[09:05] Join: km joined #corewars [09:06] MSG: Quit: Konversation terminated! [10:53] Join: fiveop joined #corewars [16:18] hrm. [16:21] MSG: Read error: Connection reset by peer [16:21] Join: fiveop joined #corewars [16:29] Join: Metcalf joined #corewars [16:30] so like, it's really active in here i guess ;) [16:30] Greetings... [16:30] yes. [16:30] it's wildly active. [16:30] we never shut up. [16:30] here i was hoping to lurk more etc [16:31] i must admit i've had a fascination with corewars for like a decade but i've never stuck with it long enough to climb the learning curve :P [16:31] also, hello Metcalf! [16:32] http://www.nvidia.com/object/tesla_s1070.html [16:32] I wonder if you could write a mars for those threads. [16:32] imagine a beowulf cluster of tho-- oh yeah, mars, right [16:34] i always thought it'd be fun to have a peer to peer sorta mars setup, where you could 'attack' people over irc :) [16:34] i can't think of a way to make it uncheatable though [16:35] and operate properly [16:36] Hi Evitable [16:40] Did anyone figure the answer to last week's question? [16:40] Which warrior scored 82% wins in a round robin tournament? [16:43] no :( [16:46] It was one of the first intelligent warriors and it absolutely thrashed it's opponents ;-) [16:56] intelligent as in? [16:56] my tiny warrior is plain stupid :) [16:56] Hi Fiveop. [16:56] hi [16:57] Intelligent as in sensing what's going on and reacting [16:57] qb followed by core clears, regardless of what happens [16:58] my tiny warriors are getting kinda scary smart :) [16:58] That's equivalent to scoring 246 on the hill, but with a legitimate warrior [16:58] I've got one that turns into a paper :) [17:00] Even certain abuse only scored 186.5 by cheating :-) [17:00] Bvowk: so it starts off as something else, and mutates? [17:00] yeah.. after it bombs itself [17:00] it's very cool [17:01] bvowk: under which conditions? [17:01] tiny [17:01] no I mean [17:01] does it change unconditionally? [17:01] can you send me a copy? :P [17:01] heh [17:01] I'll send you a copy when I'm done :) [17:02] it doesn't score so well tho.. it's alright against most everything.. but not great anywhere. [17:02] so it barely scrapes the bottom edge of the hill.. [17:02] what's your current evolver setup? [17:02] generation,selection,computers involved? [17:04] currently it's diversity hills, mutation with breeding that groups parents by instruction similarity.. [17:04] and I've got about 30 cores running [17:06] 24/7? [17:06] yeah [17:06] for about a year [17:07] :) [17:07] I really want to see some of your specimen [17:08] I'll give you the whole dump when I'm ready to drop something on the hill :) [17:08] I'm not saving everything this time tho.. [17:08] whole dump means? [17:09] could one apply your techniques to something more ... useful? :P [17:10] lol [17:10] thanks. [17:10] sure.. [17:10] I apply them to useful things all the time [17:11] like? [17:12] heh [17:12] well, I used the skeleton for my evolver to distribute GIS work not so long ago :) [17:12] :) [17:13] I'm just ignorant, sorry :) [17:13] heh [17:13] it's alright.. but wrangling piles of machines is pretty easy if your expectations are reasonable. [17:13] I did around 600 machines for a contest once.. we ran them for a couple weeks [17:14] it was fun.. mostly a collection of perl :) [17:14] basicly I just do the simplest thing that could possibly work, and then improve if needed. [17:20] hmhm [17:22] Core Painter left the nano hill [17:23] I notice I'm still on top of nano [17:40] MSG: Remote host closed the connection [17:54] Join: Neogryzor joined #corewars [17:55] hello [17:55] hey neo [17:56] MSG: Ping timeout: 245 seconds [17:57] hi bvowk, what's up? [17:57] not much.. [17:57] you? hacking anything good? [17:57] how are your computrons? :) [17:58] not much... rebuilding some sections of the web. A bit of optimization too... [17:58] whe web = my site [18:00] Join: alexander joined #corewars [18:01] wellcome Alexander [18:03] computrons are feeling well.. [18:03] every day i make more typos... [18:03] just got some new toys [18:07] MSG: Quit: Leaving [18:08] Join: alexander joined #corewars [18:08] time to go [18:08] * Neogryzor waves [18:08] Part: Neogryzor left #corewars [18:09] what's a newbie gotta learn to have some fun on the hill anyway? i mean, not just basics, but enough strategy to have some idea how to apply them or whatever [18:10] mostly just redcode. [18:13] heh, so any old thing will be competitive? :P [18:14] somehow i doubt it :( [18:14] not quite [18:14] but if you understand the warriors.. you can make them. [18:14] it's not really rocket science.. that's easy, but it's not much harder ;) [18:14] * bvowk ducks [18:14] I'm not the best person to ask about that however, I don't actually know much redcode.. [18:14] I evolve [18:15] i'm the type who has trouble when faced with too much choice [18:15] i don't know WHAT to code, where to start [18:15] also evolving is fun [18:15] well.. I'd say go through the intros first.. figure out how the basic warriors operate. [18:15] make a few of your own.. [18:15] then start picking apart someone elses code.. [18:15] look at the warriors on K [18:16] once you've figured out how things like papers and silks and qscans work.. you'll be doing alright. [18:16] the nano hill isn't a bad way to get started.. [18:16] the warriors are small [18:16] i learned the basics years ago :| [18:16] and it's a big hill [18:17] so the lower edge can be more accessible than other hills [18:17] sounds like a good idea [18:17] but you can't stay there forever.. you'll miss out on the more complex stuff that makes corewars fun :) [18:17] i know i knew what silk was [18:17] but now i forget [18:20] you know what i'd really like [18:20] i wonder if anyone has compiled a list of significant warriors over the years? [18:20] the ones that introduced new concepts and advanced the 'state of the game' or whatever [18:20] it'd be a fun excercise to start at the bottom and work on beating them from the ground up ;P [18:21] I think the hall of fame would be a pretty good list [18:21] and the K hill obviously indicates the state of the art [18:21] Join: Metcalf joined #corewars [18:21] heya met. [18:21] evitable is likely someone you want to chat with :) [18:22] met is a redcode wizard. [18:22] i recognize the name ;) [18:22] can you get autographs on irc? :P [18:23] thanks! [18:24] I'll send you an autographed postcard if you like ;-) [18:30] hah [18:30] I'm struggling to draw a flag for the USSR in 18x12 pixels :-( [18:31] Join: Neogryzor joined #corewars [18:32] Hi Neo [18:32] hi Met [18:33] i took a look at RSS's the other day [18:33] Ah, good [18:34] Are they working okay in Firefox? [18:34] Do you need to update the xml file? [18:35] before you showed up i was just asking about getting started .. not with learning redcode, but with what to do with it [18:35] i wanna have a go at doing something but i don't know what :P [18:36] hi evitable. Did you try to write a bomber? It is quite simple. [18:37] that is a struggle [18:37] just a loop [18:38] the flag i mean [18:38] not the bomber [18:38] pardon my intermittency, i'm at work [18:38] i've written a bomber, i just want to write one (or something) that's worth entering on a hill [18:39] but i don't know what that is [18:39] once i've got an idea what to write i'm sure i can hack it out [18:41] there is a beginner hill at SAL: http://sal.math.ualberta.ca/ [18:41] I might give up and just use the Russian flag :-( [18:41] If you don't fancy writing a bomber, you could always try a oneshot. [18:41] A oneshot looks for something, then overwrites it with a linear clear. The clear then continues and wipes the entire memory [18:42] oh i don't mind bombers or anything [18:43] i just want a starting point.... not a "newbie" starting point but something useful [18:43] :P [18:43] surely just a simple bomber isn't "good enough"? [18:44] for beginner-hill, it is [18:44] Hey, you can't be a "pro" from the begining [18:44] i fail at expressing myself [18:44] writing a newbie bomber isn't interesting [18:44] i just want something interesting ;) [18:44] not a winner [18:44] but not a teething ring either [18:45] and Mizcu: if that's so, then i guess i can do it and then play around [18:45] Take a look at some of the top stone/imps published and you will see how interesting a stone can be :) [18:45] stoney #3 at beginner-hill is a stone with imprings to support [18:45] though it used "staggered" table (read: Neo's launch) as an experiment [18:46] Oneshots work well on most hills [18:48] A oneshot was my first really effective warrior as a begginner [18:49] Metcalf: http://img138.imageshack.us/img138/8665/tinyflaghf1.gif 1000 hours in mspaint? [18:50] Neogryzor: stone/imp != stone ;) i've got some suggestions now though, so great [18:50] Metcalf: not great, but an idea [18:50] not gonna fit well in the corner like it should be [18:51] also thanks [18:52] evitable: thanks [18:55] Metcalf: i hope you don't plan to use that as is D: if so i'll put a little more effort into it hehe [18:58] MSG: Ping timeout: 245 seconds [19:11] Join: Metcalf joined #corewars [19:14] hum [19:18] MSG: Ping timeout: 245 seconds [20:43] Join: Metcalf joined #corewars [20:44] My best attempt: http://img79.imageshack.us/img79/2914/ussrdp6.png [20:49] MSG: Ping timeout: 245 seconds [20:50] Join: Metcalf joined #corewars [20:56] wb :) [20:59] MSG: Ping timeout: 245 seconds [21:00] Join: Metcalf joined #corewars [21:02] argh, that one was close! [21:10] Close? [21:10] i tried something at the '88 hill [21:14] time to go [21:14] * Neogryzor waves [21:14] MSG: [21:16] MSG: Ping timeout: 245 seconds [21:29] Join: Metcalf joined #corewars [21:34] MSG: Ping timeout: 245 seconds [22:02] er, what e-mail address should i send requests to? :| [22:03] http://sal.math.ualberta.ca/help.html?login=700 <- seems to say what to send, but not where [22:03] ? [22:03] yup [22:03] how about that [22:04] i don't suppose i can run something against the hill without actually entering it [22:04] i thought i recalled such an option on the koth server from reading the instructions some time in the past, but this isn't the same server anyway [22:05] nope. [22:06] oh well, we'll see what happens [22:10] k reeves? [22:10] :P [22:10] looks like about even odds [22:10] hee [22:11] it's only slightly more interesting than a .33 bomber [22:11] it looks pretty in corewin though! [22:11] :P [22:12] also yes [22:15] actually it's probably less good, i just wanted to play around :| [23:29] MSG: Quit: humhum