This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.
Messages - philetbabe
Pages: 1 ... 6 7 8 9 10 11 12 [13] 14 15 16 17 18 19 20 ... 26
241
« on: April 28, 2009, 01:51:58 PM »
Nobody has any comments at all??? :S That's odd we are voiceless in front of such a monster
242
« on: April 28, 2009, 05:25:02 AM »
Bushido's Beyatch: Someone doesn't knows to build even a decent HS... Second SFTW though Hydro's is worse.
Cube Tank: Pathetic SFTW. That is all.
[
You're a hard boy Nar (not that you look like Rocco Siffredi, i dunno), but what a mercy killer you do !
Big Bad Brother: Ungrippy wheel fail. It IS realistic
Yes indeed, i thought the disks were spinning, but looking at the showcase, they do not -> so it is reallistic -> good base for future bots.
243
« on: April 28, 2009, 02:28:00 AM »
my forecasts :
Match 1 : 5 vs 12 <-> SnSnSnS vs lol another ... : SnSnSnS should be a whipper, so lol will be hit on its side again and again until a hard death comes.
Match 2 : 6 vs 11 <-> Cube vs Cubinator : i thing Cubinator will loose its wheel giving a good chance to Cube.
Match 3 : 7 <-> 10 : Awful red thing From Outer Space (ie fos) vs Smash Down : i know awful (i've made it) : it is a very surprising bot, winning against killer bot and loosing against poor ones : so there ? i dunno, hope a win by points. Smashdown is well protected but lack of weapons.
Match 4 : 8 vs 9 : Xenophobic vs Grey matter : Grey matter should win, but i like Xenopobic, even if i am not sure this is THE good way to use those unloved sawblade. Xenophobic should have been a potential winner if it had a wedge.
Clash Cube winner : I think SnsSnSnS.. looks scary.
Best design : I like Big bad Brother, but i don't know if it is really reallistic, Naryar jobs is always well finished too.
244
« on: April 27, 2009, 09:05:46 AM »
With less components, Stock is easier to begin with, but soon, we get bored with the native bots or discouraged with the too hard bots of Starcore.
RAGE represents a very good alternative for middle level bot builder (I highly recommand it !)
DSL impresses first with its huge amount of components, but it is really not for beginner who feels quiclky lost. Good builders may feel also a little annoying with the native bots : Thanks to bbean delivery (and the futur Naryar pack ?) there is still challenging bots.
So i should vote for Stock if i was a beginner or a genious builder, and DSL as a good-but-nos-so-good builder... i did vote DSL
245
« on: April 27, 2009, 05:36:19 AM »
well, as you asked me, i also tried... and i can't : as soon as you go straight ... you bot stop spinning !
if you want this kind of bots you will have to do it with DSL and it is more a matter of components than of AI.
Anyway, i share you the result of my works : quiet unuseful : a tactics + AI that work (worst) as a Whipper but without whip zone !
Tactics (to add in Tacticts.py) :
class SnSV1(AI.Tactic):
name = "SnSV1"
# Tactic based on Charge
def __init__(self, ai):
AI.Tactic.__init__(self, ai)
self.regroupPos = None
self.regroupDir = True
self.regroupTime = 0
def Evaluate(self):
self.priority = 0
self.target_id, range = self.ai.GetNearestEnemy()
if self.target_id != None:
heading = self.ai.GetHeadingToID(self.target_id, False)
# better turn than move : priority higher if no heading
self.priority = 50 + (abs(heading) * 20)
clear_path = self.ai.IsStraightPathClear(self.ai.GetLocation(), plus.getLocation(self.target_id))
if not clear_path: self.priority -= 35
def Execute(self):
if self.target_id != None:
self.ai.enemy_id = self.target_id
heading = self.ai.GetHeadingToID(self.target_id, False)
target_loc = plus.getLocation(self.target_id)
clear_path = self.ai.IsStraightPathClear(self.ai.GetLocation(), target_loc)
distance = (vector3(target_loc) - vector3(self.ai.GetLocation())).length()
speed = self.ai.GetSpeed()
# if we're close but not moving very fast, turn as fast as possible
if (distance < 5 ) and clear_path:
self.ai.Throttle(120)
self.ai.Turn(100)
elif distance > 5 or not clear_path:
self.ai.DriveToLocation(plus.getLocation(self.target_id))
else:
# stop charging if we're near the edge!
if plus.getGameType() == "TABLETOP":
a = Arenas.currentArena
loc = vector3(self.ai.GetLocation())
# check to see if we're already over
dist, over, h = a.DistanceToEdge(loc.asTuple())
if over: return False
# now check to see if we're heading over
angle = self.ai.GetHeading(False)
dir = vector3(math.sin(angle), 0, math.cos(angle))
speed = self.ai.GetSpeed()
look_ahead = .5
loc += dir * speed * look_ahead
dist, over, h = a.DistanceToEdge(loc.asTuple())
if over: return False
# drive as fast as we can toward the target
self.ai.AimToHeading(heading)
self.ai.Throttle(100)
self.regroupPos = None
return True
else:
return False
and the AI :
from __future__ import generators
import plus
import AI
from AI import vector3
import Arenas
import Gooey
import math
import Tactics
class SnSV1(AI.SuperAI):
"SnSV1!"
name = "SnSV1"
def __init__(self, **args):
AI.SuperAI.__init__(self, **args)
self.tactics.append(Tactics.SnSV1(self))
def Activate(self, active):
if active:
if AI.SuperAI.debugging:
self.debug = Gooey.Plain("watch", 0, 75, 100, 75)
tbox = self.debug.addText("line0", 0, 0, 100, 15)
tbox.setText("Throttle")
tbox = self.debug.addText("line1", 0, 15, 100, 15)
tbox.setText("Turning")
tbox = self.debug.addText("line2", 0, 30, 100, 15)
tbox.setText("")
tbox = self.debug.addText("line3", 0, 45, 100, 15)
tbox.setText("")
return AI.SuperAI.Activate(self, active)
def LostComponent(self, id):
return AI.SuperAI.LostComponent(self, id)
def DebugString(self, id, string):
if self.debug:
if id == 0: self.debug.get("line0").setText(string)
elif id == 1: self.debug.get("line1").setText(string)
elif id == 2: self.debug.get("line2").setText(string)
elif id == 3: self.debug.get("line3").setText(string)
AI.register(SnSV1)
line in bindings.py is like this :
list.append( ("Bomb", "SnSV1", { 'turnspeed' : 1.7, 'topspeed' : 12.0, 'weapons': (8, 12, 14, 15) }) )
246
« on: April 26, 2009, 07:11:52 AM »
i would like to congratulate click for it narrative talents, reading the higlight give me the desire to look at all thoses match.
Nice.
247
« on: April 22, 2009, 12:28:59 PM »
if it is still time, i'm in. send you my bot today.
248
« on: April 20, 2009, 09:39:56 AM »
Can you explain this part of your bot :  did you made it this way to protect yout bot, by need ? by necessity ?
249
« on: April 19, 2009, 02:37:37 PM »
I've never seen this design before.
you said it was a replica ? do you have a link to the real one ?
250
« on: April 19, 2009, 02:28:04 PM »
I like this bot too.
with which kind of tactic/ai do you drive it ? like a whipper waiting for the enemy or do you try to engage battle ? Isn't it hard to move forward ?
251
« on: April 17, 2009, 04:41:38 PM »
still working (slowly) on my team Don Quichotte (many ways to write his name), here is the second LW (another model for the poor knight-errant ) : Amandis de Gaule.  titanium 1 Perm 80 3 ant battery look like to have empty space ... yes, but this keep the bot 'realistic' according to DSL. For those who have not read yet the story of don quichotte , you can get it here : http://cervantes.thefreelibrary.com/Don-Quixote
252
« on: April 15, 2009, 03:48:35 PM »
My bot is more dangerous than it looks in the splash, but I'm not very happy with it either. . With 4 motors to spin weapons, your bot is not so bad and certainly not the worth HS. As in life, there's always a part of luck in any contest, so you may go far depending of the opponents you meet : 30 % to meet another HS. according to Naryar there is about 28 % to meet a bad build bot. with such statistics, almost everyone should expect to win its first match with a little luck ... but only half will. To Natef : well, Tsunami (FTW) has been the object of few jokes, don't mind boy . Sure, it seems that you did not use the full 799Kg of motors and weapons you should have. But first, you should create a (big) surprise during your first match (why not ?), second, the simple fact to participate to a contest will improve you skills and the next time we'll see you, you will be terrific, so be happy to be here.
253
« on: April 15, 2009, 10:22:44 AM »
it is understandable ; real life is better.
254
« on: April 15, 2009, 10:16:10 AM »
In now way Redline nor Wham can won. Both will get destroyed by the first decent HS encountered...
...Same thing could happen to me... I think redline got a chance, one of the two bot may side-attack and go easily under their foe. About Wham, i don't know about its motors, but as the bot seems compact and the arena is not so wide, it may turn more easily than other bots and here again, do side-attack.
255
« on: April 15, 2009, 09:55:04 AM »
How many bots !
Best paint Job :
- StratoCaster, i like the way we feel like looking inside
- Seism 13
- Revenge of Dr Wham
Futur winner(s) : ?
- R4
- Cocytus 4
- the emancipator
-revenge of Dr Wham
- Big flips revenge
- mace in da face
I'm surprised by the similarities between Bounty hunter and Put 8, one of them may also reach the podium.
256
« on: April 12, 2009, 10:01:14 AM »
You mean this ghosty?... :
[ATTACH]296[/ATTACH]
Haven't notice this side of your art ... is it an american or british quotation of some illustrate one ? ... or is it a burp from the darkest part of your mind ?
257
« on: April 12, 2009, 08:21:09 AM »
You have 2 days after the Class and Type are announced to send in your bot. While this may be quick, it's to keep the tournament running smoothly. I won't enter because i'm not sure, because of my job, family and friends, to have enough time to build a bot in 2 days and i would not like to freeze that contest. anyway, i appreciate the idea of quick built bots. i will follow this contest as simple spectator.
258
« on: April 12, 2009, 07:38:02 AM »
I appreciate Naryar for all the jobs he does around here, but there, i'm very happy Mad goes to final :
-First because he's always give a lot in any tournament, but did not always get up the podium.
-Second, because his bot is a SOW Bot and thus he proves this kinds of bots may be good ones : beating Naryar is never easy.
259
« on: April 12, 2009, 07:23:48 AM »
Back from holidays, i rush on this thread :
First of all, the looking at the video demonstrates your inovative and challenging idea and the big work you've done for us. thank you.
2 vs1 and Beat the box were my favorite.
second : well, i've not been very sharp and my bot (solar) did not shine as much as i expect
Third : HS get the preference, and indeed, they're some of the most deadly, striking in every direction.
Fourth : I do appreciate the paint jobs of Scrap Daddy, Venko. and Jack daniels : pleasant-to-see-bots
Fifth : could you share this bots and their AI ?
260
« on: April 03, 2009, 12:59:45 PM »
Philetbabe's turret bot didn't seem to be moving in the first match against obZen. the AI was one of my firsts servo one, (this tournament starts a long long time ago) and it wotks with custom Zone so it is not as effective as the last AI likes the ones of Madiaba or the one i called LazeGuidedV2. it was effective at near range. Moreover, it has locomotion trouble : the wheels were so high (to have a good wedge) that the smallest grip problem immobilizes it : yes i had a lucky win, but it does not count at all. Anyway, i had a good time building my bot, i'll have a good one following this thread. That's the way to learn things.
Pages: 1 ... 6 7 8 9 10 11 12 [13] 14 15 16 17 18 19 20 ... 26
|
Show Posts