[00:05] MSG: Quit: humhum [00:29] that looks like one of the things i was playing with (re: elkauka's code), but it's still missing the bomb source [01:26] reading core warriors again [01:26] i keep finding ideas i had popping up in interesting places hehe [01:26] like in 90, it talks about the imp/imp gate thing [01:26] that was on my list of warriors to play with ;) [01:35] oh and there's the spl 0, }0 reference you mentioned, miz (?) [03:20] Join: elkauka joined #corewars [03:26] wtf? blastwavecomic is gone?!? [03:44] MSG: [04:06] Join: Metcalf joined #corewars [04:06] Hi :-) [04:30] hello [04:30] reading lots of Core Warriors [04:30] :P [04:55] :-) [04:55] i kinda want to write a modern qscan [04:55] i understand them in theory, but i don't really feel like the math [04:56] Join: elkauka joined #corewars [04:56] I usually write a program to do the maths [04:56] Hi Elkauka [04:56] not surprised [04:56] hiho [04:56] i'm the type that doesn't like borrowing code, though [04:56] i want to learn it by doing [04:56] :) [04:57] that's the best way [04:57] i'm kinda disappointed that the game in general is in such a late stage in its history though [04:57] i seem to be hitting on a number of interesting ideas right away, but they aren't new now :P [04:57] the first time I used a qscan, it was one I wrote myself :-) [04:57] i know .. i read that one >:) [04:59] metcalf: do you got the missing links here: http://corewars.jgutzeit.de/quickscanners/index.en.html ??? [05:01] nevermind, there are links to r.g.c in the german version [05:02] myndzi: http://corewars.jgutzeit.de/quickscanners/index.html <- probably a good point to start with quickscanners [05:31] If you'd like to have a go at writing a qscan, maybe you could try writing a qscan for the tiny hill first? [05:33] or nano ;) [05:34] The maximum length is 20 instructions. The maximum you should use for the qscan is about 15 so only 2 or 3 sne/seq pairs [05:47] lol nano [05:47] i could totally write a qscan for nano [05:47] :P [05:51] is it rude of me to keep sending so many warriors to the beginner hill? [05:52] depends: do you consider yourself a beginner? [05:52] er [05:52] i'm certainly no pro [05:52] i haven't ever submitted a warrior before like a week ago [05:52] so i guess that counts [05:53] but i consider myself to have more than a beginner's understanding of the game.... if only slightly. but i guess that depends on what counts as a beginner [05:53] "how 2 write pogram" [05:53] or passing familiar [05:53] :P [05:53] whats the points your warriors make against fsh94nop0.3 ? [05:53] what's that? :) [05:55] a set of warriors used as benchmark [05:55] http://www.corewar.info/optimax/dload/ [05:55] i'd certainly like summa that [05:55] download and use optimax while you're at it [05:56] and nano q scan can be found in nano warrior 1: http://corewar.co.uk/nw01.txt [05:56] i think i don't need to look, but it'll probably be informative anyway :) [05:57] how do i run the benchmark? [05:57] what MARS are you using? [05:57] er, i have corewin [05:58] i thought it was an optimax function, guess i'll explore a bit [05:58] add all warriors to it and choose "tournament options -> challenge by" [05:59] all in each folder? [06:01] yep [06:01] ok [06:01] it has a benchmark option [06:02] anyway [06:02] how many rounds? [06:03] optimax or corewin? [06:03] corewin [06:03] 200 [06:03] you asked the points my warriors make against it [06:03] i'm just checking under what settings that question is ;) [06:05] i guess i'm just playing around with components anyway... none of this fancy multi component warrior crap.. i bet anything gets owned [06:05] i'm curious to see what does the best though [06:05] :) [06:08] lol, 42 for this one i was just goofing with xD [06:12] i submitted warriors who made 90+ points to the standart hill. still depends a bit on whats on the hill though [06:13] 80 for bolo [06:13] i just want to see how they fare against similar warriors [06:14] i don't want to get into all that extra optimizing crap if i don't have to [06:14] but i also don't want to be discourteous .. i don't know the etiquette ;) [06:14] by similar i mean, uh, simple [06:14] ^^ [06:14] so i can play with ideas [06:15] instead of spending all my time taking ideas that already exist and trying to tweak them better than the next guy :) [06:15] then again [06:15] it doesn't seem like it's very active anyway [06:16] ho, i got an 85 :P [06:16] then give corestep a try (for bomb and scan patterns) if you dont want to optimize : http://corewar.co.uk/corestep.zip [06:17] i've got that one [06:17] that reminds me [06:17] corewars inspired me (before i actually wrote some redcode, i've known about this thing for years) to make a silly mirc script that attempts to find optimal step sizes [06:18] i mean, that's not what i wrote it for or the words i used, but it draws it all in a circle so you can watch :) [06:18] also i confess to using the idea in games of battleship [06:18] :| [06:19] i wonder how corestep calculates numbers [06:19] of it is just brute forces [06:20] but optimization is pretty much everthing in the end. vamps are back made around 40points the basic idea with handpicked constants, 110+ with optimized constants and addtional 10+ point with added qscan and boot [06:20] yeah, i'm not interested in the end at the moment is all :) [06:20] ok [06:20] i mean, look at the code i pasted above for bolo (at least, the initial idea) [06:21] it's probably highly impractical, but it was fun and i like the synergy between the two parts [06:21] but it's at 9th place lol [06:23] the last place on the beginner hill "A. K. Dewdney" <- _the_ a.k.dewdney ??? [06:23] metcalf submitted that for comparison [06:23] with his rewrite(s) [06:23] so, yes, but not submitted by him [06:24] ok [06:35] the fun thing with optimax is you end up writing some form of template warriors and let optimax figure out to build your warriors [06:35] http://www.nomorepasting.com/getpaste.php?pasteid=17604 [06:56] dont tell me you optimize the decoy's size too? [06:58] well, if the size is 0 there is no decoy maker code at all [06:59] i had some code-templates which depending on the constant would render completly diffrent strategies ;) [07:01] but most the time a set of stones and a set of end games [07:02] you might as well download the bvowk's evolver for all good that'll make [07:05] :P [07:16] whee [07:17] i made a mirc script to graph gaps in core for a given step [07:18] dunno why, but it looks kinda neat anyway [07:21] tcl/tk? [07:23] mirc script [07:23] mIRC [07:23] you run it [07:23] :P [07:24] i mean what graphic toolkit does the visualisation? [07:24] and about 90% of every other irc'ers, of which maybe 2% know about the mirc-scripts. [07:37] hrm, my file server script is gone :( [07:39] Nick Change: elkauka changed nick to elk_afk [07:40] graphic..... toolkit? [07:40] have you ever seen mirc's scripting language? [07:40] ;p [07:40] and srsly, how does anyone NOT know about mirc scripts that uses mirc? :| [07:43] i thought i know mirc scripts, but seems i only played around with sockets... [07:44] myndzi: because mirc is the choice of irc-client for people who dont understand computers. Heck, ive met a lot of people who've said "lets meet on mirc". [07:45] (also, i consider scripting on mirc to be like programming on basic) [07:45] i'm using it now because the irc gateway on my xmpp-server is down :( [07:53] MSG: Ping timeout: 245 seconds [07:55] BASIC > mirc [07:55] also mirc is also the irc-client for people who DO understand computers :) [07:55] but anyway, heh... its scripting language sucks [07:56] and is amusing at the same time [07:56] it's just really slow and silly [08:14] MSG: [09:31] Join: Metcalf joined #corewars [09:37] * Metcalf downloaded the PC port of Fuse :-) [09:47] Cool :-) [11:52] Join: elk_afk joined #corewars [11:53] Nick Change: elk_afk changed nick to elkauka [14:11] MSG: [14:54] MSG: Quit: mov.i #1,1 [20:02] Join: sh0ne joined #corewars [20:02] Hi folks [20:22] h [21:34] Join: elkauka joined #corewars [22:07] MSG: Quit: Leaving [22:13] MSG: [23:36] Join: bvowk joined #corewars [23:36] bleh.