[00:13] where? [00:27] a new person submitted something [00:27] and this J.R. guy -- he's not in here is he? [00:33] currently, not [00:38] i mean, i don't think i've seen him/met him [00:38] but he is active at least [03:13] Join: MixolHome joined #corewars [03:20] can i call you sir mixol'ot? [03:23] hehe, i do get that occasionally [03:23] the name is short for Mixolyde [03:23] Sir Mixolotl? [03:23] sometimes, Sir Mixolyde [03:24] * MixolHome likes big racks and he cannot lie. [03:24] is it bad that the first thing i thought of was rack mount servers? [03:24] here, not [03:24] just ask Bvowk about racks [03:25] ;) [03:25] not bad, but definitely informative [03:25] i once freestyled a verse of parody of that song about mecha D: [03:26] have you seen the folksy cover by JoCo? [03:26] nope [03:27] aside from Re: Your Brains, i don't think i've really taken to any of his songs [03:27] there is of course Richard Cheese, though [03:27] this is definitely the greatest cover of anything ever done http://www.youtube.com/watch?v=Ltjbnyvq_SI [03:31] ok yeah i had heard that before [03:31] hehe [03:31] damn these magnets are strong now that i have more of them [03:31] i can't set some down on the table to work with the rest, it goes ZIP! [03:32] 512 5mm (spherical) neodymium magnets [03:32] now i can make the big dodecahedron! [03:33] bunky ball time! [03:33] bucky [03:34] i am not sure you can make a bucky ball with magnets? [03:34] maybe i can, i dunno if i can make it do a hexagon [03:35] how bout that, they stick [03:35] but they don't bend as well as the pentagons [03:35] hmm [03:36] the middle one wants to come out if you put it near some others ;) [03:39] ...like gay men? [03:41] couldn't tell ya [03:53] so what's popular on the beginner hill these days? [03:53] like i would know ;> [03:53] i am still just playing around [03:54] all the warriors there are weak [03:54] :( [03:54] and theres 5 free slots after some previous warriors have been killed off by the author [03:59] well that's the place for a beginner like me then ;-) [04:00] or is nano or tiny better for a newb? [04:01] lolol [04:01] :( [04:01] nano is surprisingly brutal for such a small hill, and tiny has quite a variety of tactics not seen on the b-hill [04:01] i tried to submit a thing to tiny and nano too i think [04:01] not even close :) [04:03] gotta go to school now, ill check back later. [04:04] well bedtime for me, still reading examples and stuff. it's a lot to sift through and try to figure out where to begin [04:04] other than "take a known warrior and try to make it better" [04:05] i've been doing the opposite hehe [04:05] i don't want to take any code [04:06] starting from scratch? [04:08] something like that [04:09] i've been having fun making random weird things rather than the tried-and-true [04:09] seems kinda boring just to make the same thing as 100s of other warriors ;p [04:09] gotcha, that's cool too [04:09] yeah, the ol' RPS [04:09] but i'm learning about them too [04:10] eventually i'll try and make something good i guess [04:11] if only to beat mizcu before his warriors expires :> [04:11] warrior* [04:11] heh [04:12] hm i thought of something fun maybe [04:12] i wonder if i can make a paper to launch bolos :> [04:15] what's a bolo? [04:16] er, sorry [04:16] its one of my warriors on the hill [04:16] its kinda a hopper i guess? [04:17] the processes play leapfrog and bomb things [04:17] :) [04:21] oh wait [04:21] i'd have to carry all the code with it, that'd be a fat paper :( [05:15] Join: myndzi\ joined #corewars [05:15] MSG: Read error: Connection reset by peer [05:15] Join: evitable joined #corewars [05:18] MSG: Ping timeout: 245 seconds [08:09] myndzi\: if it wasnt for the leading dat #, it wouldnt be even hard [08:46] Join: fiveop joined #corewars [09:39] MSG: Quit: humhum [12:10] Join: fiveop joined #corewars [12:47] Join: MixolWork joined #corewars [14:41] Mornin [15:52] Mizcu: so i noticed! :) [15:52] it's not worth it anyway [15:52] Nick Change: myndzi\ changed nick to myndzi [15:52] i did something similar though [15:53] Join: CoreOld joined #corewars [15:56] its basically an 8 process paper with a bombing run, but the bombing run has a step instead of just 8 procs in a row [15:56] it still sucks :> [15:57] scores a decent amount better'n bolo though [16:01] what do you mean by the bombing run has a step instead of 8 procs in a row? [16:01] er, instead of like [16:01] * MixolWork is going to ask lots of newb/terminology questions [16:01] mov something, it appears that's what many/most papers do (?) [16:01] like an imp ring/spiral? [16:01] instead of that, it bombs 8 processes separated by a gap, more like a stone would [16:01] like a clear [16:02] there are 8 processes bombing at the same time? [16:02] yeah [16:02] and their targets are all seperated? [16:02] s/seperated/separated [16:03] yeah [16:03] gotcha [16:09] you used to play? [16:10] i modified benchmark warriors a couple of times and submitted them about a decade ago [16:10] but didn't make much progress before my gaming ADD set in and i moved on [16:11] I just started a new job a few months ago, and found out that one of my coworkers used to do assembly code for embedded systems [16:11] and that reminded me of core war, so i sent him some links and have been digging into it again [16:12] my thing is that i think i find reading docs, and strategies and figuring out how things work more than trying to write something myself [16:13] i'm not a creator by nature, more of an editor or optimizer [16:18] well it should suit you well then.. it seems like there's little left to create :P [16:21] It's not often when someone comes up with a whole new strategy [16:21] But every so often someone discovers a new kind of an existing strategy [16:21] indeed [16:21] that's more my territory, though i doubt i'll be versed enough to cover any :) [16:22] seems like the ssd pretty much owns papers [16:26] how does the scissors style work? [16:26] isn't that scanners? [16:27] http://corewar.co.uk/scanner.htm :-) [16:27] weird [16:27] glass almost never loses against this thing with the quickscanner on it, but it's like 30% otherwise [16:27] and the quick scanner never seems to help o_O [16:28] thanks met [16:28] it seems like the hill would start to tilt toward one strategy for a while and then back to another, never quite reaching equilibrium [16:29] as things go in and out of fashion, so to speak [16:30] out for lunch, bbiab [16:44] The hill balance changes as new ideas are discovered [17:08] back [17:14] i do believe i know now why i was inclined to dump this idea off [17:14] it socks [17:21] * bvowk tilts the hill [17:24] what, you tried to save your warrior from falling off and ended up tilting it? [17:28] no.. more like a pinball machine :) [17:28] how's it going btw miz. [17:28] must be starting to look like fall there too. [17:28] been raining whole afternoon [17:33] yuck. [17:36] MSG: Quit: Trillian (http://www.ceruleanstudios.com [17:39] Join: willvarfa joined #corewars [17:39] evening all [17:55] Hi Will [18:05] hi john [18:05] how goes recmath? [18:06] Not very well [18:07] I've just rewritten my inner loops to keep all the variables in registers [18:07] should be a bit faster [18:11] pure GA? [18:14] No, I'm using pure brute force :-) [18:14] Nothing clever [18:41] it poured for a few hours here in the mid-atlantic, now it's kind of nice out [18:41] still mostly overcast, but a bright gray overcast [18:41] evolving some sort of algorithm, metcalf? [19:02] MixolWork: I'm entering this http://recmath.org/contest/description.php [19:05] that is pretty damn nasty [19:08] brute force? mad :-D [19:08] It's not so bad [19:08] Approx 35 lines of Basic [19:08] QBasic? [19:08] GWBasic [19:08] I quickly hacked something up to get going [19:08] But it was pretty slow [19:09] So I rewrote it in assembly [19:09] aw, /me picks up a kidney [19:10] you are, of course, going to trounce me [19:10] About 280 lines as assembly [19:10] so whats your approach? [19:10] Oh, you're entering? :-) [19:10] got buckets of tables made [19:10] Just cycle through every possibility [19:10] there are a few possibilities? [19:11] Tables? Only two tables of primes [19:11] Yes, quite a few possibilities [19:12] I've narrowed them down [19:12] I think it's "C." [19:12] Looks like a neat Lisp problem [19:13] 1st number = odd, 2nd number = multiple of 6, remainder = multiples of 30 [19:22] i.e. 2*3 and 2*3*5 :) [19:22] the whole GA thingy is not really convincing me [19:23] given that the distribution of primes is random [19:23] i noticed that too, except it was with gaps at 3*5*7 [19:23] GAs are random [19:23] its not solving a problem [19:24] its just walking randomly over noise [19:24] GA what? [19:24] genetic algorithms [19:24] a way of searching a large space to find a good solution [19:24] but not guaranteeing to find the best solution [19:25] right [19:25] 'course, primes aren't really random [19:25] they might pass randomness tests or something but still [19:26] A GA would keep the values which are good. [19:26] Such as 2310, etc [19:26] Alternatively, you could weight your RNG towards good numbers :-) [19:27] What if you invented a time machine that found the answers in the future and then used those to seed your RNG? ;-) [19:27] then you should find the lottery-numbers [19:28] yeah, and buy all the sculptures you want [19:28] :P [19:29] If I invented a time machine, I'd have alternative plans :-) [19:29] whats her name? [19:31] or him. or some sort of alien hive-mind creature. [19:31] Hey, it's the 's [19:31] i was thinking he was thinking about Q^5 [19:31] that's a scan speed or something reference, right? [19:32] It's not about the sculpture, It's about being the best [19:32] refers to a quickscan of 5th generation (we are currently in 4th) [19:32] quickscanner [19:32] is there such a thing as a q5 yet? :P [19:32] no-one has claimed to having made one [19:33] hmm, scanning prime locations hmm? [19:35] or by fibonacci sequence? [19:35] fibonacci-bombing doesnt work, been tested [19:37] Why doesn't it work? Apart from the speed [19:37] self-hits relatively early [19:38] bad benchmark-response [19:38] mars isn't exactly a natural world simulation, either [19:38] where fib is everywhere [19:38] though if you go developing something to disprove my research, i dont care [19:39] there might be some point in combining paper with fibo, mathematically [19:40] Can't the sequence be started at an offset? [19:41] without redundant code, would be challenging [19:43] * willvarfa waves [19:44] MSG: Quit: Ex-Chat [19:45] * Metcalf waves [19:57] a replicator that replicates in positions offset by the fib sequence? [19:58] Join: CoreOld joined #corewars [20:00] MSG: Ping timeout: 245 seconds [20:10] MSG: Quit: humhum [20:11] Join: Metcalf joined #corewars [21:12] MSG: Ping timeout: 245 seconds [21:13] Join: MixolWork joined #corewars [21:23] MSG: Quit: Trillian (http://www.ceruleanstudios.com