[00:12] Join: OoS joined #corewars [00:12] 3am, and im feeling creative [00:12] where should i direct it this time.. [00:13] MSG: Ping timeout: 240 seconds [00:22] MSG: Ping timeout: 240 seconds [02:26] Join: Spnkr joined #corewars [02:27] hi [02:27] hello [02:32] Part: Spnkr left #corewars [03:10] Join: waknuk joined #corewars [03:12] Part: waknuk left #corewars [05:23] MSG: Ping timeout: 240 seconds [05:31] can I use pi to generate a sequence or will pmars not like that? [06:36] MSG: Quit: Lost terminal [08:06] Join: BlueMaxim joined #corewars [08:06] MSG: Client Quit [08:07] Join: BlueMaxim joined #corewars [08:07] Anyone here? [08:08] for what good i am [08:09] Don't be that way. [08:10] ok... HOW CAN I HELP YOU SIR [08:10] Funny. [08:12] I'm just here to get started playing CoreWar, or Wars, whatever have you.. [08:13] yeah, i don't remember either :) [08:15] I've had a look - seems like a good game. [08:15] i see you found the freenode channel, so you may want to explore the topic there [08:15] i found corewars when i was like 14, heh... but even back then i had motivation problems so i never really did anything with it [08:15] only more recently did i come here :) [08:16] it's always intrigued me [08:19] So move over to Freenode's CoreWars channel? [08:19] * myndzi shrugs [08:19] whatever you want, i suppose activity here will likely be more until the shutdown at least [08:20] i just meant the link in the topic there, though.. since it's not here anymore [08:20] The topic links back to this channel. [08:21] the blogspot link? [08:21] http://sfghoul.blogspot.com/2007/07/top-10-links-for-corewar-newbies.html [08:22] oh, right... link number 8 [08:22] heh [08:22] that page is older than this news [08:22] the freenode one "will" be the new channel [08:22] I thought so. [08:22] i can probably answer basic questions for you, but i gotta go to bed soon [08:23] Just, give me a quick rundown (while I download pMARS and CoreWin) of CoreWars. [08:23] uhh [08:23] write program, hope it wins? [08:23] :) [08:24] there's a handful of successful strategies, three major ones that have been named after rock(stone) paper and scissors because of how they interact with each other [08:24] Heh heh. [08:25] the links describe those [08:25] Replicator, bomb and imp (or something)? [08:25] scanner instead of imp [08:25] but yeah, pretty much [08:25] imp isn't really a workable strategy on its own [08:25] but it gets combined with others for survivability [08:26] So the game takes place inside a computer (from what I gather, MARS)? [08:26] also most of the complicated warriors aren't any one single thing [08:26] yeah, well a very simplified computer [08:27] kind of a mobius loop of memory [08:27] :P [08:27] all locations are relative, which is one of the interesting things about redcode [08:27] your code and theirs get loaded up, last one still executing wins [08:28] And, the idea is to attack the enemy processes to get them to shut down or something? [08:28] more or less [08:28] it's possible for them to die without your direct intervention i guess, but yeah [08:29] you want to fuck their shit up ;) [08:29] Hold on - is KOTH.org shutting down, or just the IRC channel? [08:29] "This IRC server..." [08:30] not the website i should think [08:30] It says on the Usenet group that it's going down though. [08:30] says what's going down [08:31] The website. [08:31] well then i guess it is? [08:32] i just rebooted earlier so i can't search the scrollback [08:32] http://www.koth.org/irc-logs/ logs though [08:32] 80,69429480[11:42] I suppose all good things come to an end. KOTH.org has been around for about 15 years and it's a shame to see it go. [08:32] 80,69429480[11:42] Do you think the KOTH.org hills could be hosted at SAL? [08:32] http://groups.google.com/group/rec.games.corewar/browse_thread/thread/74f28d71be1a024d# [08:32] looks for reals [08:32] sucks :( [08:33] Great. I find something, and half of it's going down the drain. [08:33] that kinda makes me sad [08:33] well, i don't know you could say that [08:34] Join: Altair joined #corewars [08:34] there's other hills [08:34] i wonder how active the koth there has been anyway [08:34] It looks like KOTH had a fair few though. [08:34] yeah, it's certainly venerable [08:35] it's certainly not the end of corewars though [08:35] the people submitting there will just submit somewhere else [08:35] I hope not. [08:36] i don't know if the hills themselves will get transported, but that'd be nice to see [08:38] It would be. [08:38] I don't like watching anything die. [08:40] I wonder if this will be like FileFront - the guys get saved at the last second. [08:43] Join: fiveop joined #corewars [08:43] sry one question. Is there a list of integers for stepsizes of scanners/bombers which have gotten good result? [08:44] doubt it [08:44] k [08:44] i don't think the step size is as important as what you're actually doing [08:44] but there are programs to find step sizes that divide up the core well [08:45] like corestep [08:45] well, Im reading a paper about the mathematics of stepsizes in corewars, but it is not that easy to udnerstand. my english isnt that good either [08:45] it'll print out a list of step sizes for you that are "good" [08:45] no need. [08:45] i can probably help "translate" :p [08:46] haven't read that one though yet [08:46] Must be a lot of depth in CoreWars. [08:46] I have corestep myself and would have used it but corestep isnt axatcly what I want [08:46] like many good games it has simple rules and complex possibilities :) [08:46] what is it you want? [08:47] well. I think I read the paper another time. it confuses me a bit ...its too early [08:47] How long have you been playing Corewars, Altair? [08:48] stepsizes which arent cannot be divided....how to say >.< corestep just calcs >.< [08:48] yeah, hell it looks really mathematical to me ;) [08:48] Altair: you want a step size that is relatively prime [08:48] that is, if it didn't kill your warrior, it would end up hitting every square in core after 8000 bombs? [08:49] I've known Corewars for a few years now but I really started ..mm...3 Months ago or so. before I wasnt able too understand redcode. [08:49] yeah myndzi, you got it [08:49] corestep -m 1 8000 [08:50] o.O? [08:50] 1 works o.o [08:50] >.< [08:50] why shouldn't it? [08:50] sry^^ [08:50] :) [08:50] by the way, core wars bombing step sizes .. back when i first discovered corewars, inspired me to create some battleship strategies lol :) [08:50] even distribution isn't best for battleship though, but still [08:51] Checkerboard is good for battleship. [08:51] yeah, but what order do you do the checkers in? [08:51] Why did you ask BlueMaxim? [08:51] first i divided the board into checkers and then i chose a step size too :P [08:51] played around until i found a pattern i like [08:51] still use it hehe [08:53] ^^ [08:54] Nice. [08:54] if it wasn't proven i'm a nerd before, well... [08:54] :P [08:54] Don't worry. [08:54] my brother has a love/hate relationship with playing board games with me [08:54] hell, i even have a strategy for playing BS [08:54] We all are at this stage. ;) [08:54] :) [08:54] well, that's a card game [08:54] anyway [08:54] Good to have strategies. [08:55] yeah, but BS is more of a social game [08:55] it feels like cheating to be all analytical [08:55] What's a card game? [08:55] ... a game played with playing cards? [08:55] *The card game you were talking about [08:56] BS.. probably has a bunch of other names [08:56] you deal out the deck to everyone, and go around in a circle [08:56] first person has to play 1 or more aces [08:56] next person deuces [08:56] threes, fours, fives, etc. [08:56] kings wrap around back to aces [08:56] you play em all face down in a pile in the middle [08:56] anyone can call "BS" on someone if they don't think they played what they said they did [08:57] then you check, and if they told the truth you pick up all the cards [08:57] but if you caught them out they pick the cards up instead [08:57] first to go out wins [08:57] Oh, Bullshit. I suck at that. [08:57] i'll tell you the cheater trick to winning [08:57] ;) [08:57] by the number of people i organize my hand by what order i'm going to play all my cards [08:57] see what i'm going to get stuck at [08:58] and either try to pick up those cards by calling BS on people and picking up cards [08:58] or sluff off everything after them before i get there [08:58] slough? probably slough [09:00] Cool. I don't have a good 'poker' face though - I get caught out almost every time. [09:02] MSG: Ping timeout: 240 seconds [09:04] Join: myndzi joined #corewars [09:04] well [09:04] this firewall is starting to get on my nerves [09:04] too bad, i wanted to like it [09:05] anyway, re: BS; as long as you plan ahead it's pretty easy to do well [09:05] make sure you end on cards you actually have [09:05] and play the ones you need to get rid of with other legit cards one at a time, spread em out [09:05] you can gamble more to win sooner [09:05] that's about it :P [09:05] Nice. [09:06] Have you tried using ChatZilla? [09:06] it's my computer that crashed, not mirc [09:07] Whoops. [09:08] Have any idea why? I'm the computer nerd around my parts. [09:08] yeah, firewall is unstable [09:09] some sort of resource leak it seems like [09:09] things start to get weird before it goes down [09:09] can't save files, audio doesn't work [09:09] Is it Windows Firewall or another one? [09:09] things start to get really laggy even though there's no cpu in use [09:09] like i was stuck for half a minute when i tried to scroll part of that page on the step sizes [09:09] Online Armor [09:10] MSG: Read error: Connection reset by peer [09:11] Join: BlueMaxim joined #corewars [09:11] Sorry, my FireFox crashed. [09:11] hehe [09:11] Don't have a clue why, either. [09:11] speaking of instability [09:11] maybe i caught iSwineFlu [09:11] and transmitted it to you! [09:11] * BlueMaxim slaps myndzi [09:12] what [09:12] my script didn't slap you back :( [09:12] I don't like iSwineFlu. [09:12] oh, cause you just joined [09:12] right [09:13] Funny joke though. [09:17] * BlueMaxim holds breath to keep from transmitting flu...but a sneeze is coming. [09:19] What's your best bot right now, myndzi? [09:19] best... bot? [09:19] Warrior, whatever you call them. [09:20] oh [09:20] i haven't actually done anything in a long time [09:20] lol [09:21] i can't really say i've been competitive [09:21] also i spent more time pursuing things that are interesting than things that are successful [09:21] Interesting...? [09:22] interesting to me [09:22] * myndzi shrugs [09:24] This community is really small. [09:26] kinda [09:26] ;) [09:26] it's sort of a niche hobby anyway [09:26] Just too bad there isn't more people around. [09:27] it sorta seems like most of the possibilities have been exhausted [09:27] Join: waknuk joined #corewars [09:27] i don't think it's likely to "break out" unless perhaps something new happens to add a lot more to explore [09:27] people seem to be having a good time playing with evolution [09:28] What was the idea with evolution anyway? I saw something on it... [09:28] er, [09:28] Join: impomatic joined #corewars [09:28] to evolve a successful warrior [09:28] Hi :-) [09:28] hi [09:28] Hello. :) [09:28] Hi [09:28] well, impomatic is a corewars rockstar, you can ask him all your questions :) [09:28] i have to go to bed [09:28] Hi myndzi, BlueMaxim, waknuk :-) [09:29] Yay! :-) [09:29] Rockstar? [09:29] In what way? [09:29] he's a successful veteran redcoder? :P [09:30] BlueMaxim: how did you discover corewar? [09:30] hell, i remember his name from when i was a kid and first discovered this :) [09:32] He must be a GOD then. [09:32] That's a first for me. [09:33] All bow down and pay tribute! [09:33] * BlueMaxim bows and begs for divine intervention [09:35] Oh, BTW - my nick is supposed to be BlueMaxima, but this seems to cut short to 9 characters. [09:35] That shouldn't be a problem once everyone moves to freenode [09:36] Fair enough. [09:38] * waknuk hadn't noticed the creation of Newton's hills. [09:45] Has anyone actually tried to improve upon the Corewar framework yet? [09:46] The games rules / instructions? Many [09:47] Because Corewar could be improved in several ways, in my opinion. [09:47] BlueMaxima: http://corewar.co.uk/opcodes.htm [09:47] Many people have interesting ideas -few implement them and without that they don't get used [09:49] Why doesn't anybody implement some of the opcode ideas? [09:50] Some have been, but never became popular. [09:51] If you take a look at ARES that's a Mars with quite a few extra opcodes. [09:51] Or the old Mars systems in http://corewar.co.uk/mars [09:51] But the extra opcodes wouldn't work in other MARS, would they? [09:52] The pmars implementation included some which were experimental and have become established [09:53] Has anyone got a copy of the pmars patch the implements IJN and a few others? [09:53] Or the pmars direct sound output patch? [09:55] Not me. Did you ever try the sound version? [09:55] Doesn't anyone have them? You'd think they'd be around somewhere... [09:55] Yes, and it made an awful noise [09:56] BlueMaxima: I've search and can't find them. I've been trying to find everything relating to corewar and archive it so nothing else is lost. [09:56] Why not ask the people who coded pMARS? [09:57] I know who coded the patches, but they no longer have a copy. [09:58] It's just one of those things, people upgrade their computer, or websites go offline and stuff gets lost. [09:58] Tried archive.org? [09:58] Yes [09:59] Join: X-Scale2 joined #corewars [09:59] Tried any other similar websites? [10:00] MSG: Ping timeout: 240 seconds [10:02] Similar to archive.org? I didn't realise there's anything else. They're definitely not on any corewar site. [10:03] Surprising that people who code stuff lose it. [10:03] Nick Change: X-Scale2 changed nick to NorthStar [10:03] :-) [10:06] Keep looking. [10:10] We believe in you. [10:10] * waknuk thinks impomatic is the google of cw [10:10] * BlueMaxim types "Corewar easy-to-understand tutorials" and clicks "I'm Feeling Lucky" [10:12] MSG: Read error: Connection reset by peer [10:12] Dangerous search, that! [10:13] Join: BlueMaxim joined #corewars [10:14] Dangerous? [10:15] Destroyed your irc connection... [10:16] No, FireFox terminated. Something's wrong with this thing. [10:18] So where did I'm Feeling Lucky send you? :-) [10:20] Down the 'The process needs to terminate' road. [10:21] A place I don't like to go. [10:34] its annoying when people fill up the backlog, i would have to, like, read it [10:35] Backlog? [10:36] * impomatic promises to speak less for Mizcu's benefit :-P [10:36] usually when i come back from school there's max fine lines of text [10:36] five [10:36] guess ill just go through regardless [10:36] Hi Mizcu - I still always see you here - helping all the newcomers [10:37] just doing my duty in the giant machinery [10:38] I'm a newcomer - any tips? [10:38] ask a lot of questions [10:39] Everybody around here's funny. [10:41] Gotta go, bye everybody. [10:42] to balance your face when you put your first "proper warrior" against a benchmark [10:42] Uh, what? [10:44] I don't care. Bye. [10:44] MSG: Quit: ChatZilla 0.9.84 [Firefox 3.0.10/2009042316] [10:44] i already had two lines of explanation written, darn [10:44] well, Altair and myndzi can recall theirs when Blue comes back [10:53] ^^ [11:03] MSG: Remote host closed the connection [11:31] Join: M0wLaue joined #corewars [12:21] are there programm which create quickscanner according to parameters like coresize , maxlength, etc. ? [12:26] or if there arent, why? [12:34] I'm not aware of anything, but it's a good idea. [12:34] Maybe one in JavaScript to run on a webpage would be a good idea [12:41] yeah. that would be nice. another question^^: has there been any quickscanner which worked well with another bomb than dat? [12:46] Not sure about working better than DAT [12:48] David Moore had one with jumps to a SPL/JMP pit. I had one with a MOV somewhere, 1 bomb. [13:10] mm.......if we have a q^x scanner x>=2, would it make sense to jump from the attack part back to the scanning part? [13:30] noones here or stupid question :D? [13:40] Not in my opinion. Early in the game, the opponent is usually a big lump at one location. [13:40] k......and would it make sense scanning in stepsizes >maxlength+1 ? [13:40] If you bomb it and miss, by the time you get another go, it'll be a small lump booted away, or multiple lumps spread throughout core. [13:41] stepsizes in a qscan, an step >maxlength should be just as good. [13:42] mm....k, thank you very much [13:42] has there been any ideas for q^5 btw? [13:42] Well, might as well jump to a normal scanner. A qscan couldn't possibly scan enough locations to have a decent chance of finding a small opponent [13:43] y [13:43] Well, I think there's been a few improvements on q^4 [13:43] yes, so called q^4.5: http://corewar.co.uk/cw/cw092.txt [13:43] I'm sure Inversed has something interesting he hasn't published yet! [13:44] ^^ [13:45] the q^4.5 contains so much labels that its not that easy for me to udnerstand :D [13:54] Most Q^? are tricky to understand! [13:55] There's also Paul Kline's qscan from last year. [14:01] mm......has it a special name? or is there an article about it? [14:06] slowQscan, linked from this article http://impomatic.blogspot.com/2008/05/improved-88-quickscanner.html [14:10] There's probably more info if your search r.g.cw [14:11] k, thx [14:38] Join: Fizmo joined #corewars [14:39] hi hi [14:56] Hi Fizmo [14:57] hi John [15:01] MSG: Ping timeout: 240 seconds [15:02] Join: OoS joined #corewars [15:03] Nick Change: OoS changed nick to impomatic [15:36] mm........what kind of q scanenr would this be? http://pastebin.com/m5f9efc14 (I know it is not very optimised atm, especially the bombing part) [15:55] http://pastebin.com/m23557428 fix some bugs but it still is not working fine [15:55] Altair: it's a bit like the original Q^2 (which was sne seq jmp), but bits from other qscans added. [15:57] >.> i dont get q^4 or q^4.5 >.< [16:10] Try Joonas's article from Core Warrior [16:13] It's in cw84 and there's a list of other qscan articles here http://corewar.co.uk/biblio.htm#quick-scan [16:16] thx. I try to fix my scan but there are a few realyl stupid mistakes [16:28] seems like I finally fix it now [16:40] does simulators replace all unused space of my warrior with dat $0,$0, so that there is 7700 positions left the other one could use? [17:03] 7801 other positions I think [17:04] o.o? [17:05] for a warrior of length 1? [17:05] If coresize is 8000, warrior 1 uses 100 cells (even if it isn't that long) [17:06] That leaves 7900 for the opponent to sit in. Because the opponent uses 100 cells, the are only 7801 start postions for it [17:06] Even if it's length 1. Because mindistance between warriors defaults to maxlength [17:08] and with mindistance? [17:09] sry, i dont get it >.< [17:27] The number of different starting positions is alway CORESIZE - (2*MAXLENGTH) + 1 [17:38] o.o [17:39] CORESIZE-(maxlength+2*mindistance) had I assumed [17:39] mm....no, for a warrior with 100 length coresize-(maxlength*2+2*mindistance) [17:55] sry, why is this wrong? [18:05] mindistance is from the start of one warrior to the start of another [18:05] So two length 100 warriors with mindistance 100 might not have a gap between them at all [18:06] Sorry for being slow to respond by the way ;-) [18:16] o.O [18:16] I tought mindistance is between the start and the end of two warriors o.o [18:20] hmm, I never way working with that [18:21] was [18:22] mm......... [18:22] at least not in detail, but part of the pmars setting [18:22] latest pmars draft says: MINDISTANCE: value of -d parameter (default: 100) [18:23] for quickscanners, is the possibility to find one everywhere equal throughtout the core? [18:23] I think it depends on the used -F number [18:24] if you let run all possible distances than it should be equal [18:26] I think the trick is to optimize the spreading of the quickscanners scanned locations [18:26] because the table based calculations gives chances for scan pairs which are quite close to each other [18:27] sry, please say that...in a non so technical way @_@ ......im a bit confused [18:28] the quickscanner scans certain locations in the core [18:28] yes [18:28] as better they are spread over the whole core as better it should be [18:29] wait a moment, back soon [18:30] k [18:31] mm......so every cell in the core has the same probability to find the other warrior there in the first scan? But would should be the distance between the scans? CORSIZE/2? [18:31] +e [18:45] MSG: Ping timeout: 240 seconds [18:48] I would think (CORESIZE-2*MAXLENGTH)/number of scanned locations [18:49] Join: M0wLaue joined #corewars [18:52] and what should the number of scanned locations be? [18:52] that the scan pattern matches cells with spaces of 100 ? [18:53] spaces of 99, sry [18:53] MSG: Ping timeout: 240 seconds [18:57] Join: M0wLaue joined #corewars [19:02] MSG: Ping timeout: 240 seconds [19:09] Join: M0wLaue joined #corewars [19:16] MSG: Ping timeout: 240 seconds [19:22] Join: M0wLaue joined #corewars [19:29] MSG: Ping timeout: 240 seconds [19:57] mm........ [20:06] that's a good question [20:06] Join: M0wLaue joined #corewars [20:06] too few compared to the hillsize and maxlength makes it less effective [20:07] too much makes it too bulky and slow [20:07] above a certain lenght it's for papers a good decoy to confuse and slow down scanner [20:08] you can read this one [20:08] http://www.corewar.info/lexicon/paper7.htm [20:10] it's difficult to say what is the perfect number of scanned locations and their pattern [20:21] mm......k, thanks very much [20:22] my q^(no idea) scanner doesnt do any good to my bomber.....mm.... [20:23] well, I just used and take advantage of the published one [20:23] but don't forget to boot away your bomber [20:24] course I did^^ [20:25] ^^ [20:48] >.< [20:48] Load error: no DPMI selectors===================== [20:48] Returncode : 27392 [20:49] close the command window and try again [20:49] k. and is it possible to skip phase 2, too? [20:49] ;optimax 34 command line says [20:50]