Robo Home | Changes | Preferences | AllPages

To be filled in shortly.
This is a fluke, but such a cool looking one:
Round 43 of 100
SYSTEM: Bonus for killing sample.Target: 15
SYSTEM: Bonus for killing sample.RamFire: 5
SYSTEM: Bonus for killing sample.Corners: 18
SYSTEM: Bonus for killing sample.SittingDuck: 3
SYSTEM: Ram bonus for killing sample.MyFirstRobot: 19
SYSTEM: Bonus for killing sample.Crazy: 10
SYSTEM: Bonus for killing sample.TrackFire: 0
SYSTEM: Bonus for killing sample.SpinBot: 19
SYSTEM: Bonus for killing sample.Walls: 6
SYSTEM: arthord.NanoSatanMelee wins the round.

I smell bug in Aristocles: Here's an excerpt from NanoSatanMelee's details, he uses a direct gun:

Opponent% of ScoreNum BattlesLast BattleExpected %PBI

Have I accidentally made a direct gun that bypasses MusashiTrick(somehow), I doubt it, so is this a bug?

-- Kuuran

You know how I always say I don't use the MusashiTrick. =) But I know what you mean with the term so this is interesting. Very interesting. I must run a few battles and see what is going on. It's not like it is the first time the little satan of yours bugs my bots! -- PEZ

Aha! It's because I don't use the MusashiTrick that this happens! You have created a gun/movement combo that fools Aristocles to believe it's never hit at GF1. Jim has experimented some with this. Your melee movement keeps you close to walls right? Aristocles thinks that as long as it is running in the same orbit direction all the time any hit it takes is at GF1. UNLESS it is wall smoothing more than just a little. NSM hugs the same wall as Aristocles is running along to pass it. Forcing Aristocles to wall smooth hard. Now it takes hit after hit from NSM without realizing those hits could be avoided. On the other hand the only way A would know how to avoid them is to abandon GF1 movement and then your head-on gun would be effective anyway. Should I have had room for the MusashiTrick in my bot I could bounce of walls instead, safe in that the trick would know if a hit isn't at GF1. See why I make such a big thing about what is and what is not the MusashiTrick? Anyway. I think I could solve this particular situation. Shouldn't cost too many bytes. I just have to find them. A doesn't have any spare ones... -- PEZ

What if you just limited the angle with which you wall-smoothed (i.e.- no fear of the walls unless you should be afraid)? -- Kawigi

I left for class right after writing that and suspected something like this might be the case on my way out the door, but didn't have time to post it. Nice to hear actual testing agreed with my suspicion (for a change ;). I guess you'll be glad to hear I will probably abandon this movement in the next version (other than killing a handful of highly ranked bots it SUCKS) ;) It's also kind of nice to be the thorn in the side again, even if only for a little while :)

(edit) There seems to be quite a bit of variance in the score, though. First test I ran gave this...

 1st: arthord.NanoSatanMelee Alpha	5413	1450	290	2642	428	457	146	31	5	0
 2nd: pez.micro.Aristocles 0.3.7	2403	250	50	1460	81	561	0	6	29	0

Aristocles also tossed...

 pez.micro.Aristocles 0.3.7: Exception: java.lang.ArrayIndexOutOfBoundsException?: 5
 java.lang.ArrayIndexOutOfBoundsException?: 5
     at pez.micro.Aristocles.onScannedRobot(Aristocles.java:94)
     at robocode.peer.robot.EventManager?.onScannedRobot(EventManager?.java:607)
     at robocode.peer.robot.EventManager?.processEvents(EventManager?.java:738)
     at robocode.peer.RobotPeer?.tick(RobotPeer?.java:1024)
     at robocode.peer.RobotPeer?.turnRadar(RobotPeer?.java:1086)
     at robocode._AdvancedRadiansRobot?.turnRadarRightRadians?(_AdvancedRadiansRobot?.java:271)
     at pez.micro.Aristocles.run(Aristocles.java:47)
     at robocode.peer.RobotPeer?.run(RobotPeer?.java:616)
     at java.lang.Thread.run(Thread.java:536)

7 times in that match.

(edit2)Argh.. Disregard that.. I left the battlefield at 1000x1000... Going to sleep.. -- Kuuran

Robo Home | Changes | Preferences | AllPages
Edit text of this page | View other revisions
Last edited May 18, 2006 22:22 EST by GrubbmGait (diff)