[02:50] Join: Shilock joined #corewars [02:50] Heyya [06:06] Join: Mizcu joined #corewars [06:27] afk [07:41] Join: fiveop joined #corewars [07:46] MSG: Quit: Lost terminal [10:45] MSG: Read error: Connection reset by peer [10:49] Join: Mizcu joined #corewars [11:01] MSG: [11:10] Join: Mizcu joined #corewars [11:38] MSG: Ping timeout: 255 seconds [14:48] Join: Shilock joined #corewars [15:50] MSG: Read error: Connection reset by peer [16:02] Yo, anyone here-- newb question. [16:03] When you split another process, where does it get added to the proccess turn order? [16:03] Like if you have proccess A B C and D, and A spawns a new one, when will it run? [16:07] sorry.. I'm not much help with redcode.. [16:07] I evolve.. I don't actually get my hands dirty with actual redcode [16:09] evolve? [16:10] yeah, I write programs to write my corewar programs [16:11] http://www.corewar.info/evolve.htm [16:11] Well, as far as the docs I'm looking at sound, the new proccess is after the one that spawns it... making it A E B C D, but the E doesn't execute until after A's next command. [16:11] Anyways I'm trying to write a Nano-core and I need to make sure my timings right so that I dont die on cycle 7 or so. [16:11] iirc, its added onto the end of the process queue, so it should run after all the rest have completed [16:12] but don't hold me to that.. [16:12] "The process at the next instruction executes before the one which jumped to a new address, which is a small but very important detail. (Many, if not most, modern warriors wouldn't work without it!) If the max. number of processes has been reached, SPL works like NOP. Like JMP, SPL ignores its B-field and its modifier." [16:13] Dang it, I want to know if my 5 lines of code will spiral right. [16:15] Well if I'm reading that, then my code should work. [16:15] Not sure its the best imp spiral, but it doesnt self kill and it should kill anything with a smaller process count then 3. [16:16] and what nano-core is going to have an impgate? [16:23] Ok, someone give me a number between 1 and 800 [16:27] erm, make that 1 and 80 [17:16] Dang, I need to find someone good with the code running... I have a unique idea, but its... an obscure one. [17:42] Join: John joined #corewars [17:42] Hi [17:44] Hey John, I haev a question for you... [17:45] :-) [17:45] Hi Shilock, fire away [17:45] JMP {0, }0 would Basically shift you back 1 line while keeping its fields at 0, right? [17:46] Yes [17:46] It's contained in a few warriors [17:47] I'm trying to write a decent nano [17:47] And what I currently have, albeit slow, only has 1 line vulnerable to .f scanners that just bombard the first non-0 it sees and then incriments from that. [17:48] and if it decrementsfrom it's first contact, then it still only has 2 lines it could see. [17:49] The problem is that its very slow progress. It bombs 1 line and incriments it's target, then it bombs again. [17:54] Can I see the code please? :-) [17:55] Erm, its a nano... there is no 'some code' its all or nothing. [17:57] Well I can trust you-- I'll private it over. [18:14] Join: Core_old joined #corewars [18:15] Hi Core_old [18:39] btw, isn't SAL that blue female chick from 2010? [18:40] Probably, bvowk names all his servers after AIs :-) [19:02] Have to go. Need to sort through my takeaway menu collection. [19:02] * John waves [19:02] MSG: Quit: mov.i #1,1 [19:05] Join: Mizcu joined #corewars [19:09] my keyboard is like Windows - need to be rebuilt every once in a while [19:31] sal is the name of the AI thats the mirror of hal in 2010 [19:32] Ha! knew it. [19:32] like john said, my machines are named after fictional AI's [19:33] hal, sal, dix, mcp, lal, roy, data, wopr, continuity, 3jane... [19:33] where's Gary? [19:34] (the AI from Red vs Blue) [19:34] never seen it.. [19:34] Oh dude, Gary is a halarious story. [19:34] Gary was the name of one of their private servers. [19:34] ...or something like that. [19:34] ...and I think that was named after a dead cat. [19:34] I had detroit and london after syndicate wars for a while too [19:35] heh [19:35] You need Rommie though. [19:35] ...we all love Rommie. [19:35] bah. [19:35] can't stand that show [19:35] actuall.. [19:36] can't stand TV for the most part these days ;0 [19:36] The early seasons or the last 2? [19:36] Cause Andromeda was good until season 4 and 5. [20:12] *yawn* [20:13] and tomorrow to wonder wtf is wrong with boss's laptop [20:16] With the GPRS-card VPN and remote desktop work fine to the server, but with cable they dont work, even though in both cases internet works fine [20:39] *groan*, this game is going to drive me insane. [20:39] ...must... get... on... hill... [20:40] insane is wrong wording [20:40] well, maybe considering you are being overflowed by information, you might be correct [20:41] Heh, I'm trying to avoid being overburdoned... thats why I'm working on making a good nano-core [20:42] 5 lines I can do... give me caffeine and 48+ hours of insomnia and i'll be able to trace 20 lines of complex back-forth with no problems. [20:43] heh, youll enjoy getting to IRC-minitournament if we are going to have anotherone somewhere in the future [20:44] heh. whats that? [20:44] 30 minutes of time to make best warrior for an abnormal ruleset [20:44] Sounds cool. [20:44] so amount of cigs or caffeine is going to get you through without sweat [20:44] so -> no [20:44] heh, in 30 minutes I can make some messed up shite [20:45] but would it work well [20:45] Most likely not incredibly well... but it might do some really odd things... [20:46] ...like have a bunch of SPL #0, 0 s followed by an incrimenting jump [20:46] as simple things can be the best in those situations [20:47] Whats wrong with creating a lot of processes and then jumping them foreward trying to land on better code? [20:47] you arent killing your opponents then [20:47] Thats part of the point. [20:48] ...lets say the ruleset is multi-warrior... [20:48] ...then jumping my processes onto other warriors might be quite a way to tie myself to the top. [20:48] but you still wont win, only tie at best [20:49] and winning is better that tieing [20:50] how exactly do multi-warriors score? when you have more then two on a hill... [20:50] winner, 2-tie, 3-tie... x-tie [20:50] winner gets 3, all tie'ers get 1 [20:51] dang. [20:51] check the scores in koth nop94-hill [20:51] lowest score is 122 - having 100% ties would be 100 points [20:53] so all warrior have more aggression than defence on average [20:53] hmm... I'll remember that. [20:54] though some warriors are more heavy on defence - if your opponent cannot defeat you, he cannot get scores from you [20:55] lol... pimp on the beginer hill has like >90% ties some times... [20:56] yes, the warriors there are are either not aggressive enough or built to resist such "trolling" [20:56] with only 27% wins, i am surprised about stride [20:57] Heh, I'm just shocked how good my first warrior did on that. [20:58] stride is built on an ultra-tough warrior, so dont expect to beat its scores any time soon [20:59] i mentioned to HWM that it might be overpowered on the beginner hill, but i guess neither did bother to do anything to it [20:59] Yeah, stride... I got 4.1 points from him with my double axe [21:00] ...which is little more then a dwarf that shoots both ways offset so that they'll eventually hit every other one. [21:00] My nano guy could probably beat it. [21:01] most probably because it uses quickscan and isnt the fastest starter [21:01] so you got a quick hit a couple times [21:03] yeah... arguably most bots would win one every now and then off of luck [21:04] which is a real tick-off when you use strategy-switching warriors [21:05] Join: pkhuong joined #corewars [21:05] hi pvk [21:06] hi [21:08] SO tell me... I can put certain contant mathematical equations in my fields and they'll be evaluated when the warrior compies, right? [21:08] like MOV <-(2*step-2), <-(3*step-2) [21:08] if you dont do anything really weird with them, yes [21:08] that will work just fine [21:11] Well, I'm going to take one last shot in hope that my improvements can earn their way onto the nano-field, and then I'm going to begin work on a larger, more complicated, and hopefully well-working thing to storm the begginers hill [21:16] Oh man... thats depressing... [21:17] Nano has been evolved very high, so dont get depressed [21:17] ...it did better then my last guy, but looking at the copy of the compiled code i got back, something in the order of operations messed up! The code was buggy and it still did better then my last effort. [21:18] hmm [21:18] * Shilock has already decided himself too much of a pureist to respect evolvers [21:18] hmm [21:19] i can debug the code if you want [21:19] and if you dont trust me, pkhuong is most likely an "volunteer" too [21:22] Nah, it was my bad. [21:22] ...What can I say, I get a 5 on my AP Calc exam, but I can't remember the distributive property when it counts. [21:27] so, no new scores yet? [21:27] hmm? [21:27] havent sent the fixed versio to nano yet [21:27] I'd find it tacky to resubmit so quickly. [21:28] if you find a bug and fix it, i see no reason not to resubmit immediately [21:28] Erm, all the people who have it set on verbose? [21:29] its their fault :) [21:29] and getting craploads of reports over the past few days of this newb trying things. [21:29] i know many have it on silent [21:31] 125... I'm getting close [21:31] But thats all for now, I have to get going. [21:31] ttyl [21:31] bb [21:40] MSG: Ping timeout: 255 seconds [21:52] :) [21:52] MSG: Quit: Trillian (http://www.ceruleanstudios.com [21:54] little experimentation and +3 points, whee [22:04] but i am unsure about its capability to enter the hill.. [22:08] o_O O_o [22:09] really weird results against old warriors [22:09] yeah, I'm from the old guard [22:09] doesn't evolve, doesn't use optimax, etc. [22:10] I likely will though, if I play again [22:10] lots of neato ideas are basically hampered by the huge search space for constants. [22:11] rare to see 51 6 43 against Rave [22:11] heh [22:11] what is it? [22:11] stone + huge spiral? [22:12] qscan into paper with different approach [22:12] k. [22:12] good QS, then :) [22:14] Not even optimized, just stolen from elsewhere [22:14] it will be interesting to see what could hours of optimization does to the scores [22:15] could -> couple [22:50] Join: pkhuong- joined #corewars [22:53] MSG: Ping timeout: 255 seconds