[00:22] what's the quickest/smallest launch for a 3 point imp ring? [00:22] numbers i mean [00:24] i didn't want to copy code so i was playing around, i don't think i've seen this used [00:24] but it's 4 lines and is set to go in 8 cycles [00:25] just 3 processes though, not one of those continuously launching things or whatever [00:34] quickest would be 3+3 cycles, smallest 3 lines (but 1 cycle slower) [00:35] this when not counting 2-point spiral (which is actually mirrored spiral, but not diving deeper into that) [00:37] actually, 3+3 is not the fastest, gimme a sec [00:38] i want to see :> [00:38] fastest 3 cycles, silly me [00:39] how? [00:39] assuming you already had three processes, you could jump two into place and then run the mov with the third [00:39] but you have to get three processes first? [00:39] both are 3 cycles, i blame 3am [00:40] jmp imp-2666 [00:40] start spl -1 spl imp+2667 [00:40] imp: mov.i #0, 2667 [00:40] end start [00:40] oh you mean 3 to make the procs and 3 to get set up [00:40] second version: [00:40] spl 2668, 0 [00:40] mov -1, *-1 [00:40] mov.i 0, 2667 [00:40] ah, that's neat [00:40] i guess i didn't do too bad though [00:41] 1 line and 2 cycles difference [00:41] not nearly as elegant though :| [00:42] 8 cycles is actually kinda slow, since in my mind it take 2 cycles to make 3 processes, and 3 to jump them into position (or 2 if the ring starts right after the launcher) [00:42] : [00:42] :| [00:42] but if you got the ring to work in the first place, then you are worth a congratulation [00:42] it was interesting [00:43] 80,69932168 spl 0, }0 [00:43] 80,69932168 jmz 2668, >2 [00:43] 80,69932168imp mov.i #10, 2667 [00:43] 80,69932168 jmp -2667, -2 [00:43] hehe, oh damn tabs [00:43] kinda funky ;) [00:44] thats kinda cool really [00:45] i'm just doing random stuff, i made a bomber that bombs the entire core, splitting it up in binary values [00:45] every 128 cells [00:45] then 64 [00:45] 32 [00:45] etc [00:45] it looks like rain [00:45] i don't have any way to use optimized steps though, i'm thinking about that right now [00:46] i think i'll try it anyway just to see what happens [00:46] spl 0, }0 is relatively new twist to process-creation, i think Fizmo experimented with it in.. er.. [00:47] i wouldn't know [00:47] it just made sense [00:47] the ony problem is how the second two processes doubled up [00:47] thus the jmz [00:47] corewarrior #90, 2004 [00:48] which originally referred to itself but that didn't work out well ;) [00:48] when "just" making an imp-spiral, i just use a vector-launch [00:49] i didn't read about that yet [00:49] i mean, i saw it but it didn't make immediate sense from just glancing at the code [00:49] idiotproof and fast in all but special-cases like the above [00:49] my imps trample the bomber D: [00:49] hehe [00:49] and the bomber can't take multiple processes [00:49] in short, vector launch places all the location where the processes go in a table, and then uses the table to put processes in position [00:50] http://www.nomorepasting.com/getpaste.php?pasteid=17022 [00:50] ^this is the silly bomber, it looks interesting though [00:50] and the bombing loop was amusing to write [00:50] i had one too many references that i needed to make it work [00:50] finally squished em all in [00:52] i could maybe stick the step sizes in the bombs instead, i did that with the other one, but then i have to work out how to detect when it's done with a run, and they all end in different places [00:52] in multiples of two, at that [00:56] lol it failed harder than the very first one i submitted [00:57] your biggest problem is the coreclear [00:57] nah [00:57] that was sorta halfassed anyway [00:57] but the big problem is it scans top to bottom i think [00:58] it got owned by everything too soon when i played with it here [00:58] even the ones i already wrote ;) [00:58] though i dunno, maybe i should submit it and see what happens without the clear lol [00:59] i just wanted to fill in the holes! :) [01:03] hm, if the step was in the bomb, then i should be able to stop when something hits at x+step away [01:04] now my head is hurting [01:04] oh yeah! [01:05] the b param is the offset, duh [01:06] MSG: Quit: humhum [01:14] the biggest problem with "halving" mod is the fact that at mod 4 or 2 one will bomb oneself [01:14] i avoided that in the paste by counting bombs, but it's pretty useless to bomb in a straight line anyway [01:15] so now i'm playing around to try and put the step in the bomb and still avoid dying [01:15] i need a separate counter it seems [01:18] well, if you throw X bombs (8000 mod X)-1 times, and bomb in a straight line, youd still need to change the place where bombing starts from [01:18] too bad i did [01:18] minus 'too bad' .. wtf [01:19] im having an idea but i kinda doubt it working [01:19] hehe [01:20] i don't want two data tables :| [01:21] oh yeah, now i remember what i was about to do [01:21] and one doesnt really want something like mov add slt jmp [01:21] right [01:21] i was thinking how useful slt would be if it didn't cost an instruction [01:21] :P [01:29] ooh, i may have it [01:29] http://www.nomorepasting.com/getpaste.php?pasteid=17024 [01:32] i'll check it in a few [01:32] first pastebin ive seen so far that requires a captcha [01:34] hehe my script automatically pastes [01:34] i never bothered to change the pastebin [01:35] the captcha is also ownable easily -- you submit the answer in a form [03:46] i succeeded but it cost me an extra instruction and i couldn't really mix up the steps all that well without interfering [03:47] oh well, it was fun messing with [04:01] what about my solution? too boring? [04:12] (its still doesnt allow wide stepsized) [04:35] it didn't split evenly, but i didn't even try to mess with it [04:35] i can't even edit my own code properly yet ;) [04:35] it looks more interesting though [04:36] i just haven't tried to figure out how it works [04:37] having them split like german marchtruppen would require spacers in the table, which is kinda useless since the result is about the same [05:00] yeah, but i already decided the idea was useless, i just wanted to do it :) [05:00] see the part where it got owned on the hill by my first warrior evar [05:00] hehe [05:00] which, ironically, is a somewhat similar construction [09:01] Join: fiveop joined #corewars [15:45] hrm. [15:45] greetings humans. [15:46] hi [15:46] how goes fiver? [15:48] I'm fine thanks [16:00] so your term must be pretty much done.. [16:00] you headed back home soon or another term starts? [17:41] Join: Metcalf joined #corewars [17:41] Hi :-) [17:41] heya metster [17:42] evitable: did you see the article in nanoWarrior #02 about imp launchers? [17:42] Hi Bvowk [17:46] Join: yoR joined #corewars [17:46] Hi all [17:47] Join: Fizmo joined #corewars [17:47] Hi yoR [17:47] Hi Fizmo [17:47] hi hi [17:47] So, how was the move? [17:47] yeah, I have finally internet again [17:47] Anything new? [17:47] Are you united now? Haven't updated the blog ;-) [17:47] move was very tired [17:48] and it took a while until I got internet again [17:48] my gf finally moved to me [17:48] so everything is very fine [17:49] That's great news [17:49] Now you just need to teach her how to play Corewar ;-) [17:54] Join: Fizmo_ joined #corewars [17:54] wait a moment, my gf needs my help [17:56] MSG: Ping timeout: 245 seconds [17:56] Nick Change: Fizmo_ changed nick to Fizmo [17:56] Heh, needs help with the imps/papers/stones? [17:56] MSG: Ping timeout: 245 seconds [17:57] first I need teach her German ;-) [17:58] And thats hard enough, I hated learning that language [17:58] yes [17:58] And I forgot everything hehe [17:59] hehehe [17:59] But if you live in Germany and hear it all day long it'll stick better en longer.. [17:59] yes [18:00] Mein Deutch ist nicht güt [18:00] she already at language school and learning very hard [18:00] :-P [18:00] Wieso nicht? Because I suck at it! [18:00] Well, I hope that I can start to do update the corewar site more regular in the future [18:01] I already prepared the rss [18:01] Join: Metcalf joined #corewars [18:01] wb John [18:01] But does she have a job now, or is it full-time studying and adjusting? [18:02] is full-time studying [18:02] that's the purpose of the visa [18:02] I always have a bad connection in the evening. [18:04] Ah ok, must be tiring [18:07] MSG: Remote host closed the connection [18:08] wait a moment. Am soon online again [18:08] MSG: Quit: ChatZilla 0.9.82.1 [Firefox 2.0.0.14/2008040413] [18:10] MSG: Ping timeout: 245 seconds [18:37] Join: Metcalf joined #corewars [18:40] Join: OoS joined #corewars [18:45] MSG: Ping timeout: 245 seconds [18:47] MSG: Ping timeout: 245 seconds [18:58] Join: OoS joined #corewars [19:02] Join: impomatic joined #corewars [19:06] MSG: Ping timeout: 245 seconds [19:47] MSG: Ping timeout: 245 seconds [19:49] Join: impomatic joined #corewars [19:56] MSG: Ping timeout: 245 seconds [19:59] Join: impomatic joined #corewars [20:39] Join: Neogryzor joined #corewars [20:40] cheers [20:42] Nick Change: Neogryzor changed nick to Neo_away [20:42] bbl [20:44] Hi Neo [20:58] dammit. [20:58] I missed fizzer. [21:06] MSG: Ping timeout: 245 seconds [21:24] Join: impomatic joined #corewars [21:31] MSG: Ping timeout: 245 seconds [21:34] Join: impomatic joined #corewars [22:13] MSG: Ping timeout: 245 seconds [22:25] Join: Neo joined #corewars [22:27] Join: Neo_away joined #corewars [22:29] that ping timeout is contagious... [22:29] MSG: Ping timeout: 245 seconds [22:30] MSG: [22:33] MSG: Ping timeout: 245 seconds [23:15] MSG: Quit: humhum