[02:03] Join: brx joined #corewars [02:04] names [02:04] hrm [02:20] MSG: [07:23] Morning all! :-) [07:43] 3039*2^280929+1 [08:02] Join: Mizcu joined #corewars [08:14] lol [08:14] SHUT UP emiclock [08:14] whee it's gone [08:14] back to anonymous? [08:14] ]/query emiclock [08:14] pah [08:15] :O [08:15] how many were comments? ;) [08:15] * Icarus waves [08:15] :( [08:15] only if you REALLY really don't have SSI/PHP/whatever [08:15] 'cos they're annoying [08:15] http://www.useit.com/alertbox/9612.html [08:17] Join: Roy joined #corewars [08:25] Nice prime JM! [10:21] Cool, just wrote a little Java applet that allows me to log on to MSN Messenger... [12:37] Why isn't Fluffy here... he is always here...! [12:37] Why is nobody talking..? :-( [12:37] I'M SO ALOOONNNEEE!!! [12:37] aww [12:38] :-) helloooooww [12:38] :) [12:38] how are you? [12:39] ...who are you? [12:39] not toooo bad [12:39] going to go drinking for St patrick's day soon [12:39] yerself? [12:39] Just bored at work, just finished that test-msn-messenger thingy [12:40] Waiting for comments from my teachers on my thesis so I can change it and deposit it next with [12:40] St. Patricks day sounds like fun... why don't we have that here :( [12:41] :( where are you? [12:41] Netherlands (Holland) [12:41] ah [12:42] Sudburry? [12:43] ? [12:44] Is that where you celebrate st patricksday? [12:45] no, I'm in London at the moment [12:45] Ah, the Geo-IP said you are from sudburry [12:46] iinteresting [12:46] no, definitely not there :) [12:47] The computer itself is to the west of london (it's a colocated box) but i'm at work atm [12:48] ah.. well, it said "Guessed location" so it was a bad guess then ;-) [12:48] :{ [12:48] * :) [13:12] MSG: [13:28] Join: raptor joined #corewars [13:37] MSG: Ping timeout: 252 seconds [13:41] Join: raptor joined #corewars [13:45] MSG: Ping timeout: 252 seconds [13:49] Join: raptor joined #corewars [14:00] Join: Roy joined #corewars [14:00] * Roy just send warriors to koenigstuhl, keep a eye out there.. [14:09] Join: Nenad joined #corewars [14:09] Hi [14:11] ... nobody here... [14:11] MSG: Client Quit [14:23] * Roy is here...HAHA FOOLED YOU! [14:23] (sorry, wan't paying attention) [14:27] MSG: [14:42] wheeeeeeeeeeeeee guinness [14:47] Join: bvowk joined #corewars [14:47] hrm. [14:47] its the day of green beers. [14:47] also the birthdate of my grandfather.. [14:48] however, I expect he likely got up at 4am and is out at the farm doing something.. [14:48] wow [14:48] wow? [14:50] hrm.... I SAID "WOW?" [14:50] *tap*tap* this thing on? [14:51] imagine being born on St Patrick's :) [14:51] heh [14:51] sorry [14:52] whatcha up to? [14:52] trying to ignore the impending clouds of guinness fog and spend an afternoon actually working :) [14:57] working on what? [14:57] * bvowk is eagerly awaiting his holiday [14:57] and going out to the farm, I'm teaching a 14 year old girl how to use unix/program.. [14:57] oh, writing Java :( [14:57] ooh. [14:58] (while my wife rides her horse) [14:58] the girl is painfully smart.. I wish I was that sharp :) [14:59] took her about 12 seconds to figure out the concept of a pointer, and we were off playing with simple assembly code [14:59] blimey [14:59] I knew people in my engineering classes who spent 2 years trying to figure out that concept [15:00] I also taught her the concept of buffer overflows.. and we exploited hello world :) [15:00] lol [15:01] when I was 14 I was writing BBC Basic [15:01] yeah, but when you were 14 you didn't have a computer professional coming to visit every weekend who was bored silly [15:02] 'tis true :) [15:03] heh.. when I was 14, the best I could do was find bored computer professionals on a BBS in the next town [15:03] :) [15:03] one guy gave me some big vax cabinets! [15:03] they were upgrading the regional control office at one of the oilfield hq's.. and it was all going ot get junked.. [15:04] sadly, I was never able to convince my parents to put in the big breakers I needed.. but it was still fun to investigate.. [15:04] how old are you Icarus? [15:05] 22 [15:06] hrm.. when exactly did I become an old man at #corewars? [15:06] this is very troubling :) [15:06] you always were [15:06] lol [15:06] heh [15:07] no, when I was here in high school.. I wasn't that old.. [15:07] I might have been grumpy... but I wasn't old :) [15:08] what about you miz? [15:08] 20 [15:08] thats what I thought. [15:08] I wonder if I could get the 14 year old into corewars.. [15:09] she's apparently been playing nethack this week, because last week I bet her $100 should couldn't ascend [15:10] lol [15:10] I suck at nethack :( [15:10] I've got a game I'm about to ascend in.. [15:10] I would be done by now, but I made a side trip to try pudding farming.. [15:10] I'd never gone pudding farming before.. [15:14] it was a bad game too.. my kitten got killed by a rock, there were no magic lamps in mine town, the minetown priest was accidently zotted by a reflected lightning bolt from a stupid gnome, and I couldn't find any good non-metallic armour until I sacked the castle.. oh.. and I found 4! yes, 4! scrolls of identify before I hit the castle wished for a damn spellbook of identify [15:15] :s [15:15] only good thing was I started with a ring of slow digestion, and an identified wand of death [15:16] and there was a gnome wearing an amulet of reflection in the dungeons [15:16] gosh? [15:16] OK, you're now confusing me. I'm logged onto #nethack over there <--- :-) [15:16] lol [15:17] I tried for a wish at the oracles level, but all I succeeded in doing was drowning my large dog :( [15:18] The generally accepted advice is that trying to get wishes from fountains is a waste of time. [15:18] yeah, but like I said, I was really hurting for some non-metallic armour.. [15:19] as I couldn't get any divine protection, and I had no luck identifying scrolls in mine town (stupid wand carrying gnomes) [15:19] oh yeah! thats the other bad thing that happened.. [15:20] there's a big shop filled with spellbooks.. [15:20] You can always buy protection off the priest of Moloch in the valley if you need to. [15:21] I'm like woohoo! so I pile them up by the door (because I had an str of 5, bad run in with orcs) and suddenly there's a bolt of fire, and I seem a bunch of puffs of smoke :( [15:21] pak: gotta make it that far first :) [15:22] I try not to descend past the 10th level until I'm at least -15 [15:22] (especially if I'm not very fast) [15:23] too many run in's with fast trolls with polearms [15:23] you play in the november tourny pak? [15:24] One ascension :-) [15:24] in the tourny? [15:24] Yes :-) [15:24] or just one ascension period? :) [15:24] very nice! [15:24] which year? [15:25] 2005. [15:25] I didn't do so well this year :/ [15:28] 254219968bvowk-Wiz-Elf-Fem-Cha ascended to demigoddess-hood.2510[2510]6Alberta(0)-1135 [15:28] did alright 2004 tho :) [15:29] 2003 I ran a server, so they pulled my scores from the list [15:32] Which is fair enough really :-) [15:32] yeah.. [15:32] I had a really good game tho.. and I wasn't really admining the box.. I just provided the hardware and did the initial setup and then handed off to my friend.. [15:33] http://groups.google.ca/group/rec.games.roguelike.nethack/browse_thread/thread/a103d97a5c592bfe/b21036b9c22574a7?lnk=st&q=bvowk+kills+everything&rnum=1&hl=en#b21036b9c22574a7 [15:35] it was still fun :) [15:35] how old are you pak? [15:37] 29 [15:37] Should be on John's list :-) [15:37] ah! [15:37] someone older than I! [15:37] wonderful. [15:38] ever wondered why sun's slogan is "the network is the computer?", because they sell bigass unix boxes, not networking hardware.. [15:38] pretty much the opposite of their slogan.. [15:38] granted, when they came up with it, they were pumping those little xterm boxes pretty hard.. [15:38] but still.. [15:38] thats more like "the hugeass network connected computer of doom is the computer" [15:40] I asked at the breakfast thingy they had a while back why all their slides had ipods on them.. given that they don't make, sell, or really support ipods on their platforms.. [15:40] and the one guy said, you know, that bugs me too.. [15:41] I better stop talking or the monologue trophy is going to go to me instead of brx. [15:41] :) [15:42] anyone planning on doing anything interesting this weekend? [15:42] watching the six nations, planning to move house (I rent),...not a lot :) [15:42] moving sucks.. I hope its not cold there [15:43] nah, not too bad. Got a couple of months before I have to move out thankfully [15:43] ahh [15:43] but need to find a place [15:43] or I'll be sleeping on park benches :) [15:43] what city are you located in? [15:43] West london [15:43] hrm.. [15:44] doesn't sound like fun to me :) [15:44] well, it's an Exciting Step Forward in life [15:44] actually, I guess it wouldn't be fun here right now either.. we're at a record low vacancy rate.. (luckily I haven't had to rent in a long while) [15:44] Exciting Step Forward (TM)? [15:44] yeah :) [15:45] let me guess, you've moving in with a marketeer? :) [15:45] heh no, probably a couple of guys from my work [15:46] actually.. thats usually pretty cool.. [15:46] yeah, but it involves breaking up with my girlfriend of two years [15:46] ah well [15:47] although, you've gotta watch out.. my one room mate got jealous when I started dating my wife (he wanted to date her I guess), and starting doing things like leaving the garbage in my room on my bed.. using my CD's as coasters... [15:47] eep [15:47] very odd. [15:48] we'd been friends for like 10 years before that... so I didn't expect that. [15:48] have a backup plan! :) [15:48] :) yeah [15:49] so what do you write in java for a living? [15:50] when I see the quality of some of the scientific software we pay astounding amounts of money for, I'm always tempted to go start a scientific software company.. [15:50] I write a content management system for mobile phone content providers [15:51] but then I remember I'd have to deal with the faculty trying to squeeze me for freebies all the time.. [15:51] heh [15:51] is it fun? [15:51] yeah, I enjoy it [15:51] thats good.. I don't think I would :) [15:51] they don't me sloping off to the pub on a friday lunch time for beer, so... :) [15:51] :) [15:51] thats not bad either then [15:52] :D [15:52] I stay here because they don't mind me burning all the wasted computrons, even if the job does suck a good portion of the time [15:52] what do you do? [15:52] sysadmin for a university math department [15:53] Ahh [15:53] I'm the network/server/security guy [15:53] that's pretty cool [15:53] we've got a really clueful windows guy now.. so thats much easier [15:53] (he should be clueful tho, I helped train him! :)) [15:53] lol [15:54] he was one of my groomsmen at my wedding.. he had been living in an oilfield camp up north... so they do a 10 hour day, and then they go watch movies and sleep (it was a dry camp, so no booze, no women) [15:55] so after a couple months of that, he told me I had to get him a job in computers again :) [15:55] :) [15:55] so I sent him a truckload of gear and books and software.. and he spent his nights on that.. :) [15:56] the previous guy.. well.. he was about as sharp as warm butter.. he interviewed really well.. but there was a tiny hamster in a broken wheel under the hood.. [15:56] aww [15:57] meh.. valuable life lesson "You cannot teach problem solving skills." [15:59] people either think about their problem, or they just keep trying random things until something works, or they really get in over their head and come looking for a bailout [17:41] Join: Roy joined #corewars [17:50] hrm.. [17:51] Now listening to Adam Green... very odd music, don't know if I like it (the humor) or not.. [17:52] A piece of the lyrics: [17:52] Goodnight to my new dead wife [17:52] Goodnight to my Nazi friends [17:52] I'm standing on the tip of my stinking ship [17:52] Hey, Princess [17:52] Over there [17:52] Why you sitting over there? [17:52] When you've got a warm spot to share [17:52] In the prince's bed, in the prince's bed? [17:52] ... stupid :) [18:14] Join: sascha joined #corewars [18:19] White Noise' end is near! [18:27] Join: DrewHK joined #corewars [18:27] hi [18:28] Hi [18:38] Join: Fluffy joined #corewars [18:39] hi Fluff@ [18:41] bbl [18:41] * Fluffy waves [18:41] Part: Fluffy left #corewars [18:44] * sascha waves [18:49] MSG: Ping timeout: 252 seconds [19:03] MSG: Ping timeout: 252 seconds [19:04] Join: John joined #corewars [19:12] MSG: [19:25] Join: Nenad joined #corewars [19:38] MSG: Ping timeout: 252 seconds [19:41] hrm [19:54] what are you hrming about? [21:19] MSG: [21:19] Join: Fluffy joined #corewars [21:19] :) [21:20] Hiya [21:21] hey fluffer [21:21] * Fluffy kicks bvowk [21:21] hi Roy [21:21] Hi bvowk [21:21] hi bkwak [21:22] All it needed was a tiny "emergency" ... [21:22] now I finally sit on a fast DSL connection [21:22] Nice. [21:23] But not for long [21:23] I have to go back to Berlin soon [21:27] Yesterday I only got time to extend the benchmark for PyCorewar [21:27] Now it contains 50 fights [21:27] with 1000, 2500, 5000 or 10000 rounds [21:28] It was really necessary, because I couldn't use the old one any more [21:28] Nice, got new speeds? [21:29] No, speed didn't change, because I didn't change the code of the library ... only the benchmark is new [21:29] http://corewars.jgutzeit.de/pycorewar/download/pycorewar-0.2.0.dev-20060316.tar.bz2 [21:29] well ... actually the speed has changed [21:30] with the old benchmark I got sth. like 6.7 MIPS [21:30] the new one is around 7.1 MIPS [21:30] hehe, with a bigger bench it'll probably be more accurate.. [21:30] Unfortunately I only could run the 1000-round version [21:31] I had no time for the 10000-round version [21:31] I had to catch a train [21:31] and now I only have a slow Pentium 133 MMX [21:31] Heh too bad, if I had a windows version I could ;-) [21:32] I've sent a copy to Sascha. Tomorrow we'll know it for sure [21:32] he has run benchmarks for a couple of days now [21:33] and the new benchmark is compatible with older versions ... so he can compare them :) [21:33] But it is kind of depressing [21:33] on my big computer I have 7 MIPS [21:34] and Sascha slowest I expect sth. like 25 MIPS [21:34] Hehe nice, I'm curious what my laptop will give [21:34] Some days ago I've tried the old benchmark [21:35] it was around 1.4 MIPS [21:35] it took ages (12 fights, 1000 rounds) [21:37] Roy: I hope, that I only have to make tiny changes to the source to make VC compile it [21:37] Seems so.. some C files compiled without problems [21:38] The fun part is, that ProcessQueue.c/h is part of the Debug-version, on which I haven't worked for ages [21:38] maybe I should remove it for a while [21:39] C is C... can't forsee much problems, only in getting the settings (VC/Python) to work [21:40] Let's hope, that you are right. [21:43] Sascha has proposed an interesting idea [21:43] instead of maintaining a process queue of size 8000 for each warriors, that wraps around [21:44] we could use a process queue of size 80000, which would eliminate the need to check for that wrap around [21:44] a lot of if-then-else could be removed that way [21:45] Ah hmm.. [21:45] Process queue == CYCLES..? [21:46] the size of the process queue (from warriors point of view) would still be 8000 [21:46] but if I would use more memory I could implement is as a kind of sliding window [21:48] Probably it is not very clear, right? [21:48] Well, I get the wrapping around stuff, but not the solution to it :) [21:49] at the moment I do the following to implement the process queue of a warrior [21:49] I allocation memory for 8000 slots [21:50] I have a pointer to the start and a pointer to the end of the current queue [21:50] If I add a new process, I write it to the end of the queue [21:50] and move the pointer on slot further [21:51] If I remove on process (from top), I only move the top-pointer [21:51] after a while the end of the allocated memory is reached and the end-pointer would point outside the allocated memory [21:51] I simply make it wrap around and point to the start of the allocated memory [21:52] and of course I check, that I don't add to many processes (i.e. <= 8000) [21:52] are you with me? [21:52] Yes! [21:53] the new solution is as follows: [21:53] there are 80000 executed instruction all in all during a fight [21:53] now I allocate memory for 80000 slots [21:54] but still make sure, that there can be only 8000 processes max. [21:54] I sill have a pointer to start and one pointer to end of current queue [21:54] but since the number of processes, which are created/removed during fight can't exceed 80000 [21:55] I don't have to check, whether I have reached the end of the allocated memory [21:55] yes, cycle amount [21:55] that removes the need to check for wraps around [21:55] Thats understandable, but well.. the system does require more mem [21:55] and which should theoretically speed things up [21:55] it is not too much memory [21:55] a pointer needs 4 bytes [21:55] yep [21:56] well yeah, its not that bad [21:56] 4 * 80000 = 320000 bytes for one warrior [21:56] not that much [21:56] even for 94x it shouldn't much [21:56] and evem for 94xm it shouldn't be that much memory [21:57] I'll try to implement that during weekend [21:57] why have a queue for every warrior? why not have a common queue? [21:57] and after that I have to steal all the good code from exhaust/exhaust-ma/fmars :-) [21:58] DrewHK: Actually I have one big queue, which splits into separate parts for each warrior [21:58] yes, but you can comb them together [21:58] but maybe that is a silly idea [21:58] no, that is difficult [21:59] thing of two warriors, which wildly add and "remove" processes [21:59] I'd have to keep track of where each process is in the process queue [21:59] and which wwarrior owns it [21:59] yes [21:59] you are right [22:00] I'd love to see your idea up and working [22:00] but it seems hard to do [22:01] my mind is not up to the problem just right now [22:02] this is your own mars aou are writing? [22:02] well ... I've taken a lot of ideas from exhaust/exhaust-ma [22:02] but I've written it myself [22:02] is it public? [22:02] yes [22:03] does it compile with vc? [22:03] the last version is http://corewars.jgutzeit.de/pycorewar/download/pycorewar-0.2.0.dev-20060316.tar.bz2 [22:03] no, not yet [22:03] I only work with Linux [22:03] and gcc [22:04] But I hope, that somebody (let's call him Roy ;-) will soon have a working version for windows [22:05] why do you have an own? for fun, or is it better than others? [22:07] I wanted to have a fast MARS with a Python interface [22:07] and the only other MARS with a Python interface is fmars [22:07] oh [22:07] which isn't exactly, what I want [22:08] where can I get exhaust-ma? [22:08] one moment [22:10] ftp://ftp.freebsd.org/pub/FreeBSD/ports/local-distfiles/pav/exhaust-ma.zip [22:10] you might want to download exmars-0.01.tar.gz as well [22:10] whats the difference? [22:10] speed [22:11] which is faster? [22:11] depends [22:11] on what? [22:11] and the warriors, which you use [22:11] to determine, which one is faster [22:12] in any case both are quite fast [22:12] around 70-95% faster than pmars [22:12] I want to check the code [22:12] you might want to start with exhaust-1.9.2 [22:13] exhaust-ma and exmars use it as a start [22:13] and exhaust's code is very easy to read/understand [22:14] okay [22:15] has it documentation, or just source? [22:16] oh i see [22:17] just the source [22:18] but you can try to ask questions here [22:19] (hmm ... I hope, you were referring to exhaust) [22:21] there is a short guide-like-thingy [22:21] (exhaust) [22:21] ah ... yes, I forgot about it [22:22] * Roy doesn't like the idea of first reading mars-code and then making one yourself.. [22:22] Roy: But it can help to avaid some problems [22:22] Its better if you try it yourself first, think how you would do it, and then look how others did it :) [22:22] Well, that too [22:23] But it boxes your mind.. [22:23] Roy: As soon as I have fasted (well second-fastest) MARS, I'll start to object [22:23] (It will be hard to become faster than fmars, but I will try) [22:24] It might not make the fastest mars, but it will improve your programming skills tackeling the problems, and you never know what new ideas you might get [22:24] Roy: I have no programming skills :) [22:24] But I'm really good at stealing other people's code [22:25] * Roy too (in Redcode that is) [22:25] How does QMars measure up? [22:25] Unfortunately all versions of gcc, that I use, refuse to compile it [22:26] and I really don't want to lay my hands on code in C++ [22:26] It is as bad as Java [22:26] Ah so you don't really know.. got any trick from it? or didn't you read it at all [22:27] Didn't look into it [22:27] Heey.. Java isn't so bad, its just a different way of programming with the object stuff [22:27] But I've only started to harvest ideas from other MARS' code [22:27] qmars makes a bold claim: "fastest simulator on planet"...so :-) [22:27] Roy: Java isn't that bad, but you have to type so many useless letters to get anything done [22:28] Faster then exhaust that is, because thats also mentioned on the site, and he still claims to have the fastest [22:29] fmars and exhaust-ma are definitely faster than exhaust [22:29] and of course PyCorewar :) [22:31] i have to check that qmars too [22:32]