Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Locked thread
Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀

Sundowner posted:

I have neither played the game nor read the previous LP. I probably should but I'm not too heavily vested in the game or series. I might go through it in the time between updates here.

I can tell you that the first game's plot will quickly be ruined by things in this game. You should really finish the other LP before this one gets too far in.

Adbot
ADBOT LOVES YOU

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀

Fedule posted:

I'm going to leave the poll open for a day or so so I can have an excuse to procrastinate or something, but here's the results up to this post:

code:
                            1         2     
                   12345678901234567890123456
Magenta/Luna       iiiiiiiiiiiiiiiiiiiiiiiii
Yellow/Tenmyouji   iiiiiiiiiiiiiiiiiii
Cyan/Alice         iiiiiiiiiiiiiiiiiiiiiiiiii

You still have to write up all of these paths eventually. No reason to wait until we're done voting to start transcribing everything. :getin:

Also, my vote is for Cyan.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
The game is kinda broken in emulating the prisoner dilemma, since there's a strategy we can employ that does not require any trust on the part of the participants. Since a betray/ally still results in a net gain of points, and a betray/betray results in no net loss of points, if the players agree beforehand to always choose betray/ally (such that the person with the fewest points chooses betray so that nobody will ever die), then eventually everyone will get enough points. If a player, for whatever reason, chooses to deviate from this, then either it will be betray/betray, in which case nothing bad happens, or ally/ally, in which case the net gain is even greater.

Dr. Stab fucked around with this message at 20:33 on Jun 2, 2013

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
From the perspective of getting out the fastest, betray doesn't make much sense. With ally, we can get out in 3 rounds. With betray, we can get out in 2. However, It's pretty easy to assume that anybody who picked betray in the first round will be subsequently betrayed, thus eliminating the possibility of getting out any faster.
Also, since we are already requiring that the opponent pick ally two times in a row to get out with betrayl, if we instead pick ally in that case, both sides will be at 7 points, at which point there is no reason to pick betray unless you really want to gently caress someone over, so the game will be effectively solved after 2 rounds of mutual ally.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
Well, phi explicitly says that you will die.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀

lotus circle posted:

At least we won't have to worry about complicated algorithm theories! Simple mathematics all the way :v:

Integer addition modulo 9 is way easier than game theory, trust me.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀

Myrrar posted:

It gets more complicated if at least one person ends up betrayed and loses points this round. Then you have one (or more) groups at 1 BP. Since their lives are on the line, they are far, far more likely to vote betray. At 1 BP, you can guarantee that you do not die by voting betray all the time. Life is a fairly powerful motivator, afterall.

Sure it would be easy right now to make an agreement with everybody to vote Ally all the time, but they intentionally were not given the chance to discuss this beforehand. Even if they discuss it afterwards, so long as one group ends up at 1 BP, can you really assume that they won't vote betray just to guarantee their safety?

I would count on an encourage 1BP groups to vote betray. Nobody can die as long as they do this. And, if I was at 6 or 5 BP and against them, I would vote ally, bringing us both to non-lethal totals. After which, we can either agree to mutually ally at no danger to ourselves, or betray/ally chain our way up.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀

Hobgoblin2099 posted:

Does it bother no one else that Zero III is calling Clover "Cleaver" considering what happened in the non-canon Axe Ending?

How do you know that wasn't the canon ending? It's not like there's anyone else from the first nonary game here. :v:

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
I think the question is less "Which characters are robots" and more "Which characters aren't robots." :tinfoil:

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
The chinese room is not literally about a man in a room looking up things in books. It's about having an algorithm which takes in input in a language, and produces output in that language. Is that the same as knowing that language? More generally: Does a computational model of consciousness constitute consciousness?

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
The system isn't necessarily memoryless.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
Then, do arbitrary computational systems have consciousness?

Obviously, the nature of consciousness is completely unknown, so we can't know either way. But, that doesn't mesh with how we currently understand consciousness to manifest.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
In the original formulation of the chinese room, it's a computer running an AI that passes the turing test, not just a series of books.

e: man I'm getting my terminology all wrong

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀

maswastaken posted:

The absurdity of the situation is really, really hard to ignore. You either have something that clearly doesn't understand Chinese because it just isn't good enough at holding a conversation or something that could lead a conversation literally anywhere it liked, having a sufficient breadth and depth of pre-written conversations to properly handle any input and context, except how it would do that is always completely arbitrary because the driving force inside it still doesn't understand Chinese. I'd grant the latter example is convincing enough that you might as well say the system understands Chinese but it's a system with incredible wasted potential because in truth it still no more understands Chinese than an encyclopedia knows things.

You're saying that it's dumb because such a system wouldn't actually understand chinese. That's Searle's point. You have a system that doesn't understand chinese, despite behaving exactly as though it does.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀

maswastaken posted:

I'm saying it's dumb because it's an actual comprehension of Chinese short of being something much more incredible.

What do you mean by that? The Chinese room isn't actually meant to perform a function. It's just a way of looking at how an AI works.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀

maswastaken posted:

Does anyone in the group even look like Akane "Zero the First" Kurashiki?

We have no idea what "K" looks like. That's a very suspicious name, don't you think?

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀

unfair posted:

Or a logical robot. Unless I missed something we haven't figured out who the android is yet. (assuming for the sake of the story that there's only one)

Is Zero Sr the same person every time you play, or does it switch around depending on who's alive? I guess potentially he/she could also be one of the "dead" people that haven't been shown as dead yet, but this should narrow down things pretty quickly.

I'm assuming that it's like 999, where each path does not constitute a whole story, but instead contributes to the overall story. Changing major plot details (like who is zero) between these paths would be a very bad thing to do.

Also, I'm pretty sure Akane dies in the 999 bad ends, so your method wouldn't work in that case.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
I was going to say that Clover killed her and Dio just stole the bracelet, then K found the body. But, the bracelet was still there, so I don't know? Dio doesn't seem like a guy to leave a bracelet behind.

Dio was with the body for an entire minute. He definitely did something.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
One thing I've noticed is that the first option has won every poll so far. Is there another polling site that makes it so that the order of the options are randomized?

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
But what about "Rival Axe" or "Axial Rev" or "A Rival Ex." Oh poo poo, that one's probably it.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
We continue to always pick the first option in every poll. I have a simple fix: From now on put Betray first in the voting. :getin:

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
Is it just me, or is there no way that that took 4 minutes? All he did was walk up to Dio and then Dio gave him the medicine. Everyone would have to be talking really slowly for that to take 4 minutes.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
Clover was supposed to be 18 in 999?

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
Maybe K is made entirely out of antimatter. That's why he needs the containment suit.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
drat, this would have been super trippy if we had done it in the other order. Oh well.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀

CottonWolf posted:

To be a classical prisoner's dilemma the value of payoffs has to be:

Betraying when the opponent allies > Allying when the opponent allies > Betraying when the opponent betrays > Allying when the opponent betrays

So you're saying that my strategy is useless?

Python code:
@name{Edge Casey}
@stratname{Check game rules to make sure we are actually playing prisoner's dillema}
@autor{Dr. Stab}
@strategy{
def decide(info):
	Facts=info.facts
	
	m = Facts.payoff.mbetray
	b = Facts.payoff.betray
	s = Facts.payoff.sucker
	t = Facts.payoff.trust
	#check for coordination game
	if m>=t and m>=s and m>=b :
		return false
	if t>=m and t>=s and t>=b :
		return true
	
	#check for betray game
	if b>=t and b>=s and m>=t and m>=s:
		return false
	
	#check for ally game
	if t>=b and t>=m and s>=b and s>=m:
		return true
	
	#check for anti-coordination game
	if (s>= m or s>=t) and (b>=m or b>=t):
		if Facts.currentround == 1:
			return randint(0,1)
		#do the opposite of what the opponent did last
		opp = Facts.opponentid
		lastround = info.GetLastXRounds(1)
		for g in lastround.games:
			for p in g.participants:
				if p.id.equals(opp):
					return not p.decision
		
	#actual prisoner game
	#always ally in prisoner game. Always betray in inverted prisoner game. I'm a nice guy.
	return b>s
}
note: I didn't test this. Probably doesn't work.

okay another dumb strategy:

Python code:
@name{Opposite man}
@stratname{Tat for Tit}
@autor{Dr. Stab}
@strategy{
def decide(info):
	Facts=info.facts
	if Facts.currentround == 1:
		#trick people on the first round
		return true
	#do the opposite of what the opponent did last
	opp = Facts.opponentid
	lastround = info.GetLastXRounds(1)
	for g in lastround.games:
		for p in g.participants:
			if p.id.equals(opp):
				return not p.decision
	#opponent not found
	raise RuntimeError("Someone has infiltrated our game and is trying to kill me")
}

Dr. Stab fucked around with this message at 17:53 on Apr 21, 2014

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀

Twiddy posted:

Yeah, after sleeping on it this became a bit more clear.

In general, having a player whose entire strategy is trying to get another specific player to win would be interesting. That's kind of the scary part of these types in general. Sometimes you get the normal person who believes in honor, or the selfish guy, or the guy who has studied game theory, and they're all relatively predictable once you nail what moral system they use. Then there's the guy whose definition of winning exists completely separate from anything you could plan for.

Similarly, I can easily imagine a computer player whose goal is to try to get everyone's score as low as possible. Or kill as much as possible.

That sounds a lot like this:

wikipedia posted:

Although tit for tat is considered to be the most robust basic strategy, a team from Southampton University in England (led by Professor Nicholas Jennings and consisting of Rajdeep Dash, Sarvapali Ramchurn, Alex Rogers, Perukrishnen Vytelingum) introduced a new strategy at the 20th-anniversary iterated prisoners' dilemma competition, which proved to be more successful than tit for tat. This strategy relied on cooperation between programs to achieve the highest number of points for a single program. The university submitted 60 programs to the competition, which were designed to recognize each other through a series of five to ten moves at the start.[10] Once this recognition was made, one program would always cooperate and the other would always defect, assuring the maximum number of points for the defector. If the program realized that it was playing a non-Southampton player, it would continuously defect in an attempt to minimize the score of the competing program. As a result,[11] this strategy ended up taking the top three positions in the competition, as well as a number of positions towards the bottom.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀

Balk posted:

I like this idea. I'm kind of surprised the jerks did so well, since Tit For Tat was supposed to be the best approach.

This is probably because of the large number of effectively random strategies. Betray is the best strategy in regular prisoner's dilemma, and if there's nobody out there to punish you, then you can keep using it. Tit for tat only beats out always betray if the rest of the actors aren't gibbering fools.

e: that gives me an idea.

Python code:
@name{Winner}
@stratname{Win}
@autor{Dr. Stab}
@strategy{
def decide(info):
	return 2+2==5
}
Python code:
@name{Helper}
@stratname{Help}
@autor{Dr. Stab}
@strategy{
def decide(info):
	opp = info.facts.opponentid
	for entry in info.scoreboard:
		if entry.id == opp:
			if entry.timestrusting>0:
				return False
			else:
				if info.facts.currentround == 1:
					return False
				else:
					return True
			
	return True
}
}
Python code:
@name{Helper2}
@stratname{Help2}
@autor{Dr. Stab}
@strategy{
def decide(info):
	opp = info.facts.opponentid
	for entry in info.scoreboard:
		if entry.id == opp:
			if entry.timestrusting>0:
				return False
			else:
				if info.facts.currentround == 1:
					return False
				else:
					return True
			
	return True
}
}
Python code:
@name{Helper3}
@stratname{Help3}
@autor{Dr. Stab}
@strategy{
def decide(info):
	opp = info.facts.opponentid
	for entry in info.scoreboard:
		if entry.id == opp:
			if entry.timestrusting>0:
				return False
			else:
				if info.facts.currentround == 1:
					return False
				else:
					return True
			
	return True
}
}
Python code:
@name{Helper4}
@stratname{Help4}
@autor{Dr. Stab}
@strategy{
def decide(info):
	opp = info.facts.opponentid
	for entry in info.scoreboard:
		if entry.id == opp:
			if entry.timestrusting>0:
				return False
			else:
				if info.facts.currentround == 1:
					return False
				else:
					return True
			
	return True
}
I could probably do a version of this that doesn't have a targeted player to propel, but instead determines whether the other player is using this same strategy and then picks an outcome which gives the win to the player with the most points but :effort:

Dr. Stab fucked around with this message at 03:10 on Apr 23, 2014

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
There's nothing that explicitly forbids file io. I say write your script.

E: you could make an ai that at the start of the game, appends its id to a file, and then betrays everyone who isn't in the file, and propels evertyone who is. Then submit 50 of those ais.

Then, since this is all publically viewable, someone else comes along who intentionally disrupts this plan by changing the names in the file.

Dr. Stab fucked around with this message at 03:48 on Apr 23, 2014

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
It's not a diphthong. It's a singular consonant. You can't have a diphthong between consonants, anyways. It's just a sound that exists somewhere in the space between r and l, as you said.

It's kinda this way with basically every sound in every language (especially the vowels) . There's not always an identical sound in your language, so transliteration can only approximate it.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀

Fedule posted:

It concerns me that I have, historically, always had an easier time maintaining a consistent LP update rate during termtime and exams than during holidays (funfact: All three of my LPs have started up during exam season). Is that weird? I think that's weird.

I'm the same way with side projects. I guess I make more progress when I have a limited time during the day in which to do things than when I can just do it whenever.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
So I guess that's the end of the LP. Congrats Fedule.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
Looks pretty clearly like the moon as seen from the southern hemisphere.

e: I think it's also been mirrored somehow.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀

zonohedron posted:

What's the big red circle in the sky, then, fellow geometric object? :colbert:

Charon.

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀
Someone smarter than me might be able to use the date and the position of the moon (a little under 2 degrees above the horizon) as well as its orientation to find out where we are.

It looks like a rising moon. Looking at this wikipedia article:

http://en.wikipedia.org/wiki/December_2028_lunar_eclipse

My guess is Saudi Arabia or Egypt.

Adbot
ADBOT LOVES YOU

Dr. Stab
Sep 12, 2010
👨🏻‍⚕️🩺🔪🙀😱🙀

Nagna Zul posted:

I can guarantee you that the game developers did not think of this nor account for it.

Source: I am a game developer.

Well, yeah, obviously. That's exactly why we, the players, need to think of it.

  • Locked thread