#COREWARS 20 OCT 2002 FIZMO: Fizmo a.k.a. ? FORDBC: Ben Ford LGRABUN: Lukasz Grabun METCALF: John Metcalf PHILB: Phil Thorne SMP10K: Ken Espiritu STEVEG: Steve Gunnell *** Now talking in #corewars hi Ben hello I'm trying to figure out what round 2, entry 2 was :) Erm, Lockdown? It was just a handy email in my inbox... *** Join: smp10k (fdfa@pool-207-68-49-48.norf.east.verizon.net) I don't save addresses, too lazy Hi Ken hey john.. oh :) new 94 hill? Not long had an IRC program, I thought it was about time! Yes. Steve Gunnell is king at the moment. looks like a hill from the mid 90's :) It is. It has been sat around in suspended animation (you could view it but not submit) JKW has killed off all the warriors which were on it and reactivated it. ah.. ok.. Looks like you may be king of '94 in a few minutes Ben :-) hopefully :) the top koenigstuhl warrior should be able to make it on I thought for round 4 there could be a non-compulsory naming theme... E.g. everyone names their warriors with names of extinct creatures, or Children's TV characters or something. Do you think this is a bad idea, I'm having doubts? 38 point lead on '94 :-) woohoo! most the warriors i submit are named by the computer these days... unless its a test :) Named by computer? I noticed you had moved away from your old one word names to two worders :-) yeah.. the optimization program just randomly generates names.. i.e. adjective and a noun Nice idea. I must admit I flip through the dictionary and choose something when I'm stuck. I always think of something a bunch of mine a names of the Netrunner CCG cards I noticed! Banpei, etc. Pattel's Virus Don't know that one. I don't really play it, I just got some out of curiousity. I named quite a few after charcters from childrens books... the only reason i started using computer generated name was it was hard to find the code window for each warrior since there are like 20 code windows... I only optimize by hand so its not a problem for me... You should try some kind of computer optimization. probably, but then I would have to write real code :) Even a small program in QBASIC which writes copies of your program to disk with random constants, then benchmark them all! yeah.. the only things that i optimize by hand are scanners and one shots sometimes.. everything else is too complex to be done by hand I'm not having much luck with scanners. Willow is a HSA type warrior with decoy (nothing original). Twinshot is a bit better. Two oneshots, as soon as one detects something, it kills the other oneshot. Helps against DAT bombers. its a wonder any of my papers last any length of time on the hills My papers don't do too well either :-( Though I'm quite pleased with my tiny hill papers. i think there are only a few type of scanner types that are effective at stopping 94 papers.. variable length spl carpets blur, fixed length spl carpets agony, hsa style carpets and linear carpets spl clears everything else is mostly ineffective... Yes. Mirage .33c scan / .33c clear doesn't seen so effective now though :-( anyone experiment with the imp-launcher I used in Olivia? yeah i've noticed that too.. no, i haven't looked at that warrior yet.. mirage/blur style has been optimized against lots I posted it a while ago I'm gonna be dominating the 94 hill! :) hehe.. you're the only one playing :) Olivia launches a colored imp-trail so a short-distance cmp-scanner will pick up the useless part of the trail more easily is there any way to get rid of pizza? it really isn't helping the community with it being broken.. People could stop linking to it, or link to it describing it as a site of historical interest only. I'm up to a 40 point lead! yeah.. that's what i was thinking... i wonder how many potential players we have lost because of it.. ah.. might be good for like a 7 point imp since there is less chance of it getting stunned.. I've just had a look at Olivia's launch. I wrote something like that about two weeks ago (my objective was to make the SPL invisible to .f scanners. I didn't think of the handy side effect. yeah, it was a side effect of making the spl invisible It makes Uninvited score better against my benchmark anyway. The SPL in Jedimp can be made invisible to .f scanners too. I've already submitted my jedimp that does that to the 94 hill Can't I come up with anything original ;-) Well apart from scanners that is! me neither *** Join: LGrabun (grabek@acn.pl) hey i come up with alot of original stuff.. it just doesn't make the grade :) Hi? most of my warriors are bits and pieces pulled from all over the place Hi. Oh Hi, John Hi all Just been talking about imp launchers. hi lukasz phew imp launchers... Hm * LGrabun don't know anything about imp launchers :) Have you tried out the ;redcode-94 hill yet at koth? nope not yet I've been working on my master thesis for the whole day We're just playing on there now. Ben is Koth. Will you send Reepicheep just to see how it handles p-spacers? What about? I could not find it on Christoph's page Okay, will send it now. The queue is about 5 long though Yes, I've noticed that Hm reepicheep is #1 nop in koen Do you know the irc command to get the list of all people that are on the channel who I just can't figure out who damn ' /who #corewars OIC yap got it thanks I am using this text version of xchat (I don't have gnome libraries installed) It is sent :-) I need to submit a warrior to the nop hills doh! wingshot cut into my lead on the 94 hill! Thanks John Yes. Have a go. How about having a go at a bomber / imp How about having a go at a bom? probably a reason those never show up This is all I see Rather than stone / imp I mean... Son of Vain would be hard to improve on incendiary bomber/imp? SoV is an ultimate warrior methinks Yes, incendiary/imp! incendiarys aren't very effective against papers.. I could give SoV my imp launcher and rename it! :) Hm guys will you be here later? I'll be back later. I gtg now Cheers. Got some work to, blast it to do, I mean ok.. see ya later.. see you *** Quit: LGrabun (Client Exiting) actually.. incendiary/imp might work if you can use a ssd type clear with it.. incendiary for the initial stun.. that would be too big I think Torch is the best scoring incendiary I could find (after removing all the decoys, qscans etc) I have a torch with better constants yeah.. and there's the problem of stunning the imp.. dragon fire or some such Torch uses a ssd clear. Draken Fire actually I used a d-clear after tho Shouldn't be stunned if you uses a d-clear and space things carefully hmm.. maybe the sd-clear may work.. gotta go and get some lunch.. i'll see you guys later.. I got hit by geometric spam from the self-battle results from everythin I submitted cheers. its a pain that especially with hotmail where your inbox is only 2meg Twinshot takes 2nd place! yeah, its not too bad tho, they are just text files dang, a point better than wingshot! wingshot++ just switched around the djn in the scan loop instead of lowering a counter, it hit one of the two scanned locations O well, completely different warriors. Twinshot will be on Koenigstuhl soon (you can find it in the INPUT folder for the 94nop Koenigstuhl) so a bit dirty for other scanners to hit and I pick it up myself when about to hit the second pass It is a pity the DJN.A in the clear can't be made into a DJN.F somehow wonder if g3-b will be published soon, must of been doing something different Have you read the code for G2? yeah firestorm knocks twinshot to 3rd ;) I have seen a oneshot with a self splitting scan loop :-) Works great, but for some reason the score isn't that good. goonie kinda does Self-splitting like a stone, with a leading spl # oh, that would be weird :) I'll have to ask if I can send you a copy. Reepicheep only 5th, for the moment scores badly against leaflet, timescape etc (as expected) hmm... have to dig up 8 more warriors to purge the hill I think leaflet was one of the tourney entries Not that I recall. oh I'm not sure what it is then, don't even know if it has had any optimization... do you remember it ever being submitted to the nop hill? Optimized to give more points to wingshot than twinshot perhaps I'll look now! You only submitted Aquafina and Wingshot++ in the last two years! lol was olivia really that long ago Sorry missed that one, 23 months ago just before I started logging properly. wow Reepicheep gives Pattel's virus a good kicking ;-) no surprising there was a glut of stone/papers when it got knocked off the pizza hill none of the components deal with a stone/paper Time to write a new warrior? Can you get a oneshot in there as well? self-modifying code is newer actually I suppose I should submit smc oh, leaflet made 10th on the no hill You killed Vanquisher. I didn't mean to And you hit Son of Vain worse than anyone else. wow I need a paper optimizing program! Bombers aren't doing very well at the moment anyway. It was the only one on the hill. imagine how good it could do if I actually gave it good constants *** Join: LGrabun (grabek@acn.pl) Hi again hi Ben Ford I killed Vanquisher Show your face here *slap slap* I just use short gwbasic program. Open a file, print a copy of the paper to it with random constants, then run a benchmark (using a couple of batch file and MTS) on them :))) He didn't mean to. Oh yes he meant to :) Leave Uninvited alone though! :-) My oldest warrior ::sniff:: Otherwise Son of Vain becomes the oldest warrior and that looks like it will be about forever I don't even know how I came up with Leaflet, it was just kinda sitting there Oh don't you say a word Murderer :))) Well guys what are your ideas for the next one next what? warrior, of course Vanquisher 14th in the HoF Well, that does not sound so bad, don't you think a incendinary/imp? :) * LGrabun thinks of anti-A&B-driven-imps-paper Neither this sound so bad A quick stone (0.6 for example, Tornado like), with leading spl, and 7-pt imps woo, SMC into 2nd on the 94 hill Since 94 hill is around still need 7 more submissions to the 94 hill Maybe it's time for deadly p-switcher just submit anything you have there I *think* I have some components Reepicheep gives SMC a good kicking too help clear out the killed warriors so its normal yeah yeah, my p-spacers don't deal with stone/papers! :P :-) They deal with older stone/paper okay. probably depends on the paper used pattels is weak against some papers *** Join: Philb (chatzilla@user-11218vp.dsl.mindspring.com) Hi phil Hello all Hi Phil John Going to have a go on the ;redcode-94 hill? submit to the 94 hill on koth, its needs real warriors Why does Reepicheep handles paper-like warriors not-so-good? Just woken up - it's moving again? Oh yes it *is* Paper ties paper, paper beats stone. We lose a little but don't win. Won't hurt us though! I think Reepicheep is a second ultimate warrior IMNSHO Just like SoV It's been published but it does not leave the top five stone/papers have strong weaknesses But what ultimate warrior will be next? paper/clear? heh SoV is a stone/clear/imp so we can't have an ultimate clear/imp :-( Ben: do you mean o/s? yeah I tried paper clear before. me too I just submitted an untoned one to the 94 hill! I think we tried it around the same time back on Pizza. I found I was better of with paper/stone. I didn't optimize it too well though. Steve made some clear/papers time ago with no succes as I recall Vanquisher is 7th Well it was not created to beat p-spacers a 94-nop warrior needs more versatility (just like SoV, though it's no p-spacer it works differently against different opponents) Vampiric paper - did that ever wrk? [I think Ben you had some?] not really but I don't tune papers very well There hasn't been an intelligent '94 warrior like Leapfrog yet has there? Argh... don't think so Irish folks decided to ratify Maastrich bill (whatever you call it) thats a possibility Phew Sorry for the OT what does that do? Leapfrog? the Maastrich bill :) Oh LOL Poland may now access the EU, I suppose It runs a stone, which at the same time detects if it has hit something. If it hit something but didn't kill it, then it runs a paper (must be fighting paper of a multi-process warrior it assumes) oh Does Poland want to enter the EU? John, please Your paper/clear seems to do well against the scanners. OK, here's my idea for next warrior Have dual-boot paper, one copy with against A-driven imps the other one with dat >2667, >5334 bomb (kill B-driven imps) PVK mentioned it three/four days ago on r.g.c Does PVK ever submit anymore? John - you have all the stats, right? I must have missed that post. Can't you get both bombs in one paper? or maybe dual-papers, each tweaked to be anti- different scanner types Paul doesn't write much these days. He was one of the very first on the '94nop hill :-) And the other PK? I think there are only 10 people or so who submit [bvowk has gone] anti-A imp bomb is something like mov.i #1, {1 or similar, can't remember that yeah One can also make two copies of paper in warrior's body something like anti-B-paper spl 1, spl 1, spl 1, ... blah blah anti-A-papr spl 1 spl 1 spl1 , mov #1 , {1 blah and in the begining make spl anti-A and jmp anti-B Ben, what were the strong weaknesses of paper/stone you mentioned? Just 8 people made it onto the '94nop since last Core Warrior. oneshot type stuff Philb Why do you want to know this? :) I think paper == huge target. Would _think_ it easy to disrupt Reep* so it papered it's own stone. John you know that Fizmo -= Schmidt? Sorry, forgot Leaflet, makes that 9 Leave Reepicheep alone! :) Ok, Purifier [but that's under control] :-) just 3 more to go on the 94 hill Lukasz significance of "-=" == or != ? No, no -= is = My irc client does not work so well Or maybe I can't use it in the proper way Anyway looking at the 94nop think of DM's hill wipe? damn Hmmm... I did suspect. Schmidt disappeared for a long time, during which Fizmo was on the hill. Fizmo's Argon in on Koenigstuhl, I haven't read through the source yet, but I think it is quite unusual... Oneshot with a paper or something :) Phil that was stupid IMO Fizmo really had me motivated - he came out of nowhere. Each day I'd wake up and "Fizmo!" JKW should disable PIN numbers I wasn't judging it just thinking Ben's almost got it made :-) I think Phooey (in the hall of fame) was Ian Oversby... Any one else use pseudonyms? I never have other than for dat 0,0 warriors :) Oh, was he? Sometimes I submit as metcalf - always try to use a new email address for him, hehe That's funny. Metcalf, Metcalf? Can't recall Argon is also #1 on the yellow begineers hill. Too strong. :) Hm is it me or Christoph just did not submit Reepicheep to open hill? I do check all those email addresses you know. One of them sends text alerts to my mobile (so I know what is going on on the hill), one is convenient for attachments and the other, I don't know. he submits to everything its valid for I think There is a holding pen of 8 for the Open hill... Reepicheep is in there, as are a few warriors I wrote which Janeczek won for round 4 and decided to publish. Well, it does not show on the hill though But I'm very interested in the results Just below the list of new warriors on the Koenigstuhl hill it mentions the holding pen. OIC have not noticed that woo, Playing with Fire back to the top of the multi hill :) www.ociw.edu/~birk/COREWAR/OPEN/INPUT lets you see the list (and source) I sent a dat 0 to the multi hill yest as it's so volatile - never knew which was best. Result = Redemption / Clock / Her Maj Multi hill? Oh the extended multi hill should either be eliminated or get a new scoring algorithm My multi warriors don't do so well these days I never bothered Don't you think it possible to break it? John you mentioned PIN - do you mean NOP supports PIN? The new ;redcode-94 hill probably supports PINs Don't like it, definitely don't like it Just Agony II to be removed now haha, Agony is beating 2 of my junk warriors :) What's so specail about TrItEFighter that it makes the holding pen? [Had the least ties BTW!] I still have 2 more in the queue Anyone interested in trying to break into the -mx hill? There are at least 4 sptst warriors there so finding a weakness against them may be enough to scrape in after which it should get easier You'll definitely be the KotH. With so many warriors... :) the -mx hill can't be broken I agree too many papers in too much code in a scoring system that promotes ties too much Pls don't say that - I only have so much time to battle the impossible.... :-) haha, my paper/clear died before Agony :) wow, Agony scores really well against twinshot Hey, Agony's moving up not down looks like I'm going to ned help removing Agony II LOL submit some anti-scanner warriors :) Are you suggesting I kill Twinshot to manipulate the hill Candy! no just some more stone/imps or p-spacers I think I have 1 or 2 more bad stone/imps if neccessary Candy is coming! Oh I *love* this s/i then after Agony is gone, you just have to wipe out all the bad warriors I submitted :) need to get Moore to submit Combatra and such stuff Uninvited is up now, giving Agony a hard time. Did we discover who prompted JKW to raise the hill? [My guess is SG] we should get the lp and tiny hills up as well I only skim the political r.g.cw posts but I believe I saw something. they are both fun I think tiny is great. I nearly posted an April 1st showing my warrior winning the 1st round of the tournament, but decided it may not be such a good idea. John maybe this should be a round for your [extended] tourney - 8 hrs to fill a new hill. Easter Egg, it even labels apri1 and foo1 in the code. Someone write some tiny warriors Or similar - anyway means that the round is over with quickly - people don't / can't spend too much time I thought about a rolling tournament. A different volunteer runs each round. It goes on forever. You score is the sum of you best 6 scores from the last 10 rounds (or something) Standings of players go in Corewarrior and on a web-page. That way you get to play too. That would shift emphasis from warriors to players. You would only have to compete in half the rounds to get a decent score. Ah deja-vu 94 hill thinks as little of knp as does 94nop heh Agony down to #20 Good. And Reep is moving on up and Candy in at #3 Great. this is kinda like when we fixed the 94 hill after the great wipe Stunning papers - is it nec. to cover the whole core? e.g. what if heavily / repeatedly stun section of core Pattel only has a 6 point lead now :-) BTW who quit after the wipe - I see references that some did but not who. pattel is really good against a lot of old-school warriors :) If you stun the paper well enough, you don't need to hit all of the papers. You'll need to clear them up with the dat wipe though. Darn. Perhaps Agony deserves to live? Erm... I think no-one actually did quit in the end. Moore retired (he's back again sometimes) nah, Agony is handicapped it has an unfair self-score Re stunning - that's what I mean - rather than having some complex circular spl carpet just work on a section of core repeatedly while slowly dating the rest. And some people didn't see the point in submitted while the hill was in the state it was. Not sure if anyone actually disappeared. Hmmm... not sure. Lets see if anyone on the '94 doesn't like bomb-dodgers then! Oh, is Fizmo definitely Schmidt, there has been another Christian around recently. Oh, isn't Reepicheep a piece of good code? :) pattel vs havenopity... 212/23/15 :) :) I'm getting a bad feeling about Piltdown... O well, worth a try. Its only a bomb dodger / scanner. My first warrior on Pizza's '94 hill. The bomb dodger does well against Wingshot though :-) Well Reepicheep has been at #1 on '94nop more than half the time since the last Corewarrior so it must be doing something right. Gee, I have just one warrior on the nop hill Time to remove the dust from my old code snippets that waited for that moment Vanquisher II for example airbag stone with dual clear or just a *stone* to be more resistant Is JKW around? You think Roy could have been mistaken/confused over which Christian? JKW's been idel for 5 days 19hours etc. use "/whois johnkw" 5 days Oh, he *does* have a patience :) I think he logs in at about 10 am US time Mondays., US time I mean Central = 6hrs behind London. I Great If I only knew what London time is now I think it's one hour later/earlier Oops, I sent a dud to the hill! ouch London 7:23 - JM can confirm Yes. It's 8:23 PM in Warsaw/Poland Not too different then. It is raining here :-( And another dud :-( Back to stone/paper - surely the stone is in danger if the paper's are deflected even slightly? And the papers make a huge target. I think so All components need to be *carefully* calibrated And it *does* take time The stone is of use very early in the game against scanners. If you ensure none of the early copies of paper (or their bombs) are too close to the stone it should be okay. Is this silly - a scanner that starts out in pieces e.g. 1 inst every 5 core locations which is then assembled later in the round? idea being that initially it will miss [regular] mod bombing Not mentioning that all boot offsets are chosen more or less randomly :) Anyone tried CIA bombs in a qscan? If you failed to kill then it lets you know if you missed your opponent (probably a booted scanner) or wounded it (a multi-part warrior). You can make use of this info and run either a stone or something else. Maybe? What are CIA bombs? If hit they decrement a known fixed location in core - the thrower can check this to see if something got hit CIA uses them. dat <-eye, Look for Anders Inver on Planars Clever But - I think - it finds its use in p-spacers only? There was an airbag CIA bomber on '94nop. Behemot lite. Oh, Behemot. One of the best bombers I've ever seen However it was beyond my skills to understand the code Does it use airbag or does it not? Well I have succeeded in pushing Agony further up the hill. Behemot does yes. I only read it carefully the other day. A spl 0, jmp -1 bomber Hey new round idea - who can push Agnony to the top... Well if I submit any more rubbish warriors I'll probably do that unintentionally *** Quit: LGrabun (Read error: 0 (Undefined error: 0)) John Behmenot "lite" ? Is that CIA style - I tried to understand the code with using pmars, hard. Behemot lite was never published and didn't last long on the hill. *** Join: grabun (grabek@acn.pl) Multi-process warriors can be hard to understand :-( Welcome back. Welcome back. Don't know what was that We do ERROR 0 :-) :) <:3_)-- No way I run Linux box And this is supposed to be...? A mouse! Oh a mouse How did I not realise it at the first look... * grabun is so silly :ppp Not many one line ascii graphics can be done. nice for a warrior name though :) Hm Guys, may I ask of a little OT? Okay Did anyone of you use kazlib? It's a library for queue handling? (no question mark at the end of previous sentence) Re Fizmo - he signs himself as Chris. CS used to sign himself as Christian. So perhaps a diff? Not me. Hm Who posted a Velvet Glove source to the ng? Compare the coding style of Fizmo and Christian and you will have the answer. it was fizmo_master@yahoo.com as Chris VG needs some nice constants :-) I thought there could be a voluntary naming theme for round 4, perhaps everyone naming their warriors after extinct animals or something? Good or bad idea? * grabun takes T-Rex :) might cause a lot of repeated names Sure, will make them easy to identify when they make the hill People could book beforehand on r.g.c if they were concerned :))) People could submit with a 2nd / 3rd alternative name, or I could let them know if the name was already in use. The scans look the same but aren't they frequently copied? Reminds me can anyone recommend a simple to add QS for '94 - I just want to see how much I miss by not having one [I use a Quick-bomber, less thinking] Second - I have Fat Expansion on the ICWS. It's a work in progress, needs a new scan. Is the one from Frieght Train the definitive '88 QS? Q^3 is easy to add I would assume so I tried a 94 one once and it made no difference - I can't remember which but I think I took it from an LG posting in which he took it from LL or maybe CS? Anyway I tried it quite a bit after this so thought maybe people were reversing it. I guess more likely my code got too easy to find. ::blush:: I can't code Qscans And always ask person who is here on the channel ATM to code it for me :) q^4 is great if you can do it, it can't be reversed Is CS on-line? I just got the fight-results from the hill I'd like to try but time is a problem - I'm amazed at KE's output. I have tried something based on FT in '88 but missed something important and never went back. To compare the '88 qscans, count how many locations they scan, and in how many instructions. I think not - verizon is US, CS is in Europe. yeah, CS submitted two warriors to the 94 hill I think Vala had the best ratio, but I may be wrong. Not checked Quicksilver '88. There is another in Stillborn Bomber and Freight train you know about already Do you find mini-Qscans more effective too? On online == attacking the hill. Yes he's ready to slaughter. :) So Fizmo is not CS, I think For a '94 qscan I always use MiniQ^3 (the one from KafuFFLe which has some tweaks which ever so slightly improve the score.) It is easier than Q^4, you just paste it in, with an ORG qGo and call the first line of your main warrior to be executed pGo. CS is CS and that's it :) Tx wil take a look. Agony has died And look who's #1 Was it datted? Good news! Yes, I know No but you don't need to DAT ... just look at the huge array. It's limited to the hill warriors. Look who's number one on '94nop! Wow, we have new KotH Christian-not-Fizmo-Schmidt :) Wonder for how long though Till the first dat I hope :) Re sending in DAT warriors - I have one [Fail?] for fun that doens't DAT - it tries to do worse than DAT by avoiding self-wins. e.g. draw vs self and then [tricky part] contrive to lose everything against hill warriors. I think that was covered in an old CoreWarrior It was interesting to find that some warriors [Behemenot] were almost impossible not to pickup points against [1 tie] Covered using PSpace but here I mean nop you actually don't want to draw at least in r.g.c postings but maybe cw - I haven't read all/remember all. scoring wise, you want player 1 to lose them all Draw gives 100pts. splitting the wins gives 150. you want player 1 to go 0-250-0 No PSPACE so can't tell Player 1 oh yeah It's tricky - new round idea for JM. Beat a white warrior by the lowest possible margin. a length 100 warrior with 99 nops and 1 jump? Try - you'll see it's tricky. Mine is something like 4 lines self-seeking scan. Interesting Crazyshot missed the 94 but it KOTH of nop Crazyshot is the real KOTH 94 isn't mature yet, it has some known to be poor warriors on it P-spacers are at 3rd, 13,16,17,18,19 or '94. The nop warriors are doing well. its been a while since people have developed strong p-warriors I never liked p-switchers anyway. bah! ...99nops + 1 jmp scores too high. Oh I *love* Reepicheep, I love it probably need something more suicidal Here comes silking - I really hate it. instead of nops, use div.x #1, #1 That risks impeding/distorting hill warriors [anti djn] only if they are suicidal if a djn hits it, next time loop goes, it will blow itself I would hope nothing that makes the hill is suicidal if the djn triggers I'm trying it. Vamps would be a problems but there aren't any for 10 jmn @0,1 rof nop 0,0 jmz -11,<-20 end argh, forget the jmn @0,1 the divs are more suicidal make it a nop okay, lets try it on the hill make it a jmp @0,1 I think After datting the nop hill Reepicheep is third See how dat gets 7.14 - next will be Loser Baby This is LB: code following dat #A, #B start: jmz: #3, } 0 sne -2 , {-1 jmp #A, #B Are you probing the un-bugness of the hill? The first dat should have '#''s I can't get on the hill so changed my challenge to failing the hardest LOL Hm This might be a challenging idea can be optimized as.... jmz #0, {0 Make a warrior 100-cell long sne 1, *-1 jmp #a, #b Whose each instruction is executed is not suicidal and scores the worse How much did your warrior score. Mine scored 6.57 :) The hill report shows 5 [this was from 06Oct ] - I don't recall the decimals but have resubmitted. Yes, I recall something similar Haven't received todays result back even though it's off the queue. The div.x version [above] scored 7 wins vs wilkies when running mts Put it next to a big decoy now :-) Loser Baby scores was 4.78 ! Nice score... Wow, this *is* impressive :) ouch depressive I wanted to say :) bah, 4.86 Am I the Peon Of The Hill? stupid vanquisher!!!!! I won a game against vanquisher??? with one tie against dry ice and trefoil LOL Thats what Imean - behemot was even worse. Crazy Hahahaha Program "low scorer" (length 3) by "Ben Ford" ORG START START JMZ.F # 0, { 0 SNE.I $ 1, * -1 JMP.X # 1357, # 2468 Oh I love this game Vanquisher lost to that??? *** Join: Fizmo (Fizmo@ACB0D562.ipt.aol.com) Vanquisher is *not* so good, you know :) Fizmo Hi Fizmo Whoa Hi all Our great unknown Fizmo ;-) We have an evening of the truth tonight . /whois fizmo Fizmo!!! :) Please say words 'bout yourself Yep to my person That's no proof. I am Christian Schmidt from germany Yap Yap We all know that 35 years old Philb Where are you from? Yeah I just came back to cw some month ago --- Ping reply from Fizmo : 0.52 second(s) ? He just *might* be from Germany :) Well, we've been doing some speculation about you, Mr Fizmo :) From? Africa via England. In Chicago USA Funny ;-) I started on my job to do something again in cw and I named me Fizmo Hi Fizmo - nice work on the hill. because I had an yahoo account calling fizmo_master I just wanted to see what's going on at the hill Well, welcome back Christian thanx Well you made the Hall of Fame under Fizmo Where do you live in Germany, exactly? I was wondering where you disappeared to all that time ;-) That's really crazy, I never thought that it would stay on the hill so long I was moving twice through germany for new jobs Lots of test ;kill test again these days Hey Fizmo - what's the story re Hunters Of The Red Moon, KOTH of the X-hill for a few hours then gone? there was not so much time for cw on these days He speaks words of Truth No much time... No time at all I remember time I tweaked Vanquisher It took more than three days to find proper constants for stone step Hunters was a p-warrior with a handshake and the two compounds from fire and ice and some more compounds, but they didn't count so much beside the fire and ice components Did you kill it or just test it off? Yeah, Vanquisher is a real beast of warrior. Congratulation. Fizmo BTW did you receive my e-mail? Yes, I did. and John's too Vanquisher *was* a good warrior. RIP - here lies Vanquisher, that is no more :) Yes I say, and Candy as well. All I'm missing is a glass of beer Hey wait a minute What all of us? crap it's gone :-) Philb who are you? Phil Thorne / waknuk Ah, ok. and smp10k? Ken Espiritu Hi Ken You're Pink Detonation beats all that s/i extremely well Ken disappeared a while back, saying he would return I don't think he's around 4hrs idle. ohh, sad. Hey perhaps Ken is Fizmo :-) 8-) Nooooo Is my english so good, that you can't see the differences??? ;-) Oh don't mention anyone's language capabilities :) I've always been frustrated when I emailed John :) *g* :-( Don't you find it a shame to learn the language for >10 years and still being uncapable of writing a well-made e-mail? :) I *do* :) Well look at me, I can only do English... And it definitely *is* enough nowadays It was very surprised that CrazyShot 2 is KotH on the 94nop. some weeks ago I was kicked off after some challanges Fizmo: Some new warriors entered the hill But it was a mistake I wanted to send it to the new 94 hill, but I forgot to change it in the header LOL Great mistake John does this qualify to the list? :) yes, the hill is now dominated by papers and stone/papers Grabun: Yes List? Aha! :-) Did you become the KotH accidentally? :) Wait for the next CW issue... :) Can't say anything more :) I will look forward something sensational? Something different! An idea for you is top-10 r.g.c postings e.g. most influenetial/ whatever. I'd add http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&threadm=6m1l0t%24hpj%241%40msunews.cl.msu.edu&rnum=2&prev=/groups%3Fq%3DMulti-warrior%2Bscore%2Bformula%2Bgroup:rec.games.corewar%26hl%3Den%26lr%3D%26ie%3DUTF-8%26scoring%3Dd Ha. aka David Moore post "Three ways to improve the hills" Hm Am I online? yes, you are Yes I have recognized, that in the past there are a lot of people from poland are playing corewar. Oh it was very popular game In the middle-nineties there was a contest It's funny I am since long time the only one from germany 50 warriors were sent by 50 different players Could you encourage some Polish players to take part in the Contest. I was 37th Hm WOW. That's a lot I think Lukasz is a little bit ashamed Do you have results from this contest? Stefan Forester? But he's gone. But I will try I am afraid not... But wait! why is he ashamed? I think there's a corewar page (in Polish though) with the results and plenty of warriors from the contest It was 86 standard or ICWS can't remember now Fizmo: don't know... but read his post about his hill he sent few days ago metcalf: I will find it I have an article also about other contest I think there were in the package one could d/l It was funny because author stated that to cease all randomness in the results the number of fights between each pair was set to ten 10 :-) 10 that is the contest was won (AFAIR) by the guy that has some warriors on ICWS'88 hill I'm using over 1000 per pair :-) gisela and all the stuff Oh, I know the ones... I use 2000 fights Bartolik? no no Maciejczak He once has a 30 point lead on the '88 hill I believe gisela is still active on the '88 hill Those giselas, same name different strategies. Those were the days :) I sent a warrior named Riffle 2 kind of bomber... and imp but I was not aware there were so many strategies (no wonder, of Internet I could dream in that time) FYI - Gunnell's hitting 94nop at present I am back in some minutes. Who mails to him to enter the IRC? Well volunteered - plus you're freshest. * grabun already did Me too! LOL Darn, I'm always off the pace... Don't worry, now we're waiting for Janeczek "just testing an idea" ? Oops, hit the top stop :) Is MJ dropping in tonight? Don't know and don't think so His account is in the school he attends to Ah, right. Is anyone logging this? [I'm using chatzilla, can't find a way] * grabun is not, unfortunately Everybody knows the trick to follow the entrance to the hill and to view the 21th place? I am, but someone will need to take over when I go View? Neat trick :-) I didn't but I do now. Handy. exit just klick the first warrior on the hill and rename the url. change the 1 into 21. that's the place for the entering warrior and after the challange for the 21th warrior Hm trying the `script' trick script damn Worked okay for me... I'll make good use of that Typing :21 ?- new to me Hm Is anyone else logging? Not me - Including me there are 7 names on my list, but I didn't know how I can check if they are still present Type "/whois name" that's the first time I do IRC ;-) I can't find the command nothing happen when i type in that command Mine's okay, I just click on a little icon at the top :-) * grabun has pure CLI text mode Ok, I found it on a different window I like Linux but sometimes it *does* upset me I'll try from mirc and log via X I use mIRC, but I am not familar with that. I have installed it shortly before I entered this chat :) I only installed mine yesterday for the sole purpose of coming here. When will be the next meeting. Next sunday? Good idea. What time suits everyone best? Evening ahh, also a newby ? *g* >8.00 PM I can make that Yes, that sounds fine to me Phil? What would it be in the USA? I'll gather MJ and LA I'm 7hrs behind Poland so if you mean 8pm Poland, should be ok or rather try to do this yes, I do mean that Did Steve give any indication he might join? [I'm having trouble finding logging] We should announce to the group then. Include times in several time zones This would be nice maybe jkw too? Being not present means ban on the koth.org LOL No indication no excuses no mercy He did reply though. He did? He's in the most difficult timezone for coordinating. I will also e-mail Roy tomorrow. Oh, yes, he is :) *** Join: steveg (steveg@i203-095.nv.iinet.net.au) Oops Look who's coming Hi steve BTW, has anyone heard something from your "oldtimer" Beppe Bezzi, Paul Kline etc. ? Good evening gentlemen. Hi Steve john how one says when the person the talk is about suddenly shows up? Hi Steve Hi Steve :) Talk of the devil! Oh, you got it Christian, ogre - did you ever "check that idea"? Steve which idea? We have all decided that you must sent to all of us a copy of Hazy Lazy * Your strategy line for Ogre on the -x And tell us more about "antidjn strams" :-) You already have a copy of HL ... I've only fiddled the constants Why am I making so much spelling mistakes :) And then Fizmo reveals Claw's code does he not, Mr Schmidt? :) Ohh, Ogre. It's long time ago. I can't remember exactly what it was. Claw ? The Claw That's really a funny thing. Guess what it is? s/i? Anti-DJN streams are eazy but I haven't found a practical use for them. perhaps in a multi-warrior... Re anti-DJN ... just joking. qscan -> mini-hsa no qscan -> bigger hsa? no ;-) It doesn't get many ties... but you are right it is a qscanner -> scanner but not with HSA it get much more points against s/i and there is enough space for improvements *** Join: pbt (pbt@user-11218vp.dsl.mindspring.com) oneshot -> dclear? it must be qscan -> hazylazy! it is miniQ^3 -> SETI SETI? Seti. Hmmm... I haven't tested that recently searh for extra-terriestrail intelligence? it is a kind of dual carpet laying scanner Ok I have a log going under pbt. [command was /set logilfe fname then /set log on] a little bit like Bishot I usually test every warrior I have against all the published hill warriors every so often. Just to see if anything unexpected does well. great Scanners I do not know well Deep Freeze X was the only one Do you maintain an internal hill as well? But - to be honest - it was a poor copy of HL The 'clear' is funny it has an add instruction with the constant (8001/21). I tried one of those clears in '88 without much success I have still adapted a d-clear but it isn't still ready for sending it to the hill I've not had much success with them either. you can usually ignore 7pt imps. All the more reason to use them. Wintermute, Jedimp and Freight Train do. steveg: you said it. Does anyone else have trouble getting a d-clear to stop imps? Since I've been makig many paper-like warriors these days I do not use d-clears personally 7pt imps are not a problem for d-clear I prefer: spl #0 mov bomb,>gate djn.f -1,{gate bomb:dat >1,10 end lets you use spl #0 bombs with your scanner instead of spl 0 SETI uses spl -1 and it works much better than spl #0 I like these `intelligent' bombs like spl #0, }0 Spl -1 works well with limited size runs - like say agony or medusa. Anyone tried spl -1 with HSA? Yes, but it wasn't very successfull as far as i remember Self colouring bombs work best with a pure scanner ... not needed if you are segueing to a clear BTW, Hi Ben hi with your qscan -> Hazy Lazy you was very close It's Blade ;-) :) Hi ben ... Claw looks more like an f-scanner than a cmp-scanner It sneaks really all kind of oneshots :-) Kudos to Steve for Hazy Lazy *** Signoff: metcalf (Ping timeout) Not my skill ... I'm just the tuner. call it a collaboration if you will. *** metcalf (metcalf@modem-68.dwarf-lion-fish.dialup.pol.co.uk) has joined Welcome back Did I miss much? Well, it seems it's time I went Good-night. I am going to work tomorrow okay ... bye lukaz Me too, at 5.30am good night By you can skip work for corewars! 6.30 am But it's almost midnight here until next sunday until next Sunday, then CU guys bye Lukasz, John I'll be here, see you then *** Signoff: grabun (Client Exiting) Bye bye john Oh, am I going? Will be soon. ;-) ??? - you seemed to suggest that 8-) I like the paper in Wallpaper Thanks Saying good-night to Lukasz... I don't need much sleep. It has a bit of CCPaper and the Head or Tail paper I need to write a good scanner :-( Sorry all the good scanners are taken write a bad one then reverse every instruction Try maybe with SETI. That's at the moment my no. 1 Or a good bomber it has some potential for improvements I admire the dedication of the people who developed SoV. Have they no lives? I think I could have fun tweaking it. That could almost be a contest ... tune this! SoV is really a nice pice of code So is reepicheep. *** Signoff: fordbc (Ping timeout) Stone/clear/imp, wonder how it scores if you get rid of the stone It was fun to write my clear - imps based on Dust dont quite make it onto the hill but they just make my test suite I was in the last weeks working on an improved version of digitalis do tell? I saw Digital Fire Breathing Monster 2002 look to digital firebreatingmonster 2002. I changed the imps back to that one in digitalis. Is there anyone who can explan the maths of the Q^4 scan for Corewarrior? Digitalis scores better against my test suite though (composed of published hill warriors) I understand it, but I would find it difficult to explain adequately. It would be nice to have a qscan history along with it. the new digital firebreathingmonster 2003 is more like digitalis, but boots also the imp-launcher uff, I didn't mach this too I'll have to get digitalis for my zoo. Just in case DFBM 2003 runs amuck.. 8-) I will publish it soon ;-) I tried to make a tiny/clear imp without much success. *** fordbc (x@pcp984954pcs.northw01.in.comcast.net) has joined channel #corewars Hi again. you have a lot of stuff on the new hill ben. Imps a quite weak on the tiny hill. BTW. Thanx for tweaking Tiny Bishot. I now remember Willow and Twinshot are in Koenigstuhl's inbox if you want to grab them to optimize against I was just purifying the hill Twinshot sounds very interesting the hill was set up with a lot of kill-ed warriors, was just clearing them I must send Trefoil to the Koeningstuhl Guys, I'm off soon. Suggest someone post re next week meeting. Also anyone underscore 4.78 on -94nop I'd like to know. Will leave "pbt" to log - and mail it onto John? It is okay, it was a test of concept, so the oneshot I used wasn't that brilliant. I didn't score under that. Okay, if you mail me the log, I will join the two together. bye I'll post a meeting announce on Wednesday, unless someone does it before. bye Phil Cheers. *** Philb has left channel #corewars But you got it on the hill bye I am looking forward to see how it works Yes, it got on the hill. Take a look its in the OPEN Koenigstuhl input http://www.ociw.edu/~birk/COREWAR/OPEN/INPUT/ As is pre75, a .75c prescan. I was quite pleased with the .75c scan (also in Mischief, Clockwork and Clock strikes twelve) If the scanned location is in the a-field, the scan loop automatically moves it to the b-field. It took ages to get that right Mmhhh, if you let run two. Maybe one can add airbag-technologie .... .75c that would be a cmp a jmn and two adds? Yes. And carefully alighted pointers. Sounds somthing like my Aoshi which has two jmns and an add. and swaps pointers between the a and b fields So, guys. I will go to bed. I hope to see you all next sunday. Byee. bye See you next week *** Fizmo has left channel #corewars You use add / jmz 2 / mov.x jmz -3 ? Airbag is a bit dicy with scanners .... keeping the processes synchronised ... brr not quite add jmn 3 jmz -2 mov.x You could have an airbag scanner then one process elsewhere which doesn't slow it down too much. Aha, the same as willow. It is in the Koenigstuhl input box too The .75c scan needs no mov.x :-) yep but you have a gap in the process sequence for a while. Maybe an air-bagged bomber with a seperate scanner to point the bomber? My airbag didn't score too well (a bomber) new idea... airbag bomber or stone + a scanner Did anyone ever write a bomber which directed a clear? Moore suggested it way back like the one in IIRC Freight Train? I once replaced the clear in a scanner with a stone which self bombed to become a clear. Not that great. Stone/scanner sounds good. the bombs/scans can be aligned so you will never pick up your own bombs That '88 airbag Vamp would have done a lot better in round 3 if there weren't so many small stones. yep ... a cmp scanner and a mov mov stone with identical spacing I haven't tried bombing from the scan loop recently, except in oneshots. Must try that. I ought to get going. It is getting pretty late. :-( This from the writer of zooom? Okay ... g'night Thats clearing from the scan loop (well thats how I describe it) I mean: bomb:dat 10,10 loop:mov bomb,@ptr mov bomb,*ptr ptr: sne *100,*120 if you see sort of what I mean? Ahhh ... yeah Must go anyway. See you next week? probably not Im not usually up this early Mon mornings Okay, cheers everyone. Bye bye well ... any thoughts ... I havn't had my first cup of coffee yet so I'm a bit fuzzy. *** Signoff: metcalf (Ping timeout) I'm thinking I'm hungry and its dinnertime :) Okay maybe i'll close down too... so I'm heading out, later bye *** Signoff: fordbc (Leaving) *** Signoff: steveg (Read error: 0 (Undefined error: 0)) *** Signoff: smp10k () IRC log ended Sun Oct 20 18:18