Over the last couple weeks I tried to craft 11 BIS items with dragon eggs using all mithral assets (60% chance of tier 3, 40% of tier 2). Of the eleven, only three ended up being tier 3, which is under 30%. Am I the only one noticing this?
Here's a thing about statistics and sample size. To get CLOSE to the exact % of anything, you need a very large sample size to see it.
If you flip a coin 4 times and only got head once. That's 25%, and not 50% like it's suppose to. Even if you do it 10 times, you won't be guaranteed to get heads 5 times.
Have the gear and skills, but lack the friends to play with? Come and apply for Essence of Aggression. We have been here and strong since beta. (Immature, rude, and arrogant people will not be accepted)
I only brought this up, well honestly because I'm pretty unhappy about it, but because the probability of this occurring is about 3% (with a fair amount of uncertainty with that small of a sample size). The probability of failing 7 times in a row is much lower, less that a tenth of a percent using a binomial distribution. I was curious if there were any other rarities, but it is mainly just me venting
to put it simply, what ever they % stated is the % each time you try. So when you open a skill node you have a 25% chance to fail every time you try to open it. Fail the first time, you have a 25% chance to fail the second time and so on. Yes it would be awesome if your tries stacked but in fact they don't.
I will admit that the RNG does seem to be off a little bit lately as more often then not I seem to need multiple tries to open a skill node. So much so that im half expecting to fail a 100% chance to open one,
I spent time and money to get a profession on 3 of my characters to level 25 with purple masters and tools. Then I find out that even with the highest level and best workers/tools, resources I still only have a 60% chance of getting the top stufff. And that is the problem, like many others I have tried crafting many thing, some I have had terrible results, others better. But bottom line, would you go to any craftsman in RL to have something made or something repaired and except it if they told you you only had a 60% chance of getting it done right. Come on this is taking the gambling part of this game to new lows. If you go to a master craftsman, using best resources and best tools you expect a superior product, if not they will either keep workig on it or give you a discount. I can except some chance there will be a blemish, a 40% chance is totally unacceptable. Given the cost of dragon eggs I have completed stopped trying to craft. I have the ablity to craft under armor gear in leather, Chain and plate for our guilld, and it like all the crafting now just seems like poorly designed system.
It has been said before, the problem is not in the chance of success or failure. But in the streaky nature of number generator. No one knows what data Cryptic is using to generate the RNG seed. But when you get a good or bad seed, the effect is similar to being Wi Flagged. The only solution is to break the streak by doing something different. Change enough conditions, and you might just get a new seed, breaking the streak and resetting your odds.
to put it simply, what ever they % stated is the % each time you try. So when you open a skill node you have a 25% chance to fail every time you try to open it. Fail the first time, you have a 25% chance to fail the second time and so on. Yes it would be awesome if your tries stacked but in fact they don't.
I will admit that the RNG does seem to be off a little bit lately as more often then not I seem to need multiple tries to open a skill node. So much so that im half expecting to fail a 100% chance to open one,
Lately? The RNG has been like this since open beta. The 25% failure chance has been more around 65% for a long time now.
I for one will suspect that things are not working right until we see a roll with every drop in game. If I have a 75% chance at success doing a skill roll. I want to see a roll giving me a number showing me that I failed. This should be for all skill nodes, upgrades, unlocked chests from leadership or lockboxes....Everything.
That would be the only way I could trust Neverwinters RNG calculator.
to put it simply, what ever they % stated is the % each time you try. So when you open a skill node you have a 25% chance to fail every time you try to open it. Fail the first time, you have a 25% chance to fail the second time and so on. Yes it would be awesome if your tries stacked but in fact they don't.
I will admit that the RNG does seem to be off a little bit lately as more often then not I seem to need multiple tries to open a skill node. So much so that im half expecting to fail a 100% chance to open one,
Lately? The RNG has been like this since open beta. The 25% failure chance has been more around 65% for a long time now.
I for one will suspect that things are not working right until we see a roll with every drop in game. If I have a 75% chance at success doing a skill roll. I want to see a roll giving me a number showing me that I failed. This should be for all skill nodes, upgrades, unlocked chests from leadership or lockboxes....Everything.
That would be the only way I could trust Neverwinters RNG calculator.
I disagree, before mod 6 it was rare for me to fail multiple times on a skill node. That's not to say it didnt happen, just that I succeeded more often then failed. Now its the complete opposite, I seem to fail at least once on every node.
The way the Game is Coded I wouldn't doubt they have RNG Tables that they use as a look up and there isn't actually an RNG just a seed that sticks you to a certain set of tables
Typically it's an iteration algotithm, where you use the last roll as seed.
...like:
ROLL = ( Last ten digits of ( LASTROLL x 283467298473847873 ) ) / 10000000000
Sure, the actual algorithms are often a wee more complex. And there are quite some of those, certified for randomness and yadda yadda, used in computerized gambling slot machines. Bally is e.g. one company that does writing and certing of those...
The results given are technically repetitive, and hence not truly random, but pseudo-random.
BUT: Streaks are actually a characteristic of better randomness than that evening out many people expect to happen. Also, with the "n failures in a row"-thing... ....if that's balanced by n x chance successes on first try, all is fine in the state of Denmark.
There's quite a bunch of (googleable) informative and enlightening webpages, vids, and discussion threads if you care to learn more.
Comments
If you flip a coin 4 times and only got head once. That's 25%, and not 50% like it's suppose to. Even if you do it 10 times, you won't be guaranteed to get heads 5 times.
Have the gear and skills, but lack the friends to play with? Come and apply for Essence of Aggression. We have been here and strong since beta. (Immature, rude, and arrogant people will not be accepted)
I failed 7 times on one skill node.
Each of fail is 25%
So pro above plz give me the possibility on 7 fails in a row?
I will admit that the RNG does seem to be off a little bit lately as more often then not I seem to need multiple tries to open a skill node. So much so that im half expecting to fail a 100% chance to open one,
I for one will suspect that things are not working right until we see a roll with every drop in game. If I have a 75% chance at success doing a skill roll. I want to see a roll giving me a number showing me that I failed. This should be for all skill nodes, upgrades, unlocked chests from leadership or lockboxes....Everything.
That would be the only way I could trust Neverwinters RNG calculator.
...like:
ROLL = ( Last ten digits of ( LASTROLL x 283467298473847873 ) ) / 10000000000
Sure, the actual algorithms are often a wee more complex. And there are quite some of those, certified for randomness and yadda yadda, used in computerized gambling slot machines. Bally is e.g. one company that does writing and certing of those...
The results given are technically repetitive, and hence not truly random, but pseudo-random.
BUT: Streaks are actually a characteristic of better randomness than that evening out many people expect to happen. Also, with the "n failures in a row"-thing... ....if that's balanced by n x chance successes on first try, all is fine in the state of Denmark.
There's quite a bunch of (googleable) informative and enlightening webpages, vids, and discussion threads if you care to learn more.