PDA

View Full Version : Pokerstars shuffle sources


frostbrn
08-25-2005, 08:10 PM
On their website, Stars indicate that they take "user input" into their shuffle algorithm. They mention mouse movements and data collected from client software. Does this mean that the way in which I move my mouse has at least SOME effect on the cards I am dealt?

OrangeKing
08-25-2005, 10:22 PM
[ QUOTE ]
On their website, Stars indicate that they take "user input" into their shuffle algorithm. They mention mouse movements and data collected from client software. Does this mean that the way in which I move my mouse has at least SOME effect on the cards I am dealt?

[/ QUOTE ]

Probably, but you wouldn't know what impact unless you first knew exactly what kind of input every other player was making, and knew exactly how it would effect the RNG. I wouldn't worry about it. /images/graemlins/smile.gif

daryn
08-26-2005, 03:45 AM
yes, if you lose with AA vs. KK all in preflop, you should curse yourself for having moved or not moved your mouse.

Rosie5
08-26-2005, 07:10 AM
[ QUOTE ]
On their website, Stars indicate that they take "user input" into their shuffle algorithm. They mention mouse movements and data collected from client software. Does this mean that the way in which I move my mouse has at least SOME effect on the cards I am dealt?

[/ QUOTE ]

yes but the equation for grabbing a random number from your mouse movements is probably so complicated that you couldn't even fathom it

partypoker is worth 10 billion, I'll go out on a limb and say they're probably a little more advanced than something you could predict/change with any consistency

VivaLaViking
08-26-2005, 12:46 PM
All PRG's have idiosycracies that make them non-random. ALL. As far as mouse movement, it will have the physical constraints of the window bounds that adds to the idiosychracy. I believe they would be better served using the cumulative times that users resond as the seed value for the PRG.

VivaLaViking
08-26-2005, 09:21 PM
I guess a better way to explain this is you cannot put physical constraints on even a purely random process without anomolies. A pseudo random generator, PRG, makes these anomonies more apparent. for example if the window size was 100, all results will contain an anomoly due to modulo 100 values. Yes, I recognize they may use the x and y dimensions but the same principle applies.

magiluke
08-26-2005, 09:22 PM
[ QUOTE ]
All PRG's have idiosycracies that make them non-random. ALL. As far as mouse movement, it will have the physical constraints of the window bounds that adds to the idiosychracy. I believe they would be better served using the cumulative times that users resond as the seed value for the PRG.

[/ QUOTE ]

First off, you are thinking too that it is only taking the position of the mouse as input. It could be taking ANY kind of user input. The amount of people that fold between shuffles, sentences that people type could be sent through a complicated algorithm that determines the multiplier and add-on to the seed. Each button (Call, Raise, Fold) could add or subtract a different (possible random) amount to the seed each time anyone clicks it. It could be even more complicated than that!

Whatever the case, I consider the process to be random. You don't have to detail to me how computer programs aren't random, cause I know; I've written some (functional, although basic) number generators. The reason why I (and many others) consider it random is that the seed itself (and probably the multipliers and add-ons) are taken from unpredictable human action.

If you want to get technical, nothing is random EVER. They have built machines that will flip a coin and land on heads every time. People have practiced flipping coins so they can land heads every time. People have practiced so that they throw the numbers they want in craps. What makes thigs seem random is the fact that things are done slightly differently every time they happen--every time you flip a coin, you position it differently on your thumb, you flip it in a slightly different direction, you apply more or less joules.

I'm tired of typing now.

sakki
08-26-2005, 09:37 PM
[ QUOTE ]
If you want to get technical, nothing is random EVER.

[/ QUOTE ]

I don't think so. See random.org (http://www.random.org/essay.html) for starters.

VivaLaViking
08-26-2005, 09:46 PM
[ QUOTE ]

First off, you are thinking too that it is only taking the position of the mouse as input


[/ QUOTE ]

I am not making any assumptions. The only reason the mouse was brought up is the origonal post referenced it. I was told by a Poker Stars representative that they are using a new Intel chip set, the only thing I can say is that Intel is no guarantee of reliability (they had a recall of Pentiums for a math problem). But after all your typing we reached the same conclusion, it is not random.

magiluke
08-26-2005, 09:58 PM
[ QUOTE ]
I don't think so. See random.org (http://www.random.org/essay.html) for starters.

[/ QUOTE ]

I'm aware of using noise and radiation to generate random numbers (I've never heard about lava lamps though... That's cool). The reason that they are "random" is because they are very unpredictable. If there was ever a way for us to take into account every single thing that would influence these measurements, we could predict the outcome. In most cases, we just can't do this yet (See: Dr. Heisenberg).

sakki
08-26-2005, 10:07 PM
[ QUOTE ]
If there was ever a way for us to take into account every single thing that would influence these measurements, we could predict the outcome.

[/ QUOTE ]

Yes, that's true. But then the measurement wouldn't anymore be random. The definition of the word includes the uncertainty.

I guess a more important point is that no algorithm is capable of producing truly random numbers.

magiluke
08-26-2005, 10:23 PM
[ QUOTE ]
Yes, that's true. But then the measurement wouldn't anymore be random. The definition of the word includes the uncertainty.

I guess a more important point is that no algorithm is capable of producing truly random numbers.

[/ QUOTE ]

No algorithm, or anything!

Anyway, I think you are missing the point here. Just because you don't know the measurements doesn't mean it's random.

Let me steal some stuff from Wikipedia (http://en.wikipedia.org/wiki/Random) (I love it!):

[ QUOTE ]
The word random is used to express apparent lack of purpose or cause. ... Note that the bias that "everything has a purpose or cause" is actually implicit in the expression "apparent lack of purpose or cause".

[/ QUOTE ]

Here's the top link (http://wetzel.psych.rhodes.edu/random/mainbody.html) on the bottom of that Wikipedia article. It's kind of neat.

Also, here's a definition given by a link I found in Random.org (the source you provided) (link) (http://www.lavarnd.org/faq/chaotic_src.html):

[ QUOTE ]
A chaotic source is a physical system that is dominated by chaos. A small change in the physical system now, will increasingly impact future conditions within the physical system over time. The impact of a small change compounds over time in such a way that correlating between the disturbed state and future conditions becomes intractable.

[/ QUOTE ]

As according to this source, "a digital snapshot of a physical chaotic process is obtained" to use as a seed for the random number. But as according to the definition above, any "digital snapshot" is directly influence by previous states of that source.

Also, here's a link to Wikipedia's article on determinism (http://en.wikipedia.org/wiki/Determinism), because that is important too.