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 - R0B0SH4RK
Pages: 1 ... 3 4 5 6 7 8 9 [10] 11 12 13 14 15 16 17 ... 104
181
« on: January 05, 2012, 11:58:11 PM »
The release of Hax Mode has to be a great moment here too.
*High fives*
182
« on: January 05, 2012, 11:47:28 PM »
Fewest nominations for an award --> R0B0SH4RK
Hell yes.
183
« on: December 15, 2011, 03:32:02 PM »
Actually, Hax Mode was initially discovered by Scrap Daddy and Sage. I figured out what they did and wrote the guide with them.
184
« on: December 10, 2011, 02:28:14 PM »
Who cares if it's a clone? As long as he doesn't claim it's his own original design, it's no big deal. Less drive, more armour.
This isn't DSL, you can't have any less drive than that :3
185
« on: December 06, 2011, 04:39:04 PM »
Determined to defend my winning streak from Home And Away, I teamed up with Jack Daniels to make the best fighting machines known to man.
(Image removed from quote.)
I'll remake the splash when I can get to the computer with my bot on it.
I'd say this is now ready for signups.
Well, you've certainly won something. I'm not sure exactly what, but you won it.
186
« on: December 06, 2011, 01:39:26 PM »
Kind of, but not really. The rationnel behind the razors is that you can technically weld a whole bunch together in real life to form a row, but a complete 100% overlap of anything (like the middle sledgehammer in this case), including razors, has always, always, always been a violation of the DSL realistic rule.
187
« on: December 05, 2011, 09:52:43 PM »
Well built and all, but that's not even DSL realistic. Your middle sledgehammer is entirely stacked within the other two. As lax as the realistic rule is, that's a pretty blatant violation.
188
« on: December 02, 2011, 06:21:05 PM »
The term "really good in battle" is relative. It's "really good" in relation to what opponents exactly?
189
« on: December 01, 2011, 07:09:40 PM »
I come from a long line of department store Santas. I don't plan on continuing this tradition.
190
« on: December 01, 2011, 05:41:33 PM »
*Is disappointed that a 16 bot Stock tourney hasn't filled up yet*
191
« on: November 30, 2011, 05:33:07 PM »
Where do you see a problem ? Exactly as it did for stock between 2003 and a few years ago (where stock stopped evolving a lot), natural building evolution towards more refined, efficient designs in DSL is a given.
The problem is that DSL strived for increased realism upon conception and release. Eventually it became about efficiency at all costs, virtually abandoning the concept of realism all together.
192
« on: November 30, 2011, 02:20:42 PM »
And now for something completely different:
HO HO HO HO HOOO HO
HO HO HO HO HOOOOOO HO
HO HO HO HO HOOO HO
HOOH HOOH HO HO HO
193
« on: November 30, 2011, 02:16:41 PM »
Hey, at least I've acknowledged that my building helped advance the trend towards unrealism just as much as anyone else. Just because I'm part of the problem doesn't also mean I can't help provide solutions too.
194
« on: November 30, 2011, 01:13:43 PM »
Not too sure that you can apply the "overrated" label to relatively obscure bands with narrow fanbases. Even if they do hail said band as brilliant, that's more of appealing to a very distinct set of fans rather than being overrated IMO. What's funny is that I prefer much of Hunters to their previous works.
And damn you for guessing what I was going to post :P
But I would disagree with "Most Thrash Metal". While Metallica has been on the wain for a very long time and Slayer's best records are the ones nobody actually listens to (Seasons In The Abyss and South Of Heaven), I feel that bands like Testament, Exodus, Overkill, Kreator and Destruction (at least pre 2003 in the case of the latter) all have had a resurgence and are comparatively unknown compared to modern metal bands and all of them successfully mix melody(especially Testament) and technicality(especially Destruction). Then you have all those bands like Heathen, Toxik, Tankard, Metal Church and so on which are all extremely good (at least in my opinion; I know you're more of a sludge/doom/stoner kinda guy; I still dig the good Cathedral song occasionally :P)
And Mustaine and Hammett suck. Really, while I liked the former a while back I dislike how overrated they are compared to guitarists like Jeff Loomis or Rob Jarzombek. (@R0B0: Did you buy Joel McIver's book on the subject of best metal guitar players?)
The "most thrash metal" comment applied to the fact that I find that thrash seems to be the most lauded and popularized genre of metal, mainly because Metallica, Slayer, Anthrax and Megadeth seem to get a ton of (unjustified) love. TBH, it's a bit of an unfair comment to make because I don't have the depth of knowledge on the genre to really make a statement like that. But my god I hate Metallica. And no, I have not read that book. Was it the one that listed Mustaine as the greatest metal guitar player ever?
195
« on: November 29, 2011, 11:42:33 PM »
Also, I am the creator of DSL realistic ! WTF no you're not :O You weren't even around when it was created!
196
« on: November 29, 2011, 03:15:07 PM »
Well, if you're not, you're admitting it's not legal for BB5 use 
197
« on: November 29, 2011, 05:12:45 AM »
Building on 123's point, can you tell me the difference between these two bots, just by looking at their AI lines?
1) list.append(("----","Omni",{'invertible':True,'nose':math.pi/2,'range':30,'radius':0.1,'topspeed':99,'throttle':130,'weapons':(1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30)}))
2) list.append(("----","Omni",{'topspeed':99,'throttle':130,'radius':0.1,'range':99,'weapons':(1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23)}))
Besides some very rudimentary stuff (# of weapons, nose orientation), it's impossible to tell that these two bots rely on completely different strategies to defeat their opponents. So, an AI, seeing only "Omni.py" will adapt how? Will it know that the first bot is completely different than the second? By the way, these are bindings lifted straight out of Starcore v4.0. The first robot is Spin Doctor, the second is NWB, but there is no way you would have known that because there are no discernable markers in the AI that would tell anyone the difference in approach a "smart" AI would need to take in order to gain a tactical advantage.
While the idea of looking for markers is good in theory, it's a nightmare in practice, because it requires a complete standardization and codification of all types of AI in order for the proposed "AdvancedOmni.py" to even have a hope of functioning. It would require all rammers to be AI'ed with a Rammer.py, all gut-rippers with a gut-ripper.py, all spinners with a spinner.py, etc, etc, just so advanced AI's can see this and devise tactics accordingly.
And what if you have two advanced AI bots fighting one another, like a pair of rammers? Surely, the best tactic for rammers is to flank and attack the hopefully unguarded sides, so both rammers would theoretically just try to outflank one another, leading to a stalemate. I'd be willing to argue that not attacking one another for the duration of the match is even more unrealistic than charging headlong into one another. And what if one rammer is slower than the other? Getting into a flanking war, as tactics would dictate, means death by stupidity for the less fleet bot as it would inevitably be outflanked. As humans, we can see this, and devise a strategy to compensate in an instant (here, having the slower rammer taking a defensive stance as to avoid being out-flanked), and continue to devise strategies and counter strategies ad infinitum. Hard coded AI (with a finite number of coded tactics) can't do this, so you're always going to wind up in situations where you either a) smash straight into each other like you would have anyways or b) enter a tactical stalemate since neither competitor is smart enough to devise new tactics on its own.
I'm not saying that having "advanced" AI is a bad idea, but I am saying that it won't work within the framework of RA2. We'd all like our bots to behave more intelligently in tournaments, but it's worth putting into perspective that this game has existed for the better part of a decade now, and we've only ever seen one AI with the capacity of altering a bots strategy (Seism 13 in BBEANS). Even then, this was no more than "if A) happens, then fire B)," it was highly specialized and written for use with that one specific bot, this adjustment had to be made post-match, and it didn't actually alter the tactics that the bot used. It's also worth noting that some very smart and code savvy guys have spent heaps and heaps of time playing with the AI code in RA2, and have yet to produce more than routines that have bots function like "rammers with spinny things attached." If it was really practically do-able, don't you think that someone would have pounded out the framework to something by now? I mean, it's not as if nobody's ever thought of having complex AI before.
---
Just as an aside, I'm growing increasingly more skeptical of how Seism 13 actually functioned in BBEANS 5, as in why did it have to wait until post-match for it to realize that it had to change the position of its wedges? Also, how was it able to differentiate between being outwedged and simply entering the air from some other event? Would a random havok explosion sending the robot airborn influence its wedge position in the next match? If it did, would the wedge AI have to be manually reset, and isn't manually tinkering with the AI mid-round as not to gain/lose an advantage cheating? SO. MANY. QUESTIONS.
198
« on: November 28, 2011, 01:34:06 PM »
#4 because MINI SFTW YES.
Probably would've voted for #3, but I didn't because Click probably built it the splash is too dark for a holiday-themed bot.
199
« on: November 28, 2011, 01:08:33 PM »
Ohmygod, "defenestrate" was an inside joke between my friends and I all through high school. I lol'ed.
200
« on: November 28, 2011, 03:30:53 AM »
More IRL stuff while I try to build myself out of a DSL builders block. HW flipper. Judge burst, NPC fasts and the like.  Someone made a bot that looked like Plunderbird (yet again, I forget who) so it inspired me to make this. I'm surprised that I haven't really seen any IRL-style crusher like Razer yet. Judge burst, so it's really more of a glorified hammer than a crusher. 
Pages: 1 ... 3 4 5 6 7 8 9 [10] 11 12 13 14 15 16 17 ... 104
|
Show Posts