From news-rocq.inria.fr!jussieu.fr!math.ohio-state.edu!howland.reston.ans.net!newsfeed.internetmci.com!in1.uu.net!spcuna!news.stormking.com!d91andiv@und.ida.liu.se Mon Dec 4 10:55:33 1995 Article: 3182 of rec.games.corewar Path: news-rocq.inria.fr!jussieu.fr!math.ohio-state.edu!howland.reston.ans.net!newsfeed.internetmci.com!in1.uu.net!spcuna!news.stormking.com!d91andiv@und.ida.liu.se From: Anders Ivner Newsgroups: rec.games.corewar Subject: Pspace - worth dying for? (Round 7) Date: 2 Dec 1995 07:19:24 -0500 Organization: Storm King Ind. Inc. Lines: 88 Sender: server@news.stormking.com Distribution: world Message-ID: <9512021214.AA21704@su9-8.ida.liu.se> Reply-To: d91andiv@und.ida.liu.se NNTP-Posting-Host: valhalla.stormking.com Originator: corewar-l@stormking.com Well, this is my entry for round 7. It contains a strategy that seems, at first, to be madness. Let me take a few moments to explain the reasoning behind it. 1. Noone will submit a non-pspace program. Such a program would behave the same in all rounds, and lose BIG. Pspace manipulation is usually done directy at startup. 2. The area 0-500 is a dangerous place to be, so most people will boot away to the relative safety outside this area. 1 and 2 implies that most warriors in this round will have a section of bootup code of length at least 6 instrucitions. Probably more. If I can find this code and bomb it first thing every round, I will win the remaining rounds. So my first three instructions drops two bombs in the area 4000-4500. The bombs are JMP's to a brainwashing pit. If I catch something, the same positions will be bombed again the next round. If not, I step up my pointer in pspace. What should I do next? If I caught something I just wait for the opponent to suicide, but If I didn't? My answer is: At all cost, don't get brainwashed! That would reset my bombing position, rendering my initial attack useless. There is (to my knowledge) only one foolproof way to avoid being brainwashed, and ensure that my pspace is intact at the next round. It is (and here comes the part that, even to me, seems like madness): DAT 0,0 Yes, that's right! My program will valiantly kill itself to protect its secrets! Give up its life for the good of the cause! Rather die than be captured by the foul enemy! [..3 pages of romantic metaphors deleted to save bandwith :-) ] So, is my reasoning sound, or have I gone completely out of my mind? /Anders ;redcode-94 ;name Victory or death ;author Anders Ivner ;strategy Catch opponent at startup by throwing two jumps to a brainwashing ;strategy pit at him. If he is not caught, change target positions, and ;strategy suicide to protect pspace from brainwash. ;strategy If opponent has a startup longer than 6 instructions, I should ;strategy be guaranteed to catch him before round 50. ;strategy I have no idea if this will work...nothing to test against. VSTEP equ 74 s i for 145 mov <-i,500 jmp -3, >-2 bait1 jmp @4000-250, -250 bait2 jmp @4000 b dat 0,1 start ldp #250, baitpt mov bait1,