I just played 3 levels of T2 with 1 other person and this was the random distribution of materials:
Them:
20
Me:
9
They got over twice as many materials as me, anyone know how that is random?
With 2 people you think it would be 14 and 15 materials as you have a 50-50 chance of getting it but it doesn't seem that it works that way.
Hey vhett, this is indeed random. In a random system, the distribution will not necessarily be equal at every individual point of measurement; however, over time the distribution will even out.
Flip a coin 10 times and keep track of heads and tails. Will you always, always get a 5/5 split? No, you actually have a fairly good chance of a 4/6, and a decent chance at 3/7. Who knows, you might even pull a 1/9.
If you want proof of the randomness of the system, you should keep track of the distribution of materials every time you venture down into the clockwork.
On a less technical level, who could possibly benefit from rigging the system to screw you over? What could anyone gain from an elaborate conspiracy to give you less mats? Did your partner pay off the devs to get 11 extra Blue Shards and Scrap Metals?
And if you're too sensitive on mats distribution (and not too lame to share when you're on the other end of the random) then consider making a party with friends / guild mates. That way you can always ask them if you can get a certain drop that "random" assigned to them. If they're really friends/mates they'll share. :)
For those who like probability, the odds of getting 9 or fewer drops out of 29 is 3%. Unlucky, but not bizarrely so.
Is that assuming the distribution is normalized? If do, I'm assuming you used a CDF examination? :P
Why are all these kids posting these ignorant issues. If you learn anything in mathematics or statistics, you would learn that random doesn't always end up that every number is chosen equally. It will only look like it's even if your party collected a lot more mats than that. When you are dealing with random in small quantities, it can be lopsided. However, if you deal with random in a much greater quantity, it will eventually iron out and be close to even.
And this may be a bit too complex for you kiddies but since this is a java based game, I suppose its based on java.util.random. So trust me, it's random.
Schrödinger's Cat programmed the loot distribution.
So unless you can accurately tell the rate at which theoretical sub-atomic particles you cannot see deteriorate, you have the same chance of receiving an item as not receiving it. Or more accurately, you have both received and not received the item, until it is distributed.
Until it's distributed, it is a Quantum Item.
Wikipedia on Binomial Distributions
I ran the numbers, just cause I was curious what the chances of a 20/9 split are. Assuming the chance each material will go to you is exactly .5 or 50%, it's uncommon, but not as unlikely as you might guess. (~1.87%)
The odds that the split would have been more lopsided in your partner's favor? 21/8, 22/7.... are at about (~1.21%)
If you're wondering how often a party of two would see a split as lopsided as yours or worse (either in your favor or against you), the chance is about (6.14%) or more than 1 in 20 runs.
Rotten luck for sure, but no evidence of a bug or anything like that. Take comfort in the fact that the odds of a 29/0 split are about 1 in 537 million.
It's possible that OOO uses a different formula that tries to balance out the material distribution, in which case you were just that much more unlucky. But there's no way to know if that's the case without gathering distributions for a large number of runs.
Why don't they just alternate who gets each material? Well, that system would be easy to game. Lets say you see a 5* mat like sunsilver drop along with a scrap metal. If you remember who got the last material, you could easily ensure that you get the sunsilver while your partner's busy fighting monsters (exactly the sort of situation this random material assignment was designed to avoid). Anyway, with any sort of random assignment, some people are bound to get unlucky. In the long run though, it should balance out.
Let me just reiterate something. This is a JAVA game. It means that it's coded in java. They use the class Random which comes from java.util.random. There are already algorithms used to create the class Random. If you really have a problem with "Random", blame Oracle, not Spiral Knights.
"For those who like probability, the odds of getting 9 or fewer drops out of 29 is 3%. Unlucky, but not bizarrely so."
"Assuming the chance each material will go to you is exactly .5 or 50%, it's uncommon, but not as unlikely as you might guess. (~1.87%)"
Just to reconcile these two numbers: 1.87% is the chance that you will receive _exactly_ 9 out of 29 drops. 3% is the chance that you receive _at least_ 9 out of 29 drops. I don't find the first number as useful -- you don't have a better than 14.4% chance of getting _any_ specific number of drops, but (as you might expect), you have a 50% chance of getting at least 14 out of 29 (one of the two of you is going to get more loot, and it's equally likely to be either of you).
I absolutely love that I didn't have to fly into this thread as the Mathamagician and explain the concept of "random." This makes me feel all warm and fuzzy.
I am not being sarcastic.
Let me just reiterate something. This is a JAVA game. It means that it's coded in java. They use the class Random which comes from java.util.random. There are already algorithms used to create the class Random. If you really have a problem with "Random", blame Oracle, not Spiral Knights.
I don't know of a developer quote about SK, but in another game that OOO has (puzzle pirates), the developers have very strongly hinted that they don't use the default Random class that comes with Java. This is good, because the default Random class is a really really bad RNG. Why Sun chose to use it is beyond me, that class of RNGs has been known to be bad since the 1970s, and they didn't even choose very good parameters for that it. Vastly better RNGs can be implemented in not much more code and be almost as quick.
And, I too am thankful that several people understand probabilities in this thread.
Did this person started in Moorcroft Manor too? Maybe he/she was going all the way up from Haven and you just joined later (on T2). That way she had listed mats gained during T1.
i just some-what started and i got the basic idea of this game, but what the hell is up with the "clockwork" stuff?
Well the idea is the world is a giant clockwork mechanism, twisting and turning constantly. That's why the gates rotate on us, and why each level of the strata may switch as you go down the floors.
Lets have more studies with sample size 1! this is great science!