From corewar-l@stormking.com Sat Jun 8 22:13:22 1996 Received: by couchey.inria.fr (5.57/Ultrix3.0-C) id AA04474; Sat, 8 Jun 96 22:13:22 +0200 Received: from uu3.psi.com (uu3.psi.com [38.145.250.2]) by nez-perce.inria.fr (8.7.1/8.7.1) with SMTP id WAA21369 for ; Sat, 8 Jun 1996 22:13:14 +0200 (MET DST) Received: by uu3.psi.com (5.65b/4.0.940727-PSI/PSINet) via UUCP; id AA17319 for ; Sat, 8 Jun 96 16:05:48 -0400 Received: from (server@localhost [127.0.0.1]) by valhalla.stormking.com (8.6.12/8.6.12) with SMTP id PAA27122; Sat, 8 Jun 1996 15:52:30 -0400 Date: Sat, 8 Jun 1996 15:52:30 -0400 Message-Id: <4pbldc$a04@sparcserver.lrz-muenchen.de> Errors-To: tuc@stormking.com Reply-To: guenzel@extern.lrz-muenchen.de Originator: corewar-l@stormking.com Sender: corewar-l@stormking.com Precedence: bulk From: guenzel@extern.lrz-muenchen.de (Bjoern Guenzel) To: Multiple recipients of list Subject: Re: Help improving warrior X-Listprocessor-Version: 6.0b -- ListProcessor by Anastasios Kotsikonas X-Comment: Usenet News "rec.games.corewar" rmorganl@fred (Ross Morgan-Linial) wrote: >Bjoern Guenzel (guenzel@extern.lrz-muenchen.de) wrote: >: further hints... >: try to make the b-field of the bomb a 1, it's a much better attack on >: clears (and almost every warrior contains a clear...) - dat >2667,>1. >: I think your step is mod 10, have you tried a smaller one? I think >: with your current warrior you could achieve mod 4. Unfortunately, if >: you make it self-splitting it becomes a bit more complicated to switch >: to the clear. I have no really good idea for that, I don't know if the >: following possibilities work good: >Someone sent me a *MUCH* better coreclear, so I replaced the old one. What is it like? The problem I mentioned is not the Core Clear, but how to get to the Core Clear (at least for me it's a problem, I am always happy about new ideas). >Actually, my step size is 5. It looks like 10 because I have two step-10 >bombing runs offset by 5... but it hits every 5th location. It's a simple >optimization. Yes, sorry, I discovered that in the meantime, too :-) >: c) some complicated self-modification (I don't know if it is possible, >: you would have to change the bomb to spl sometime in your run, perhaps >: by bombing a pointer or something...) In the meantime I made up an example for variant (c), because I always wanted to test a bomber with imp-bombs. It doesn't score good (my guess, couldn't reach pizza yet), but perhaps if I replace the bomb by dat >1,>1 or mov.i #1,2 ,it's ok (guess I'll have to wait till monday :-( ) Will pizza be down every weekend now? I thought the rewiring was only last weekend? Planar is having a lot of work with me these days... :-) (I hope it's ok...) Bjoern ;redcode-94 ;name Imperator ;author bjoern guenzel ;strategy bomber with imp bombs ;assert CORESIZE==8000 ;p-class: stone,clear ;release 7.6.96 ;kill Imperator shift equ -3000 step equ (10*21) gate dat 0,0 for 8 dat 0,0 rof bptr dat imp,1 ;bomb here to change to spl-bomb steps spl #step,>-step ptr mov.i hit+step,bptr-5*step mov.i *bptr,*ptr add.f steps,ptr hit jmp.b ptr,>gate ;bomb here to change to clear mov.i clbomb,gate imp mov.i #1,1 ;used as bomb clbomb dat >1,>gate boot mov.i clbomb,*booptr for 9 mov.i {boot,{booptr rof booptr spl @imp+1+shift,>1000 mov.i #0,-1 dat >2000,>1500 z for MAXLENGTH-CURLINE spl #z*103,z*(-101) rof end boot