flat assembler
Message board for the users of flat assembler.
Index
> Heap > Lottery tampering... 
Author 

AsmGuru62
Well, well... our Canadian Lottery insiders are at it again:
http://www.msn.com/enca/news/canada/lotteryinsidersbrotherarrestedinjackpotfixingscandal/arBBrroC0?li=AAggNb9&ocid=mailsignout And here I thought that numbers are drawn by a balls in a rotating wheel. And they actually use RNG, which some guy added code to. Cool! 

06 Apr 2016, 18:05 

sleepsleep
cool!,
maybe it is more easier to manipulate using computer algorithm =P 

08 Apr 2016, 06:15 

sleepsleep
i always have a thought like this,
if i got a dice, and i roll it 10 times, 4, 1, 3, 1, 6, 4, 2, 6, 1, 2 is the probability to draw any number is equal 100% / 6 for my next throw, or based on past results, i could have more probability on certain number? 

14 Jul 2016, 16:56 

revolution
sleepsleep: It depends upon your die and your rolling technique. Being a physical device it is sensitive to physical differences. If the die is slightly out of balance then that will skew the results, but the amount of skew may only be negligible and perhaps undetectable without some extreme testing regime. If you always begin your roll with the same number facing up and try your hardest to perform precisely the same hand motions each time then you may be able to influence the outcome. It would likely take a lot of practice to but is theoretically possible to control the outcome.


14 Jul 2016, 23:59 

YONG
sleepsleep wrote: if i got a dice, and i roll it 10 times, http://www.softschools.com/math/topics/theoretical_vs_experimental_probability/ If, after A LARGE NUMBER of trials, the experimental probability still significantly differs from the theoretical probability, we can confidently conclude that EITHER the die is loaded OR some external factor(s) is/are influencing the outcomes, for example, the rolling technique pointed out by revolution. 

15 Jul 2016, 02:39 

revolution
YONG wrote: http://www.softschools.com/math/topics/theoretical_vs_experimental_probability/ Quote: As more trials are conducted, the experimental probability generally gets closer to the theoretical probability. 

15 Jul 2016, 05:13 

YONG
revolution wrote:


15 Jul 2016, 05:26 

revolution
The problem is how people interpret the word "closer". There is a very important detail that is not mentioned, i.e. are we talking about absolute values or relative values? Understanding the difference is vital to a proper understanding of probabilities.


15 Jul 2016, 05:32 

sleepsleep
i thought about the following,
how lottery or gaming result numbers get draw? i assume they use the ball with number and air pressure, revolution website link me to the following http://entertainment.howstuffworks.com/lottery3.htm Quote: They are designed and proven using statistical analysis to produce random combinations of numbers. is a nonskew result equal to some sort of pattern? so if you discount the possible skew result, probability for nonskew result would be higher, am i right? 

15 Jul 2016, 07:11 

revolution
What do you mean "if you discount the possible skew result"? Are you suggesting to throw away outputs that "look" nonrandom. If so then you would be making a big mistake. Either the output is random, or it is not. There can't be any inbetween state that you can somehow decide which are random and which are not and then filter it.
Alternatively there are techniques that can be used to remove bias in results. But bias is not the same a nonrandom. For example, most coins are biased to one side because of the differences in the material on each face, but you can still get random results from coin tossing. 

15 Jul 2016, 07:59 

sleepsleep
i think the word "random" needs definition,
Quote:
the definition might be different in mathematics, idk, but what kind of input we need to determine random? a set of results, then we need to figure out how big or large the result we need, in this case, assume we use the lotto machine, we choose the high quality, the best one we could get from market, (of course, i am not sure how we gonna define a good one vs a not so good one) assume we eliminate 100% of bias, or it is impossible? idk but what kind of result is considered good random? assume 0 to 9 is a good random means, +10% occurrence of each digit in a 10 thousand chosen digit? or, we will call it bias if 4 got a 30% occurrence? 

15 Jul 2016, 13:04 

revolution
There are many methods that are used to test for randomness, my website has the details. But no method is perfect, so we use approximations and estimates. That is the best we can do right now.
There are some physical processes we can use to gather randomness, and to the best of our knowledge these are indeed genuine randomness generators (as long as we use them properly). Once again, my website has the details. 

15 Jul 2016, 13:16 

sleepsleep
maybe complex mathematics are required to understand good randomness,
i could only understand it the simple way, +10% occurrence of each digit is good randomness for 0 to 9 in [you fill in size] of result, so anything that don't conform to this +10% is bad randomness they will probably call it "bias" too if a smart monkey punch "typewritter", 

15 Jul 2016, 13:31 

YONG
revolution wrote: ... my website has the details. 

16 Jul 2016, 05:04 

YONG
sleepsleep wrote: maybe complex mathematics are required to understand good randomness For example, consider the following sequence of nonnegative integers: 7, 3, 2, 0, 5, 0, 8, 0, 7, 5, x Could you predict the value of x based on the first 10 numbers? Now, consider the following sequence: 7, 3, 2, 0, 5, 0, 8, 0, 7, 5, 6, 8, 8, 7, 7, 2, 9, 3, 5, 2, y Could you predict the value of y based on the first 20 numbers? Once again, consider the following sequence: 7, 3, 2, 0, 5, 0, 8, 0, 7, 5, 6, 8, 8, 7, 7, 2, 9, 3, 5, 2, 7, 4, 4, 6, 3, 4, 1, 5, 0, 5, z Could you find any pattern so far? Could you predict the value of z based on the first 30 numbers? We can repeat this process indefinitely. You will never find any pattern. You will never be able to predict the value of the next number based on the previous numbers. That is what we call "the lack of pattern or predictability". Refer to: https://en.wikipedia.org/wiki/Randomness http://apod.nasa.gov/htmltest/gifcity/sqrt3.1mil 

16 Jul 2016, 05:55 

< Last Thread  Next Thread > 
Forum Rules:

Copyright © 19992020, Tomasz Grysztar. Also on YouTube, Twitter.
Website powered by rwasa.