AG RNG and the math isn't adding up.

245

Comments

  • GroundedWisdomGroundedWisdom Posts: 25,817 ★★★★★
    Thanks, I might check them out.
  • liquidkarmaliquidkarma Posts: 80
    thank you for your replies @DrZola and @GroundedWisdom i am new to the forums and wasn't sure what has already been "debunked."
  • GroundedWisdomGroundedWisdom Posts: 25,817 ★★★★★

    thank you for your replies @DrZola and @GroundedWisdom i am new to the forums and wasn't sure what has already been "debunked."

    No worries.
  • Cpt9ToesCpt9Toes Posts: 32
    Also remember, when figuring probabilities, you need a HUGE sample. These four pulls are a fraction of what has been pulled. With millions of players popping millions of crystals daily, you would have to see a data mine that just isn’t available.

    Trying to simplify probability and then apply it to a single person just won’t work in this format.

    This is why you should only focus on drop rates, and not compare crystal to crystal...
  • DNA3000DNA3000 Posts: 11,235 Guardian
    DrZola said:

    DNA3000 said:

    DrZola said:

    https://www.mathsisfun.com/data/probability-events-independent.html

    This is, in theory, how it should (sort of) work. How it works in practice, however, is speculation by anyone who has not seen the actual engine that runs the game’s pRNG. But we are told by company representatives it is “random.”

    Dr. Zola

    In theory, particles obey quantum mechanics. In practice, only the creator of the universe actually knows. And a lot of people on the internet don't believe in quantum mechanics and have their own theories that they believe more.
    It was a simple point demonstrating how randomness usually works. Not sure I would put phone games and company spokesmen on the same footing as quantum theory, Nobel laureates and recognized quantitative/scientific geniuses, but you do you.

    Dr. Zola
    I didn't put phone games on the same footing as quantum theory. I put people putting their own gut instinct over hard data on the same footing regardless of the domain they are out of their depth on. Since you can replace "quantum mechanics" with "random number generators" or "game design" or "giraffe dentistry" this isn't a statement about any of those things per se.
  • DrZolaDrZola Posts: 3,677 ★★★★★
    edited March 2019
    DNA3000 said:

    DrZola said:

    DNA3000 said:

    DrZola said:

    https://www.mathsisfun.com/data/probability-events-independent.html

    This is, in theory, how it should (sort of) work. How it works in practice, however, is speculation by anyone who has not seen the actual engine that runs the game’s pRNG. But we are told by company representatives it is “random.”

    Dr. Zola

    In theory, particles obey quantum mechanics. In practice, only the creator of the universe actually knows. And a lot of people on the internet don't believe in quantum mechanics and have their own theories that they believe more.
    It was a simple point demonstrating how randomness usually works. Not sure I would put phone games and company spokesmen on the same footing as quantum theory, Nobel laureates and recognized quantitative/scientific geniuses, but you do you.

    Dr. Zola
    I didn't put phone games on the same footing as quantum theory. I put people putting their own gut instinct over hard data on the same footing regardless of the domain they are out of their depth on. Since you can replace "quantum mechanics" with "random number generators" or "game design" or "giraffe dentistry" this isn't a statement about any of those things per se.
    Wait. So now you’re comparing Kabam Miike to a giraffe dentist? :o

    Dr. Zola
  • DshuDshu Posts: 1,356 ★★★★
    5 awakening gems and all 5 were mystic
  • SknZnSknZn Posts: 426 ★★★
    Stop cry, legacy got so many sentry so far enough.
  • Suros_moonSuros_moon Posts: 118
    Has kabam ever officially confirmed that it isnt weighted to favor any gem for any reason. Just pulled my 6th science and they havent releqsed a science champ worth awakening in months ofc.
  • Suros_moonSuros_moon Posts: 118
    5th not 6th
  • DeliDeli Posts: 272 ★★
    Every person I know has pulled multiple gems of the same class (I pulled 2 cosmics and 3 tech) however if you look at the whole community and everyone is pulling the same class gem multiple times it still can equal a random drop rate. I pulled 3 blue, you pull 3 red, numerically it still evens out 50/50
  • Zeronaut81Zeronaut81 Posts: 200
    I would be much less skeptical of their RNG generator if the outcomes were more consistent in-game. I’m assuming that all rng in the game are subject to the same system, so rng odds for combat should be in line with rng odds for items. I’d assume that implementing a different system each for rng used in combat rolls and item rolls would be inefficient from a cost perspective.

    It’s hard to not be skeptical about Kabam’s process when a 3% chance is expressed so differently in certain outcomes. Why 3%? That’s the same chance Black Widow’s evade has as I have pulling a 5* from a fgmc. I have bit of a hard time believing that both 3% chances are weighted evenly. And yes, I understand that each outcome is an independent result, unconnected to any other rng roll. When I pull a 5* champ on 4/12 fgmc rolls (similar to a recent fight I had with BW) I’ll happily leave my doubts behind.

    It’s hard not to be skeptical when the stated odds seem to consistently be at odds with reality. Without using a 3rd-party plug-in or at least allowing some insight into the rng engine, Kabam should expect skepticism from their player base. Especially with having a sketchy-sounding exploitative patent that sure does seem to mimic a lot of in-game occurrences.
  • GroundedWisdomGroundedWisdom Posts: 25,817 ★★★★★
    I've had this debate before, and it all boils down to where the pools are located. The Crystals access pools on the server. Those are set codes, and the Crystal accesses the specific pools on the server when you open them. It's not even a process that registers what you have. Account A has a Crystal that accesses Pool B, and it generates outcome C based on the possibilities in the pool. Same pool for everyone.
    The RNG with Champs is different, and is programmed into content. It's entirely possible to increase the occurrences in content to match a desired level of challenge. Now, I'm not trying to generate any conspiracy or anything, but it's always been obvious to me that when it comes to harder content, they are more likely to proc more than their indicated percentages. I've had Fights in War, Master, and UC that triggered much, much higher than their suggested Abilities. In many ways, this is necessary because in order for content to provide more challenge, it needs to be stronger. Then there's the AI behavior, which I won't get into because I'm adamantly against varying speeds and reaction times above and beyond what our controls can do, but in terms of the extra procs, I have no issue with that perse. If it's AI, and it's strong, it's going to trigger more, and I've always maintained that, even back when they called me a Heretic for saying it. LOL.
  • Hammerbro_64Hammerbro_64 Posts: 5,680 ★★★★★
    I got a tech and a science. #debunked

    But nearly all of my 4* gems are Cosmic or science...
  • liquidkarmaliquidkarma Posts: 80
    @Zeronaut81 regarding your black widow evasion, it is fallacious to compare the two, as you are the receiver of many more hits from your opponent when fighting with black widow then the amount (i assume) of FGMCs that you open.
    in other words, the numerical value of 3% of incoming attacks aimed at black widow will be much higher than the # of fgmc's opened. you can compare the percentages of each to themselves, and they will probably be close to 3%, but it would be inaccurate to assume that 3% of each number would be the same, since *that* number will be different. if that makes sense.

    as far as 3rd party plug ins for stat purposes alone, not a good idea for obvious reasons and i would suggest just keeping a spreadsheet of your own to track if you are curious.you can pick your own parameters and as long as you don't forget to input any of them you will still have your stats.
  • mostlyharmlessnmostlyharmlessn Posts: 857 ★★★
    All 4 of my last 5* AG's have been mystics.

    There is quite a few possible reasons for the apparent batch like behavior of these pulls like not regenerating the seed for pRNG... But I'm going with Kabam wired in a Bambleweeny 57 Sub-Meson Brain with an Atomic Vector Plotter suspended in a Hot Cup of Tea.

  • DNA3000DNA3000 Posts: 11,235 Guardian
    DrZola said:

    DNA3000 said:

    DrZola said:

    DNA3000 said:

    DrZola said:

    https://www.mathsisfun.com/data/probability-events-independent.html

    This is, in theory, how it should (sort of) work. How it works in practice, however, is speculation by anyone who has not seen the actual engine that runs the game’s pRNG. But we are told by company representatives it is “random.”

    Dr. Zola

    In theory, particles obey quantum mechanics. In practice, only the creator of the universe actually knows. And a lot of people on the internet don't believe in quantum mechanics and have their own theories that they believe more.
    It was a simple point demonstrating how randomness usually works. Not sure I would put phone games and company spokesmen on the same footing as quantum theory, Nobel laureates and recognized quantitative/scientific geniuses, but you do you.

    Dr. Zola
    I didn't put phone games on the same footing as quantum theory. I put people putting their own gut instinct over hard data on the same footing regardless of the domain they are out of their depth on. Since you can replace "quantum mechanics" with "random number generators" or "game design" or "giraffe dentistry" this isn't a statement about any of those things per se.
    Wait. So now you’re comparing Kabam Miike to a giraffe dentist? :o

    Dr. Zola
    Am I allowed to do that on the forums?
  • DrZolaDrZola Posts: 3,677 ★★★★★
    DNA3000 said:

    DrZola said:

    DNA3000 said:

    DrZola said:

    DNA3000 said:

    DrZola said:

    https://www.mathsisfun.com/data/probability-events-independent.html

    This is, in theory, how it should (sort of) work. How it works in practice, however, is speculation by anyone who has not seen the actual engine that runs the game’s pRNG. But we are told by company representatives it is “random.”

    Dr. Zola

    In theory, particles obey quantum mechanics. In practice, only the creator of the universe actually knows. And a lot of people on the internet don't believe in quantum mechanics and have their own theories that they believe more.
    It was a simple point demonstrating how randomness usually works. Not sure I would put phone games and company spokesmen on the same footing as quantum theory, Nobel laureates and recognized quantitative/scientific geniuses, but you do you.

    Dr. Zola
    I didn't put phone games on the same footing as quantum theory. I put people putting their own gut instinct over hard data on the same footing regardless of the domain they are out of their depth on. Since you can replace "quantum mechanics" with "random number generators" or "game design" or "giraffe dentistry" this isn't a statement about any of those things per se.
    Wait. So now you’re comparing Kabam Miike to a giraffe dentist? :o

    Dr. Zola
    Am I allowed to do that on the forums?
    May get you double secret probation. Even I haven’t stepped in it that deep.

    Dr. Zola
  • Frivolousz21Frivolousz21 Posts: 416 ★★
    Hen said:

    Hamin said:

    First, I'm totally aware drop percentages are independent of each other; i.e. - you have a 16.7% chance of rolling a certain class. Rolls aren't cumulative and every single time it's 1 in 6.

    I'm not talking drop percentages. I'm talking about probabilty of rolling the same number multiple times.

    As it stands, I've rolled 5* Cosmic 4 times in a row. If I'm doing my math correctly, that's a .00077% chance (1/6^4).

    This isn't adding up.

    I feel your pain, just opened my 4th mutant gem, it’s total BS!
    There is a YouTube video, Proff Hoff I think, where they’ve done a lot of research that shows Kabam often reward you with the same class because surprise surprise it encourages you to spend more money to get the one you want :(
    That's nonsense.


    There are hundreds of thousands active players every week.

    And millions in a month+.

    It can be assured some guy on YouTube has no evidence of anything to disprove or prove some RNG bias.


  • Frivolousz21Frivolousz21 Posts: 416 ★★
    Most of the responses in this thread are embarrassing.

    So many people think the universe revolves around them.
  • Zeronaut81Zeronaut81 Posts: 200
    @liquidkarma it’s not really fallacious. Both events are listed within a single framework with the exact same odds of 3%. I should expect consistent results. Each time I attempt to strike BW, she has an independent roll with really low odds of evading. Over time, with tons of independent chances to normalize to a 3% rate, I shouldn’t notice a difference between her evade rate and crystal redemption rate. 4* rate drop rate on phc is at 3%, too. I’ve had a pretty good sample size of these. I track my phc openings, and they fall in line with expectations. They normalize over time, and I don’t notice any huge difference between the stated odds and the realized results.

    I shouldn’t notice any huge difference between BW’s 3% chance to evade and her realized results, but I do. With more chances of seeing her evade, the opposite should occur. Her evade procs should fall in line with her listed odds. Combat rng should behave like prize redemption rng. There should be consistency across the platform in how odds are calculated or realized. A 3% chance during one event should behave like a 3% chance in another event. If there are noticeable discrepancies, Kabam should expect skepticism. And if I could take the 3% chance of BW’s evade and apply that to my next dozen fgmc, I’m pretty sure my roster would be a lot stronger than it is now.
  • JaffacakedJaffacaked Posts: 1,416 ★★★★
    Jaded said:

    I’ve gotten 4 cosmic, 2 science, 2 tech, 1 mystic.

    Only need skill and mutant. But can’t get them. So the rest sit and collect dust.

    Edit:it doesn’t add up

    I've got had 4 cosmic, 3 skill, 2 science. Every damn class I don't need lol
  • KattohSKattohS Posts: 623 ★★
    You get what Kabam wants you to get.

    Which is not what you want to get
  • GroundedWisdomGroundedWisdom Posts: 25,817 ★★★★★
    KattohS said:

    You get what Kabam wants you to get.

    Which is not what you want to get

    Kabam has no hand in it, other than creating the code with 6 equal possibilities.
  • GroundedWisdomGroundedWisdom Posts: 25,817 ★★★★★

    @liquidkarma it’s not really fallacious. Both events are listed within a single framework with the exact same odds of 3%. I should expect consistent results. Each time I attempt to strike BW, she has an independent roll with really low odds of evading. Over time, with tons of independent chances to normalize to a 3% rate, I shouldn’t notice a difference between her evade rate and crystal redemption rate. 4* rate drop rate on phc is at 3%, too. I’ve had a pretty good sample size of these. I track my phc openings, and they fall in line with expectations. They normalize over time, and I don’t notice any huge difference between the stated odds and the realized results.

    I shouldn’t notice any huge difference between BW’s 3% chance to evade and her realized results, but I do. With more chances of seeing her evade, the opposite should occur. Her evade procs should fall in line with her listed odds. Combat rng should behave like prize redemption rng. There should be consistency across the platform in how odds are calculated or realized. A 3% chance during one event should behave like a 3% chance in another event. If there are noticeable discrepancies, Kabam should expect skepticism. And if I could take the 3% chance of BW’s evade and apply that to my next dozen fgmc, I’m pretty sure my roster would be a lot stronger than it is now.

    Theoretically, and on paper, I agree. However, those are two different RNGs, with different purposes. One is a set code on the server that regulates drops. It's the same across the board for everyone, and is indicated appropriately. It doesn't change, unless they specify otherwise.
    The other is embedded in content, but is also affected by other things like Nodes, differences in PI, etc. I haven't pinpointed the reason, I had a decent conversation with DNA a long while back about how I hypothesized that it was differences in CR and came to the conclusion that there's nothing about CR that would affect proc. Nevertheless, if AI is stronger than us, and combined with Nodes and other factors in content, it will proc more. Case-in-point, Mordo Act 6. While it may seem unfair, it's their creative prerogative to increase whatever they want in order to provide the challenge they want. I agree that there is a responsibility for Champs in our Roster to perform as intended, but when it comes to content, it can really be whatever they want. Which is really a good thing when you consider how easy it would be if they didn't exceed what's on paper. We all would have been bored long ago, and things wouldn't be as challenging as they are now. Either way, it's probably worth noting that 7% for AI means very little because it will Evade when it wants to. ;)
  • GroundedWisdomGroundedWisdom Posts: 25,817 ★★★★★
    Although, that example probably wasn't the best, that was a Node. I meant more in terms of their Abilities. Don't even know what to say about that Mordo, other than "AI". LOL.
  • ReferenceReference Posts: 1,441 ★★★
    I got 4 4* cosmic AG too...strange RNG
  • Vdh2008Vdh2008 Posts: 966 ★★★★
    It's not just AG RNG... I opened 5 Cav crystals yesterday... 3 of them were 5 star Quake. The odds of that are astronomical.
  • DrZolaDrZola Posts: 3,677 ★★★★★
    Vdh2008 said:

    It's not just AG RNG... I opened 5 Cav crystals yesterday... 3 of them were 5 star Quake. The odds of that are astronomical.

    Got a screenshot? I would love to see that.

    Dr. Zola
  • Dmonster69Dmonster69 Posts: 14
    I pulled my 5th skill five star awakening gem, in a row. There’s no way they are completely random, it should not even be possible to get 5 of the same class five star awakening gems in a row, unless kabam has some hidden algorithm that gives you a much higher chance to pull a 5* ag that’s already in your inventory, and it wouldn’t surprise me at all......
Sign In or Register to comment.