Path: ibmpcug!wendy.ibmpcug.co.uk!pipex!news.sprintlink.net!howland.reston.ans.net!gatech!swrinde!emory!nntp.msstate.edu!nntp.memphis.edu!nntp.rhodes.edu!harlie.mathcs.rhodes.edu!graham Newsgroups: rec.games.corewar Subject: Comp-Hunter Message-ID: <1995May8.121310.2710@rhodes> From: graham@harlie.mathcs.rhodes.edu (Randy Graham) Date: 8 May 95 12:13:09 -0500 Nntp-Posting-Host: harlie.mathcs.rhodes.edu X-Newsreader: TIN [version 1.2 PL2] Lines: 46 OK, here is my last shot at Comp-Hunter. I have played with step sizes, bomb runs, this, that, and the other. I don't think I can take him any further in his current form. I am now going on to try to get a good paper on the hill. Anyway, except for trying to get on the big hill, maybe with a QScan, and trying to add in a double scan like Taking Names, I believe I am giving up on Comp-Hunter. He's been good to work on, and I've learned a lot about corewar from developing him from his first days as a b-scanner in '88 too and f-scanner in '94 to a cmp-scanner. There actually are a couple of other changes that could now be made, but I'm no going to make any more of them I think. So, anyone wanting to see a competent (but not outstanding) scanner, here he is: Randy ----------------------------- ;redcode-94 ;Kill Comp-Hunter v2.0 ;Name Comp-Hunter v2.2 ;Author Randy W. Graham ;assert 1 ;strategy cmp-scanning carpet bomber based on Rave. ;strategy v2.0 Better djn-stream. Changed spltrap bomb. ;strategy v2.2 Try a new imp-gate/step size. OPT equ 66 ;mod 2 LEN equ (fini-adder) DIFF equ 12 BOMBS equ (DIFF+2) ;how many bombs to lay OFF equ -402 ;djn stream offset for long run ;for OPT 7934 BEST equ -402 ;best I've found so far in length of run ENDAT equ 78164 ;how long BEST runs adder add.f spltrp, cmper ;adjust our compare sites cmper cmp.i adder-DIFF-OPT, adder-OPT ;see if they differ slt.ab #LEN+BOMBS, cmper ;different, only bomb if not us djn.i adder,