Bug in Blackjack with splitting a hand
Hi,
Discovered some new aspects of this bug.
Splitting a hand is not considered playing 2 hands in Blackjack. The bet is just doubled.
If you double the first hand again, the total bet is doubled. (And if you also double the second hand, the total bet is doubled again.) You can increase your bet to 8x the initial bet this way.
This isn't good as you should play hand1 at 2x the initial bet and hand2 also at 2x the initial bet. (The math is correct when you lose one hand and win with the other -except you pay too much for the jackpot- but losing both hands loses 8x and winning both hands wins 8x, while it should only be 4x.)
On this site, when busting hand1, you are not allowed to play hand2 (I think, because the game thinks you are busted). But, if the dealer busts and you win hand2, you are paid correctly (= play even, minus addition to Jackpot). But losing both hands doesn't lose another bet, strangely.
A solution may be to seperate all bets and not add them together in one stack as it does now.
- When the dealer shows an ace and the player decides to insure, place the half bet on the insurance line. Win this bet and the winnings (= 2x) is put next to the half bet and then paid to the player, but the original bet is lost and moves to the dealer. Losing the insurance means the bets on the insurance line are shoved to the dealer and players can continue playing.
This (the mathematics) is done well now! But the animation may be unclear. The half bet is added to the main stack and vanishes when the dealer doesn't hit Blackjack and the dealer does hit Blackjack, the player wins his original bet minus the addition to the jackpot.
- When the player splits his hand, make 2 seperate chip stacks at each hand. (Since you allow only 1 split, the player can only have 2 hands maximum and doesn't need extra space.) You can then treat each hand as a seperate hand and the split-doubling to 8x and the forced pass after busting the first hand will be avoided.
Hope this helps.
Regards,
SiuYi.

Thanks, fixed in b164.