Dice Complaints and Various Suggestions to Fix Them

[AndyDufresne]

If you could suggest some other random number generators, by all means do. But for now, random.org works rather well.


--Andy

[Metaphore]

I'm not suggesting a different generator, I'm suggesting getting every die from the generator and not de-randomizing it with a shrinking list of dice rolls.

As an example, picture this:

You pull a line out of the file at random. That line has 5 numbers, A1, A2, A3 and D1, D2. Even if (big IF) the numbers within the line are random, say they are sorted - A1 largest to A3 smallest and D1 larger of D1, D2. Even if all the numbers are randomly generated, the sort will throw rolls needing less than 5 dice way off. If there is only 1 defender, then you're not rolling 1 die, you're rolling 2 and picking the larger. I don't know if that is how it works, just giving an example how an algorithm BASED on a random number can be not random.

RANDOM.ORG in this instance is nothing more than name dropping - you're getting 1 random number and generating all others yourself. At best, this algorithm is unproven. At worst it is responsible for the number of 6's rolled by defenders with 1 army.

[BaldAdonis]

Or let's start with How Should the dice work?

Check the wikipedia Risk entry, there's a chart with all the rolls and the probability of winning each one. Using these and some simple Markov chains, you can work out the odds of winning at any time of any battle. For example, say that all you want to do is take over one territory, and you don't care how many men you lose. If the defender has 3 armies, then you have
a 2.7% chance with 2,
a 20.61% chance with 3,
a 47.03% chance with 4,
a 64.16% chance with 5,
a 76.94% chance with 6 and
a 85.69% chance with 7, courtesy of Open Office Calc.

[AtomicSlug]

Or let's start with How Should the dice work?

Check the wikipedia Risk entry, there's a chart with all the rolls and the probability of winning each one. Using these and some simple Markov chains, you can work out the odds of winning at any time of any battle. For example, say that all you want to do is take over one territory, and you don't care how many men you lose. If the defender has 3 armies, then you have
a 2.7% chance with 2,
a 20.61% chance with 3,
a 47.03% chance with 4,
a 64.16% chance with 5,
a 76.94% chance with 6 and
a 85.69% chance with 7, courtesy of Open Office Calc.

So basically, any player willing to count should shut up about the dice, right? I think so.

[ParadiceCity9]

i think the dice analyzer is somewhat no good, because it shows just the dice that have been rolled....but not what they were up against.

[Wild_Tiger]

I'm not suggesting a different generator, I'm suggesting getting every die from the generator and not de-randomizing it with a shrinking list of dice rolls.

As an example, picture this:

You pull a line out of the file at random. That line has 5 numbers, A1, A2, A3 and D1, D2. Even if (big IF) the numbers within the line are random, say they are sorted - A1 largest to A3 smallest and D1 larger of D1, D2. Even if all the numbers are randomly generated, the sort will throw rolls needing less than 5 dice way off. If there is only 1 defender, then you're not rolling 1 die, you're rolling 2 and picking the larger. I don't know if that is how it works, just giving an example how an algorithm BASED on a random number can be not random.

RANDOM.ORG in this instance is nothing more than name dropping - you're getting 1 random number and generating all others yourself. At best, this algorithm is unproven. At worst it is responsible for the number of 6's rolled by defenders with 1 army.

That is an interesting theory, I hope we can look deeper at that,
lack?

[AndyDufresne]

I think it's fine.


--Andy

[hulmey]

i could suggest a random generator for you but it costs money!!

[AndyDufresne]

I think we should look at free possibilities. We don't want lack charging up the premium price


--Andy

[max is gr8]

way back in my early games I won a battle against 80 with 2
I never complain anymore
even though my dice analyzer says I have 40% 1s

put it this way That one roll won me the game

[NESconqueror]

way back in my early games I won a battle against 80 with 2
I never complain anymore
even though my dice analyzer says I have 40% 1s

put it this way That one roll won me the game

WHOA! That is enough to put me out of the waters.

[max is gr8]

which part 40% 1s
or 2 v 80

[reptile]

We need to do something about new dice.. something that is a little more fair for everyone!

too many turns to i go 0 for x. already like 5 times this week i have had more than double the amt of armies as my opponent to take his bonus (two of wich were 9 vs 2 and 7 vs 1) and i lost. Every time i didnt attack with less than 3 dice! then i will watch them take their turns and they turn around and they not only have the balls to attack my army that has more than theirs and destroys it when they dont lose a single army or very few...

i know i dont get good enough rolls to ever ... ever try some of this stuff that is always working for my opponents against me.

So i dont think that everyone gets these bad of rolls... just by the way ppl play in games i watch. Very respectable players will recieve 4 armies to deploy... split them 2 in one area and 2 in another area and take two of my territories, or 3 in one area and 1 in another and with the 3 try and kill a 3 or 4 of my armies. And these are GOOD players. I wouldnt even dream of splitting the deployment up. i woudnt chance my rolls vs theirs to not deploy all 4 of my armies on the same area and just hope i can take one of their territories. This doesnt even bother me when i still dont kill a territory. But it does when they take thier turn RIGHT after i finish and i sit there and watch them take ridiculous amounts of armies with way fewer... that is when i realize how much we need a better "random" dice system for CC. here is an example, i accidentally made a game i was going to play against my buddy public. I am around the 2400 to 2500 point area, and a pfc comes in and joins the game. i take a look at the damage... sure enough with a deployment of 4, he takes 4 territories on his first turn!!!!!!!!!!!!!!!

no shit.... 812181

In this istance i am STARTING a game down 8 territories and only get to deploy 3 instead of 4.
Of course not to mention that is an automatic what? 30 points gone before i even start the game?

Can we please get a new "random" dice system?

i would do a vote, but i personally think that there are players that get really good rolls and others like myself that are consistently screwed. So i know my vote wouldnt pass.

Is it at all possible to switch our random dice server? or if not new dice at least make it to wher everyone in the game starts out with the same # of armies to deploy on the first turn, even if someone takes your territory # down below whatever they had. this will fix a lot of things!

let me know please.. cause too many games do i start out knowing there is no possible way i can win. i dont mind a little bit of luck to make things fun, but when the game is over before you start... where is the fun in that?

Serious question!

[misterman10]

Ok, make a new dice generating system, upload it to Conquer Club and quit your whining.

I hate little girls who whine about the dice

[Hyronimus]

The dice are based on high quality random numbers from Random.org. The numbers are read from a large file containing columns of numbers from 1 to 6, in the format A1 A2 A3 D1 D2. When the dice are rolled, the game engine reads a line from the file and discards it. The appropriate numbers are used and the others are ignored. The file contains 500,000 lines of dice rolls and is re-loaded when all the lines are used up. As of November 2006 we consume 125,000 lines of dice rolls per day.

Sounds pretty random to me.

Actually, it doesn't... The original random number might be random, but the latter playing with it, especially preventing lines from repeating and reducing the sample set in the process is highly suspect.

A long time ago in a distant laboratory I coded what's known as a Chi-squared test for randomness and distribution. All I remember from those days is if you f*ck with a random number it ceases to be random in most cases.

The right thing to do would be to get 5 random numbers. Generating 5 random numbers using an algorithm would be better and surely much faster than getting 1 from random.org plus file i/o anyway.

I would never have believed such a cock-eyed system would be used !!

if you were to do this in the flesh in a game of R**K with a sack containing strips of paper with sets of numbers on in place of dice. The complete un-randomness (and thus unfairness) would immediately be pointed out. 1. the strips in the sack need not necessarily be numbered randomly and 2. the poor bloke drawing his strips from the last few in the sack can only draw from what remains not from a random even spread of numbers or even number combinations !

infact anyone drawing a strip other then the first strip drawn is drawing from an incomplete set of values that is assuming the 500,000 cover evenly every number combination to start with !!

I must also agree with the guy that predictable rolls keep coming up ie 3V1 and the defender gets a 6 first time almost every time . I keep getting a pattern when the defender rolls a stupidly un random number of 6,4 either their is a glitch in the system or more likely as it seems Metaphore points out the system used sucks.

[Hyronimus]

The dice are based on high quality random numbers from Random.org. The numbers are read from a large file containing columns of numbers from 1 to 6, in the format A1 A2 A3 D1 D2. When the dice are rolled, the game engine reads a line from the file and discards it. The appropriate numbers are used and the others are ignored. The file contains 500,000 lines of dice rolls and is re-loaded when all the lines are used up. As of November 2006 we consume 125,000 lines of dice rolls per day.

Sounds pretty random to me.

Actually, it doesn't... The original random number might be random, but the latter playing with it, especially preventing lines from repeating and reducing the sample set in the process is highly suspect.

A long time ago in a distant laboratory I coded what's known as a Chi-squared test for randomness and distribution. All I remember from those days is if you f*ck with a random number it ceases to be random in most cases.

The right thing to do would be to get 5 random numbers. Generating 5 random numbers using an algorithm would be better and surely much faster than getting 1 from random.org plus file i/o anyway.

I would never have believed such a cock-eyed system would be used !!

if you were to do this in the flesh in a game of R**K with a sack containing strips of paper with sets of numbers on in place of dice. The complete un-randomness (and thus unfairness) would immediately be pointed out. 1. the strips in the sack need not necessarily be numbered randomly and 2. the poor bloke drawing his strips from the last few in the sack can only draw from what remains not from a random even spread of numbers or even number combinations !

infact anyone drawing a strip other then the first strip drawn is drawing from an incomplete set of values that is assuming the 500,000 cover evenly every number combination to start with !!

I must also agree with the guy that predictable rolls keep coming up ie 3V1 and the defender gets a 6 first time almost every time . I keep getting a pattern when the defender rolls a stupidly un random number of 6,4 either their is a glitch in the system or more likely as it seems Metaphore points out the system used sucks.

[Malkithe]

way back in my early games I won a battle against 80 with 2
I never complain anymore
even though my dice analyzer says I have 40% 1s

put it this way That one roll won me the game

WHOA! That is enough to put me out of the waters.

Random made up statistic
Sarchastic statement
Random made up statistic
Yawn.

[Itrade]

I say instead of complaining about the dice we come up with a game mode that eliminates them completely. Like a real-time rock-paper-scissors against these so-called random numbers. Now if we can just figure out how to make auto-attack compatible with this...

[Levezinho]

The admin of CC should analize the dice rolls, because i don't believe they are randomize like it says in the instructions... They just look forward to new maps, maps, maps, maps... We, members, think that allready exists suficient maps to play the game. Now, the dices sucked...
Are they forgoten that we are paying to play???
I think this deserve a resolution!!!

[Visaoni]

Yes. Because you have not gotten the dice you wanted, there is clearly a problem with the game. As much as you may try to get the whole "we pay" etc etc thing going, you forgot one major flaw.

Guess what? The person your attacking wants good dice too!

So, going by that logic, we should make everybody happy, and everybody should roll 6's every time! Sounds great doesn't it?

Oh wait... wait... I think I may be getting the feeling there is a problem with that... Oh yeah! The defender wins the tie! Wait. Why does that matter. I know! Games would go nowhere! Nothing would happen, ever!
/flame

On the serious side, if you really are determined to see this through, and have intelligence way beyond what you showed in your post, go get a sample size large enough to prove it is not 100% random. Good luck with that by the way.

Edit: Oh yeah. Also, as I did suggest you find proof of the dice not being random, I might as well ask you find proof of your other claims. You may want to PM every member of the site and ask them if they are happy with all the current maps and do not wish for anymore before the changes you want make the game impossible. Make sure to get everybody's responses too.

[john1099]

This will be moved in:
5
4
3
2
1
SHIP IT~!

[Rocketry]

We really need a dice complaints forum

Rocketry

[turtle32]

yes, actually this should be a asuggestion, then no one would have to look at this crap in other forums

[wcaclimbing]

We, members, think that allready exists suficient maps to play the game.

Sorry, but i LOVE new maps.
If they stopped making new maps, i would most likely leave this site.


And the dice are fine the way they are. I stole all of your good dice so you only get bad ones now.

[Syzygy]

I'm not complaining about the dice - what site do we use again though?