Robo Home | Changes | Preferences | AllPages

1v5, Try to get the highest score percentage against the sample team.

My rule ideas: 1000x1000 Field Standard sampleteam vs your entry 500 Round battle

I think that this will do alot for target prioritization and probably some BombSheltering.

Bot Name Author Score Bot Score Sampleteam Score perc 1st perc
Some experiment Jonathan 366810 79261 82.231 80.6
BlindDroid? Dummy 258308 64269 80.076 67.0
SandboxLump 1.52Paul Evans 317742 88304 78.253 72.6
Walls Mat 234058 70436 76.868 70.8
SimpleFire? 0.1nfwu 233345 71872 76.452 70.2 Extends Robot!
FleetingWisps? Kinsen 247708 110626 69.128 36.0
Tron 2.02 ABC 261916 133729 66.199 48.0
Tron 3.11 ABC 261260 135049 65.923 47.2
Shadow 3.59.9bABC 261152 139841 65.126 45.2
Walls Mat Nelson 197328 106450 64.958 54.8
DoBoh Krabb 185649 147172 55.780 27.4
GrubbmGrb 1.1.0 GrubbmGait 183732 158314 53.716 27.4
Gruwel 0.2 GrubbmGait 178692 165284 51.949 9.4
Akim 1.12 Krabb 108035 200476 35.018 3.7

Just entered Gruwel to set the first horde. I just wish I could hit that leader. Frankly any adequate meleebot should be able to win against sampleteam. --GrubbmGait

I guess Tron's movement is still the best for this kind of battle. Maybe I'll do some experiments with Shadow's melee movement, when I get the time... -- ABC

This is stupid, my OneOnOne bot does better than my meleebot, because it can hit the leader sometimes. --GrubbmGait

Well, I thought I'd point out how an old melee bots do this pretty well (SandboxLump and Walls). I disagree, however, that any adequate meleebot should be able to win against sampleteam, I'd call FloodHT, Coriantumr and Shiz adequate in melee, and they get their butts handed to them by the sample team. I think it'd be more accurate to say that a good melee bot that has good head-on avoidance of bots that probably wouldn't be targeting it in a real melee battle should be able to win. -- Kawigi

What is the 1st percent column? -- Kinsen

1st place percent :) -- Krabb

Thanks, I'm running it right now and I should have it ready by the time I have to go. -- Kinsen

Seems that SandboxLump outperformed everything :D This just shows, how irrelevant this challenge is.--lRem

I've found the current winner: sample.Walls. Is it fun anymore, or should I tweak it to have 90% score? ;] --lRem

Two times Walls in the list? What about Dummy's BlindFighter?, that should do it quite well too. --GrubbmGait

Someone must have run it on smaller battlefield. I've tried it 3 times and it never got under 70%. --lRem

Ah, yes, I might have accidentally run SandboxLump and Walls on 800x600. Walls has already been fixed, I ran SandboxLump again on the right size of battlefield. Best survival yet! -- Kawigi

I'm not planning to post that experiment, but trust me, it did it. Maybe I'll post something when it's cleaned up. -- Jonathan

Robo Home | Changes | Preferences | AllPages
Edit text of this page | View other revisions
Last edited December 19, 2005 10:55 EST by Nfwu (diff)