[01:03] I run redcoder, corewin on Vista Business x64. haven't tried pmars, but I expect it to work [03:47] my poor head [03:47] combinations, permutations, yuck [03:52] I have three strings [03:52] for simplicity let's say, "a", "bb", "ccc" [03:52] i am drawing a blank on how to generate all strings of length N containing combinations of these [03:52] (efficently) [03:53] i don't know what words i want to use (combination, permutation, something else?) [03:53] but i just want to know something like [03:53] for N=6 [03:54] i can have 6 a's [03:54] or 4 a's and 1 bb [03:54] or 3 a's and 1 ccc, [03:54] etc [05:27] got that one, now have a different problem lol [05:27] ;) [09:31] Join: fiveop joined #corewars [10:28] Join: M0wLaue joined #corewars [18:08] Join: Altair joined #corewars [18:09] hi [18:12] Could anyone help me to understand optimax^^? sadly optimax.exe itself wont run on this PC but I have tried on another. I wanted to test out the stepsize of my bomber so : size equ !(10-7900) and giving severeal options (I just used the ones of the example) but .....well it didnt work. Sry to ask, it is maybe just of stupid mistake I made [18:14] source of the warrior ready to be optimaxed, please, so i can see the settings [18:18] http://pastebin.com/m1b96d553 [18:19] just a stupid bomber, I wanted to try out optimax [18:19] you need to add a lot more than just step step-line [18:19] see the bottom of http://www.corewar.info/optimax/faq/ [18:19] as I said I used the optimax instructions of the example [18:21] there's an example.red in the directory [18:21] so I just copied these to see whether anything happens [18:21] optimax isnt exactly the most userfriendly program [18:23] this isnt an excuse for my stupidness [18:23] so, any idea what got wrong? [18:23] you got the benchmark in its directory, perl installed, every ;phase has been defined? [18:27] installed active perl. and executed it in the optimax dir; the phases were allready defined for the example in phase3.lst? [18:28] look at the link i gave, look at the bottom example. I see no phase3.lst [18:29] and I also tried the example.red and it started beautifully :D [18:29] ;optimax phase3 20;optimax phase3 0%;optimax phase3 top15;optimax phase3 phase3.lst [18:29] that's not the example I used [18:30] http://pastebin.com/m5be95db8 [18:30] this one [18:31] i just copied the entire header from the example and modified it [18:31] :P [18:32] bah, i dislike optimax anyway [18:32] :D [18:33] yeah, it never really did much good for me [18:33] it's fun though, i like to see the patterns and interactions from different step sizes and stuff ;) [18:35] mm........so at least it seems I've done everything right? [18:35] i don't know? [18:35] if it's not working, obviously not [18:35] ^^ [18:35] i didn't mean i modified the headers [18:35] i kept all those i think [18:35] i just modified the code [18:36] it worked for the example [18:36] the headers are touchy [18:36] yeah, I just copied the headed for optimax [18:36] yeah, well, what did you change from the example to your code? [18:36] file in the same directory and everything? [18:36] open example, replace code, save with new name, run [18:36] :P [18:36] just instead of the warrior of Zapf one of mine, the bomber I send you. yeah, all in the right dir [18:37] guess it just hates you [18:37] solution: rewrite the perl code [18:43] mm.....well, I test it again another time [18:49] Join: Fizmo joined #corewars [18:49] hi hi [18:52] Altair: regarding optimax, to use a paper in phase 2 against a stone is a bit unluky. I'm sure that nothing got past to phase 3 so that you could see results [18:53] maybe modify like: [18:53] ;optimax phase2 fsh94nop0.2/scn/hsa.red [18:55] also use instead of phase3.lst the following line [18:55] ;optimax phase3 cds:clr:scn [18:56] if perl works fine you should get things to phase3 and phase4 [18:57] I think you could also remove the: [18:57] ;optimax phase1 cyclemin 80% [18:58] the stone shouldn't make suicide an any time during the game [18:58] * Fizmo hopes that this helps to solve the issue with optimax [18:59] okay thanks, I try it out. Unfortunatly this PC has some problem with the optimax.exe so I have to go to another pc [19:00] I exprerimented with windows und somehow its crippled now :D [19:00] did you get some errors? [19:01] did you start it from the cmd dos windows? [19:01] yes...DOS does not seem to work fine. optimars.exe -> file not found. optimars.exe cannot be executed [19:02] you should go into the optimax directory and use the following command: [19:02] oh, I remember. I deactivated the DOS name support [19:02] perl optimax.pl -o example.red [19:03] can't remember the original .pl name as I renamed my one :-P [19:03] you need to look in the optimax directory for the right pl name [19:03] I tried this before:http://pastebin.com/m1d84698b [19:04] but like I said, this is only a local problem [19:04] with "wperl. remote.pl" you can start a separate window with in-process changes of the phase parameter [19:05] hmm, strange [19:06] didn't saw this error message before [19:06] yeah, my windows is simply crippled [19:07] maybe try the changes I suggested [19:07] especially to remove the phase3.lst [19:07] that can make problems [19:11] k, on my way^^ [19:19] lol: http://pastebin.com/m33556cc0 [19:20] this appears after phase 2 [19:23] found the error [19:24] optimax does not like" ;optimax blabla" there was a space before ; [19:26] $rounds [19:26] seems something with rounds [19:28] keep something like [19:28] ;optimax rounds 1 100 200 200 [19:34] maybe try out something like [19:34] ;optimax 1234 [19:34] ;optimax work Stoner [19:34] ;optimax rounds 1 250 250 250 [19:34] ;optimax suite fsh94nop0.2 [19:34] ;optimax phase2 fsh94nop0.2/scn/hsa.red [19:34] ;optimax phase2 20 [19:34] ;optimax phase3 top25 [19:34] ;optimax phase3 cds:clr:scn [19:34] ;optimax phase3 20 [19:34] ;optimax phase4 top25 [19:34] ;optimax phase4 100% [19:34] if this doesn'T work then there is an incompatibility between Windows, Perl and optimax [19:37] ok, I'm away for a while [19:37] * Fizmo is back later [19:37] MSG: Quit: ChatZilla 0.9.84 [Firefox 3.0.10/2009042316] [20:45] hell, it needs time [20:58] ;optimax phase1 cyclemin nn% - the warrior must survive at least nn% of max. cycle to be passed to phase 2 [20:58] is it possible to do exactly the opposite [20:58] ? [20:59] you want it to die as fast as possible? [20:59] for some reason, yes [20:59] bomber and spiral [20:59] why do you want the spiral to die? [20:59] but im only testing the bomb part [20:59] just ignore the cyclemin? [21:00] m......well, could be bad if the bomber survives [21:01] for example, the bomber is hit by a stun of the scanner [21:01] so I would prefer to know the bomb part dying [21:02] or isnt it a stupid idea? [21:05] well, how does that have to do with the time of survival? [21:05] so you dont say anything cause it is stupid :D [21:06] it has to pass a minimum, but also not exceed a maximum; I want to know the bomb part is dead cause it could be a weak spot [21:06] no, i just dont want to push my way of thinking on to you [21:07] nono, I always have some kind of stupid ideas............if your arguments are better than mine I would like to hear about how you see it [21:08] the more bombs a bomber throws without hitting itself, the longer it lives [21:09] but I dont want it to live anymore. why should it? [21:09] after a certain amount of cycles it should hit the same positions it bombed before [21:10] and I dont want to omptimize it against silks [21:10] just anti scanner [21:10] Hi all, in light of http://groups.google.com/group/rec.games.corewar/browse_thread/thread/74f28d71be1a024d I propose irc.freenode.net #corewars [21:10] if you use random stepsizes, then it becomes hard to predict where the bomb will hit [21:11] damnit [21:11] !{A-B} and this way? [21:11] you will probably end up getting the stepsize of 3039 [21:13] mm...... [21:13] when finding a bombing-pattern, you usually need to "push" a pattern so it hits exactly where you need it [21:15] so if you have something like add.ab #step, 1 | mov bomb, step then mov bomb, step+1 will make the final bomb, maybe hit the instruction following where it would "normally" hit [21:16] Topic Change: johnkw sets topic: This IRC server will be shutting down soon - please move to irc.freenode.net #corewars [21:17] mm....yes....... [21:18] so I end up with this: http://pastebin.com/m645855be [21:22] ..should i bother trying to run it.. [21:22] sry^^ [21:24] i am not offended, or anything, so no need to say [21:26] does it score better in the benchmark? [21:27] because i shouldnt try to analyze the bomber too deeply when we are only after scoring right now [21:29] mm.......I think I had better scores this afternoon with two dirrent values [21:29] *different [21:35] well, thank you very much. [21:35] probably I try some primes or so [21:35] primes wont work [21:36] ._. [21:36] xD [21:43] Topic Change: johnkw sets topic: This IRC server will be shutting down the 30th or 31st of May - please move to irc.freenode.net #corewars [22:35] MSG: Quit: humhum [22:46] I advocate the change to freenode [22:46] but we don't have logs there yet [22:51] MSG: [22:51] Join: Altair joined #corewars [22:52] MSG: Client Quit