Pseudorandom number generators

Drpepper · Post by **Drpepper** » Tue Sep 25, 2012 11:20 am

Anyone heard of these before??

Apparently a FOBT machine can ever have a RNG or a PRNG, with a PRNG being at least in part influenced by the previous spin(s)...

"Category B

Category B games are divided into subcategories. However, the differences between B1, B3 and B4 games are mainly the stake and prizes as defined in the above table. Category B2 games – Fixed odds betting terminals (FOBTs) – have quite different stake and prize rules. FOBTs are mainly found in licensed betting shops, or bookmakers, usually in the form of electronic roulette.

The games are based on a random number generator and thus the probability of getting the jackpot in each game is independent of any other game, and these probabilities are all equal. If a pseudorandom number generator is used instead of one that is truly random, the probabilities are not truly independent, since each pseudorandom number is determined at least in part by the one generated before it."

betchrider · Post by **betchrider** » Tue Sep 25, 2012 12:05 pm

Oh oh!!

Noels Beard · Post by **Noels Beard** » Tue Sep 25, 2012 12:50 pm

I think you'll find that this is to do with the maths of how the numbers are generated. Whilst this may add grist to the mill of any conspiracy theorists, it has little real world influence. I have very limited maths knowledge but let's attempt a working example from first principles.

In this example we will attempt to randomly create a number between 1-10. What I would do is start with say a very large number, say 17689486 and then apply a series of calculations to it to find the next number. I'm just making this up as I go along, but say we're going to find the square root of this number and then times that by 9.

I make this next number 37,852.984 and the next 1,751.0259. If we just take the last digit of these numbers I'm pretty sure they're close to random. Obviously there are certain processes that are going to produce more random results than others, simply dividing by ten is not sufficient. However, you could pretty quickly do some statistical analyses to show whether the results are trending in any direction. And before anyone pulls holes in the example above, I'm not sure whether the processes I've applied here would work, it's just a proof of concept.

As for the idea that you could use this knowledge to unravel the next numbers....fat chance. They're being thrown out at such a high rate it would be impossible to get your bet in at the right time, add to this the complexity if reverse engineering the formulae. You'd be better off applying for a job at NASA.

AMK · Post by **AMK** » Tue Sep 25, 2012 12:56 pm

This needs to be fully analysed by Plum.

Plumy2k · Post by **Plumy2k** » Tue Sep 25, 2012 1:54 pm

Simply by square rooting a number and multiplying by nine then using the last digits will not give a random number as there is a process behind how to get it. A number can never be truly random however I would imagine it would be close if you had 1 billion identically weighted balls with numbers between 0-36 and for one ball to be picked blind out o the bag as long as there were the same number of each number.

Noels Beard · Post by **Noels Beard** » Tue Sep 25, 2012 2:12 pm

Well obviously that's better. But I was trying to explain it as a mathematical process. There is a whole branch of maths to do with it. For those interested....this seems to back me up nicely :-)

http://en.m.wikipedia.org/wiki/Pseudo-r ... _generator

harry2 · Post by **harry2** » Tue Sep 25, 2012 2:21 pm

Some people spend their whole lives tying to approve/disapprove whether systems can generate a true set of randomly generated numbers.

Me, I'll just go down the pub.....

......life's too short.

Noels Beard · Post by **Noels Beard** » Tue Sep 25, 2012 2:39 pm

Well there is quite a legitimate argument that no event at scales above the atomic level is truely random. Both reassuring and strangely depressing!

maverick69 · Post by **maverick69** » Tue Sep 25, 2012 2:45 pm

FOBTS are rigged end of story ....that is it

blackmogu · Post by **blackmogu** » Tue Sep 25, 2012 3:07 pm

True RNG's are expensive and relatively slow. PRNG's are fast and good enough for most applications. Take the Mersenne Twister for example, which is the PRNG of choice in several computing applications, with a period of 2^219937 − 1.

Simo7 · Post by **Simo7** » Thu Sep 27, 2012 1:09 am

In hills the other day see two terminals next to each other with identical history for the last 7 numbers took a pic of them numbers were 27 28 8 22 29 1 35 while not beyond the realms of possibility at all it just seems strange. When i tried to work on a possible error or exploit I was quickly 170 down
Luckily the conference held up as the best league for both teams to score and won losses back plus profit. http://www.random.org has a lot of info about rngs and also has what they say is a true rng

Cf · Post by Cf » Thu Sep 27, 2012 2:31 am

If they are generating their own spins then it IS beyond the realms of possibility that they have the same last 7 numbers.

mr lugsy · Post by **mr lugsy** » Thu Sep 27, 2012 6:29 am

you could compare that to 2 simultaneous independent lotteries throwing out the same numbers in the same order. a bit fishy.

Ruler of The World · Post by **Ruler of The World** » Thu Sep 27, 2012 8:20 am

What's the mathematical probability that Frankel wins the Champions Stakes at Ascot?

Simo7 · Post by **Simo7** » Thu Sep 27, 2012 1:58 pm

Yeah it is very very fishy I didnt see who had playing either machine or what sort of bets they were doing and these 2 terminals were next to each other. Its just bollocks that could happen yet on just a quick trip to put some footie bets on in the bookies there it was when I started betting both machines were just giving out all sorts of scattered numbers and generally robbing me.
I rang up global draw a year or so back basically saying the results the machines display are not random and they tottaly fobbed me off in a cuntish way. Ended up speaking to some manager guy at ladbrokes also after ringing them he said it would be "commercial suicide" to rig the games. It just pisses me off when I see stupid historys on the machines that theyd find hard to explain but also they know doesnt prove anything against them