Guild Wars 2

Buttercup.5871:

Thanks Mirta, for your elaborate thesis on why pseudo RNG is not the problem. Did you read anything in this thread before you started posting? Because I’ve never implied any “tampering” anywhere. If you have the knowledge to explain to me why a pseudo RNG system is not problematic for the manner in which this game uses it, I’m all ears, I truly am. But if not, please stop trolling.

okay, how about, I did games design and predetermined is not how you do it? You put at least 3 layers of random on it to begin with and I made a simple example in past pages of the thread. Please give me the code to GW2 now and show me their RNG systems and why you believe it’s pre-set.

Mirta.5029:

Buttercup.5871:

Thanks Mirta, for your elaborate thesis on why pseudo RNG is not the problem. Did you read anything in this thread before you started posting? Because I’ve never implied any “tampering” anywhere. If you have the knowledge to explain to me why a pseudo RNG system is not problematic for the manner in which this game uses it, I’m all ears, I truly am. But if not, please stop trolling.

okay, how about, I did games design and predetermined is not how you do it? You put at least 3 layers of random on it to begin with and I made a simple example in past pages of the thread. Please give me the code to GW2 now and show me their RNG systems and why you believe it’s pre-set.

ah my idea was validated to be possible yay

sorry for going off topic a bit

predetermined is silly because with enough data people would figure it out and then know exactly how to get items leading to a ton more than ANET intended to be in circulation

I want to understand how the word “unfairness” gets involved in a strict gambling situation.

My main issue with the whole deal is that this is a game which after everything is said and done is supposed to be fun for those that engage with it. The current implementation of RNG allows for some users of the game to be basically be “kitten” upon with no regard. That is not fun and goes against the core ideology of the game. That is a major flaw imo, one that definitely warrants correcting.

Hence the HoT task-based alternative that’s been in discussion/development almost since the day the game launched.

Using a cumulative binomial calculator, the odds of getting 1 or more pre from 350 attempts is 50%….

Far too many players actually believe this “math”. The 50% is horrifically misleading and the unfairness is that the OP believes he has found some magic number of Mystic Flushes to get a Precursor and in reality he already KNOWS the odds…..0.2% when tossing Rares. How many times you toss does NOT change that (very low) possibility.

It’s an issue of wishful mathematics that causes all these “Mystic Forge hates me!” posts.

Sam Gem.1246:

Whatever. I’m staying the course.

There is no “course”.

Your previous attempts have not changed your likelihood of success in the future by any percentage. Zero. Non-relevant. Discrete events. User-error detected. Gambler’s Fallacy wins again.

Just like having a precursor drop doesn’t magically reset the number of attempts counter before your next one.

The other day, I ended up with Storm after throwing in only 4 random rares. RNG is RNG.

They’re changing the way precursors are dropped now.

Seems to me you just got a gambling fix you wanna shake off pal, not an actual hate for the game

That a variable number of people are accessing the same table/function between requests for a singular user also adds a largely “real” level of variability.

Nike.2631:

Sam Gem.1246:

Whatever. I’m staying the course.

There is no “course”.

Your previous attempts have not changed your likelihood of success in the future by any percentage. Zero. Non-relevant. Discrete events. User-error detected. Gambler’s Fallacy wins again.

Just like having a precursor drop doesn’t magically reset the number of attempts counter before your next one.

yes, probability is instaneous,

the chance of him getting it before attempts were made may have been 90%
but with every fail the probability changed.
the chances of me losing the next 499 times are..
the chances of me losing the next 498 times are..
the chances of me losing the next 3 times are..

basically, for those attempts he fell into the unlucky category, for the next attempts he is in the who knows but its fairly unlikely category.

mtpelion.4562:

Draknar.5748:

Mirta.5029:

throwing a coin, your chances are 50% heads, 50% tails, however you could easily get heads 20 times in a row before you see tails.

I wouldn’t say easily. While each independent flip technically is a 50/50 chance, the probability of getting 20 in a row is (1/2)^20, which is a reeeeaallly small chance.

The probability of getting any specific combination of 20 in a row is the exact same. The issue is that we’ve deemed some specific combinations to be preferable to others and thus are unhappy when those specific combinations do not appear.

Well obviously. The 20 <insert coin side> in a row is a specific combination we are looking to achieve. It’s the same probability of getting HTHTHTHHTHTHHTHTHTHT in that exact order, or any order so long as it is a pre-determined order you are wanting, as you said.

Sounds like we are on the same page. Saying one could easily see 20 heads before a tails implies getting 20 heads in a row, which is highly improbable and certainly not easy.

Just got another Dawn from the forge, my 9th pre since christmas.

I must be ridiculously lucky then, I have not tried my hand with the mystic toilet much.

But I have managed to get 2 pres from it so far, and both happened within the first couple tries.

First was when I got dawn, I threw in roughly 20 rare GS into the forge and got it on my 2nd try.

Then dusk, I bought 50g worth of exotic GS and got it on my 4th try.

http://blog.minitab.com/blog/understanding-statistics/difference-between-probability-and-cumulative-probability

Also check out the links at the bottom of the blog. Especially the one on gamblers fallacy.

By my (probably flawed) calculations, 1100 tries would give you ~88% chance of obtaining a precursor. Assuming you spent 1.5 gold per try, you could have probably bought the precursor.

Dont use mystic toilet, they made that for addicted gamblers so they can earn money.
Didnt you watch south park ?

And why would you even gamble with something that has so little chance…
Gamble only if you have 50% or more chance. Or be smart and never gamble.

Biohazard.7523:

Dont use mystic toilet, they made that for addicted gamblers so they can earn money.
Didnt you watch south park ?

And why would you even gamble with something that has so little chance…
Gamble only if you have 50% or more chance. Or be smart and never gamble.

I’d like to see a casino that gives a 50% or higher odd in any game, I’d play there in a heartbeat. Even roulette isn’t 50/50 when playing black/red (0 and 00 are green spaces).

Had a friend the other day throw in a stack of 250 hammers and pulled a precursor out of it, he sold it and went to try again with pistols for the hell of it a couple hours later. Another stack of 250 pistols and he pulled out the precursor, sold that and decided to try staves, ended up with a lot of exotics.

While everyone will say RNG!, the rng in this game always seemed messed up in some way. I wouldn’t be surprised if they had an account reward level attached to the accounts much like fractals and some accounts were marked up a lot more than others.

Xenesis.6389:

While everyone will say RNG!, the rng in this game always seemed messed up in some way. I wouldn’t be surprised if they had an account reward level attached to the accounts much like fractals and some accounts were marked up a lot more than others.

just one short little question. why should they do this? what is the benefit of it?

Buttercup.5871:

“PRNGs are typically also periodic, which means that the sequence will eventually repeat itself. While periodicity is hardly ever a desirable characteristic, modern PRNGs have a period that is so long that it can be ignored for most practical purposes.”

The length of the periodic sequence is huuuge, and it’s nothing that makes the results predicable so easily. Especially if another randomicity like order of execution is put into play.

The reason why there are applications for true random generators is because some use cases like cryptographic key generation have strict requirements as some algorithms may be weak if fed with weak keys (i.e. keys with a low so called entropy). I know this, since I am a software architect in a team that creates firmware for hardware security modules, we use true and pseudo random generators.

Don’t let yourself be set on the wrong horse. Everything here is working as intended and from the figures presented I cannot see a problem with the RNG implementation.

Draknar.5748:

Biohazard.7523:

Dont use mystic toilet, they made that for addicted gamblers so they can earn money.
Didnt you watch south park ?

And why would you even gamble with something that has so little chance…
Gamble only if you have 50% or more chance. Or be smart and never gamble.

I’d like to see a casino that gives a 50% or higher odd in any game, I’d play there in a heartbeat. Even roulette isn’t 50/50 when playing black/red (0 and 00 are green spaces).

At least casinos state odds and will most likely comp your room or something along those lines if you lose big. What conclusions can we draw…….Casinos have a bigger heart than Anet? lol

In order to beat RNG, you just need to throw more attempts at it. On TC, there’s a guy that got over 300 Precursors from the MF. I can’t imagine the amount of Rares and Exotics that went in for that amount.

Xenesis.6389:

Had a friend the other day throw in a stack of 250 hammers and pulled a precursor out of it, he sold it and went to try again with pistols for the hell of it a couple hours later. Another stack of 250 pistols and he pulled out the precursor, sold that and decided to try staves, ended up with a lot of exotics.

While everyone will say RNG!, the rng in this game always seemed messed up in some way. I wouldn’t be surprised if they had an account reward level attached to the accounts much like fractals and some accounts were marked up a lot more than others.

How many people have done the same as him and succeeded/failed? People always focus on the successes/failures for specific individuals rather than collectively across the entire player base.

All accounts are treated the same. What I mean by this is that there is no account specific variable that influences your outcome.

Sam Gem.1246:

I’m a logic person. After buying 3 precursors, I have done the math, using statistics and the experience of other people to persuade myself to use the mystic forge instead of buying the precursors. Basically, the odds of getting a pre is .2 % from rares. Using a cumulative binomial calculator, the odds of getting 1 or more pre from 350 attempts is 50%….
Well, after 1100 attempts, I’ve convinced myself to quit. I know.. 100 trolls will say good ridden. Whatever. I’ll never buy an Anet game or all other affiliates. You have wasted my time enough already…

You did your math wrong. You do not have 50% chance of getting one after 350 tries, you have 0.2% 350 times. Your chance doesn’t go up because you do it more often.

The thing players don’t take into account is that your individual attempts aren’t special and separate from everyone elses. By this I mean, that while you are making these attempts, so are others, and that all rolls into the bigger picture of the odds of getting the lucky bounce. It’s not calculating individual odds, it’s odds overall, so while you may contribute 1000 attempts, that 1000 is merged into all the others, so now maybe there’s 500,000 attempts that those .01% success rate is distributed amongst them all.

People tend to look at it as if the odds are specifically only going to be focusing on their attempts alone in the distribution of successful results, and that’s not the case.

The thing is that individual attempts SHOULD matter.

This is a game. You can’t have indivual players not having fun and ejoying it because of some ‘big picture’.

The game can’t be fun just for some, and the rest just there to watch those few have the fun.

The very moment you can’t get something regardless of skill and dedication, that’s something to fix.

RNG cannot exist alone. Every item must be acquirable with at least either skill or dedication, and RNG being there just for nice extras. RNG must NEVER be the one and only way to acquire anything. This includes indirectly by allowing players to sell somthing they got from RNG in the first place. Something being tradable is never an excuse for something being RNG-only.

Raw RNG alone without filtering and control should not exist in any game. Because it makes things as boring as a slot machine. There must always be some way to weight chances, like specific spots to have a higher chance for something to drop, or being able to salvage and craft what you never get.

There’s people who are being fooled into enjoying things like gambling slot machines, but that’s just a flaw in the human brain’s reward system that gambling exploits.

Hopefully the pre-legendary masteries with the upcoming HoT will address this issue, by adding a way to aquire pre-legendaries through dedication and skill.

Ayrilana.1396:

How many people have done the same as him and succeeded/failed? People always focus on the successes/failures for specific individuals rather than collectively across the entire player base.

All accounts are treated the same. What I mean by this is that there is no account specific variable that influences your outcome.

There would be no point to it, it’s just one of those things that make you go hmmm on a daily basis. Especially when you walk into a fractal with a friend and she pulls out 5 exotics and an ascended from drops and you walk out with rares even though same personal level and 30% more mf, that’s happens to her often and for me it doesn’t at all.

Just one of those things that make you think something is up with the rng and magic find in the game. I heard you like rng so we at arenanet have put rng on your rng under your rng in your rng for your rng.

Sam Gem.1246:

It’s like I said. It’s possible to never get a pre, but at 0.0018 chance from other people’s experience, the odds were in my favour with 1100 attempts.
90% chance is nothing to sneeze at…

10% fail chance is however still substantial.

Buttercup.5871:

Now, with your experience, can you explain how what you say aligns with what I read here:

“PRNGs are typically also periodic, which means that the sequence will eventually repeat itself. While periodicity is hardly ever a desirable characteristic, modern PRNGs have a period that is so long that it can be ignored for most practical purposes.”

What I’m ready here is two things: 1) PRNG’s are very long and 2) PRNG’s consist of a sequence that eventually repeats itself.

There’s one thing that is here that you didn’t get (that period is so long that it can be ignored for most practical purposes). There was also one thing that wasn’t there that is also important: that there’s another rng layer added to it. You see, you assume that whenever you make an Mf attempt, you draw a number from PRNG sequence. That is true. What you seem to miss is that other rng rolls for you also draw from the same sequence, and, what’s more important, other players draw from it as well. There is really no way to tell how many rng rolls were made in the game between your two MF attempts, and each of those rng rolls moves the sequence. The end is likely a random generator of better quality than you throwing the dice.

OP knows about the RNG involved in mystic toilet → proceed to try his luck → complains about result

Am I the only one amused by the fact that during the conversation the OP keeps trying to make very logical, numbers based arguments, but then concludes his account is cursed?

So, which is it, are we dealing with supernatural circumstances, or cold statistical accounts?

phys.7689:

yes, probability is instaneous,

the chance of him getting it before attempts were made may have been 90%
but with every fail the probability changed.

the chances of me losing the next 499 times are..
the chances of me losing the next 498 times are..
the chances of me losing the next 3 times are..

You know the chance of the run failing to produce a success GOES UP for every failure, not down, right?

Your odds of getting a precursor in the course of 500 tries are never better than on the first try. Because after the first try, you’re just trying to get one in a run of 499 attempts. Then in a run of 498 tries.

You are losing chances, not driving the non-existent God-of-Fairness to cut you a break.

MithranArkanere.8957:

The thing is that individual attempts SHOULD matter.

This is a game. You can’t have indivual players not having fun and ejoying it because of some ‘big picture’.

The game can’t be fun just for some, and the rest just there to watch those few have the fun.

The very moment you can’t get something regardless of skill and dedication, that’s something to fix.

RNG cannot exist alone. Every item must be acquirable with at least either skill or dedication, and RNG being there just for nice extras. RNG must NEVER be the one and only way to acquire anything. This includes indirectly by allowing players to sell somthing they got from RNG in the first place. Something being tradable is never an excuse for something being RNG-only.

Raw RNG alone without filtering and control should not exist in any game. Because it makes things as boring as a slot machine. There must always be some way to weight chances, like specific spots to have a higher chance for something to drop, or being able to salvage and craft what you never get.

There’s people who are being fooled into enjoying things like gambling slot machines, but that’s just a flaw in the human brain’s reward system that gambling exploits.

Hopefully the pre-legendary masteries with the upcoming HoT will address this issue, by adding a way to aquire pre-legendaries through dedication and skill.

and exactly this they will do with the addon and maybe the thought of “i want this pre or cannot enjoy this game” is a flaw in some brains too

Nike.2631:

phys.7689:

yes, probability is instaneous,

the chance of him getting it before attempts were made may have been 90%
but with every fail the probability changed.

the chances of me losing the next 499 times are..
the chances of me losing the next 498 times are..
the chances of me losing the next 3 times are..

You know the chance of the run failing to produce a success GOES UP for every failure, not down, right?

Your odds of getting a precursor in the course of 500 tries are never better than on the first try. Because after the first try, you’re just trying to get one in a run of 499 attempts. Then in a run of 498 tries.

You are losing chances, not driving the non-existent God-of-Fairness to cut you a break.

yes that was exactly what i was pointing out. he started out at 90%, and with every fail he was further down the road of being in the 10% that lose given that many tries

Then you did know. Carry on .

Buttercup.5871:

Halvorn.9831:

Buttercup.5871:

“PRNGs are typically also periodic, which means that the sequence will eventually repeat itself. While periodicity is hardly ever a desirable characteristic, modern PRNGs have a period that is so long that it can be ignored for most practical purposes.”

The length of the periodic sequence is huuuge, and it’s nothing that makes the results predicable so easily. Especially if another randomicity like order of execution is put into play.

The reason why there are applications for true random generators is because some use cases like cryptographic key generation have strict requirements as some algorithms may be weak if fed with weak keys (i.e. keys with a low so called entropy). I know this, since I am a software architect in a team that creates firmware for hardware security modules, we use true and pseudo random generators.

Don’t let yourself be set on the wrong horse. Everything here is working as intended and from the figures presented I cannot see a problem with the RNG implementation.

So there is a predetermined sequence. If you have one sequence per day and every millisecond represents a number, of course that number will be huge. The point is, when does that sequence reset, and at what moment(s) your sequence gives you the best odds of pulling a precursor? If you can find that out, and from what I’m reading from you it can be done, RNG is merely a pool of data rather than a roll of the dice. I’m also wondering whether the pRNG system your company uses has the low level of complexity required to accommodate tens of thousands of users without bogging down the servers to a complete standstill. Apart from the fact that your company is probably not stuck with pRNG architecture from 2008, cause from what I’ve read a whole lot has changed since then.

just to think about, you have a segment of predetermined numbers, and you have another segment with predetermined numbers, but shorter, it will reset befor the other. now they will take numbers from both as a variable and generate a new number. nearly invinite possibilities. i really think you underestimate programers, there is no 1 segment of numbers that will repeat after a while, normaly they use prng only as a variable in another equation, thats called a hybrid random number generator

http://www.appliedmathgroup.org/pubs/hrng.pdf

Buttercup.5871:

If you have the knowledge to explain to me why a pseudo RNG system is not problematic for the manner in which this game uses it, I’m all ears, I truly am. But if not, please stop trolling.

- Any high quality pRNG system has an extraordinarily long period in comparison to the segment of it used; we’re talking you could be pulling a million billion numbers from it per second and you’d reach the heat death of the universe before you got through only a tiny fraction of its period. Repeating is really a non-issue.

- In a multi-player game like GW2 you don’t have a personal pRNG stream for every player, but a shared thread that multiple players draw from whenever needed – and as the numbers you personally observed are not sequential due to the human inputs and you don’t have any knowledge of those inputs, the numbers you personally observe are very close to truly random.

- The random numbers being pulled aren’t for a single purpose, but for many purposes – critical hit chances, damage variances, item drops, AI, forges. All of these are drawing random numbers, interspersed, which makes observation of any one in isolation extremely non-deterministic, for the same reasons as multiple users.

- Furthermore, GW2 needs to draw enough random numbers that there are almost certainly multiple random number generating threads running in parallel so that the game never has to ‘wait’ on the number generator or pulls the same number twice when two different users want one at the same time. When you personally want to generate a number, you could be pulling from any of the random number generators – you won’t know which.

- The previous three points inject so much uncertainty into the output that it doesn’t even matter if the underlying random number generator is a good one. Even a really terrible, short cyclical table of numbers is going to kick out truly random numbers if you have enough users querying it at uncertain intervals.

pRNG has known flaws to randomness, but those flaws are massively overstated by the hobbyist, and are largely mitigated in a multi-user environment. I don’t doubt that your concerns are sincere, but the examples of concerns you were offering really are hyperbolic caricatures that really aren’t applicable at all to real systems.

[Tl;DR] RNG is not fair or unfair. It merely gives results you want or fails to do so. You are not guaranteed a desired result no matter how well you feed the toilet.

Perceived unfairness is perceived.

Essence Snow.3194:

My main issue with the whole deal is that this is a game which after everything is said and done is supposed to be fun for those that engage with it. The current implementation of RNG allows for some users of the game to be basically be “kitten” upon with no regard. That is not fun and goes against the core ideology of the game. That is a major flaw imo, one that definitely warrants correcting.

Mystic Forge is a completely optional part of the game and so are legendary weapons. No one is forcing you to use it or penalizing you for not using it.

Complaining about bad luck and quitting due to it is akin to someone complaining he didn’t win the mega jackpot lottery after buying a whole bunch of tickets and committing suicide because he spent all his savings on it and thought life was specifically unfair to him.

Taku.6352:

TL;DR: I was a clown that gambled my money and now blaming Anet for being dumb and not using the sure methods already provided in-game.

I’m one of those clowns that played the forge to attempt to get my precursor (which is the ONLY method of acquiring one other than random drops… LOL) Anet designed a punitive system that OP hates and so do I.

If anyone is a clown it’s whoever designed the RNG

Glad I saved this.

Naqaj.6219:

Herre you go, OP. This thread is going nowhere fast, but atleast you can learn something useful:
http://en.wikipedia.org/wiki/Gambler%27s_fallacy

interesting information

Photonman.6241:

Lotta numbers in this thread… but I think it boils down to ’DON’T LEAVE! We love you!’

very caring player

Hmmm. I think your experience shows that the odds of getting a precursor from rares is less than .2%.

If they didn’t announce precursor masteries I wouldn’t even consider buying the expansion. My experiences with luck in the game are that bad. It’s a game forum so I get why some of you act like children but this game’s RNG beats some players like myself to a pulp.

I ran COE today and a sub 1500 ap elementalist with Bifrost asked what he should do with 1300g as he just got another Legend from the forge. He’s had amazing luck which made me want to kick him, I didn’t, but I wanted to.

The over use of RNG in this game is “fair” to the general playerbase, but unfair to individual players. Most players will fall really close together on the luck bell curve, but the outliers are unfairly rewarded or punished though no fault of their own.
John Smith talk abut solutions about this problem in his RNG thread, BUT NONE ARE BEING IMPLEMENTED IN GAME. Then added to the fact there are account only item that you can’t even buy your way out of the RNG trap. Every day I kill teq, and the game doesn’t care how large my spoon collection (100+ with no horde drop) gets, I still get the same odds as the guy who is killing him for the first time to get the horde. In Anet eyes that it fair.

I wonder how the game would be if legendaries never existed….

Serophous.9085:

I wonder how the game would be if legendaries never existed….

It would be an exotic game.

Guys, after 35 attempts today, I got the spark.
Final results are for the dagger:
Attempts: 1088
Exotics received: 218
Legendaries received: 1
Exotic drop rate: 20.04%
Legendary drop rate: 0.09%

Annoyingly, I had prepared all the mats to go to war. Now I’ll have to sell some of it and keep the inscriptions…

Using my jsfiddle, and with the values of found here: http://jsfiddle.net/a6qpsso5/13/
4000 copper for rare value
32000 copper for exotic value

I’m in the money apparently. My spark was 50 gold less than the TP one.

Posts like this make me laugh. I used to deal blackjack and it always amused me when someone would lose then get more money to “win it back” then lose again. Then they would get all upset at me and/or the Casino. I always wanted to tell them that there is no one behind them holding a gun to their head.