Robo Home | Changes | Preferences | AllPages

Difference (from prior major revision) (no other diffs)

Added: 58a59
** If you don't care for out-of-bounds checks for every step, you can do it without trig functions in the inner loop. You project your own location in relation to the enemy's heading and iterate using the enemies displacements (+dx +dy), then you convert the angle back to get the firing solution. That's how I do it ATM, it's pretty fast. -- ABC



Robot Author ScoreSurvival 1sts BDam / round
Shadow 3.71 ABC 592204 497 454.498
Reach 1 Tim Foden 586018 486 453.894
Gruweltje 0.6 GrubbmGait 583520 482 440.300
Gruwel 0.9 GrubbmGait 570668 482 422.004
Phoenix 0.41 David Alves 561749 486 408.21
Slartibartfass 0.5 Krabb 536697 488 368.22
WeekendObsession Melee 1.0 Simonton 525853 430 369.32
Griezel 0.4 GrubbmGait 52250 476
Freya 0.82 Loki 500590 316
Smash Starrynte 476620 366 314.232
WallsPoetHaiku PEZ 447007 223
Girl Kawigi 416375 301

Extends Robot:

Robot Author ScoreSurvival 1sts
WallsPoetHaiku PEZ 447007 223
Girl Kawigi 416375 301

Thought there needed to be something to start this out, and I'm not too surprised to see PEZ outdo me just slightly here. I am surprised that he did it with quite a bit fewer firsts. -- Kawigi

408.21 average bullet damage inflicted / round. --David Alves

Edit: whoops, didn't read this completely. Splitting it into two categories.

As I expected, circular targeting and staying relatively close pays off against the samplebots. The result is a yet undisputed 422.004 BDam/round. -- GrubbmGait

I don't think that's bad for a nano. -- Simonton

I think the result of Reach is proof that this challenge doesn't closely rate how good a bot is overall, as it would get slaughtered by most of the other bots here. :) -- Tim Foden 28Jun07

None of the challenges on this site are ever the "whole picture", but they're still useful and interesting for measuring a certain aspect of your bot. Whatever the case, that's a really impressive score, nice work! You gonna enter it in the MeleeRumble?? Oh yeah, and welcome to the wiki. =) -- Voidious

Yes, I agree. I've tried some of the other challenges too, and so far Reach is pretty poor. Now I've figured out how to add stuff to the wiki, I may post my results in them. I added Reach to the RoboRumble and MeleeRumble? just before I noticed your post. It'll be interesting to see how it does. -- Tim Foden 28Jun07

Someone can remove the older Shadow if they want, I would think. Anyway, watching Shadow in this challenge let me realize why using a play-it-forward instead of a GuessFactor technique for aiming would be superior in Melee. The revelation occurred when I say him hit Walls and SpinBot with perfect accuracy. Lots of simple Melee bots run little patterns, which play-it-forward inherently predicts better. Looks like whenever I take DynamicClustering to the Melee field, I'll be following Shadow's lead on that one. -- Simonton

Thanks for running Shadow 3.71 in this challenge, I forgot that the changes I did to my base gun class would influence the melee gun. The next version will be all about melee gun changes. I'm testing something really cool, here is my latest test result: Score: 607630, Survival: 500, BDamage/round: 478.44. :) -- ABC

Robo Home | Changes | Preferences | AllPages
Edit text of this page | View other revisions
Last edited December 4, 2007 15:13 EST by ABC (diff)