives on forum. 2012-01-03T07:02:29 how should I add my own bot to that list of standart bots? 2012-01-03T07:02:29 It's hard to calculate because there were very big breaks between developing. 2012-01-03T07:03:33 your bots filename must satisfy mask *Bot*. You can put your bot in any place in Ants directory. 2012-01-03T07:04:04 It is written on Delphi, I can give you source code if you are interested. 2012-01-03T07:04:23 I don't know delphi 2012-01-03T07:06:19 hehe, i can test now all my previous versions of bot at once. thank you 2012-01-03T07:07:17 I thought Ibefore shared it with you already. Pity mistake, I'd like you to have it during contest time. 2012-01-03T07:08:36 Yes, me too.. 2012-01-03T07:10:02 I think there will no need in major changes for the next contest. And I already have a good experience in doing it. So if I'll have a little time, I'll port it to he next challenge 2012-01-03T07:10:58 would be fine 2012-01-03T07:12:42 *** ikaros has quit IRC (Quit: Ex-Chat) 2012-01-03T07:19:02 morning 2012-01-03T07:19:39 *** Jak_o_Shadows1 has joined #aichallenge 2012-01-03T07:19:49 *** cyphase has joined #aichallenge 2012-01-03T07:19:53 *** X-Scale has quit IRC (Read error: Connection reset by peer) 2012-01-03T07:20:07 hi 2012-01-03T07:20:27 mcstar: seen my link? 2012-01-03T07:21:31 did you change the algorithm? 2012-01-03T07:22:56 *** Jak_o_Shadows has quit IRC (Ping timeout: 268 seconds) 2012-01-03T07:23:12 not really. Changed a flow to make it more clare: 2012-01-03T07:23:44 was next_permutation calls find_key, find_key calls swap_and_rev 2012-01-03T07:24:24 now next_permutation calls find_key, find_key returns left, key, right. next_permutation calls swap_and_rev. 2012-01-03T07:24:24 *** X-Scale has joined #aichallenge 2012-01-03T07:24:45 *** X-Scale is now known as Guest33325 2012-01-03T07:25:26 And rewrote print_int_list to what you've suggested 2012-01-03T07:26:08 *** ahow984_ has joined #aichallenge 2012-01-03T07:29:01 mcstar: I've showed our codes to my friend (who doesn't know haskell or ocaml) and he said 1st is more clare (1st was mine ;p) 2012-01-03T07:29:13 *** dorisabayon has joined #aichallenge 2012-01-03T07:29:32 Let's do an experiment. 2012-01-03T07:29:36 *** Fandekasp has quit IRC (Ping timeout: 240 seconds) 2012-01-03T07:30:33 (my findLesser function is stupidly named) 2012-01-03T07:30:49 sashaSochka: Which one scares you more: 2012-01-03T07:31:17 sashaSochka: http://codepad.org/prXLnKQk or http://codepad.org/yuD18B1n 2012-01-03T07:31:33 second 2012-01-03T07:31:43 UncleVasya: http://codepad.org/D7CfUrD9 2012-01-03T07:32:29 i tend to agree 2012-01-03T07:32:31 ok, we need bigger control group :) 2012-01-03T07:32:34 :D 2012-01-03T07:33:30 * ahow984_ wakes up 2012-01-03T07:35:17 ahow984_: hi. The question above is for you too. 2012-01-03T07:35:34 but he is biased now 2012-01-03T07:35:40 I think the first one looks better, because it's easier to read and shorter 2012-01-03T07:35:42 he heard the answer of sochka 2012-01-03T07:35:51 mcstar: nahhh :P I gave reasons to justify myself 2012-01-03T07:36:31 just going to say though, C++ would beat both of them on readability most likely ;) 2012-01-03T07:36:37 mcstar: you did it only to help me with size or you like that kind of placing statements? 2012-01-03T07:37:00 i like that formatting 2012-01-03T07:37:04 makes more sense to me 2012-01-03T07:37:10 and i got rid of a lambda 2012-01-03T07:37:20 (fun x -> string_of_int x) 2012-01-03T07:37:25 do either of them rely on whitespace and indentation? 2012-01-03T07:37:29 like PY 2012-01-03T07:37:30 this is equivalent to string_of_int 2012-01-03T07:37:46 haskell does, but not like python 2012-01-03T07:38:28 you can delimit haskell too, so it becomes freeform, with {} 2012-01-03T07:38:37 ah, ok 2012-01-03T07:38:49 brackets + whitespace ftw 2012-01-03T07:39:31 haskell doesnt care about how much you indent, but you have to be consistent for a given level of indentation, cause another level introduces a new block 2012-01-03T07:39:43 yeah, ok 2012-01-03T07:40:05 Is this that algorithim you were talking about yesterday mcstar? 2012-01-03T07:40:21 i dont know 2012-01-03T07:40:25 it was yesterday 2012-01-03T07:40:31 ok 2012-01-03T07:40:34 :D 2012-01-03T07:40:35 I'll check the logs 2012-01-03T07:40:35 ahow984_: this is SEPA for next pemutation. 2012-01-03T07:40:43 yeah, thats the one 2012-01-03T07:40:52 ty 2012-01-03T07:40:58 *** codetiger has joined #aichallenge 2012-01-03T07:42:46 ok, mcstar, you have won :D 2012-01-03T07:43:04 2:1, it isnt significant 2012-01-03T07:43:43 so what use is Haskell or Ocaml? 2012-01-03T07:43:48 2:1? 2012-01-03T07:43:57 what do they do better then any other language? 2012-01-03T07:44:14 UncleVasya: didnt you say your friend liked yours better? 2012-01-03T07:44:15 ahow984_: they scare you better :) 2012-01-03T07:44:27 UncleVasya: yes, thats true :P 2012-01-03T07:44:42 ah, I thought you are about some another battles. 2012-01-03T07:45:08 Like your 1 is for this one. And my 2 for something else. 2012-01-03T07:45:57 my opinion still, is that those 2 codes shouldnt reach any publicity 2012-01-03T07:46:08 they are both ugly and not easily understandable 2012-01-03T07:46:19 *** codetiger has quit IRC (Remote host closed the connection) 2012-01-03T07:46:38 mcstar: I agree 2012-01-03T07:46:54 that said, I don't know haskell or ocaml so I can't judge really 2012-01-03T07:46:55 *** ikaros has joined #aichallenge 2012-01-03T07:49:14 same here 2012-01-03T07:50:01 mcstar: http://acm.timus.ru/problem.aspx?space=1&num=1349 2012-01-03T07:51:58 i dont have time for another problem :( 2012-01-03T07:52:21 *** grc48 has quit IRC (Quit: This computer has gone to sleep) 2012-01-03T07:53:01 Just tell me first solution that came to your mind. 2012-01-03T07:53:12 Without coding. 2012-01-03T07:53:28 i think you should use http://en.wikipedia.org/wiki/Fermat%27s_Last_Theorem theorem 2012-01-03T07:53:41 This is a tricky problem. And funny. 2012-01-03T07:53:44 yep. 2012-01-03T07:53:47 there are no integers if n>2 2012-01-03T07:53:55 you should try projecteuler.net 2012-01-03T07:54:27 that was a test for mcstar-s mathematic knowledge :) 2012-01-03T07:54:32 heh 2012-01-03T07:55:22 btw, I didn't know about that theorem and decided to make brute force. I don't know math :) 2012-01-03T07:55:41 Lucky I decided to google this formula before coding just in case 2012-01-03T07:57:50 is it too late to say that i didnt read the link? 2012-01-03T07:59:53 A man of high cultural traditions reads the linked page before he says 'I do not have a time to read it' >< 2012-01-03T08:00:50 i said i had no time for another problem 2012-01-03T08:01:01 you linked in timus problems before 2012-01-03T08:01:08 so theres was no need to check :D 2012-01-03T08:02:09 yep, but if you'd read you's swa that you need 5 seconds to give an answer: 2012-01-03T08:02:19 1. I will use Fermat Last Theorem 2012-01-03T08:02:26 2. I will use 3 loops 2012-01-03T08:02:32 no offence 2012-01-03T08:02:39 *** Jak_o_Shadows1 has quit IRC (Remote host closed the connection) 2012-01-03T08:02:57 I don't know, maybe you are really busy now. 2012-01-03T08:03:26 as busy as on any other day 2012-01-03T08:03:51 3 loops? 2012-01-03T08:03:55 Keep learning c++? 2012-01-03T08:04:09 4? 2012-01-03T08:04:11 :) 2012-01-03T08:04:14 0 2012-01-03T08:04:27 do you need to loop over n? 2012-01-03T08:04:42 ah, ok thats given 2012-01-03T08:04:48 bleh, whats the number you take away from a char to get the number back? 2012-01-03T08:05:00 '0' 2012-01-03T08:05:02 like (char)9 - x = (int)9 2012-01-03T08:05:09 >_< 2012-01-03T08:05:16 I was thinking ascii codes, ty mcstar 2012-01-03T08:05:48 you don't need any loops- if n=0 or n>2 answer is -1; if n=1 answer is 1,2,3; if n=2 answer is 3,4,5 2012-01-03T08:06:11 :( 2012-01-03T08:06:17 you dont follow sashaSochka 2012-01-03T08:06:42 had you not known fermats conjucture, how many loop would you require 2012-01-03T08:06:48 e 2012-01-03T08:07:13 UncleVasya asked if i had looped over a,b,c 2012-01-03T08:07:15 2 2012-01-03T08:08:50 UncleVasya: btw, that wast a particularly tricky question because i think every mathematician and physicist heard of it 2012-01-03T08:09:42 im busy again, got to wash the dishes 2012-01-03T08:11:07 *** Guest48473 is now known as yy 2012-01-03T08:25:04 ok, i reflected on this, i should have read the link, after you put up with my problem for so long 2012-01-03T08:25:47 i hope one day you will forgive me... 2012-01-03T08:26:16 I hope you are kidding and do not take it so seriously. 2012-01-03T08:26:51 i dont know how seriously YOU take it 2012-01-03T08:26:59 if you dont, i wont 2012-01-03T08:27:03 I wasn't serious when was talking that you'd should read link :) 2012-01-03T08:27:15 ok, then all is good 2012-01-03T08:27:54 And about 'put up with my problem for so long' - it was interesting, I'm glad I was involved. 2012-01-03T08:29:00 mcstar: yesterday you said you wrote your haskell code without using debug statements. Why? 2012-01-03T08:29:36 because you saw everything in the interpreter? 2012-01-03T08:30:01 *** foRei has joined #aichallenge 2012-01-03T08:31:28 *** Guest33325 has quit IRC (Read error: Connection reset by peer) 2012-01-03T08:31:42 no 2012-01-03T08:32:11 because of haskell's purity 2012-01-03T08:34:07 hmm, I think I've got Sasha's point about 2 loops. You loop only throw a and b, the c you get when (a^n + b^n) ^ (1/n) returns integer value. 2012-01-03T08:34:23 through 2012-01-03T08:34:26 correct 2012-01-03T08:34:32 yup 2012-01-03T08:35:05 but its probably faster to loop than to get the nth root 2012-01-03T08:35:17 maybe not 2012-01-03T08:35:36 you dont have to loop over the full range 2012-01-03T08:35:45 *** Guest33325 has joined #aichallenge 2012-01-03T08:36:43 UncleVasya: did you understand my answer? 2012-01-03T08:36:51 or do i need to elaborate 2012-01-03T08:37:43 but in n-th loop you will need to get the n-th root too 2012-01-03T08:37:52 I didn't understand but you can not to bother with further explanation. 2012-01-03T08:38:00 not root but power** 2012-01-03T08:38:14 UncleVasya: why? 2012-01-03T08:38:26 it is an interesting question 2012-01-03T08:38:43 if so, go ahead 2012-01-03T08:38:56 ill check your ocaml code for an example 2012-01-03T08:39:33 ok 2012-01-03T08:39:41 in find_key 2012-01-03T08:39:53 in the first version 2012-01-03T08:40:08 you have several print statements and after that, you return a list 2012-01-03T08:40:32 you cant do that in haskell 2012-01-03T08:40:47 that print statement is a side-effect 2012-01-03T08:41:05 it affects the outside worl(otside wrt the function) 2012-01-03T08:41:13 Understood. 2012-01-03T08:41:34 so every pure haskell code is like nesting 2012-01-03T08:41:41 you nest function until you get bored 2012-01-03T08:41:44 s 2012-01-03T08:42:18 but you can have several monads, in these monads you can have a sequential program 2012-01-03T08:42:34 executing pure or un-pure code one after another 2012-01-03T08:42:56 and the return values are not necessarily the input values of another function 2012-01-03T08:43:37 but that function, that uses this IO monad(one of several monads) must have a different type signature than a pure function 2012-01-03T08:44:10 so if i had to debug that code with print statements id need to write the whole thing in a monad 2012-01-03T08:44:25 the main function in my code is un-pre 2012-01-03T08:44:26 pure 2012-01-03T08:44:46 "do" introduces the un-pure part 2012-01-03T08:45:02 so this is my short answer 2012-01-03T08:45:25 i dont have any idea yet, how to debug in haskell with prints "comfortably" 2012-01-03T08:46:48 I was lucky to avoid haskell in contest :) 2012-01-03T08:48:41 why you decided to code next_permutation? 2012-01-03T08:50:34 ok, for the last time 2012-01-03T08:50:58 there was rwest's problem, i did it in haskell, was slow, and had a recursive permutation function 2012-01-03T08:51:10 after that i did a mucgh better version in c++ 2012-01-03T08:51:16 but it used the next_perm funciton 2012-01-03T08:51:23 and i wanted to do the same in haskell 2012-01-03T08:51:28 thats the whole story 2012-01-03T08:52:25 (it wasnt slow because of the permutation funtion, but for a whole different reason, i sed a recursive graph-walker in it, but with c++ mutable arrays i could do a waaay faster one) 2012-01-03T09:03:38 maybe we can ask on some haskell, ocaml or f_prog channel to show us a nice solution? 2012-01-03T09:04:37 *** dorisabayon has quit IRC (Ping timeout: 248 seconds) 2012-01-03T09:05:00 UncleVasya: go ahead, but i want to write a nice solution myself wo help 2012-01-03T09:05:31 so you will try again? 2012-01-03T09:06:09 im reading c++, but if i have a new idea with SEPA ill try it out 2012-01-03T09:14:21 *** sigh has quit IRC (Remote host closed the connection) 2012-01-03T09:30:03 *** epicmonkey has joined #aichallenge 2012-01-03T09:36:00 *** tobiassjosten has quit IRC (Read error: Operation timed out) 2012-01-03T09:39:31 *** tobiassjosten has joined #aichallenge 2012-01-03T09:42:47 how can I tell if I've an integer overflow with gdb? 2012-01-03T09:44:01 Possible cause: the monomorphism restriction applied to the following: 2012-01-03T09:46:07 try adding an assert or something, if your new value is less than the original, you had an overflow 2012-01-03T09:46:16 (if you only increase the said variable) 2012-01-03T09:46:36 yeah 2012-01-03T09:46:36 (i have no idea about gdb) 2012-01-03T09:46:40 kk 2012-01-03T10:05:46 *** |UncleVasya| <|UncleVasya|!kvirc@46-133-212-114.dialup.umc.net.ua> has joined #aichallenge 2012-01-03T10:05:47 *** UncleVasya has quit IRC (Read error: Connection reset by peer) 2012-01-03T10:28:40 *** nha_ is now known as nha 2012-01-03T10:41:31 fixed it btw mcstar 2012-01-03T10:41:34 ty 2012-01-03T10:43:30 *** dapplegate has joined #aichallenge 2012-01-03T10:59:50 *** Guest33325 is now known as X-Scale 2012-01-03T11:11:27 *** TheLinker has joined #aichallenge 2012-01-03T11:29:23 *** |UncleVasya| <|UncleVasya|!kvirc@46-133-212-114.dialup.umc.net.ua> has quit IRC (Ping timeout: 252 seconds) 2012-01-03T11:29:25 *** JorgeB has joined #aichallenge 2012-01-03T11:38:28 *** Okke has joined #aichallenge 2012-01-03T11:41:47 *** McLeopold has quit IRC (Quit: Leaving.) 2012-01-03T11:48:02 *** Okke has quit IRC (Quit: Page closed) 2012-01-03T11:59:13 *** raemde_ has joined #aichallenge 2012-01-03T12:02:13 *** raemde has quit IRC (Ping timeout: 248 seconds) 2012-01-03T12:19:25 *** GeorgeSebastian has joined #aichallenge 2012-01-03T12:24:02 *** treeform has joined #aichallenge 2012-01-03T12:26:29 *** GeorgeSebastian has quit IRC (Ping timeout: 240 seconds) 2012-01-03T12:26:29 *** alehorst has quit IRC (Ping timeout: 240 seconds) 2012-01-03T12:26:57 *** mviel_ has joined #aichallenge 2012-01-03T12:30:33 *** mviel__