User talk:Jlm0924

Care to introduce yourself? » Nat | Talk » 08:20, 28 April 2009 (UTC)

Hi Justin - I saw you entered a new Melee bot and got afraid, so I went to check its scores and saw a bunch of very low scores from me. It's because my clients are 1.5.4. I'm getting this error:

justin.DemonicRage 1.0: Throwable: java.lang.NoSuchMethodError: justin.Module.getGraphics()Ljava/awt/Graphics2D;
java.lang.NoSuchMethodError: justin.Module.getGraphics()Ljava/awt/Graphics2D;
    at justin.movement.MovSim.futurePos(MovSim.java:70)
    at justin.movement.MovSim.futurePos(MovSim.java:43)
    at justin.movement.MovSim.futurePos(MovSim.java:37)
    at justin.movement.Hawk.move(Hawk.java:136)
    at justin.Module.executeBehavior(Module.java:96)
    at justin.Module.run(Module.java:83)
    at robocode.peer.RobotPeer.run(Unknown Source)
    at java.lang.Thread.run(Thread.java:619)

I should just update my clients, as this happened with another bot recently. But as of now, 1.5.4 is still allowed for Darkcanuck/RRServer, so there's nothing stopping others from running it and hitting this problem. I've stopped my clients for now and will update them, so if you want to re-post as 1.0b, you can get the "real" scores for it... Sorry about that. Also, I want to know about your bots - post some details! =) --Voidious 22:38, 25 May 2009 (UTC)

Also, it looks like you must have the wrong participants URL in your MeleeRumble client. I'm seeing a bunch of battles for 1v1-only bots in the MeleeRumble that came from your client... I've updated my clients and tested DemonicRage, so whenever your re-post it, you'll have my clients contributing battles. --Voidious 00:33, 26 May 2009 (UTC)

Hey, haven't look at the source yet, but is this robot use FuturePosition library from Abert and HawkOnFire movement from rozu? » Nat | Talk » 03:20, 26 May 2009 (UTC)

Hey Nat, I've been attempting to use Alberts prediction to add wave surfing results into a minimum risk type movement. (unsuccessfully so far) I'm hoping robocode vr 1.7.2 bug fixes help. hawkOnFire's movement was a influence and reference when making Demonics initial movement. Though Demonics movement has grown 10 times the size of hawks entire bot (at least) Hawk is a strong Melee bot. I wouldn't be surprised if Hawk out moves Demonic with head on targeting :) -justin

Thanks Voidious, I'll make sure 1.1 doesnt have same issue.. -justin

Signature

Please use --~~~~ to sign your posts, it will automatically create a link to your userpage with nice timestamps. You can configure your signature in your preferences page. Thank you. » Nat | Talk » 06:24, 30 May 2009 (UTC)

Rumble package

Hey dude - I'm curious why you're sticking with "DemonicRage" as your top level package name? The norm is definitely to have all your bots under the same package, for the sake of clarity and rumble flags and such. Some people do use separate packages for bots (abc.tron3.Tron, wcsv.PowerHouse.Powerhouse, etc.), but not at the top level. --Voidious 14:12, 13 August 2009 (UTC)

Hey Voidious :) I've been meaning to remove the extra package :) I guess i rushed into poking around with demonic again with a different computer and haven't got to it yet.. Congradulations by the way on Diamond ! :) He's kicking butt! :) I got a couple questions for yah...I need to find a better way to test... lately I've made some changes and to test I just through him back in the the rumble.....( I hope no-one minds, and it doesn't sway the stats...) I would like to test at 35 rounds with 10 bots (representitive of the field) multiple times to average out variance and get an Idea if my changes are for the better .. Is there a way to do that ? .. How do you test Diamond ?

 -Jlm0924 21:52, 14 August 2009 (UTC)Justin

Yeah, I understand your testing dilemma. I felt quite like that when I started with Diamond, since I was very used to 1v1, where we have lots of challenges to benchmark with and RoboResearch support. You could try instead running 1,000 round battles, though obviously that does not seem as reflective of the Rumble. I use a modified version of RoboResearch to run batch battles, which I posted at Talk:RoboResearch#Melee (along with instructions / caveats) if you want to grab it. (Simonton and I are discussing getting it merged into the main code branch, but it isn't yet.)

I also just uploaded the .rrc file (defines a specific test bed) and .cfg file (specifies bot, test bed, and number of seasons) that I'm currently using, which might help as examples for using RoboResearch. The test bed includes each bot twice, in different groups. I actually package each test version into a .jar for RoboResearch, that's why the version is like 1.27x.5. It can be a little confusing to setup, but RoboResearch is an amazing tool once you get it going, so I'd highly recommend it. Good luck!

--Voidious 21:58, 14 August 2009 (UTC)

Wow !! sounds great !! thx Voidious , I'll give it a go :) -justin -Jlm0924 23:11, 14 August 2009 (UTC)

Robocode Installation

OK, OK, before I go nuts... I'm trying to install robocode on a friends computer...though nothing is installing... it only creates an empty directory and shortcuts... why is this???? (it's not the robocode install jar)..???? -Jlm0924 16:28, 18 August 2009 (UTC)

If it isn't robocode installer, which do you used? » Nat | Talk » 16:36, 18 August 2009 (UTC)

I tried robocode 1.7.1.3 setup.jar and the two previous versions -Jlm0924 18:18, 18 August 2009 (UTC)

I'm not sure how much configuring the installer does. I know at least once, in the past, I just unzipped the .jar (and maybe another .jar within that) and used that as a Robocode directory. I don't think it has the roborumble files in the right place, but for just running Robocode, it might work... What OS is it? Windows? If it's some kind of Unix, I'd check the permissions on the directory you're trying to install to (though if it can create a dir, that seems unlikely to be the problem). --Voidious 18:24, 18 August 2009 (UTC)

Yeah, as Voidious said, actually you can just extract .jar two times. It is all in right places (at least for 1.6.1.4, haven't tried this method with 1.7 yet) You can't run Robocode with installed shortcut, right? » Nat | Talk » 18:29, 18 August 2009 (UTC)

yep, I had to use unZip the setup jar file ,,, and I had to go back into the file a second time to grab the libs files, for what ever reason they did not extract as the other files did.. back in business,, thx guys :P -Jlm0924 21:07, 18 August 2009 (UTC)

DemonicRage v2.0

I spent alot of time lately working on Demonics movement. In the process I lost the source to some older versions like 1.4 while tring to re-invent the wheel. but I'm happy with the out come, and think it has alot of potiental. I noticed after posting that it got booted from a rumble match while fighting with shadow and sandBox for skipping turns. I will have to fix. --Jlm0924 06:12, 8 September 2009 (UTC)

v2.0b, is funtionally the same, the disabled wave surfing / tracking processing was commented out to try and speed it up a bit.. if you notice skipped turns I'd appreiciate the heads up.. (Dame slow bot :) I guess I should do everyOne a favour and learn kd trees. Any 'simple' code examples of them implemented to start learning from ??

This has a nice applet on k-d tree and quadtree. Try looking at Bucket PR k-d tree as it is faster than original kd-tree, also it has been used by this community =) Simple implementation... Voidious' kd-tree probably most readable 'simple' kd-tree code on this wiki but the Simonton's tree is probably more easier to understand, but with confusing code style. This tree was where I (and probably Voidious) used to learn from. Don't try Rednaxela's kd-tree if you are looking at anything simple =) PS. Could you please at least write about yourself and your robot? » Nat | Talk » 12:59, 8 September 2009 (UTC)

Those applets are nice, but what I found most helpful when I was first learning was this. Also, pfft, this older revision of my tree is in my opinion the very most readable one out there Nat. PS, I agre with Nat, it would be nice if you make a page for your bot some time :) --Rednaxela 13:22, 8 September 2009 (UTC)

For what it's worth, the only thing I used for reference was the Wikipedia entry. =) Keep in mind that the tree won't magically solve your skipped turn issues. In fact, optimizing a brute force method might give better performance. Where the trees excel are in guns in longer 1v1 matches - you only get ~5,000 gun scans of each enemy in a 35-round Melee match (unless you start creating lots of extra gun waves, like I did recently...), and far less on the Wave Surfing side. Then again, Rednaxela's tree is really fast. Anyway, good luck. And yes, tell us about DemonicRage, already. =) --Voidious 13:40, 8 September 2009 (UTC)