The site's dice are accurately pseudorandom.

ive shown my calculations several times... in 3 vs 1 441/1296 times the defender wins the roll its 1296 possible rolls 6^4 defender wins if he plays 6 everytime 6^3 =216 defender wins if he plays 5 5^3=125 defender wins if he plays 4 4^3=64 defender wins if he plays 3 3^3=27 defender wins if he plays 2 2^3=8 defender wins if he plays 1 1^3=1 216+125+64+27+8+1=441 normal (or Gaussian) distribution when you say every single dice roll you mean not conquering? or doesnt win a single troops to opponent? if the first it might be well possible because newbies attack 3vs3 and 2vs3 etc. if second its not very reasonable that opponent not lose 1 single troop in 3 games... i think once again im beeig misinterpreted i want to do positive criticism... theres a site that shows all your rolls since you joined and you can actually see the number of times you rolled 1,2,3,4,5,6 and the oppo stats, and there you know for sure it tends to all of them beeing equal in the long run... and this is really my last post dont bother asking anything else, sorry about disagreeing with you...

Say you are going up against 1 army, and you have 8 armies in the territory next to it. You attack, you lose an army and now have 7. Here is a fun fact: should you attack again, you have the same chance to win against your opponent as you had when you had 8 armies. In general, it is a unhealthy train of thought to consider the past attack and think that you are more likely to win. 3 dice v 1 followed by 3 dice v 1, is more likely that the attacker will win; only because there have been two attacks, not because the chance to win increases.

very good point Glanru

here is some good dice for you (its copyied from the game log so read from bottom to top) ------------------------------ Bolivia (marcoxa) attacks Argentina (Mike) conquering it, killing 1 troops, losing 0. 4 May, 22:00 Bolivia (marcoxa) attacks Chile (Mike) conquering it, killing 6 troops, losing 3. 4 May, 22:00 S. Central Brazil (marcoxa) attacks Bolivia (Mike) conquering it, killing 6 troops, losing 2. 4 May, 22:00 N. Central Brazil (marcoxa) attacks S. Central Brazil (Mike) conquering it, killing 6 troops, losing 0. 4 May, 22:00 N. Central Brazil was reinforced by marcoxa with 10 troops. --------------------- thats some good dice right there if you ask me

I know the dice work out in the end but holy snotburgers, Batman, do they need to be so lopsided one way or the other? Very hard to come up with an effective strategy. It makes kamikazee play more attractive...

As long as the dicegenerator is behaving like this here on this site i will never pay for this site. This dicegenerator is really abnormal and is against ALL ODDS. I know they all come up with I-don't-know-howmany explanations to proof their dicegenerator is 100% random and that there is also something they call clumping. But I tell you the clumps here are huge !!!!!!!!!!!! In fact if a picture is taking of the dicerolls here it will be full of dark black spots. And that goes especially for the attacker. In one game severals clumps of 15 to 2 and the 15 are lost and the 2 are still there. Come on man; that's impossible; it's like you win the jackpot-lottery 5 times in a week !!!!!!!!!!! It's possible but very very very very unlikely. And I really mean what I say: as long as the dice are so weird here I will never pay for it. I played several risk-games on the internet and there is NO SITE were there is so much complaining as here regarding the dice. Nor do I want to win if I attack with 15 and the other has 20 and he looses all and I dont loose any nor do I want to win if I defend with 10 and the other has 30 and he looses all and I am still there with 10; for that happens also here; and not once a while but repeatedly. It just kills all the tactics in the game. I want to win a game due to my tactics and not because of impossible-against-all-odds good dice of mine or impossible-against-all-odds bad dice of the others. It's a pity ...........

There is a plan for having the option to choose biased dice, who will give less of these 'extreme' results, like on some other sites. However, real dice just give these kind of results, wheter you like it or not. I agree with you that sometimes it ruins a game. But to me it also adds beauty.

I personally agree with you tryitonce, I hate real dice. I only want to play with fake biased dice that are only half as random. The reason I started this thread is to prove that the dice are perfectly random. Most players seem to be arguing otherwise when what they should be arguing is that they don't like perfectly random dice.

I'll point out that the game is called "Risk" and not "Certainty"

In my opinion strategy should play a major part in this game and not the randomness of the dice. What use is it to create ranks if luck plays a major role in this. I think it's good to have the choice of how the dice will be work out. So strategy can play a more important role in the game. I will see when and how the biased dice will work out in practice.

I have a solution: take out the "random" dice and replace them with the fake biased dice, only don't tell anyone. I bet no one will complain ;)

I'm trying not to be bias, which is a process I don't think anyone is truly capable of, but I am trying to learn more about this topic. After doing some research(and yes I will continue to do more because you can't believe everything you read), I came across a site called "Random.Org" and it shows the difference between a Pseudo Random Number Generator(PRNG) and a True Random Number Generator(TRNG). In the opening post Vexer explains that a PHP is used to generate the randomness. According to this site(and it explains why) a PHP is a PRNG and not a TRNG. On a side note, according to several other sources the PHP used here - you said the mt_rand - is one of the best PRNG's but it is still not a TRNG. Let be clear, though. I am NOT a computer whiz and I don't pretend to be. I don't even really know how to post a link to this site without asking my kids for help:) I just came across this info while using 'the google.'

Yes, Elysium, what you are saying is true, but it is impossible to write a True Random Number Generator with code only, because there is no way to ask the computer for "give me a random number". If I remember correctly, the Random Generator of Random.org uses a think that counts particles caught by some kind of sensor and the amount of particles sent out it more or less a random process that we can find in our universe, meaning that the Random.org generator is in essence a True Random Number Generator. Well then, why are we not using that then? There is a very simple explanation for that: speed. This site has used random.org as random generator for a while. The consequence of this however is that attacking was incredibly slow all the time. You sometimes had to wait at least a full second for the dice to be rolled and this is of course a big problem when playing live games or having a game where troop numbers have reached hundreds. It was therefore decided to use the PRNG mt_rand. I am not completely sure how it works, but in short it calls a special function that has a seed as input and has a nearly-random output. The seed is usually chosen as the current time. Using the sine function for example it is fairly easy to create one of these functions yourself (though I can't remember the exact function). Using the mt_rand function has hugely sped up the attacking during games. The mt_rand function is indeed not Truely random, but believe me: it is very close. If you are interested I suggest looking up some information about mt_rand and while I am not really into the subject, I am pretty, pretty sure that it comes very close to a TRNG, so close that you won't notice any difference. EDIT: On [url=http://en.wikipedia.org/wiki/Mersenne_twister]this page[/url] about Mersenne twister (the algorithm that mt_rand uses), you can find more information if you are interested, but I especially want to point people to the "advantages", where it states that this algorithm passes almost every random test.

Thanks, Cireon. I was wondering if there was a possibility that the amount of 'clumping' referred to might have to do with the small amount of numbers (1-6 I think I read, but I really don't know how it works) in the random generator. The explanation lead me to believe that the more input that was used in the generator, the more it would allow a more accurate 'randomness'. I believe that the numbers do work out pretty close in the end but I also believe that there is an extra amount of 'clumpiness' due to the fact that it is a small amount of numbers that are input into the generator and that could be a contributing factor in the amount of 'clumps' that occur seemingly more often than not than when using real dice. Maybe that is why people seem to think the dice are screwy because the 'random clumps' happen more often than they would if it were truly random. And I have no idea how to suggest this could be fixed. Does anyone with programming knowledge have any suggestions regarding how the input could be used to lessen the 'clumps?'

We could make the input larger (say... choose a number between 0-599) where we look at the number modulo six (where 0 will result in a die roll of 6). Theoretically this shouldn't make a difference and like I said, I did not really look into it a lot, so I have no idea how these clumps work...