[00:30] MSG: Remote host closed the connection [04:05] Part: Mannerisk left #corewars [04:48] Join: Mannerisk joined #corewars [06:21] MSG: Ping timeout: 240 seconds [06:38] Join: Mannerisk joined #corewars [06:38] MSG: Read error: Connection reset by peer [07:00] Join: Mannerisk joined #corewars [07:06] Join: myndzi\ joined #corewars [07:10] MSG: Ping timeout: 240 seconds [08:58] Join: fiveop joined #corewars [09:28] MSG: Ping timeout: 240 seconds [09:44] Join: sh0ne joined #corewars [09:44] Hello [09:56] MSG: Ping timeout: 240 seconds [11:52] Join: ehird joined #corewars [12:20] MSG: Quit: Leaving [14:06] Join: yoR joined #corewars [15:18] MSG: Remote host closed the connection [17:19] Join: myndzi joined #corewars [17:55] Join: ehird joined #corewars [18:00] MSG: [19:23] Join: Fizmo joined #corewars [19:51] Greetings Dr Fizinhiemer. [19:52] Greetings bvowk [19:52] How are things? [19:53] fine fine, thanks [19:53] u? [19:53] Quite well in fact. [21:03] Join: yoR joined #corewars [21:30] MSG: Quit: humhum [22:09] Join: Pat joined #corewars [22:10] hi Pat [22:10] G'day, Fizmo. [22:10] I remember your name from the mailing list, actually [22:10] ;-) [22:11] Been watching it for ages without saying anything much [22:11] ohh, really? [22:11] Yeah [22:11] Like... Well over five years, offhand [22:12] hehe [22:12] Since I first discovered Corewars, reading old Scientific American issues [22:12] woahh, cool [22:12] I originally joined because it looked interesting, but then I realised it was all way over my head [22:12] Finally decided to have a shot [22:13] I've been googling and such, and I still have a question: Is there a name for the instruction "JMP 0"? [22:13] I heard first time of corewar in a german translation of a nature or science article [22:13] (like MOV 0,1 is an imp, of course) [22:13] hmm [22:13] maybe "sit and wait" [22:14] but there is no special name for it [22:14] I'm calling it a 'slug', but I thought I'd check if there's already a name [22:14] Not up on all the papers and stones and such [22:14] :-P [22:14] That instruction is kinda pointless, but the slug gun I wrote last night is actually performing pretty well on my systerm [22:15] Just against a basic datbomber or two, imp, imp gun, and Skybuck's JackTheSkipperIV (since I happened to have that handy) [22:15] so you throw that as a bomb around the core [22:15] And split to it each time [22:15] The Horcrux theory of immortality, if you're familiar with the Harry Potter series [22:16] if you add a "spl 0" in front of the "jmp 0" you would strongly increas your offensive power, as you start to collect the opponent processes [22:16] Though JackTheSkipperIV does kill it sometimes [22:16] Ooh, good point [22:17] to further improve bomb with: [22:17] I'll roll that into a v2 or something. Itching to set this thing loose on a hill... :P [22:17] spl 0 [22:17] jmp -1 [22:17] then you hand back the process to the spl [22:17] But that would tie up a bunch of my processes, all running side by side [22:18] ohh, no no [22:18] I mean you need to bomb with this pair [22:18] so you collect the opponents process doing nothing than wait [22:18] If they accidentally jump wrong, y'mean? [22:19] that' [22:19] s [22:19] Or if I overwrite them, true [22:19] the trick to choose the constans well [22:19] yes, sure, at the end, shortly before the battle is finished you need to overwrite all the core with dat's [22:20] then the opponent will start kill all his processes by himself ;-) [22:20] Yeah, I noticed JackTheSkipper overwrite the whole core with DATs [22:21] That's the only thing that's killed the slug gun so far, in its protected little garden ;) [22:22] Hi there Pat [22:22] honestly, Skybuck's warriors are horrible to read [22:22] Good, so it's not just me not having a clue :P [22:22] Though that'd be a factor [22:22] Hi, yoR [22:24] if you want get some better idea of cw than I suggest not look to detailed in Skybuck's code :-P [22:24] * Pat sends in the original slug gun, to watch it get massacred [22:24] I've actually blocked out Skybuck completely [22:25] loool [22:25] really? [22:25] Main reason I was looking at Skybuck's is because I just told him how to optimise that warrior a bit. Which surprised me, that there was a flaw I could see :P [22:26] well, no surprise for me [22:26] Yeah well, he has no clue and is constantly spamming this (and much more) mailing lists [22:26] How long does it usually take before I get a reply to my email? The FAQs I read suggest only a couple of minutes, but I'm not sure if they're current [22:27] Heh, it'd make sense that I've noticed him then [22:27] Most of his replies to other mailinglists are Dutch (I can read that too) and are pretty offensive [22:27] because he spend too much time in cloging r.g.c. than in writing reasonable programs [22:28] heheh, he also complains about my dynahill, grrrrrr [22:28] yoR: Charming [22:32] Join: not_away joined #corewars [22:32] g'night [22:32] evening i mean [22:32] I did wonder [22:32] Hi, by the way [22:33] hi Pat, new here? [22:34] Kinda new. Longtime lurker on the list, but this is the first time I'm actually doing anything [22:34] And by "longtime", I mean 5-10 years :P [22:35] http://groups.google.com/group/alt.ufo.reports/browse_thread/thread/f940fac5a1d038d5?q= [22:35] Come on... [22:35] Hi Neo [22:35] ow, yeah. Thats a long time [22:35] hi Fiz [22:36] Nick Change: not_away changed nick to Neogryzor [22:38] Good grief, yoR [22:40] So, what are you planning on doing now Pat? [22:41] yoR: First thing is figuring out why my warrior didn't compile :P [22:41] Beyond that, just mess around with redcode [22:41] Heh, thet errorcodes aren't clear? :) [22:42] Ah, apparently '88 doesn't allow JMP.A [22:43] I'm using the B field on that instruction to store data, and CoreWin automatically turned JMP into JMP.B [22:43] '88 daesn't allow modiffiers [22:43] Ah [22:43] * Pat checks the warrior for other modifiers [22:45] * Pat fixes, tries again [22:46] no >, *, { and } also [22:46] Cool, I haven't used any of them [22:47] Blegh, I'd miss them :( [22:47] Seriously though, warriors now seem to be all hybrids juggling the importance of different strategies. Since I don't know most of the strategies, I'm going to mess around for a while trying to come up with new strategies. Like the slug, for instance, even though it's likely to get hammered against real warriors [22:47] *slug gun [22:48] Pat: Did you try the beginners hill at SAL? [22:48] True, but still, once you get the hang of the language the warriors are readable [22:49] Neogryzor: I tried to try the beginner's hill at Pizza, but got no reply - is that server gone now? [22:49] its down since long ago [22:50] yoR: Yeah, just got to get used to the language. [22:50] Today's beginner hill is at SAL hills: http://sal.math.ualberta.ca/ [22:50] koth.org has other hills also [22:51] Neogryzor: Someone might want to update a bunch of the FAQs - everything I read pointed me at Pizza for the beginner's hill. Moot point for me now (and I'll go for SAL), but it could block new people from getting into Corewars [22:51] Yeah, I know [22:51] true... [22:51] I've been watching the rankings for years, mostly because I didn't know how to unsubscribe from them <.< [22:52] I think we all are too respectul and don't want to modiffy others work [22:52] lol! [22:52] :P [22:56] By the way, SAL is a tad misleading - after I (successfully) set my password including nonalphanum characters, I found out that passwords are supposed to be alphanum only. Heh, whoops :P [22:57] Heh, slug gun is winning battles on the beginner hill at SAL. That makes me happy [22:58] Whoops, actually the '88 koth hill [22:58] Silly me [23:00] heh [23:00] Still got spanked, of course - score of 38 or something, and the lowest on the hill is 120 or so [23:01] * Pat gives it a shot in a melee [23:04] * Fizmo waves [23:04] already late here [23:04] byebye [23:04] MSG: Quit: ChatZilla 0.9.84 [Firefox 3.0.5/2008120122] [23:04] you can put the code here if you want. Maybe we can suggest some improvements [23:05] * Neogryzor waves to Fiz [23:05] ;redcode-94m [23:05] ;name slug gun [23:05] ;author Pat [23:05] ;strategy Shoot slugs (Horcruxes, anyone?) [23:05] ;version 1 [23:05] ;date 4 February 2009 [23:05] ;assert 1 [23:05] start MOV $slug,@slug [23:05] SPL @slug [23:05] SUB #137,2 [23:05] JMP start [23:05] slug JMP 0,#-7 [23:07] Might look a tad bizarre, since a few decades of corewars development have passed it by :P [23:09] why do you send processes to JMPs? it delays your code [23:09] Makes me harder to kill [23:10] This is almost purely a passive warrior [23:11] oh. Tried imps? The are the usual deffensive component [23:11] Yeah, I've tried them [23:11] First warrior I tried, just like everyone else :P [23:12] You can drop dat bombs to kill opponents and use the imps to tie oif the bomber gets destroyed [23:15] Yeah, I've written datbombers and imp guns - didn't bother submitting them, since they've been tried before :P [23:15] I was using an imp gun and a couple of datbombers to test this one [23:16] if you are looking for something different you can try papers [23:16] By the way, how can I have a bunch of non-zero scores on the '94 Multi hill and get an overall score of zero? [23:16] Oh, and what are papers exactly? Is there a reference anywhere explaining all the jargon? [23:16] i think only wins give points [23:17] Program "Fluffy Paper VI" (length 15) by "Jens Gutzeit" [23:17] (contact address ""): [23:17] ;strategy paper [23:17] Number of wins with zero, one, two, .., 10 other winners and losses [23:17] 42 40 26 20 13 24 27 21 24 8 2 253 [23:17] Total score: 42 [23:18] I'm getting similar (but not quite as good) results with the other warriors, but my overall score is 0. I'm confused [23:19] btw, the newsgroup guide, it also explains the jargon: http://homepages.paradise.net.nz/~anton/cw/corewar-faq.html [23:20] Do you get any wins with sero other warriors alive, (the first number)? [23:20] sero=zero [23:20] 42, yes - you can see that :P [23:21] Oh... unless that's that warrior's score [23:21] aw, that war your score [23:21] Whoops, I'm an idiot [23:21] * Pat rereads [23:21] isn't :P [23:22] thats why Fluffy get 42 points [23:22] Yes, I see now [23:22] I'd been reading results from one-on-one challenges at the '88 hill, and didn't adjust. My bad <.< [23:22] Program "slug gun" (length 5) by "Pat" [23:22] (contact address ""): [23:22] ;strategy Shoot slugs (Horcruxes, anyone?) [23:22] Number of wins with zero, one, two, .., 10 other winners and losses [23:23] 0 7 14 14 9 21 25 20 22 8 2 358 [23:23] Total score: 0 [23:23] Yes, I understand now. [23:23] * Pat feels silly, but guesses he's not the first to go through pretty much the same stage [23:27] Neogryzor: Is it considered spamming to submit a bunch of warriors to a hill in quick succession? Assuming that they're not making it onto the hill, of course [23:29] KOTH can have problems if you send a lot of warriors in a shot time [23:29] It is considered spam though [23:30] Can you give me a guideline on what would start to annoy people? [23:34] Well, more then one a minute, or having 10+ warriors in an hour might be a bit of a max [23:34] IMHO anyway :) [23:34] crashing the server would be a bit annoying certainly. I suggest to send your entries one after another, (once you recivied the result) [23:34] Ah, thanks. That's beyond what I'm doing [23:34] Oh, and I wouldn't send a warrior while I'm still waiting to hear back about a previous one [23:35] You can always test against other known warriors [23:35] Neogryzor: Your suggestion is pretty much what I was hoping would be OK, but I figured it'd be a good idea to check. Don't know the etiquette yet [23:35] It is suppoused to be possible but KOTH is a bit fragile i think [23:35] The bottom of the infinite hill has a lot of bad(ish) warriors [23:35] Infinite hill? [23:35] What's the redcode-XXX for that? [23:35] Koenigstuhl, it just holds all the warriors, non is pushed off [23:36] Ah [23:36] It has 88, 94, 94nop [23:36] http://users.ociw.edu/birk/COREWAR/koenigstuhl.html [23:36] Thanks :-) [23:36] * Pat checks his email to see the results of his simple bomber [23:36] Koenigstuhl is not automated though, results will not come inmediately [23:37] Ah [23:37] But it does contain all the sources, so you can run it yourself :) [23:37] Ah, I might do that [23:37] Just download some warriors and test against those on your local machine [23:37] And I don't think anyone is surprised that a simple bomber scores very poorly on the '94 Multi hill at koth :P [23:38] Then you can try to beat them one at the time from the bottom up, also looking what they are doing and learning in the mean time [23:38] *meant: "at the same time" [23:39] Yeah [23:41] * Pat is reading the glossary [23:41] Pat: No one will laugh at your scores at the hills, don't worry [23:41] Neogryzor: :) [23:42] I wasn't too worried about getting laughed at - you've seen newbies before [23:42] hi! [23:43] hi [23:43] hi, ehird! [23:45] Also 'pro' redcoder often fail the enter a hill. I have a long experience at this. :P [23:45] lol [23:45] lol [23:45] * yoR seconds [23:45] Reassuring :P [23:45] The hill is very hard.. [23:46] Yeah, I did know that the chances of a simple warrior using a new and untested strategy with no offensive capability... not so high :P [23:54] MSG: [23:56] core clears usually do well on the multiwarrior hill [23:56] Simple bomber with increment 1? [23:57] no need too use ADD, the adressing > is enought so the code is shorter [23:57] Did you see the D-clear i post at the newsgroup? that's it [23:58] How long ago? [23:59] mmhh, let my look, it got buried under Skybuck's