[06:15] *** jkw_koth changes topic to 'My virus's better than up your virus! My virus's better than yours!' [11:43] *** fordbc (fordbc@2Cust42.tnt2.austin2.tx.da.uu.net) has joined #corewars [13:14] *** Strike (strykeslam@dhcp212.dhcp1.muskingum.edu) has joined #corewars [14:58] *** Strike has quit IRC (Leaving) [15:33] *** Lucky (ufoludek@148.81.93.178) has joined #corewars [15:34] Hi, there! [15:34] Nice topic P-) [15:38] hello [15:38] Hello! [15:38] !!! [15:39] ? What's so strange? [15:39] you used a lot of exclaimaiton points, I had to catch up [15:40] So... what would we talk about? [15:41] don't know =) [15:41] It is a lot of time to the meeting [15:41] yes, bout an hour 15 [15:41] but I think I can't stay long. [15:42] What time is it in your place now? [15:42] 3:45pm [15:43] You're in America? [15:43] yup [15:43] *** Strike (strykeslam@dhcp212.dhcp1.muskingum.edu) has joined #corewars [15:43] last I checked [15:43] hi [15:43] hi [15:43] I'm from Poland and it's 15 to 11pm here/ [15:43] Hi Strike [15:43] thats not too late [15:44] Yes, but I have to wake up early tomorrow. [15:44] that sucks [15:44] yeah, that can be a problem :/ [15:45] It's nice that you decided to get here earlier too. :) [15:45] well, I'm watching football so I just logged on [15:46] I'm expecting to be somewhere else when the meeting is suposed to start [15:46] but I still want to show my support :) [15:46] Strike, I did /whois and saw David M. David Moore? [15:46] last week I was just watching football at the meeting time and I forgot to log on [15:47] nopers [15:47] David Morgan [15:47] * Lucky doesn't have TV-set. [15:47] So nice to meet you, David. [15:48] [turns off his TV since his roommate isn't watching anymore] [15:48] Nice to meet you too, Lucky [15:48] My name is Lukasz (Luke in English). [15:48] you prefer Luke, Lukasz, or Lucky? [15:48] But maybe better we use nicks? [15:49] easier to remember if we use nicks, IMO [15:49] Right. [15:49] It's amazing that people using IRC never use their real names. [15:50] I am =) [15:50] heheh :) [15:50] Oh, right. You're the exception from rule. [15:50] I don't IRC much, but maybe part of it is there are so many with the same name? [15:51] I know there are *tons* of Davids :s [15:51] I have never met Lukasz :) [15:51] I think most people just like to use some cool 'hacker-type' mickname [15:51] There are 100s of Luckys. [15:51] yeah, thats probably part of it too. [15:52] there's also a lot of other "Strike"s, but I almost never bump into them [15:52] plus it might give it away if those messed up guys who like to pretend to be females used their real name [15:53] LOL!! [15:53] I hate pretending to be females. [15:54] yeah [15:54] I kow they are doing it for fun, but it's a lie. [15:54] But maybe we start talking about CoreWar? [15:55] that would be good, heheh [15:55] whats corewar? =) [15:55] :p [15:55] :S [15:55] maybe jkw will explain you :) [15:56] you guys should submit some warriors to move mine into first place =) [15:56] heheh I have to make that'll even get on the hill first! :< [15:56] I already had to suffer looking at developement color scheme jkw was using on the new matrix displays [15:56] I agree: you guys should submit some warriors to move mine into fistr place :) [15:58] heh, what place is it at now? [15:58] none [15:59] will be hard to move it up then [15:59] I DIDN'T GET ON ANY HILL!!! (yet) [15:59] same here (sigh) [15:59] I'm still exploring the basic strategies, I guess [16:00] just rename a good warrior and submit it =) [16:00] Till now I was only testing some kind of replication loop. [16:00] That wouldn't be fair [16:00] I was just kidding [16:00] I know [16:01] most of my warriors are modified versions of other warriors tho [16:01] thats not a bad way to go [16:01] but I think I learn better if I start from scratch instead of modifying other stuff [16:02] perhaps [16:02] Sometimes it's hard not to copy someone else idea, but I'm still trying to think up something new. [16:02] It's LOT OF FUN! [16:02] I learned a lot by figuring out how existing warriors worked [16:03] That's right, but sometimes it's written so complicately. [16:03] :( [16:03] especially with the self-modifiying warriors [16:03] well, thats why you learn a bunch when you finally figure it out [16:03] yeah, half of the modern stuff has too much self-modding [16:04] pretty much has to to survive [16:04] I tend to give up if I can't figure it out after a minute or so [16:04] but what I do figure out I try to use in my next warrior [16:04] You must be specialist in it :) [16:05] & then I can understand a little more next time I look at the others' [16:05] especially when I was starting, I could only start to understand things but running the code with the debugger [16:06] I was starting with a paper (real, not warrior:) and pencil :) [16:06] ack! [16:06] yeah, I used that when I first found corewars several years ago [16:06] I've tried that, too hard to move/change instructions [16:07] * Lucky has a lot of imagination. [16:07] heheh :) [16:07] I like Notepad that comes with Windows [16:07] that's what I'm using now [16:07] Sometimes I think about my warrior using nothing but my imagination. [16:07] didn't have it back then though, & i did most of my work during class [16:08] I use Notepad to create WWW pages. [16:08] isn't it versatile? :) [16:09] thats how I made my WWW pages too [16:11] Right now I'm working at Polish page about CoreWar to make this game more popular in Poland. [16:11] cool! [16:11] yeah, cool [16:12] However there were 3 Polish Championships in CoreWar now it's rather dying. [16:13] :( [16:14] One of Polish Computer Magazines have decided to make those Championships, I took a part in one of it, but it was several years ago. [16:14] how did you do? [16:15] There are some Polish MARSes too. [16:15] How I did? Don't ask... :( [16:16] Actually I tried to organize one more championship, but totally fails. [16:18] Now I'm trying to do something like KOTH, but smaller. [16:18] I wish I had more time :) [16:18] hopefully that will work [16:18] sounds interesting- what do you mean by 'smaller'? [16:18] Thanx [16:19] smaller - only one, maybe two hills (standard and multiwarrior) [16:19] cool [16:20] and I will have to do battles by myself (I don't have KOTH scripts or something like that) [16:21] hm.. that could become very time consuming [16:21] Maybe jkw could present it to me? [16:21] I know, Strike. [16:21] I think there is a utility that can handle something like it [16:22] ? [16:22] probably in ptools.zip [16:22] How does it work? [16:22] you can give it a list of filenames of warriors and it will fight them all against each other [16:23] and give you the results [16:23] well, ptools has something that will round-robin a list of warriors, if thats what you mean [16:23] yeah [16:23] He'd still have to set it up & send the results out himself though [16:24] OK, but can it get warrior straight from my e-mail? [16:25] no, it doesn't have any email support (that I know of, at least) [16:25] you'd have to get scripts from jkw or someone for that [16:26] What a pity... :( [16:30] ok, here's the problem I'm having with my latest warrior attempt: [16:30] "calling all stations, does anybody hear me" [16:30] * Lucky is listening to the radio. [16:30] [anything good on?] [16:30] [Genesis] [16:31] go on Strike [16:31] I need a way to have no more than two processes running at a time, [16:31] each one in its own little block of code [16:32] just SPL block1, 0 ; JMP block 2, 0 [16:32] I can't just have each block SPL to the other, 'cause then I get too many processes [16:32] & that makes it work wrong [16:32] but I need to keep a process going in each block [16:33] so if one dies, I want to spawn a new process in its block [16:33] why? that would mean that something in the block broke... [16:33] self-repairing code! :) [16:33] so make 3rd process, which "takes care" about two others? [16:34] I don't have much failt in self-repairing strategies [16:34] hm.. might work, & I don't have to worry about speed too much ATM [16:34] well, after seeing my bomber not work against imp-rings, I want to mess w/ self-repair for a bit [16:35] I see. [16:35] ahh... have to fall into a gate or bomb with anti-imps bomb [16:36] I don't know of any successful self-repair stratgies except for imp-rings [16:36] Can you explain me what's the difference between gate, bomb and anti-imp bomb? [16:36] yeah, what's an anti-imp bomb? [16:37] bombing with DAT <2667, <5334 [16:37] what's the meaning? [16:37] that way if the imp executes it, it messes up the next two imps instructions [16:37] [16:37] it has decrements to mess up the next instructions of the imp [16:38] means you can't use the bomb for aiming though.. [16:38] Oh, I see. [16:38] usually use the line doing the bombing for the aiming [16:38] Have you heard about torpedo? [16:38] can also use DAT <2667, XXXX if you have to use the field for something [16:38] torpedo? don't think so [16:38] you mean tornado? [16:39] MOV $0, <2 [16:39] ah! I was just messing around with that! [16:39] you can also use MOV #1, {1 as anti-imps bombs [16:40] but with those you have to use a core-clear afterwards [16:40] I have never tried to use it, but I think it's not good weapon if the warrior must be fast. [16:41] waitasec, that torpedo thing isn't quite what I thought it was... [16:41] What were you thinking about then? [16:41] still interesting- I'll put it in my (nonexistant) list of imp modifications [16:42] I was thinking of MOV $0, 2 MOV $0,2 (thats 2 lines) [16:42] Remember, it's easy to stop torpedo: DAT $0, $-1 [16:42] I think linear imps are too weak, you really need to use rings [16:43] What's the use? All imps tie! [16:43] rings don't always tie [16:43] unless you use them as bombs... [16:44] yeah, rings don't have to tie [16:44] linear imps are unseen in modern warriors [16:44] which is why they're unexpeceted [16:44] all I have to do is get one to work right... [16:44] nope [16:45] linear imps can only tie and will pretty much die at the gates the kill normal imps [16:46] What is gate? [16:46] it wouldn't be the warrior's only strategy.. [16:46] a location in memory that you continuously increment/decrement to prevent imps from getting through [16:46] useually used in combination with a core-clear [16:46] Is it working? [16:47] working? [16:47] gates? yeah, they work pretty well [16:47] I mean what if imp is faster than gate? [16:47] most imps try to slow themselves down with enough processes so they never reach the gate [16:48] then there is a chance for the imp to slip through on the off-cycle [16:48] slow imps? sounds kinda wierd... guess its the only way to run an offensive imp though [16:48] a common gate used is in d-clear, it increments the gate 100% of the time so imps can't get through [16:50] a bunch of scanners on the hill fall into a d-clear after scanning all the locations they can [16:51] what is a d-clear? [16:51] Right, I used this strategy in some of my warriors. [16:51] My worst problem was pretty and fast changing scanning step. [16:51] you can download d-clear from planar's archive [16:52] ok, I'll go hunt it down sometime [16:52] just a clear with a 100% gate [16:52] which forced the stone-imps to start using slower imps so they wouldn't die on the gate [16:53] increment? decrement? [16:53] increment -> +1 decrement -> -1 [16:53] I know, but which is used in d-clear? [16:53] increment [16:53] oh, sorry [16:54] B-field I guess [16:54] forward clears are generally stronger than backward clears [16:54] why? [16:54] you have to kills imps from the back forward [16:54] otherwise it will self-repair [16:55] so the forward type is slightly better at killing imps [16:55] ah [16:55] you just have to clear faster than the imp runs, which isn't a real problem now that we're using slow imps [16:56] heh, thats almost impossible with a linear clear [16:56] you have to hit locations N and N+2667 to be able to kill it in one sweep [16:57] What do you think about that: [16:57] JMP $0, <1 [16:57] [afk for a bit] [16:57] so unless the imp is insanely slow, N+2667 will execute, keeping the imp alive, before you get to it [16:57] for what purpose? [16:58] JMP $0, <-2 [16:58] with two processes. [16:58] as gates? [16:59] Those two instructions are decrementing B-field of whole the core without killing themselves. [16:59] not worth it, won't kill anything [17:00] It won't, but may disturb. [17:00] I wouldn't dedicate processes just to doing that like that [17:01] many programs with self-splitting loops use a djn-stream for the loops to decrement both fields in the whole of core [17:02] *** lipo (lipo@bhe-200-251-187-187.homeshopping.com.br) has joined #corewars [17:02] Hi lipo! [17:02] Hi all! [17:04] hello [17:04] I think there's something wrong with my connection. [17:04] Very interesting are the matrices at Koth. [17:05] Yes, and very colour. :) [17:05] yeah, lots of useful info all at one glance [17:05] I am reading you Lucky. [17:06] That's good, but it's getting late and I think I should finish my discussion. [17:06] Are you talking about somethiong when I arrived? [17:06] It was nice to talk with you, guys. [17:07] About many things, lipo? [17:07] lipo. [17:07] Bye Lucky. Go sleep it is past midnight im Poland. [17:07] later Lucky [17:07] I hope jkw is saving logs and everything will be available on KOTH. [17:08] Ok. [17:08] I know lipo, I know... [17:08] Bye then. [17:08] *** Lucky (ufoludek@148.81.93.178) has left #corewars [17:10] Well I connexcted but I have no sugestion for a topic. Any ideas? [17:10] nope, I was just giving some advice to Strike and Lucky [17:11] Strike is David Moore? [17:11] no, he doesn't need advice from me =) [17:12] :Yes. I think it is easy he govive advice to us than we give advices to then. At least in corewars. [17:12] heh =) [17:12] Last Sunday, David was talking about an interesting projetct. [17:12] I think David Moore uses davmoore as a handle anyways [17:13] about corewars? [17:14] Yes. About corewars. But only experienced players like him can take the project going. [17:14] what is the project? [17:15] Maybe he want to talk about it again. Are you listening David? [17:15] Strike isn't David Moore, he is a different David [17:16] Oops... Sorry for my mistake. [17:17] The project is about to take a sample of published warriors and divide the corewars history in erasages. [17:18] Noweadays, begginners like me read the current warriors and make modifications to put warriors into the hills. But I know little of old strategies and corewars ewvolution. [17:19] yeah, same ere [17:20] The idea is to heave becnhmark warriors by eras. So a beginners can learn old strategies and try his own warrior using that starategie agaisnt warrior take make sense fot r that strategy. [17:21] that would be cool [17:22] This is a project that takes much time. David doesnt have sufficient time now. [17:22] yeah, sounds almost like a full-time job to do it [17:23] Even if made by a team, this project can only be taken by few current players. Few have the experience necessary. [17:23] yeah, I know I certainly don't [17:23] I've only been playing a bit over a year [17:24] Neither mye. I started in Nopvember 1998. [17:24] beat me by a month =) [17:25] Besides the becnhmark, David thinks of making a tutorial also, to teach the strategies and the evolution of corewars. [17:25] there are some of those already [17:26] a bunch of tutorials are for 88 code tho which ends up teaching beginners obsolete coding [17:27] Yes. [17:28] Even the glossary in the FAQ is a bit old. The ere are new terms and strategy variations. [17:29] tell the guy who maintains the FAQ then, if you give him new info, I bet he will update it [17:30] piriv [17:30] Excuse me the noise. Yes, I will make this. [17:31] Or maybe I will open a thread in the list about this. [17:31] should make a thread about the -xm hill on koth also, its useless =) [17:32] [back] [17:32] unless you like watching papers tie and tie and tie [17:33] :Yes, appears that other kinds of warriors dont have chance. [17:34] But I am not sure if that is true. I still hope that will appear a warrior that will break the equilibrium. What do you think? [17:34] scoring system there rewards tying too much over risking trying to win [17:34] no one warrior can kill 20 papers into 80,000 instructions by itself [17:35] I think you are right. I forgot about the scoring system. [17:36] I even tried a vampieric paper so that would suck other papers into self-destructive core-clears... no luck [17:37] Well. The way is to change the scoring system or the parameters. [17:37] yeah [17:37] maybe make it a lp-multi-hill or a tiny-multi-hill [17:39] LP will exclude true replicators. I voto e for tiny. [17:40] heh, papers don't do well on the standard multi-hill, don't think they would do well on tiny either [17:41] They are able to enter the hill, but they dont live long. [17:42] hopefully dracula's cape will change that =) [17:43] It was a paper that borkrroke the equilibrium of the warrior on the standadrd mul ti hill when David filled it with his handshaking warriors. [17:43] true [17:44] stuff over-specialized in order to kill of david's warriors tho [17:45] Weell , i acessed koth.org anbd Dracula is with age 2. Lets wait and see... ;-) [17:45] hehe =) [17:45] submit some stuff to age it [17:46] I am not writing much redcode these days. [17:46] me neither, so much to do, so little time [17:47] Me too. [17:48] When I learn to write a decent vampire I will submit it to Multi. Jonhs's Quiva is doing an excelent job. [17:48] quiva isn't a vamp [17:48] its misnamed =) [17:48] What is it? [17:49] he was throwing vamp-bombs at 1c orginally in testing [17:49] he figured that just throwing dat-bombs was just as good [17:50] so now its just quick-bomb into a core-clear [17:50] privmsg #corewars :Ok [17:52] Well maybe [17:52] heh [17:52] irc doesn't like your delete key [17:53] Bracu [17:53] I see. [17:54] I hope thhe log donjt stay ugly because my deletes. [17:54] can always pretty up the logs manually [17:54] Yes. ;-) [17:56] I see you are conecting form Texas. Right? What time is it there now? [17:56] just about 6 [17:57] I am with no words again. Do you sugest another topic? [17:57] hmm... getting irongate off of the 88 hill, its too old =) [17:58] any comments on using imps as bombs? [17:58] Wait a minute. I will look the Koth matrices to see what perform good against IronGAte. [18:00] anti-imps make decent bombs [18:00] the MOV #1, {1 or MOV #1, <1 [18:01] both are backward-filling? [am I reading them right?] [18:01] I remember you saing about this in the list. But I had no time to play with it. [18:01] if they hit an imp, its ends up overwriting the imp [18:02] until it gets to the front edge of the imp and then it just falls off, killing the imp [18:03] how does it get to the front? [18:03] well, if loc X executes it, then X+1 is mov 0, 2667 [18:04] so the mov #1, <1 gets the 2667, decs it to 2666 so it gets moved to X+1+2666 or X+2667 which is the next step of the imp [18:04] the #1 part forces it so the mov statement itself is moved [18:05] when it hits the end, then X+1 is dat 0,0 so the mov statement makes it dat 0, -1 and the mov just overwrites itself, leaving empty core for the imp to fall onto [18:05] ok, I think that makes sense [18:06] I understand. The problem is that IMP are more popular in 88 hill than imn 94 hills. But your idea is good and it is worth trying. [18:06] on the 88 hill you have to use MOV 0, <1 tho which is slightly worse but still ok [18:07] Yes. [18:08] so many imps on the 88 hill, should be worth a lot of points to whoever tries it first [18:09] Looking the Koth matrices I see that my sIMPly.Red is what performs better against IronGate. It will be the main warrior to loose point with your idea. sIMPly.red uses a huge IMP. [18:10] odd... iron gate devoles to a hyper-perfect gate [18:11] must be the nature of the stone that beats iron gate [18:11] sIMPly.Red is a glue Stone/IM
I think so. [18:12] well, I have to go places now :s thanks for the input, I'll see what I can do with it [18:12] But I will analyze to see how the stone alone and the imp alone scores agains IronGate. [18:12] ok, good luck core warriorering [18:13] thanks, good luck to the both of you as well [18:13] Bye Streike. [18:13] *** Strike has quit IRC (Leaving) [18:14] Well Ben, I think I will leave too. It is 22:15 here and I my Monday will begin at 7AM. [18:14] ok, seeya later [18:15] Bye, it was a pleasure to talk with you. [18:15] *** lipo (lipo@bhe-200-251-187-187.homeshopping.com.br) has left #corewars [18:15] likewise [18:26] *** fordbc has quit IRC (Leaving) [20:01] *** BrianH (haskin@199.176.148.72) has joined #corewars [20:02] Hi [21:22] *** Strike (strykeslam@dhcp212.dhcp1.muskingum.edu) has joined #corewars [21:22] *** Strike has quit IRC (Leaving) [21:22] *** Strike (strykeslam@dhcp212.dhcp1.muskingum.edu) has joined #corewars [21:22] *** Strike (strykeslam@dhcp212.dhcp1.muskingum.edu) has left #corewars [22:52] *** Janzert (haskin@199.176.148.72) has joined #corewars [22:52] *** BrianH has quit IRC (Read error: 54 (Connection reset by peer))