From news-rocq.inria.fr!univ-lyon1.fr!jussieu.fr!math.ohio-state.edu!uwm.edu!chi-news.cic.net!newsfeed.internetmci.com!in2.uu.net!spcuna!news.stormking.com!morrell@math.utah.edu Mon Dec 11 13:02:25 1995 Article: 3268 of rec.games.corewar Path: news-rocq.inria.fr!univ-lyon1.fr!jussieu.fr!math.ohio-state.edu!uwm.edu!chi-news.cic.net!newsfeed.internetmci.com!in2.uu.net!spcuna!news.stormking.com!morrell@math.utah.edu From: "Steven C. Morrell" Newsgroups: rec.games.corewar Subject: This is even worse than my other post Date: 9 Dec 1995 03:03:03 -0500 Organization: Storm King Ind. Inc. Lines: 176 Sender: server@news.stormking.com Distribution: world Message-ID: Reply-To: morrell@math.utah.edu NNTP-Posting-Host: valhalla.stormking.com Originator: corewar-l@stormking.com XXXXX XXXXX XXXXX X X XXXXX X X X XXXXX XXXXX XXXXX XXXXX XXXXX X X X X X X XX X X X X X X X X X X X X X X X X X X X X X X X X X XXXXX X XXXXX X X XXXXX X XX XXXXX XXXXX XXXXX X XXX XXXXX X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X XXXXX X XXXXX X XXXXX X X XXXXX X X X XXXXX XXXXX A pathetic attempt to give Corewar fans something to read on a weekend when no CoreWarrior is being published. Dedicated to decreasing the signal-to- noise ratio of the group. Steven's Page is published whenever I feel like it. If you complain about this one, I may write another. You know how I am at writing regular features! For information about the ASCII art used in the title, contact me, and I will cry. --- Well, since they said stop lurking... Redcode is obfuse. Assembly language in general is not very enlightening, but when all memory is addressed relatively, things get weird. It's fun figuring out all the details of a warrior sometimes, but it has it's limits. My sorting program from round 4 is a perfect example of what can happen if a warrior gets too complex. It really is a simple quicksort stolen from Kernighan and Ritchie's C reference version 2 (near-ansi). But actaully seeing this from the code is ridiculous. People often say that redcode would be a lot better if it weren't for those dumb bombers keeping the smart programs off the hill, but I think this is the only deterrent from code bloat, and beleive it or not, I don't have much fun reading more than a couple of pages of redcode to find out how a program works. I'm not looking for intelligence, just cleverness. This is one big reason I am against storing complete instructions in P-space. It is a stupid place for code to be between rounds, considering I can't even touch the code pMARS uses to load your warrior. And the only other argument that has been given for it is that it could make stronger, deadlier paper. I shudder to think of what Die Hard would transform into. --- The pizza hill is furiously churning. We keep pushing off warriors just before they reach 200, so nobody can get on the What's Old List. Meanwhile, the three old-timers keep advancing, and at the request of B-Panama X, I would like you to push them off, too. The beginners hill is empty, thanks to the recent request that everyone get their excess warriors off. If it weren't, wouldn't I be getting mail? Does Frontwards still think it's 20th on the Hall of Fame? I was suprised to see it on, because it was bumped off the hill before I received the previous CoreWarrior. It did quite well, considering it entered at 19th place. Oh, and don't forget to look at my other post for the source to Frontwards v2. In coresize 55440. Which makes it useless to me. Until someone can make a 55440-cell visual display. Preferably for a Mac Classic. (Okay, maybe not.) --- Anyhow, for the mathematicians, I think I can prove that optima numbers come in imp-pairs. For instance, in coresize 8000, 3359 and 3039 are optima numbers, and a 3359-point imp (Gee, I haven't seen many of those lately!) would need a step size of 3039. --- There has been a lot of talk about symmetrical access to P-space. This is not a good idea if you ask me because it disguises the differences, the most important being absolute vs. relative addressing. Symmetry in syntax should reflect symmetry in semantics. Increasing access to P-space will not do what you hope. My code can access your P-space, simply by having your processes run it. Increased access would make my job of destroying your P-space MUCH easier. The only proposal to this end that I would prefer is eliminating STP and LDP and using addressing modes [x] P-space indirect, -[x] pre-whatever P-space indirect, and [x]- post-whatever P-space indirect. Think about it. :) --- Now that 25 slots are implemented on the pizza hill, I think displaying all warrior results against each other -- upon request -- would be a cool idea. I didn't like it before because I thought it would shorten a warrior's lifespan. But it only shortens the lifespan of the average warrior by 5, so it would be a perfect counterbalance. And while I'm at it, do assertions like that seem more plausible when they are published in a newsletter? --- Actually, I am sure I would like P-space a lot more if I weren't still running v6.3 on my computer. But, since I am too cheap to buy a real computer, I suppose I shall have to continue opposing it. I think my biggest problem is that P-space pushed my stupid bomber off the hill, and I have already discussed the merits of that... --- The math department changed news servers, and the new one doesn't carry rgc. Now why would a server have the entire alt.binaries hierarchy and not have corewars?! Is it because corewars comes from the same people who invented computer viruses? Is it because Brant annoyed someone? Or is it just that I have a 2400-baud modem? On a related note, it seems that RARS has fewer participants than it did when it was in comp.programming.contests. --- We can't get channel 4 clearly on our TV unless the modem is off-hook with a carrier. Does anyone know how to fix this? (Okay, I know this has nothing to do with Corewar, but if you've read this far, you don't have very discriminating tastes.) --- Here is some useful code: ;redcode-94 quiet ;name Scorekeeper ;author Steven Morrell ;strategy P-space marvel ldp #0,#0 jmz -2,-1 nop 0,0 end To my knowledge, it is the only warrior to date that scores perfectly on the Pizza Hill. But, that was a couple of months ago. I wonder how it would fare now? --- And speaking of interesting code, did anyone notice the warrior that was submitted to the '88 hill a couple of months ago that scored 50/50/0? I would really like to know how to do that! --- Nandor's bogus Tournament Rules were very funny. Ha ha. I showed my wife, the artist, that posting, and she said, "Gee, that looks sort of like math." I can get a little more accurate, in that it looks very Cohomological or K-Theoryish. Should I have covered that in my Algebraic Topology class? --- Well, that about wraps up this issue. Next time I catch the flu twice in one month and stop eating solid foods for a week, I may write another one. 99% of it is just your state of mind. (When I woke up last Monday, I was very upset that someone had cut a large square hole out of the bedroom wall. A few minutes later, I discovered that it was just a mirror.) --- Steven Morrell morrell@math.utah.edu --- Finally, a newsletter all about ME!!!