Explain this about the random number generators

[mc1]

Lee Jones recently stated that PokerStars at least are preshuffled before the hand starts.

[puckboy]

isn't this the way it should be????(preshuffled) since that would most mimic a B&M room. The dealer isn't constantley shuffling the cards throughout the hand. However, The constant shuffle premise actually makes the hand "more" random then the B&M preshuffle.

Either way however the shuffle is random and NOT rigged which is all i ask for personally.

[Yads]

[ QUOTE ]
However, The constant shuffle premise actually makes the hand "more" random then the B&M preshuffle.

[/ QUOTE ]

How so?

[ QUOTE ]
[ QUOTE ]
However, The constant shuffle premise actually makes the hand "more" random then the B&M preshuffle.

[/ QUOTE ]

How so?

[/ QUOTE ]


Any human shuffling is technically not completely random. This is because we don't shuffle the deck perfectly. Professional dealers probably shuffle better than the donk in your friend's home game. For example, 2 Q's might might stay next to each other in the deck even after multiple human shuffles. In reality, this minor non-randomness doesn't make much of a difference; it's close enough that we don't care. The RNG comes as close to truly random as you can get.

[ QUOTE ]
Any human shuffling is technically not completely random. This is because we don't shuffle the deck perfectly. Professional dealers probably shuffle better than the donk in your friend's home game. For example, 2 Q's might might stay next to each other in the deck even after multiple human shuffles. In reality, this minor non-randomness doesn't make much of a difference; it's close enough that we don't care. The RNG comes as close to truly random as you can get.


[/ QUOTE ]

True, but not the answer to the question. puckboy was suggesting that the constant shuffle was "more random" than the one-time shuffle, but both are with RNGs.

[puckboy]

yup that's exactly what i was trying to say, both are random, one might be slightly better than the other, but they are both close enough to make me feel comfortable

Yeah . . . I'm still wondering why one is better than the other.

[ QUOTE ]


Quote:
--------------------------------------------------------------------------------

However, The constant shuffle premise actually makes the hand "more" random then the B&M preshuffle.


--------------------------------------------------------------------------------



How so?

[/ QUOTE ]

I'd really like to see the answer to this one, if someone's got a moment. Though obviously, you get a more random shuffle with a more unique RNG.

[Rudbaeck]

[ QUOTE ]
isn't this the way it should be????(preshuffled) since that would most mimic a B&M room. The dealer isn't constantley shuffling the cards throughout the hand. However, The constant shuffle premise actually makes the hand "more" random then the B&M preshuffle.

[/ QUOTE ]

It already is many orders of magnitude more random than what any human dealer can accomplish before the heat death of the universe either way. But both continuous shuffle a pregen decks are equally random.

The downside to shuffling the deck once and keeping it for the hand is that it encourages hacking. As far as I can tell there is very little extra computation required to get a new shuffle for each card, so the sites should go with the slightly better alternative. If nothing else it soothes the nerves of paranoids.

[Greg Miller]

[ QUOTE ]
The downside to shuffling the deck once and keeping it for the hand is that it encourages hacking. As far as I can tell there is very little extra computation required to get a new shuffle for each card, so the sites should go with the slightly better alternative. If nothing else it soothes the nerves of paranoids.

[/ QUOTE ]

Actually, not running the PRNG for cards that are never dealt saves CPU load. PRNGs are semi-expensive.

As you said, generating the whole shuffle at once is a security issue, potentially, because PRNGs always generate the exact same sequence of numbers (cards) given the same starting state. If there's a problem with entropy collection or some other issue that allows someone to know the starting state, then the entire deck can be accurately predicted. If, on the other hand, the PRNG output stream is sampled at unpredictable times, how do you predict the next card?