Without question the "quality" of players here is poor and you will frequently lose with excellent pocket cards. While I don't profess to be a professional or even reasonably accomplished player I can tell when to fold and when to play. However, as noted by others if there were real money at hand, play would certainly be difference as I have witnessed when playing at pay sites. So lesson with GPokr, fold the pocket aces and all with the 2, 7 offsuit!

Scalba, if the average standard of gpokr players is well below your own I look forward to seeing you in the top 10 at the end of the month.

LOL

A random num gen sucks
If you want a good realistic shuffle you have to create an array to hold the cards. Now create two arrays each will contain half the cards. Into a fourth array you will alternate cards from the two half deck arrays based on even or odd through a random num gen.

If the game is not doing this to shuffle, it's not like real poker.

kilo - your a bitch. shut up

holdme: trying to simulate a real shuffle like that doesn't make it more random... in fact restricting yourself to that method could lead to errors.

The benefit of electronic shuffling is that it isn't restricted physically. It's the physical restrictions that cause a deck to not be shuffled well. Why try and simulate this?

I should add...


Using your version of shuffling you're still using a random number generator. Any other convoluted algorithm you add on top I can pretty much guarantee won't approach the randomness of a good random number generator. There's no reason to think this generators are poor. They are serious business for computation and are used extensively for security schemes.

To add a bit of transparency to the GPokr shuffle: The deck is held in an array. Each card in the deck is swapped with another card in the deck based on a random location in the array. No card remains in it's original location and ordering correlation between cards is random.

The gpokr deck should be given to a chimp who throws it into the air, a second chimp could then be used to eat some of the cards if necessary. Perhaps even a thrid chimp could be used but i dont want to get silly.

Surely this would make it very random, but i assume Ryan is implementing this as i type.

You said:
> No card remains in it's original location
Doesn't this constraint speak against a real randomness? When do you "shuffle" the deck, once before each round? Cause then I could always know for sure which card would not be in a certain stack position.
Would it be a problem for you to publish just the shuffling algorithm in pseudocode or any programming language you want?

there's a 1/52 chance that a card is in the same position

Perfect then, guess I misunderstood this. I'm not one of the people complaining about bad shuffling here btw, everything perfectly random in my opinion. The big difference here is that you can play many hands in a short period, while in real life 20 hands in an hour would be a lot, any in real money poker site people hardly see a flop with a bad hand preflop. And these facts combined make people think it would be rigged here.
An example would be AK, people have the feeling it would lose more often than in real life. It actually does, but that's not because the cards are shuffled wrong. The big power of AK is that you dominate so many "good hands". And if you raise preflop there should only be good pockets or ace or King and a very good kicker in it. But in gpokr people call with 89 suited or even worse stuff, the problem of play money.