IPB

Welcome Guest ( Log In | Register )

5 Pages V  < 1 2 3 4 5 >  
Reply to this topicStart new topic
> Cracking The Rabbit Races, Uncovering the algorythm
Ishman
post Oct 10 2006, 02:52 AM
Post #41


Ee san
Group Icon

Group: Nobles
Posts: 506
Joined: 4-July 06
Member No.: 126



To me, it seems like this is what you're doing...

For the sake of illustration, I will use n trials of a coin toss. As may be apparent, H means that the coin landed on Heads and T means that the coin landed on Tails.

H, H, T, H, H, T, H, H, H, T, T, H, H ,H ,T, T, H, H, H, H, T, T, T, H, ..., T, T, T

Observing the first twenty-three and the last three results [trials (n-2), (n-1), and (n)] of n trials, one might look at the results and say, "There's obviously a pattern to the outcome of the coin toss!" While it may appear to be so at face value, that's nowhere near correct. In this example, we can say that after 23 tests, there is certainly a distinctive pattern. However, that pattern is only valid for trials 1 to 23. You cannot prove that this pattern will hold for every test afterward. This remains true for one thousand tests, for one million tests, even for 10^(10^(10^10)) tests. Obviously, the more tests you perform, the more reliable your data will be; but, you would need to perform n tests to show that your pattern is, indeed, valid.

Suffice to say that even after a century of data analysis, it is unlikely that you will have found anything even resembling a pattern. At least, mathematically speaking, the odds are stacked against you.

In other words, unless you want to be playing with rabbit races for the rest of your life, I suggest you stop and find a different approach, such as the suggested probability approach. I'm sure as hell not going to waste my time with the method of exhaustion -- I have goals in life, you see... laugh.gif


---
Edited: Illustration was poorly conceived, text was very hard to follow. Adjusted accordingly. Please bear with me, as my eyes are closing themselves and my A.D.D. is kicking in. Maybe I should sleep. I'll check on this tomorrow. smile.gif
Go to the top of the page
 
+Quote Post
PIEMINISTER
post Oct 10 2006, 09:03 AM
Post #42


Sa san
Group Icon

Group: Banished
Posts: 1,388
Joined: 31-July 06
Member No.: 531




QUOTE
Then let's put it this way, if you already know the "pattern" and there are "few other instances" then go in, follow it UNTIL your pattern comes up, and then bam, figure out the new one. Simple.


ROFL.

I don't know "the pattern", I know A pattern. It's like I know the first 6 digits of pi but everything else is oblivious to me...for now...


Ishman: The difference between a coin toss and a rabbit race is that the rabbit race is built on a randomizer which is built on an algorythm, meaning a randomizer is not RANDOM. There are way more variables in tossing a coin then there are of running a randomizer. If you create a device to flip a coin, in an enclosed area, with an undamagable surface you could easily find a pattern in a coin toss, simply because of the reduction of variables.
Go to the top of the page
 
+Quote Post
Ishman
post Oct 10 2006, 04:41 PM
Post #43


Ee san
Group Icon

Group: Nobles
Posts: 506
Joined: 4-July 06
Member No.: 126



I think you misunderstand how a pseudo-random number generator works, so I'll explain the basic components.

Entropy for the pseudo-random number generator is seeded from the hardware interrupts of the computer. The algorithm, which is not very well documented for the Linux RNG, refreshes the generator's "state" every once in a while and "outputs pseudo-random bits" (Gutterman 1).

These are some basic properties of all pseudo-random number generators:

QUOTE(Gutterman @ 6 March 2006)
Pseudorandomness. The generator's output looks random to an outside observer.

Forward security. An adversary which learns the internal state of the generator at a specific time cannot learn anything about previous outputs of the generator.

Break-in recovery / backward security. An adversary which learns the state of the generator at a specific time does not learn anything about future outputs of the generator, provided that sufficient entropy is used to refresh the generator's state. (Gutterman 1-2)

The appearance of randomness is good enough if an attacker has no information about the state of the generator, which, by the way, cannot be gathered without at least access to the machine's operating system. If such information is gathered, then, ideally, the attacker cannot learn anything about past or future states of the generator due to forward-security and backward-security measures (2).

On KRU's servers, there is more than enough sufficient entropy to create good pseudo-random output. That is why your method of data analysis will not work.

-----
QUOTE
Acknowledgements:

Gutterman, Zvi, et. al. 2006. Analysis of the Linux Random Number Generator.
On-line. Available from Internet, http://eprint.iacr.org/2006/086.pdf,
Accessed 10 October, 2006.




P.S. Because the first, oh, SEVERAL TRILLION digits of π show no pattern, I'm certain that there won't be one found. laugh.gif Just illustrating what a bad example that was.
Go to the top of the page
 
+Quote Post
Toshimaru
post Oct 10 2006, 09:41 PM
Post #44


Ee san
Group Icon

Group: Nobles
Posts: 452
Joined: 7-July 06
Member No.: 207



ok wait a second


EVERYONE READ THIS...


Say you do crack the rabbit races.

Money will flock into the kingdoms.
Everyone will have TONS of money.
Ok thats fine and dandy.

but with the mass ammount of gold in the game, All the prices will raise for everything.

Because alot of people read these forums.

I havnt been on for a bit but just say For example


A big axe

Say it goes for 200k ( I dunno anymore )

And after the rabbit races

200k will be NOTHING

It will raise to..

5-6 mill?

Whats the point? El Seriously?


--------------------
I don't care about your forum but I'm lurking anyway.
Go to the top of the page
 
+Quote Post
Ishman
post Oct 10 2006, 10:34 PM
Post #45


Ee san
Group Icon

Group: Nobles
Posts: 506
Joined: 4-July 06
Member No.: 126



It would obviously be a stupid idea to "crack" the race and let everyone know how to do it. Any person with a basic grasp on economics knows that.

Personally? I don't really care about the outcome. I just like discussing it. If this thread, in whole or in part, was harmful to the game, I'm certain the moderators would have taken action long ago. <3

I stand by the fact that "cracking" the rabbit races is virtually impossible based on my reasons mentioned above. Using probability to increase your odds of winning is a more practical approach, much like with any other gambling.
Go to the top of the page
 
+Quote Post
Nagnag
post Oct 13 2006, 12:20 AM
Post #46


Ee san
Group Icon

Group: Retired Staff
Posts: 381
Joined: 3-July 06
From: Kugnae forever! Weapon: ** Black Thorn ** Armor: ** Black Tuxedo ** Shield: ** Dagger **
Member No.: 4



QUOTE
I stand by the fact that "cracking" the rabbit races is virtually impossible based on my reasons mentioned above. Using probability to increase your odds of winning is a more practical approach, much like with any other gambling.


Exactly. And you know what, if let's say you did find a series of patterns based on a certain algorhythm, they'd change the variables based on the algo. So in essence, you'd lose more than you win, because as soon as you crack it and announce it, they'll fix it. However - if you discover it, don't announce it ;D

Certain people I know have figured certain get-rich techniques and don't share their tricks, like Newhook, but there's a select few, maybe one day I'll tick them all off and reveal a few tricks (none illegal, just time / patient consuming / knowledge based)


--------------------
"They must find it difficult...those who have taken authority as the truth, rather than truth as the authority."
Go to the top of the page
 
+Quote Post
stabme
post Oct 13 2006, 01:10 AM
Post #47


Sam san
Group Icon

Group: Nobles
Posts: 907
Joined: 4-July 06
Member No.: 66



QUOTE(Sephyrson @ Aug 2 2006, 07:17 PM) [snapback]5850[/snapback]

lol... its not hard to make money off the tracks at all

check this out, get 3 friends

5 rabbits race
you have a 3/5 chance to win
the payout is 4x the original bet
so each time one of you wins, you split the 4x payout 3 ways
you make money
its simple math
yes, Its after work, I'm tired, and not speaking in complete sentances
but trust me, you can easily make 2mil a day, per person, just by doing that while doing hw, writing a term paper, playing xbox, watching tv, whatev...

-seph

this is the most sensible thing. its the only way to beat chance: there will be no pattern, since its random. so you ned to stack chance in your favor.

but anyway.. isnt there a flaw to the random function in programming, how its not TRULY random? isnt it based onf a clock cycle etc? ... although its so insanely had to time that it might as well be random.


--------------------
IMEDIAT: I am the Lixir! I speak for the Wars!
I no longer Elixir or draw Imediat. I now help make music free for all: Grooveshark
Go to the top of the page
 
+Quote Post
Ishman
post Oct 13 2006, 02:40 AM
Post #48


Ee san
Group Icon

Group: Nobles
Posts: 506
Joined: 4-July 06
Member No.: 126



As far as the random() function goes...

QUOTE

The random() function uses a non-linear additive feedback random number
generator employing a default table of size 31 long integers to return
successive pseudo-random numbers in the range from 0 to (2**31)-1. The
period of this random number generator is very large, approximately
16*((2**31)-1).

The random() and srandom() functions have (almost) the same calling
sequence and initialization properties as the rand(3) and srand(3) func-
tions. The difference is that rand(3) produces a much less random
sequence -- in fact, the low dozen bits generated by rand go through a
cyclic pattern. All the bits generated by random() are usable. For
example, `random()&01' will produce a random binary value.

Like rand(3), random() will by default produce a sequence of numbers that
can be duplicated by calling srandom() with `1' as the seed.

The srandomdev() routine initializes a state array using the random(4)
random number device which returns good random numbers, suitable for
cryptographic use. Note that this particular seeding procedure can gen-
erate states which are impossible to reproduce by calling srandom() with
any value, since the succeeding terms in the state buffer are no longer
derived from the LC algorithm applied to a fixed seed.


It looks like srandomdev() is the only function that actually utilizes the random number generator. rand() is the function that I would never use for anything, ever, but it appears that random() improves on it. tongue.gif

( **sigh** Sadly, when I took my C++ class, my teacher was still showing everyone to use rand()... That guy was, perhaps, one of the reasons I quit pursuing Comp. Sci. I mean, aside from the fact that the degree is worthless...)

I digress. If these functions were being used, I suppose if one knew the seed, it wouldn't be terribly difficult to figure it out...? As usual, the problem remains that we don't have enough info. Probability is your best bet.
Go to the top of the page
 
+Quote Post
Adam
post Oct 16 2006, 01:27 AM
Post #49


Sa san
Group Icon

Group: Nobles
Posts: 1,555
Joined: 8-August 06
From: Saskatoon, Saskatchewan, Canada
Member No.: 574
Characters: Jesus Christ, Allah, Jehovah... And Satan.



Nerds.




















Haha, sike. PIEMINISTER, I'd be willing to help, as long as it doesn't entail more than 5 minutes of watching pixelized rabbits run.


--------------------
I'm too legit. Ask your nieces and your nephews, I'm the shhh... Be quiet! Cursing's not allowed.
Go to the top of the page
 
+Quote Post
Gokku
post Oct 16 2006, 04:14 AM
Post #50


Ee san
Group Icon

Group: Nobles
Posts: 322
Joined: 4-July 06
Member No.: 79



QUOTE(KaN @ Oct 16 2006, 08:27 AM) [snapback]14863[/snapback]

Nerds.

Dumb wink.gif


--------------------

Swift sonGokku ..........Panoply Gokku .............Mage Kenzo

Retired from Nexus!

Repeat after me: Nexus is only a game ... Nexus is only a game ... Nexus is only a game ...Nexus is only a game! ...
Go to the top of the page
 
+Quote Post
Nagnag
post Oct 18 2006, 06:44 AM
Post #51


Ee san
Group Icon

Group: Retired Staff
Posts: 381
Joined: 3-July 06
From: Kugnae forever! Weapon: ** Black Thorn ** Armor: ** Black Tuxedo ** Shield: ** Dagger **
Member No.: 4



Oh wow 1-liner responses. Right...

So here's what I suggest we all do. I suggest we hold a huge rabbit races event where 50-100 people just bet and have fun. And then tally up the results to see how much total money was won vs lost, and who bet on who and which won.

If you can organize something like that - that would rock. Because then you'd have a massive amount of people to compare numbers around with.


--------------------
"They must find it difficult...those who have taken authority as the truth, rather than truth as the authority."
Go to the top of the page
 
+Quote Post
Euphoria
post Oct 18 2006, 07:27 PM
Post #52


Sam san
Group Icon

Group: Nobles
Posts: 828
Joined: 20-July 06
From: Humboldt county
Member No.: 446



I'd come smile.gif


--------------------
Go to the top of the page
 
+Quote Post
Ishman
post Oct 18 2006, 10:03 PM
Post #53


Ee san
Group Icon

Group: Nobles
Posts: 506
Joined: 4-July 06
Member No.: 126



I'd breathe heavily. smile.gif
Go to the top of the page
 
+Quote Post
spinkns
post Oct 21 2006, 02:54 PM
Post #54


Peasant


Group: Peasants
Posts: 1
Joined: 5-August 06
Member No.: 559



HEY RANGENS AND RANDOS THE SEED IN YOUR EQUASION IS BASED ON SERVERS SYSTEM CLOCK PROLLY TO >MILISECOND GOOD LUCK
Go to the top of the page
 
+Quote Post
Ishman
post Oct 21 2006, 07:27 PM
Post #55


Ee san
Group Icon

Group: Nobles
Posts: 506
Joined: 4-July 06
Member No.: 126



-sigh- Do you people actually read the thread?!
Go to the top of the page
 
+Quote Post
stabme
post Oct 29 2006, 02:29 PM
Post #56


Sam san
Group Icon

Group: Nobles
Posts: 907
Joined: 4-July 06
Member No.: 66



QUOTE(Nagnag @ Oct 18 2006, 07:44 AM) [snapback]14959[/snapback]

Oh wow 1-liner responses. Right...

So here's what I suggest we all do. I suggest we hold a huge rabbit races event where 50-100 people just bet and have fun. And then tally up the results to see how much total money was won vs lost, and who bet on who and which won.

If you can organize something like that - that would rock. Because then you'd have a massive amount of people to compare numbers around with.

even though we've already established it's RANDOM and you can already prove mathematically how much money everyone will lose. rolleyes.gif and despite we've ALREADY ESTABLISHED the most effective way to make money . . . . i can effectively say you only read the last two post and not a single other post on this page. so yes, THANK YOU FOR READING. and then making an idiotic contribution.

Don't bash!

Ally


This post has been edited by Rachel: Nov 2 2006, 01:34 AM


--------------------
IMEDIAT: I am the Lixir! I speak for the Wars!
I no longer Elixir or draw Imediat. I now help make music free for all: Grooveshark
Go to the top of the page
 
+Quote Post
solman
post Nov 8 2006, 10:56 AM
Post #57


Oh san
Group Icon

Group: Nobles
Posts: 2,560
Joined: 4-July 06
Member No.: 104



QUOTE(Ishman @ Oct 10 2006, 01:52 AM) [snapback]14499[/snapback]

To me, it seems like this is what you're doing...

For the sake of illustration, I will use n trials of a coin toss. As may be apparent, H means that the coin landed on Heads and T means that the coin landed on Tails.

H, H, T, H, H, T, H, H, H, T, T, H, H ,H ,T, T, H, H, H, H, T, T, T, H, ..., T, T, T

Observing the first twenty-three and the last three results [trials (n-2), (n-1), and (n)] of n trials, one might look at the results and say, "There's obviously a pattern to the outcome of the coin toss!" While it may appear to be so at face value, that's nowhere near correct. In this example, we can say that after 23 tests, there is certainly a distinctive pattern. However, that pattern is only valid for trials 1 to 23. You cannot prove that this pattern will hold for every test afterward. This remains true for one thousand tests, for one million tests, even for 10^(10^(10^10)) tests. Obviously, the more tests you perform, the more reliable your data will be; but, you would need to perform n tests to show that your pattern is, indeed, valid.

Suffice to say that even after a century of data analysis, it is unlikely that you will have found anything even resembling a pattern. At least, mathematically speaking, the odds are stacked against you.

In other words, unless you want to be playing with rabbit races for the rest of your life, I suggest you stop and find a different approach, such as the suggested probability approach. I'm sure as hell not going to waste my time with the method of exhaustion -- I have goals in life, you see... laugh.gif
---
Edited: Illustration was poorly conceived, text was very hard to follow. Adjusted accordingly. Please bear with me, as my eyes are closing themselves and my A.D.D. is kicking in. Maybe I should sleep. I'll check on this tomorrow. smile.gif



Oh I love the coin toss debate. No pattern eh your right in that aspect but you can predict with over a 90% accuracy on getting this right. Theres 3 major things you have to keep in mind when deciding in a coin toss. 1 is the weight of a coin. Meaning lets just take a quarter for example. The head side of the coin is actually much heavier then the tails side. So its going to constantly land in a truely even flip tails 100% of the time. This is due to the head side being heavier will tend to pull and have a higher gravitation pull on that side of the coin making this face down.

2. The way the coin is tossed. Is this coin flipped or is it spinned? This would have a huge bearring on the outcome. Again due to weight complexities the spinning will give you a greater percentage towards tails. Where as with it being flipped (by flipped i mean truely flipped with many rotations) your going to have a higher increase of it being the same face value as it being started out as. Meaning this.. Place the coin on your thumb as if your going to flip it. Whatever face value its showing either it be heads or tails will most likely be that same result.

3rd is power. A higher power in the spin or the flip will increase or decrease the rate of it being heads or tails. Now this part will have to go into alot more details in explaining however you start to get the idea from these points. If you can add all these concepts and incorporate them, you can have around a 90% sucess rate on picking heads or tails. I believe I had an 87.9% using these methods.

So in this case its not the patterns of the results but rather the results of the start. So theres more then just 1 way to figure out a pattern. Hope that helps a little in figuring out how to crack a few things within nexus.


--------------------
Time to do the naked Pee Pee dance.

Go to the top of the page
 
+Quote Post
PIEMINISTER
post Nov 8 2006, 01:22 PM
Post #58


Sa san
Group Icon

Group: Banished
Posts: 1,388
Joined: 31-July 06
Member No.: 531




Also, to add to Solmans post if you set up a machine and make sure there are no variables (even temperature, and air pressure) you can continually get a coin to flip on a certain side 100% of the time)
Go to the top of the page
 
+Quote Post
SawSan
post Nov 9 2006, 10:59 AM
Post #59


Level 5
*

Group: Peasants
Posts: 7
Joined: 3-November 06
Member No.: 1,106



I see people advising to get 3 more friends (total 4) to make a better winning chance....

Why not bring your other chars and bet?

i never done it before, dont like bets anyways.

just a thought.
Go to the top of the page
 
+Quote Post
Ishman
post Nov 9 2006, 02:08 PM
Post #60


Ee san
Group Icon

Group: Nobles
Posts: 506
Joined: 4-July 06
Member No.: 126



I see your points on the coin issue, but I think you all missed my point. The purpose of the coin toss illustration was to demonstrate that just because there looks to be a pattern doesn't mean there is one, as one miss will break the entire pattern. It was NOT intended to illustrate that a pattern cannot be devised using artificial methods; though, ideally, the objective of artificial apparatus is to eliminate entropy altogether (i.e. try to achieve only one result 100% of the time), so of course you can predict what the outcome will be. wink.gif

Moreover, recall, you don't have those options available to you in the rabbit race. With the rabbit race, assuming it utilizes the pseudo-random number generator as previously mentioned... It's more similar to going to a crowded city street, blindfolding yourself, handing someone a coin and saying "toss it," asking them to tell you what it landed on, and writing down the result. From there, you proceed to spin yourself around one hundred times and down a few pints or something, pick up the coin and walking around until you find a different person to hand the coin to. You tell them, "toss it," asking them to tell you the result, writing down the result. You repeat this process until you get tired of it (or until you vomit on someone else's shoes.) Time passes, you recover yourself, and you decide to use your data from your past trials to predict the outcome for future trials.

I'm arguing that it would be impractical to do such a thing, because for one, time passed, meaning there are different people present. Two, each person will not flip it the same way, even if you run into a person you encountered during the previous trials. Finally, the amount of entropy you put into the system changes as well; for instance, maybe you won't drink so much this time to avoid getting sick. It's just impractical, if not impossible, if I'm looking at everything correctly, and assuming that my hypothesis (utilization of the PRNG) is correct.

This illustration matches with the way the PRNG works, because you don't know the state of the generator (inc. which person you handed it to, how they flip the coin, etc.), and the state changes based on how much entropy is put into it (inc. how many times you spin yourself, how many pints you down, how much time has passed between trials, and how long you walk around until you bump into another person to toss the coin). Furthermore, you can't really know anything about the past or future states of the generator -- being blindfolded and tipsy, it's unlikely that you'll bump into a previously encountered person any time soon (and if you do, you won't know it); and of course, you won't know who you'll end up bumping into in the future.

The only thing on your side is probability -- you will always have a 50% chance of guessing correctly.

(Oh, by the way. Assume that, for the sake of illustration, you won't get arrested, robbed, or beaten to death with a tire iron by doing such a profoundly stupid thing.)
Go to the top of the page
 
+Quote Post

5 Pages V  < 1 2 3 4 5 >
Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 

RSS Lo-Fi Version Time is now: 28th March 2024 - 09:13 AM
Nexus Forums is part of: Nexus Atlas © all rights reserved.