Russian Roulette........

[ob]

yes, it doesnt matter how many numbers there is, the variation would not change....

btw for any of you s16 punters those things are like a GIANT roulette wheel with 37 million numbers rather than 37, each number corresponding to a specfic payout (in most cases £0!).

[Mystery_Plum]

The only reason I can think of is to try to allow greater variation in the outcomes somehow.

but why on earth would it show ANY variations in the outcome? If the odds on any one outcome are the same, then the standard deviation of the results will remain the same.

Yep, so it's a waste of time trying it.

[theoak]

btw for any of you s16 punters those things are like a GIANT roulette wheel with 37 million numbers rather than 37, each number corresponding to a specfic payout (in most cases £0!).

I too was thinking people may be getting confused with an s16s RNG. Obvioulsy in roulette all numbers correspond to the same odds. In an s16 if £500 was as liekly as £0 we'd all be laughing!

[chav666]

your all missing the point. I can gaurantee you that roulette RNG's DO NOT go from 1-37! can anyone explain why? The point I am making is that it may appear to you that you are getting the same number again and again - but in fact the machine is giving different results -but DISPLAYING the same number. Sure, OVER TIME it will pay the same odds as a R/T wheel. But in the VERY SHORT TERM - IE 1-10 spins - you WILL see different patterns that those generated from a R/T wheel - certainly LOTS more number repeats etc. Hence MANY a tale of rip offs.

If anyone can give me any evidence of these machines operating from a 1-37 RNG I will start watching eastenders again - and believe me - thats the last thing I want to be doing!

[chav666]

[quote="&quot"]yes, it doesnt matter how many numbers there is, the variation would not change....

This is totally untrue. Check your maths. The more number there are the more spins it will take for the statistics to even out.

For Eg: RNG of 1-37 - generate 370 numbers and youwill have a roughly even spread of results will a HIGH liklihood that every number will have been picked at least once. Check lotto stats - (RT 1-49)

RNG 1-3700000 - generate 370 numbers do you think you'll get the same spread - NO WAY. you will have to spin many more to gaurantee every number is picked. Imagine if 1-10000 = 1 2-20000=2 and so on. Your only picking 370 mumbers. The RNG could pick shitloads of high numbers and toatally miss low numbers - IN THE SHORT TERM!!!

Come on - someone on this forum has the intelligence to see what I am saying here - Its really not that complex

[theoak]

are you assuming a RNG of just 37

a) works in sequence from 1-37
b) chooses only one number a second?

why WOULD it need a bigger base than 1-37 if it shifts through these numbers 'randomly' and at a rate of hundreds a minute?

[theoak]

RNG 1-3700000 - generate 370 numbers do you think you'll get the same spread - NO WAY.

God this is so wrong. YES 361928486239746234 is a different number to 236547653465654 but if they both REPRESENT a 17 on the roulette wheel, then its JUST the same as a 1-37 spinning two 17s. The Odds of ANY number coming in will ALWAYS be 1/37 REGARDLESS of how big your RNG is. so its just as likely to get 10million zeros in a row on a 37billion number generator as a 37 one, even if behind the seens the 10 million zeros allll have a unique RNG number.

[chav666]

I dont think I can explain it any better than in previous posts. All I can say is those roulette machines do not work on a RNG of 1-37. And oak - you fail to take into acount the fact that I am stating a larger RNG range will have an affect in the short tern only. I am not stupid enough to think that 37-1 is different from 3700-100! I know my maths. Most computer programming languages include functions or library routines that purport to be random number generators. They are often designed to provide a random byte or word, or a floating point number uniformly distributed between 0 and 1.

Heres a section from the following link: - http://en.wikipedia.org/wiki/Real_time_clock


Such library functions often have poor statistical properties and some will repeat patterns after only tens of thousands of trials. They are often initialized using a computer's real time clock as the seed. These functions may provide enough randomness for certain tasks (for example simple video games) but are unsuitable where high-quality randomness is required, such as in cryptographic applications, statistics or numerical analysis.
I suggest you investigate the subject ... power....

[chav666]

sorry - wrong link! I obviously aint that clever!

Heres the correct one: -

http://en.wikipedia.org/wiki/Random_number_generator

[Cardinal Sin]

Right. This is quite interesting, but it is deviating somewhat from the original argument.

Namely, are these FOBT machines "rigged"? Just cos it is got a shabby RNG doesn't mean it will have a payout percentage - or prevent a player from winning a big prize. If anything, it would work out in the player's favour.

[TheMission]

Come on - someone on this forum has the intelligence to see what I am saying here - Its really not that complex

Someone does have the intelligence: me. As I have indicated before, I've done RNG software for machines. Anyway as CR said, somewhat straying off topic.

[mjd]

Anything but 12 / 24 / and 10........................................... 12.

Trebling Up on Red..... Black Black Black Black Black Black

[Drpepper]

Don't worry... i can see what you're saying.

I'm doing further maths at college and in stats we're looking at stuff like this... we have too much time on our hands lol.

I can't remember if it was here or fruitforums, but there was certainly a large discussion questioning the randomness of RNG's used nowadays, may be worth a look to some people.