[00:10] MSG: [00:11] MSG: Quit: humhum [03:03] MSG: Ping timeout: 252 seconds [06:30] Join: datagram joined #corewars [06:54] Join: fiveop joined #corewars [10:33] MSG: Read error: Operation timed out [12:23] Join: Mizcu joined #corewars [14:29] MSG: [14:33] Join: Fluffy joined #corewars [14:36] :) [15:31] Join: sf joined #corewars [15:31] Hello [15:31] No bvowk today? :( [15:31] no, bvowky seems to be away [15:31] Hi sf [15:32] Hi Fluff [15:32] What are you up to? [15:32] Started to write warriors again [15:33] Who is hwm! [15:33] Now in 3rd place on nano [15:33] Maybe you should ask him/her and not me ;-) [15:34] I would if SAL showed email addresses [15:34] Then ask He-Who-Must-No-Be-Named [15:35] Ah, I thought that yesterday. Is it true? [15:35] Why is he on the beginner hill too? :) [15:35] Maybe because he's a beginner? [15:35] He-Who-Must-Not-Be-Named isn't a beginner [15:36] huh? [15:37] hwm = he who must (not be named)? [15:37] Don't think no [15:37] *so [15:38] 3rd place with a oneshot! [15:39] What kind of warrior are you making? [15:40] One with many Redcode instructions [15:40] More than 5?! :( [15:40] Yes [15:40] less than 100 [15:41] For the nop hill or draft hill? :) [15:41] ... and man "dat.f $ 0, $0" [15:41] *many [15:41] 94nop [15:41] I haven't looked there for a while [15:42] Good, I'm still there. My oldest 94nop warrior ever [15:42] I'm always on verbose on 94nop. It is easier to keep an eye on the hill that way. [15:42] Nice :-) [15:43] hmm ... I've just realized, that Fizmo has already an 420 warrior on the hill. Seems to become as old as Arrow [15:45] Kline's tests help :) [15:54] Join: Quinn_Sto joined #corewars [15:54] Nick Change: Quinn_Sto changed nick to Quinn [15:54] Hi Quinn [15:54] hi Fluffy [15:54] It's cool to see that corewars isn't totally dead [15:55] Huh? When was it dead? [15:55] well, it seemed that interest had died out years ago [15:55] Hi Quinn_Sto [15:56] So have you played Corewar before? [15:56] Or just interested? [15:56] Nah, not yet, but a friend of mine is totally obsessed with VM-based games, so I had to look into it [15:56] :) [15:56] I'm looking for a good redcode simulator to run on linux [15:56] pmars? [15:56] doesn't seem to have a linux port [15:57] http://sf.net/projects/corewar [15:57] thanks [15:58] It is quite usable, if you want to debug your warriors, but you should use other MARS, if you want to optimize your warriors [15:58] wow, pmars's last release is pretty old...maybe I should update it [15:58] that'd be a fun project, especially finding cool ways to visualize it in openGL [15:59] :-) [15:59] Writing a decent and portable debugger would help more [15:59] redcode is such a strange language to debug...what's lacking in current systems? [16:00] As far as I can say the learning curve is the problem [16:00] that's true for any assembly language I would think [16:01] and an HLL would defeat the purpose especially since it'd never be as efficient as directly coded warriors [16:03] Does anybody know which issue of the CoreWarrior dealt with one-process-stones? [16:06] Check the bibliography ;) [16:07] * Fluffy sighs [16:07] If you mean like silver bullet, see cw65 :P [16:07] Yes, that's it [16:07] thanks [16:08] Found it in http://corewar.co.uk/biblio.htm [16:09] * sf makes a note of what Fluffy's new warrior might contain [16:10] * Quinn giggles [16:10] sf: You already know that [16:11] Those stones are really nice. They all score > 200 against fsh94nop-0.3/scn [16:12] Quinn: have you had a google look around at the corewar websites? [16:12] a bit, I've had trouble finding ones that didn't stop updating around 2000 [16:12] The first two in the topic :) [16:12] btw, pmars compiles & runs fine [16:12] http://corewar.info [16:12] cool [16:12] sf: As usual you forget to mention the most important homepage ;-) [16:13] Only because it isn't in the topic Fluff ;) [16:13] sf: Repair you 'y'-key! ;-) [16:13] well this channel is now in my autojoin on konversation [16:13] *your [16:14] Theres nothing wrong with it [16:14] Quinn: I hope, you won't be disappointed by it [16:14] I don't think I will be [16:14] what should a beginner do first? [16:14] Read the beginner's guide to redcode [16:14] sf: Do you read the logs? That stupid emiclock/brx-bot now has started to use your nicks when joining :-( [16:15] http://vyznev.net/corewar/guide.html [16:15] Fluff: best to ignore it and it might go awa [16:15] I know :-) [16:16] Then try sending some programs to the beginner hill [16:16] you should install NickServ [16:16] ok, cool [16:16] http://sal.math.ualberta.ca/hill.php?key=94b [16:16] hmm ... I've thought, that those one-process-stones would be what I need, but when I convert them to multiprocess ones they become quite usual :-( [16:17] Fluffy: try putting processes in so they run forward, instead of in reverse [16:17] Join: Core29 joined #corewars [16:18] Hi Core29 [16:18] hi [16:18] yeah I did read part of that [16:18] I guess I should try coding a simple warrior [16:18] and sending it off to the hill [16:19] Quinn: You will even enter the hill, because currently it has one free slot :-) [16:19] heh [16:20] Hey, what happend to the 94b-hill? [16:20] You're lucky there wasn't a beginner hill when I first started [16:20] Did He-Who-Must-Not-Be-Named finally implement the age limit? [16:20] Yes, 50 [16:20] why 50? [16:20] there doesn't appear to -be- a beginner hill on the site [16:20] Don't know. 100 would have been better [16:20] http://sal.math.ualberta.ca/hill.php?key=94b [16:21] http://sal.math.ualberta.ca/index.php are all hills at SAL [16:21] 50 is too low for a hill with 25 warriors on it. A warrior will retire every other challenge [16:21] And reaching age 100 seems more of an achievement [16:22] yes [16:23] When a warrior retires, another will get on in the free slot, then another is ready to retire. Almost everything will get on :) [16:23] Hopefully HWMNBN has only change one value to increase the age limit [16:24] *has to change [16:25] Someone probably has a good arguement why it should be 50 [16:27] I wonder, what VL-7-4 is. Even an imp would score better [16:28] instant suicide warrior? [16:30] lol [16:30] ok, I'm trying to figure out where to start with writing a warrior [16:31] How about writing a warrior which bombs each location in core one by one [16:31] sequentially [16:31] that sounds like an easy way to lose...IIRC from reading about it that's a core-clear strategy? [16:31] at http://www.ociw.edu/~birk/COREWAR/koenigstuhl.html you will find plenty of examples [16:31] Yes that's a core-clear [16:32] And they are included in many warriors [16:32] yeah [16:32] well maybe I should write one as an exercise [16:32] On some hills the score well on their own [16:32] I see [16:32] Then send it to the beginner hill, the see how you can improve it [16:32] yeah [16:33] Quinn: Don't forget to use a good name :-) [16:33] ... for the warrior [16:33] heh [16:34] Just think, what happens, if your warrior becomes the worldbest one and you've given it a lame name like "Warrior 1" ;-) [16:35] lol [16:35] Or like "Harmless Fluff Bunny" [16:36] that's a good name [16:36] sf: That name is already trademarked and patented by me! [16:36] :) [16:36] And already on Koenigstuhl [16:36] * Fluffy kicks sf for suggesting that name ;-) [16:38] In any case "Harmless Fluffy Bunny" is a far more better name than "Son of Vain" ;-) [16:38] well I mailed one out [16:38] lol [16:39] You' [16:39] Quinn: I hope, that you've registered at SAL first before sending your warrior [16:39] ll get a mail back saying if it compiled ok or not [16:40] Fluff: I never registered [16:40] Fluffy: yeah I registered [16:40] in fact I sent a @auth in my mail just to be sure [16:41] What did you send? [16:41] ugh, warnings and errors [16:41] apparently my @'s were in the wrong place somehow [16:41] :( [16:41] and it also is complaining that I put no ;assert in [16:41] use ";assert 1" [16:42] that always works [16:43] ok I edited and tried again... [16:43] Good louck [16:44] my warrior is "SupperBomberMan" and yes I misspelled that on purpose [16:44] Are you hungry? [16:44] lol yeah, I recently had dental surgery and eating is...not easy right now [16:45] hmm ... now there is a "Unknown" by "Anonymous" on 94b [16:46] Did you use: [16:46] ;name SupperBomberMan [16:46] ;author Quinn [16:46] ? [16:46] yeah [16:46] err [16:46] ;author QuinnStorm [16:47] strange [16:49] no it works now [16:49] the errors were because I put the @ in the wrong place [16:49] its in now [16:49] oh! [16:49] I see [16:49] "unknown" by "anonymous" [16:50] Try without the @ commands to see it that helps [16:50] Join: bvowk joined #corewars [16:50] ok [16:50] I don't use them in anything I send [16:50] Hi Bvowk [16:50] I have to kill the old one then [16:50] Hi bvowky! [16:50] greets [16:50] how's it going fluffster [16:50] sf? [16:50] * sf throws a wasps nest in bvowk's direction [16:51] * Fluffy joins sf and kicks kicks bvowk [16:52] * sf sprays a soft drink all over bvowk to attract the wasps [16:52] hmm ... my score against fsh94nop/scn never becomes > 168 with multiprocess stones :-( [16:53] ow [16:53] why the abuse? [16:53] ugh now I have TWO "unknown" by "anonymous" warriors! [16:53] just because I'm an arrogant asshole doesn't mean you need to abuse me :) [16:53] * sf mocks bvowk and points at the nano hill! :) [16:53] uh oh.. [16:53] that's a bad sign [16:53] bvowk: We all love you! [16:54] holy shit! thats awesome! :) [16:54] * sf hugs bvowk! [16:54] Quinn: If you post here, what you send, we might help [16:54] I'm down to 4 warriors on the hill.. [16:54] ok, no flood problems? [16:54] bvowk: Does you evolver still not work [16:54] All the old ones have gone bvowky [16:54] yeah.. [16:54] Quinn: Just join again then [16:54] nicely done :) [16:54] fluffy: it does work, but I'm not running it until I get the new code all happy skippy. [16:55] I should hustle on that huh? [16:55] ;redcode-94b [16:55] ;name SupperBomberMan [16:55] ;author QuinnStorm [16:55] ;kill Unknown [16:55] ;assert 1 [16:55] ;strategy Yeah I know, I misspelled that on purpose. [16:55] ;strategy It is a bomber. It bombs. [16:55] ADD #4, 3 ; execution begins here [16:55] MOV 2, @2 [16:55] JMP -2 [16:55] DAT #0, #0 [16:55] end [16:55] that add is actually tabbed in [16:55] hmm ... looks good [16:55] its on nano, of *COURSE* it bombs :) [16:56] spaces are better than tabs [16:56] I'll try removing all blank lines, and spacing instead [16:56] did you send as plain text? [16:56] I can [16:56] of course I did [16:56] 't see a problem [16:56] one more try with no blank lines & tabs->spaces [16:56] bvowk: Does SAL accept warriors from any mail address (registered or not)? [16:56] yeah [16:56] I'll have three identical supperbombermans on 94b [16:57] it'll accept anything thats formatted like a warrior [16:57] I hope the ;kill Unknown works [16:57] ok someone or something killed the Unknowns [16:57] hmm [16:58] bvowk: Can you help? [16:58] ah there it is [16:58] it worked [16:58] supperbomberman is on the hill [16:58] I can't help.. [16:58] That's better :) [16:58] all I can do it restart it sometimes.. [16:58] no blank lines is the problem [16:58] bvowk: ok, problem already solved itself ;-) [16:58] he who must not be named has all the magic, I keep myself out of the loop so there's no issues with me playing on sal [16:59] Quinn, now can you improve it? [16:59] bvowk: Good idea [16:59] well shouldn't I let it run a bit and see how it fights? [17:00] Quinn: you'd be waiting for other players to send new warriors [17:00] ahh [17:00] ok then its time I edit [17:02] sf: The age-limit-problem is about to start! [17:02] :) [17:03] Fluffy: I be we could take over the entire hill with dat warriors ;) [17:03] By the time we'd filled the empty spots, they'd be more empty spots. [17:03] sf: We could, but we'd probably beaten up by HWMNBN ;-) [17:03] Then we'd just be pushing off other dats. [17:03] Until everything else had retired ;) [17:04] sf: We should not push around dat's, but keep on sending on strong warrior [17:04] Someone one predicted they'd be a hill in the future containing only dats! [17:04] No beginner would be able to age the hill ;-) [17:05] now to see if my update works [17:05] Quinn: :-) [17:05] I may have b0rked the code, I dunno [17:05] * Fluffy is back in 1 min ... [17:05] MSG: Quit: Leaving. [17:06] hmm, that's what I was afraid of...I think it ;kill'd itself [17:06] Join: Fluffy joined #corewars [17:07] It looks like we could empty the hill too with test versions. [17:07] MSG: Read error: Connection reset by peer [17:07] We'd still get killed by HWMNBN [17:08] hmm ... running away wouln't solve that problem ;-) [17:08] hmm it seems to have gotten worse [17:09] Quinn: Maybe it kills itself [17:10] doesn't look like it in pmars [17:10] maybe my attempt to do better just did worse [17:10] What did you change? [17:10] well, it tries to both bomb every 5 spaces, and copy itself forward...though I realize now it never executes the copy [17:11] Sounds quite complicated [17:11] How does it do that? [17:11] will show in a moment [17:11] k [17:11] Join: sf joined #corewars [17:11] Hi sf! Already taken over the b-hill? [17:12] ADD #6, 3 ; execution begins here [17:12] MOV 2, @2 [17:12] SPL 3 [17:12] JMP -3 [17:12] MOV 0,1 ;never executed anyway [17:13] DAT #0, #0 [17:13] MOV -6,6 [17:13] MOV -6,6 [17:13] MOV -6,6 [17:13] MOV -6,6 [17:13] JMP 2 [17:13] hmm ... the first two instructions don't seem to be right [17:14] OH it should be add #6, 6 [17:14] then mov 5,@5 [17:14] d'oh [17:14] Quinn: because you are using a step of 6, when the bombing wraps around core it will bomb itself [17:14] it should have SPL'd ahead by then tho [17:15] but that might be a good thing to fix too [17:15] I should use a step of 8 [17:15] To make things easier, you should use labels in your code. [17:15] sth like: [17:15] labels? [17:15] Quinn: you can use labels [17:16] add.ab # 4, bomb [17:16] mov.i bomb, @ bomb [17:16] Join: Mizcu joined #corewars [17:16] jmp -2 [17:16] bomb: dat.f 0, 0 [17:16] That way you don't have to compute all values by yourself [17:16] Hi Mizcu [17:17] never fear, the unofficial newbie-mentor Mizcu is here.. [17:17] Mizcu: Then you should change your nick so that everybody knows ;-) [17:17] :) [17:18] Quinn: Do you know now how to use labels? [17:18] well ha-ha [17:19] Mizcu: No, "ha-ha" isn't a good nick to make you instantly recognizable as a cw-tutor ;-) [17:19] yeah I do now [17:19] :-) [17:19] Cant fit more into topic =/ [17:21] well I've sent another one [17:21] :-) [17:22] MSG: Ping timeout: 252 seconds [17:24] well now its doing -really- bad [17:24] the source is on the site, or I can paste it, its gotten longer [17:24] how long? [17:24] 16 lines or so [17:25] though part of that is meaningless filler [17:25] then leave out the filler :-) [17:25] start: ADD #step, bomb ; execution begins here [17:25] MOV bomb, @bomb [17:25] SPL split [17:25] JMP start [17:25] then three mov 0,1's [17:25] bomb: DAT #0, #0 [17:25] split: MOV -step,step [17:25] MOV -step,step [17:25] MOV -step,step [17:25] MOV -step,step [17:25] JMP newstart [17:25] then two mov 0,1's, a dat #0,#0,and label newstart: [17:26] hmm ... and what do you think, that it does? [17:27] Don't get me wrong. I usually write warriors, which do everything but not what I thought they would do ;-) [17:27] well the first half bombs every 8 spaces, the second half, jumped into by the spl split, copies the first half forward to the next open spot, which then executes...though it does spl after that into undefined space [17:28] so maybe I should mov something else in place of that split when I copy it forward [17:28] designing a warrior is harder than I thought [17:28] mov start, newstart | mov start+1, newstart+1 .. [17:28] that would be better [17:31] the real trouble I am having is...essentially the problem that plagues the author of a quine...how do I copy the program in fewer/the same # of steps the program takes [17:32] Quinn: The solution is called "paper" :-) [17:32] * Quinn reads up on paper [17:32] http://www.ociw.edu/~birk/COREWAR/94/HILL/timescape10.red [17:32] You cannot go belowe the number for steps, and only imps can make it in same number [17:32] that's an "easy" example of a paper [17:33] not easy in the meaning of understanding, but in ease of writing [17:33] ouch, paper is complicated [17:34] I think I need another tactic, heh [17:34] Quinn: Yes, but look at pmars as see, what it does [17:34] Making a self-copying program is a stepping-stone everyone will have problem going over [17:34] with "pmars -e warrior.red" you can step through the execution [17:35] hit "s + ENTER" to execute the current instruction [17:35] yeah [17:36] Unless one just takes timescape -style paper and play around with it, that is the easiest way of making a strong warrior. But if you dont know how it works, you probably cannot make it better than just plain good [17:36] I'll have to learn all the extra modes, etc. to do that [17:36] Quinn: that might help ;-) [17:36] can I do #(newstart-start) btw? [17:37] yes [17:37] ok [17:42] Mizcu: Oh great tutor, do you know a good warrior against pwi? [17:42] (no scanners, please ;-) [17:42] Recon2 [17:42] working on my warrior still [17:42] ... [17:43] Quinn: We will wait :-) [17:43] Fluff: There are non [17:43] Mizcu: Are you sure? [17:44] You need an aggressive scanner, Blur-clone with short scan-run or anti-imp oneshot [17:44] Join: NanoKing joined #corewars [17:44] hmm ... then I have to develop a new alternative ;-) [17:44] hi Roy [17:44] Hi Roy! [17:44] :) hiya [17:44] maybe, just maybe a d-clear/imp [17:45] Mizcu: At the moment I have some problems with clears, because my papers usually don't like to be paired with clears :-( [17:46] they never like to be paired with clears (exception, the one Inversed did) [17:46] ok wtf am I doing wrong now? [17:46] Hmm, that one was cool.. was that uhm Blotter..? [17:46] Nick Change: NanoKing changed nick to Roy [17:46] let me /join a non-channel for those interested in my warrior...it'll be #sbm [17:47] I'll paste it there so as not to flood here [17:47] hehe ok, it doesn't really matter... [17:48] well, I just prefer good netiquette [17:48] Join: sf joined #corewars [17:49] Hi sf.. :-P [17:50] * Roy prods sf about nano [17:52] Hi Roy [17:52] IRC isn't working very well for me today [17:53] Who is that new guy..? (third place) [17:53] Nice entries on nano [17:56] Join: sf[2] joined #corewars [17:56]