the dice, quit kidding yourselves and open your eyes

[canderson92]

there is no way you can tell me the dice are completely random. ok i understand losing `12 to 1 maybe once. but just listen, you are rolling one die and i am rolling three, ok so you beat me 9 times in a row, fine but when you do it again on my next turn and when you do it more often when the game is almost over and your losing then i say that someone has a stupid idea to try to make this stupid game of his more interesting. I dont win the lottery every other time i play although some people win the same odds on here every week. if i knew where you lived i would come show you what random violence means.

[hecter]

If you can provide me with a good case as to why the admins of the site would decide to make the dice not-random, then I will believe that they aren't.

[canderson92]

exacty that, to try to make the game more interesting. i admit that the dice are not random but bad both ways as long as the game is close but i have had more games that have ended in me trying to finish two or three armies with a dozen or more and it ended with me beating my head against the wall

[wcaclimbing]

exacty that, to try to make the game more interesting. i admit that the dice are not random but bad both ways as long as the game is close but i have had more games that have ended in me trying to finish two or three armies with a dozen or more and it ended with me beating my head against the wall

if you can provide some kind of statistical analysis of the dice, not just a
"I attacked ## with ### armies and lost ## three times in a row!!!11!!!one!"
Then you might be able to convince a few people.
The mods/lack have said before, if you want it changed, you need to prove that it needs to be changed. A whiny "the dice suck" post does not count as proof.

[canderson92]

ok so i wrote it down. when i read the paper it says, "the dice are obviosly not random, if you cant see that you are an idiot"

[hecter]

ok so i wrote it down. when i read the paper it says, "the dice are obviosly not random, if you cant see that you are an idiot"

Well, you see, that isn't going to convince anybody. Provide me with statistical proof that, over a good length of time, the dice will come out favouring (by your estimates, greatly) the defender. See, I have been doing that... I've been recording thousands of roles with a nice little tool called the "Dice Analyzer". My conclusion (though my data isn't very conclusive) is that the dice are in fact, random. See for yourself:
http://i188.photobucket.com/albums/z138 ... ure3-2.png
As you can clearly see, where ever I have collected a good amount of data that it comes out as what it should be (or VERY close to it) if it were random. So, perhaps you should maybe get some proof of your own before calling people idiots just because you decide to rant about the dice and they tell you otherwise.

[demon7896]

there is no way you can tell me the dice aren't completely random. i understand losing `12 to 1 once, cause that's just probability. but just listen, you are rolling one die and i am rolling three, ok so you beat me 9 times in a row, fine then. I just got sucky luck. SUCK IT UP CUPCAKES!! I dont win the lottery every other time i play, and all the dice analyzers show that people have the same odds on here every week. means.

fixed

[jecko7]

I agree with this guy.

I think that there is a seperate program that the admins made just for him. It makes the maximum roll for him a 3, and the minimum for his opponent a 4. It took some time, and effort, and the entire structure of the server for CC was altered, but the admins knew it was worth it because finally, canderson92 would get bad dice in some of his games.

They house canderson92's server in Area 51, and all dice rolls come from a sadistic android in Atlantis. This is all a seperate deal for him, but it came cheap since the sadistic dice-rolling androids work for magic beans.

OPEN YOUR EYES PEOPLE! (I think I had too much caffeine...)

[demon7896]

I agree with this guy.

I think that there is a seperate program that the admins made just for him. It makes the maximum roll for him a 3, and the minimum for his opponent a 4. It took some time, and effort, and the entire structure of the server for CC was altered, but the admins knew it was worth it because finally, canderson92 would get bad dice in some of his games.

They house canderson92's server in Area 51, and all dice rolls come from a sadistic android in Atlantis. This is all a seperate deal for him, but it came cheap since the sadistic dice-rolling androids work for magic beans.

OPEN YOUR EYES PEOPLE! (I think I had too much caffeine...)

lol

[Vyk]

The dice analyzer may show the dice are random over time, but that's not how it should be, the dice should be accurate game to game... Don't you see that? Thousands of games shouldn't matter because the problem with the dice are streakiness... the numbers shouldn't even matter when looking at the bigger picture... Maybe the dice numbers are the right %s over thousands of games, but how many times have you lost 10 in a row, rolling nothing higher than a 2 or 3???

What good is it that the correct %s are rolled over thousands of rolls when it comes down to you rolling 20 3s and 4s in a roll and your opponent rolling nothing but 5s. Period.

It's annoying and shouldn't happen. I just auto rolled 18 vs a 2 for no other reason then to get my card in an esc game, and I lost 15 and didn't kill 1. You're right though, the dice are perfect. *cough*

[Snorri1234]

The government is behind this conspiracy!!!11

Isn't it a little strange that just before I joined this site, George Bush's brother bought it? And then he profited from something related to making dice not random in a way that I can't explain right now because it's a secret.

CHECK THIS VIDEO!!!1






The truth is out there people! Wake up!




SAY NO TO THE NATIONAL ID-CARD!!! THE BEAST IS COMING!!




VOTE RON PAUL!!! FOR THE CONSTITUTION!!

[jecko7]

It's government censorship!

Think about it...the only people that get non-random dice are the ones that complain about it in the forum! That means that GW Bush is using the Minority Report system, predicting who will bitch about the dice BEFORE THEY DO, and then giving those people bad luck! It's 1984 all over again!

I think that lack should refund all points to all people that lost a game (or more), because the only way people lose games is when they lose a dice roll, therefore since those people lose more than the rest of us they must get worse dice than the rest of us!

I'm not sure which of these 2 scenarios makes less sense, but I'm pretty sure the suggestion that the dice are non-random makes negative sense. If that makes sense.

[KoE_Sirius]

Dice thread again...ROFLMAO

[sully800]

exacty that, to try to make the game more interesting. i admit that the dice are not random but bad both ways as long as the game is close but i have had more games that have ended in me trying to finish two or three armies with a dozen or more and it ended with me beating my head against the wall

if you can provide some kind of statistical analysis of the dice, not just a
"I attacked ## with ### armies and lost ## three times in a row!!!11!!!one!"
Then you might be able to convince a few people.
The mods/lack have said before, if you want it changed, you need to prove that it needs to be changed. A whiny "the dice suck" post does not count as proof.

and/or provide a better system to implement!

If you can prove that there is a flaw with the dice then lack will be forced to search out a solution. In two years of complaining from many people no one has ever been able to prove there is a problem. The other idea is to research a better way to generate random numbers than taking them from random.org. If you find a system that can prove to be more random then I am sure lack would be willing to switch to it.

[jangler3]

I've given up worring about the randomness of the dice. You win some, you lose some. On another thread someone said "Think about all the dice that spin at one time one." It's probally as random as your going to get.

[Top Dog]

For example a couple games ago I took europe on my first turn no losses

In a current game the dice are going horrible and I'm playing a private! So I hope I can last the game without a demotion at the end cuz of poor dice!

[Twill]

I KNOW the dice are random, I have the numbers for the whole thing so I know what the distribution is...now, I DONT know if there are streaks...

If you know how to visually or statistically represent "streakiness" let me know and I'll run the numbers on it to see if there is actually a streak.

[lancehoch]

I wonder if people have actually looked at "streaks" in dice that are hand-rolled. I mean, no one is rolling 75,000 dice like hecter did, but actually rolling dice would probably come up with similar results and similar streaks. Also, if there were no streaks, then the coding, or however it is done would not truly be random. Without streaks, the dice would be predictable and then people would be able to manipulate the system.

[Bean_]

Oddly enough, although the methodology used here is not correct, it may actually result in *less* streakiness than true randomness would suggest. Here is a thought experiment:

Consider a sequence of flipping 20 coins. Under true randomness, there should be a 2^-20 chance of having 20 heads, or 20 tails. Alternatively, the last 2 coin flips should have a 25% chance of 2 heads.

What CC does in effect is to run a random number generator once, and lock in the file. By analogy, we would have a "coin file" that has 20 flips, and let's say 10 are heads and 10 are tails. [The first time the file is generated, it is random; but once it's reused, the data set becomes fixed, and randomness is lost.]

Using this coin file, we could *never* get the result of someone having 20 heads. Similarly, the chance of the last 2 flips being heads is not 1/4, but actually 9/38. As the coin file grows larger, however, the chance of the last 2 flips being heads approaches 1/4. (Technically, assuming n flips, with n being even, the number of possible outcomes is n C (n/2), rather than n^2, and the chance of the last 2 flips being heads is [(n-2) C (k-2)] / [n C k], where k = n/2. This last formula reduces to (k-1)/2(n-1).)

Twill -- here is a way to test one aspect of streakiness of your file. Convert the A1A2A3D1D2 dice file into a simple yes/no test of whether that line wins a 3v1 attack. (This doesn't test D2, but is much easier to code than a 3v2. Basically, if max(a1,a2,a3) > d1, it's a win, otherwise it's a loss.) Then count up how many streaks of say, 6, 7, 8, 9, 10, 11, 12 losses there are, starting on each 20th line. Compare the number of streaks you get to the number expected from probability. Since there are 25,000 unique sequences of 20 lines, the expected number of 6+-loss streaks is 38.81, 7+-loss 13.21, 8+-loss 4.49, 9+-loss 1.53, 10+-loss 0.52, 11+-loss 0.06, and 12+-loss 0.02.

[Snorri1234]

What CC does in effect is to run a random number generator once, and lock in the file. By analogy, we would have a "coin file" that has 20 flips, and let's say 10 are heads and 10 are tails. [The first time the file is generated, it is random; but once it's reused, the data set becomes fixed, and randomness is lost.]

This would be true if it was. However, CC generates a new series of dice when the old ones are all used. (Well, they update the file after a few re-uses.)

And it is still basically random. I mean, the file is huge so it approaches true randomness in the best way possible.