• Announcements

    • IMPORTANT - REACH US IN THE NEW FORUM   05/04/2017

      Ladies and gentlemen ATTENTION please:
      It's time to move into a new house!
        As previously announced, from now on IT WON'T BE POSSIBLE TO CREATE THREADS OR REPLY in the old forums. From now on the old forums will be readable only. If you need to move/copy/migrate any post/material from here, feel free to contact the staff in the new home. We’ll be waiting for you in the NEW Forums!

      https://community.blackdesertonline.com/index.php

      *New features and amazing tools are waiting for you, even more is yet to come in the future.. just like world exploration in BDO leads to new possibilities.
      So don't be afraid about changes, click the link above and follow us!
      Enjoy and see you on the other side!  
    • WICHTIG: Das Forum ist umgezogen!   05/04/2017

      Damen und Herren, wir bitten um Eure Aufmerksamkeit, es ist an der Zeit umzuziehen!
        Wie wir bereits angekündigt hatten, ist es ab sofort nicht mehr möglich, neue Diskussionen in diesem Forum zu starten. Um Euch Zeit zu geben, laufende Diskussionen abzuschließen, könnt Ihr noch für zwei Wochen in offenen Diskussionen antworten. Danach geht dieses Forum hier in den Ruhestand und das NEUE FORUM übernimmt vollständig.
      Das Forum hier bleibt allerdings erhalten und lesbar.   Neue und verbesserte Funktionen warten auf Euch im neuen Forum und wir arbeiten bereits an weiteren Erweiterungen.
      Wir sehen uns auf der anderen Seite!

      https://community.blackdesertonline.com/index.php Update:
      Wie angekündigt könen ab sofort in diesem Forum auch keine neuen Beiträge mehr veröffentlicht werden.
    • IMPORTANT: Le nouveau forum   05/04/2017

      Aventurières, aventuriers, votre attention s'il vous plaît, il est grand temps de déménager!
      Comme nous vous l'avons déjà annoncé précédemment, il n'est désormais plus possible de créer de nouveau sujet ni de répondre aux anciens sur ce bon vieux forum.
      Venez visiter le nouveau forum!
      https://community.blackdesertonline.com
      De nouvelles fonctionnalités ainsi que de nouveaux outils vous attendent dès à présent et d'autres arriveront prochainement! N'ayez pas peur du changement et rejoignez-nous! Amusez-vous bien et a bientôt dans notre nouveau chez nous

2% droprate vs 1% means half of the work or twice the drops

80 posts in this topic

Posted (edited)

this is for anyone interested in probability, and also to "anyone" (YOU KNOW WHO YOU ARE @Le_Flemard ) who has ever doubted my probability-related knowledge and dared suggest that they're probabilistic knowledge is greater than my own via less-than-agreeable tones:

the person who has a 2% drop rate instead of the 1% drop rate will get a drop twice as often. it will reduce the work that you have to do to half of its previous value. that is pretty significant.

first, i'll justify the second half of that statement. we'll call the  number of monsters killed (or runs, or whatever) y.  so the chance of getting a drop after killing y monsters is

P = 1 - (1 - .01)^y  = 1 - .99^y (where the .01 represents the 1%)

you can think of this like:

  • every time i kill a monster, i have a .01 chance of getting a drop, or a 1 - .01 = .99 chance of not getting a drop
  • so if i kill two monsters, the chance of me not getting a drop is .99*.99, if i kill 3 monsters, the chance is .99*.99*.99, four is .99^4, etc.
  • the chance of me getting a drop is just 100% minus the chance of me not getting a drop, so 1 - .99^y

now, what happens to y if i want to keep P the same, but change the droprate? let's say I want to take it from 1% to 2%. first, let's solve for y:

1 - .99^y = P

=>

.99^y = 1 - P

=>

log_{.99}(.99^y) = log_{.99}(1 - P)

=>

y = log_{.99}(1 - P)

=>

y_1 = ln(1 - P) / ln(.99)

now, let's increase the droprate. recall that the .99 above comes from 1 - .01, so 1 - .02 would simply give us .98. therefore:

y_2 = ln(1 - P) / ln(.98) when the drop rate is 2% instead of 1%

that is, the only difference that matters is the 1/ln(x) part. we can easily quantify this with a ratio. what is the ratio of between these two values?

y_2 / y_1 = ln(.99) / ln(.98) = .497475... ~ .5 = 1/2

what's that, you say? to maintain the probability of getting a drop, the person with a 2% droprate instead of a 1% droprate has to kill half as many monsters? yes, this is, in fact, true. (truf)

let's look at the first part of the statement, now. specifically, "the person who has a 2% drop rate instead of the 1% drop rate will get a drop twice as often." here, we'll just use the expected value of the binomial distribution, which quantifies the number of successes in independent trials with the same success rate (such as killing monsters). it's expected value is yp, where y is the same as defined above, and p is just the droprate.

so the expected value after y kills with a 1% drop rate is E_1 = y*.01, and the expected value after y kills with a 2% droprate is E_2 = y*.02.

again, we can just take the ratio to quantify the relationship: E_2 / E_1 = 2

THIS MEANS THAT SOMEONE WITH 2% DROPRATE IS EXPECTED TO GET TWICE THE DROPS AS SOMEONE WITH 1% DROP RATE

http://imgur.com/a/deMJa

DO YOU FEEL WHAT PAIN I'VE BROUGHT TO YOU, @LE_FLEMARD ????

DO YOU SEE WHAT KNOWLEDGE CAN BE CONJURED FROM THE DARKEST DEPTHS OF HATRED INFINITE!?!? 

DO YOU NOW KNOW THE POWER THAT I STRUGGLE TO WITHHOLD, POWER THAT CONSTANTLY TESTS THE WILL OF THE ETHEREAL SEAMS OF THE MIND THAT KEEP IT FROM SPILLING FORTH INTO THIS WORLD AND DOOMING US ALL?????

say you're sorry right now

right now i'm not even kidding

(p.s. - this post is mod approved if any other mods want to step up to me DO IT
I CAN BACK UP MY WORDS

YOU SEE WHAT I'VE JUST BROUGHT)

Edited by AmagicalFishy
5 people like this

Share this post


Link to post
Share on other sites

Posted

this is a game forum. not a mathclass

4 people like this

Share this post


Link to post
Share on other sites

Posted

<3

Share this post


Link to post
Share on other sites

Posted

this is a game forum. not a mathclass

this is my thread and it's whatever i say it is

AND RIGHT NOW IT'S A MATH CLASS DO YOU HAVE A QUESTION, SIR, OR WOULD YOU LIKE TO GO TO THE BATHROOM?

1 person likes this

Share this post


Link to post
Share on other sites

Posted

@Luciferia

is this what you wanted me to see eariler, before i start reading

Share this post


Link to post
Share on other sites

Posted

ok seriously you guys need to tell me how to link people's names and also link pictures

Share this post


Link to post
Share on other sites

Posted

Missing the mic drop.

I change my grade of this post from 11/10 to 2½/100 :(

1 person likes this

Share this post


Link to post
Share on other sites

Posted

this is my thread and it's whatever i say it is

AND RIGHT NOW IT'S A MATH CLASS DO YOU HAVE A QUESTION, SIR, OR WOULD YOU LIKE TO GO TO THE BATHROOM?

Mathclass when it should be computer science? They are not the same. By the way, if you keep up the goofy stuff they will just delete the post. I'm willing to prove you wrong if it's worth my time.

Share this post


Link to post
Share on other sites

Posted

Mathclass when it should be computer science? They are not the same. By the way, if you keep up the goofy stuff they will just delete the post. I'm willing to prove you wrong if it's worth my time.

naw, i already talked to gm; the post is fine (unless you're saying goofy responses will mean the thread gets deleted? that'd be werid)

also you can only prove wrong what is wrong

and, sir, i am not wrong

(but please tell me how to link pictures why is no one giving me this information)

Share this post


Link to post
Share on other sites

Posted

@Luciferia

is this what you wanted me to see eariler, before i start reading

Roughly speaking, this is the poster and a similarly formatted post.  However it is missing the "HOW DO I EVEN INSERT PICTURES" comment and some other content.  Is it a good readD?

Share this post


Link to post
Share on other sites

Posted

naw, i already talked to gm; the post is fine (unless you're saying goofy responses will mean the thread gets deleted? that'd be werid)

also you can only prove wrong what is wrong

and, sir, i am not wrong

(but please tell me how to link pictures why is no one giving me this information)

chrome_2017-03-29_16-27-34.thumb.png.f12

you're wrong because you're disregarding applied mathematics.
Applied being, you're disregarding how computational mathematics work in regards to, you know, computers.
It's not entirely the same as your textbook shit, there are limitation variables that you're leaving out of your equation, to put it simply.

Roughly speaking, this is the poster and a similarly formatted post.  However it is missing the "HOW DO I EVEN INSERT PICTURES" comment and some other content.  Is it a good readD?

idk i haven't read it yet but he posted some shit about binomial crap as i scrolled down, my eye caught it. I've already debunked it once so i might just quote that.

Share this post


Link to post
Share on other sites

Posted

1% or 2% of rng = rng no matters what xD

Share this post


Link to post
Share on other sites

Posted

Pfft, next you'll be telling us 10 is twice as big as 5...

2 people like this

Share this post


Link to post
Share on other sites

Posted

this is for anyone interested in probability, and also to "anyone" (YOU KNOW WHO YOU ARE @Le_Flemard ) who has ever doubted my probability-related knowledge and dared suggest that they're probabilistic knowledge is greater than my own via less-than-agreeable tones:

I'll start off by saying (I reply as i'm reading, i don't read the entire thing before i respond) if your education in probability isn't inclusive of computational probability, everything i'm about to read is going to be a waste of my time, which will be unfortunate. I'll read it though.

Share this post


Link to post
Share on other sites

Posted

I'll start off by saying (I reply as i'm reading, i don't read the entire thing before i respond) if your education in probability isn't inclusive of computational probability, everything i'm about to read is going to be a waste of my time, which will be unfortunate. I'll read it though.

xD who do you even think you are

and also why do you have the option to insert images and that is nowhere to be found on my interface

 

 

Share this post


Link to post
Share on other sites

Posted (edited)

luckily i don't have to read it all. (I feel like this might come back to nip me in the butt)

first, i'll justify the second half of that statement. we'll call the  number of monsters killed (or runs, or whatever) y.  so the chance of getting a drop after killing y monsters is

P = 1 - (1 - .01)^y  = 1 - .99^y (where the .01 represents the 1%)

Incorrect.
You're using textbook mathematics. That being said, that quote was also wrong, i think. it didn't really make much sense to me either way. Anyways;

I'll try to be as clear as possible.

Computers only do what we tell them to.
If you don't define a pool based probable outcome to define success and failure limitations in code on any level, using such as an example will only get you screwed over.

I made the same mistake back in college. It's a misconception for people who don't actually know how computers work. This is to say; Theory. Not knowing how many pins on each processor, but what the pins on a processor/socket does and their place in the entire system that makes up the engine that allows us to do what we do now.

If you can bring me backed evidence that the game uses this type of probability engine, i'll be all for it. Until then, nope. Sorry. I don't think PA would go out of their way to code in this measurement or feature if you will while still incorporating the one that is currently in place- chance by chance success and fail. Computers, process this chance by chance success or fail without emotion or faith or hope, they forget what happened last unless we tell them not to. That is why binomial bull-----ery does not apply unless explicitly told to apply in these cases.

Does that make sense? Or do i need to explain more.

xD who do you even think you are

I'm a nobody.

and also why do you have the option to insert images and that is nowhere to be found on my interface

Dunno. Do you have any warning marks for any reason? Lol.

Edited by War
I mean, i guess computers drive. In the case of tesla.... but still.

Share this post


Link to post
Share on other sites

Posted

luckily i don't have to read it all. (I feel like this might come back to nip me in the butt)

Incorrect.You're using textbook mathematics. That being said, that quote was also wrong, i think. it didn't really make much sense to me either way. Anyways;

I'll try to be as clear as possible.

Computers only do what we tell them to.
If you don't define a pool based probable outcome to define success and failure limitations in code on any level, using such as an example will only get you screwed over.

I made the same mistake back in college. It's a misconception for people who don't actually know how computers work. This is to say; Theory. Not knowing how many pins on each processor, but what the pins on a processor/socket does and their place in the entire system that makes up the engine that drives what we do now.

If you can bring me backed evidence that the game uses this type of probability engine, i'll be all for it. Until then, nope. Sorry. I don't think PA would go out of their way to code in this measurement or feature if you will while still incorporating the one that is currently in place- chance by chance success and fail. Computers, process this chance by chance success or fail without emotion or faith or hope, they forget what happened last unless we tell them not to. That is why binomial bull-----ery does not apply unless explicitly told to apply in these cases.

Does that make sense? Or do i need to explain more.

whoa, whoa, whoa. it makes sense, in that i understand the words that you're saying and why you're saying them—but it looks like you're making a "theory / practice" distinction when what it seems you're actually proposing is: PA coded drops that are not independent. (i.e. - drops that are not "chance by chance success and fail") that is a pretty bold claim...

... but then you say "Computers process this chance by chance success or fail without emotion or faith or hope, they forget what happened last unless we tell them not to..." but that's exactly what independence is, and that's exactly what is requisite for the binomial distribution to be relevant (because it requires independence)...

... and then you say "binomial bullshittery" ???

 

Share this post


Link to post
Share on other sites

Posted (edited)

 PA coded drops that are not independent. (i.e. - drops that are not "chance by chance success and fail") that is a pretty bold claim...

? I am claiming that drops are chance by chance success and fail. I'm more than positive they were.
I was reading your equation as if you were claiming they weren't, was I incorrect? If so, basing droprate based off an amount of kills makes no sense in that case. You can't use binomial distribution to describe what isn't there to describe. In order for that system to work in place on a computational level, you'd need to program it in. Aka, go out of your way to program it in ontop of the system already in place then recursively program in success pointers based off of what was just programmed in that overlap the system already in place.

... but then you say "Computers process this chance by chance success or fail without emotion or faith or hope, they forget what happened last unless we tell them not to..." but that's exactly what independence is, and that's exactly what is requisite for the binomial distribution to be relevant (because it requires independence)...

... and then you say "binomial bullshittery" ???

refer to the above.
There is a reason probability theory and computer science are not the same thing. Computers can't exactly do everything, yet.
You're leaving the limitations of the CPU & friends out of your mathematics, assuming you're implying that THAT system is the one being used.

Your computer is not going to just give you a success because you've tried 1000, 10000, 100000000 times. Computers do not think like humans do. They do what they are told, and forget everything else. Only follow the pattern they are given, meaning that all that you apply extra to what the computer does is faith and hope for a better world tomorrow.

Technically it's obscene to even say that you will get something, when those limitations are taken into piratical consideration as well. But hey. Anything can happen.

bd is a textbook theory.
It has to be programmed for it to be a computational reality.
In that case, very CPU will have a different outcome as well, as every CPU processes differently due to it's own instructional & design limitation. But that's beyond the game, and just computer theory.

Edited by War

Share this post


Link to post
Share on other sites

Posted

? I am claiming that drops are chance by chance success and fail. I'm more than positive they were.I was reading your equation as if you were claiming they weren't, was I incorrect? If so, basing droprate based off an amount of kills makes no sense in that case. You can't use binomial distribution to describe what isn't there to describe. In order for that system to work in place on a computational level, you'd need to program it in. Aka, go out of your way to program it in ontop of the system already in place then recursively program in success pointers based off of what was just programmed in that overlap the system already in place.

refer to the above.There is a reason probability theory and computer science are not the same thing. Computers can't exactly do everything, yet.
You're leaving the limitations of the CPU & friends out of your mathematics, assuming you're implying that THAT system is the one being used.

Your computer is not going to just give you a success because you've tried 1000, 10000, 100000000 times. Computers do not think like humans do. They do what they are told, and forget everything else. Only follow the pattern they are given, meaning that all that you apply extra to what the computer does is faith and hope for a better world tomorrow.

Technically it's obscene to even say that you will get something, when those limitations are taken into piratical consideration as well. But hey. Anything can happen.

What you're accusing the OP of describing is Hypergeometic distribution, when he is merely calculating the probability of an independent system (a WITH REPLACEMENT system) which is done using Binomial distribution. 

I think you're losing here. 

Share this post


Link to post
Share on other sites

Posted

? I am claiming that drops are chance by chance success and fail. I'm more than positive they were.I was reading your equation as if you were claiming they weren't, was I incorrect? If so, basing droprate based off an amount of kills makes no sense in that case. You can't use binomial distribution to describe what isn't there to describe. In order for that system to work in place on a computational level, you'd need to program it in. Aka, go out of your way to program it in ontop of the system already in place then recursively program in success pointers based off of what was just programmed in that overlap the system already in place.

refer to the above.There is a reason probability theory and computer science are not the same thing. Computers can't exactly do everything, yet.
You're leaving the limitations of the CPU & friends out of your mathematics, assuming you're implying that THAT system is the one being used.

Your computer is not going to just give you a success because you've tried 1000, 10000, 100000000 times. Computers do not think like humans do. They do what they are told, and forget everything else. Only follow the pattern they are given, meaning that all that you apply extra to what the computer does is faith and hope for a better world tomorrow.

Technically it's obscene to even say that you will get something, when those limitations are taken into piratical consideration as well. But hey. Anything can happen.

i'm not sure why you linked a professor's page; did you mean to link something else?

the whole point of a lot of probability is that you don't have to explicitly program it into something. when you flip a coin, nature isn't giving you a heads because you flipped it 100 times. or when you roll a dice, the chances of you getting a 6 after 100 rolls are such because not because mother nature somehow remembers what you rolled.

you don't always have to explicitly program limitations of a certain distribution into a computer in order for it to follow said distribution. for example, for the binomial distribution, all you need are the most default assumptions: that the condition be success/fail (drop or no drop), the trials are independent, and the chance of dropping a particular item doesn't fluctuate.

all of my equations above, and the binomial distribution, necessitate independence. i'm not sure why you read them assuming they weren't independent, because then none of it would work.

i think you might've misread something, dude, but you're totally wrong here

Share this post


Link to post
Share on other sites

Posted

Pfft, next you'll be telling us 10 is twice as big as 5...

I am blown away! 

Share this post


Link to post
Share on other sites

Posted

i'm not sure why you linked a professor's page; did you mean to link something else?

the whole point of a lot of probability is that you don't have to explicitly program it into something. when you flip a coin, nature isn't giving you a heads because you flipped it 100 times. or when you roll a dice, the chances of you getting a 6 after 100 rolls are such because not because mother nature somehow remembers what you rolled.

you don't always have to explicitly program limitations of a certain distribution into a computer in order for it to follow said distribution. for example, for the binomial distribution, all you need are the most default assumptions: that the condition be success/fail (drop or no drop), the trials are independent, and the chance of dropping a particular item doesn't fluctuate.

all of my equations above, and the binomial distribution, necessitate independence. i'm not sure why you read them assuming they weren't independent, because then none of it would work.

i think you might've misread something, dude, but you're totally wrong here

I kinda think he's trying to troll you, unless he's just super confused. He's propped up a straw man that has nothing to do with what you said and is viciously attacking it - ultimately arguing that what you are doing in the OP is correct unless BDO has some fancy probability system that increases the chance each time you fail, which we all seem to agree is unlikely (in which you might use a hypergeometric distribution to determine the probability). 

But just in case, @War, do you understand the concepts of with replacement vs without replacement in this context? Binomial distribution calculates probability with replacement, which means it is independent of any previous success or failures. You, along with the OP, are both assuming that BDO's probability would be independent or with replacement, meaning you could calculate probability with binomial distribution... So uhh, who the heck are you arguing with? 

Share this post


Link to post
Share on other sites

Posted

I kinda think he's trying to troll you, unless he's just super confused. He's propped up a straw man that has nothing to do with what you said and is viciously attacking it - ultimately arguing that what you are doing in the OP is correct unless BDO has some fancy probability system that increases the chance each time you fail, which we all seem to agree is unlikely (in which you might use a hypergeometric distribution to determine the probability). 

But just in case, @War, do you understand the concepts of with replacement vs without replacement in this context? Binomial distribution calculates probability with replacement, which means it is independent of any previous success or failures. You, along with the OP, are both assuming that BDO's probability would be independent or with replacement, meaning you could calculate probability with binomial distribution... So uhh, who the heck are you arguing with? 

i mean... i don't think he's trolling. earlier, someone specifically pinged him to "check my math," so i assume he has some kind of math background.

@War , you seemed pretty prepared to pull out the "pft, these book-stuffy theorists don't know anything about how the real world works," and i think it's coming around to bite you in the butt, man. one thing kind of confuses me, though—the binomial distribution depends on independence; like, it wouldn't work without it, but you said a multiple times stuff like "computers don't remember what you're doing [independence], so this binomial bullshittery doesn't work," which... is totally the opposite of what actually goes on. IS THERE SOMETHING FISHY GOING ON HERE?

ALSO MORE IMPORTANTLY WHERE IS @Le_Flemard ????

Share this post


Link to post
Share on other sites

Posted

the binomial distribution depends on independence; like, it wouldn't work without it, but you said a multiple times stuff like "computers don't remember what you're doing [independence], so this binomial bullshittery doesn't work," which... is totally the opposite of what actually goes on. IS THERE SOMETHING FISHY GOING ON HERE?

^ That's why I think he's trolling you. His overall argument is in agreement with yours aside from claiming that binomial distribution is not independent. With or without a math background this is not in the slightest difficult math. 

Share this post


Link to post
Share on other sites

Posted

the whole point of a lot of probability is that you don't have to explicitly program it into something.

This is my point.
We do.
You are aware that everything computers do, they are explicitly told specifically to do, right? You need to stop assuming, and take a class, preferably an applied mathematics class.

^ That's why I think he's trolling you. His overall argument is in agreement with yours aside from claiming that binomial distribution is not independent. With or without a math background this is not in the slightest difficult math. 

That's not what i'm implying at all. I'm implying that it's independent of computer science unless specified and out defined.
None of these concepts are native to computer code on any level unless they are defined to, example being matlab, which was created specifically to handle programmatic mathematics like the textbook ones you're using.

Basically what i'm implying is that you math nerds are speaking from the wrong field, into one that you've clearly no education in.
Proof?;

the whole point of a lot of probability is that you don't have to explicitly program it into something.

He goes on to explaining coins and shit, without acknowledging that us developers have to create those coins in a programmatic manner (every aspect, as any left out leaves an unclear and malformed anomaly) on top of hardware limited by what us engineers are limited by on a technological level. 

You kids act like using an RNG function tells the secrets of the universe, when the engineers that developed the entire foundation it sits on don't know the secrets of the universe themselves.

Forgive me for not being as elaborate as I probably should, but i'm not here to teach a class.
I linked that professor due to much of his work, some that you two may learn from, as you're clearly educated in mathematics, but not computer science & engineering.

Share this post


Link to post
Share on other sites