Topic: Distributed trustless poker (Read 5757 times)
Would love to see your progress.
Another thread about this topic
https://bitcointalksearch.org/topic/poker-and-the-shared-pot-at-the-table-in-a-decentralised-network-1487
https://bitcointalksearch.org/topic/poker-and-the-shared-pot-at-the-table-in-a-decentralised-network-1487
We need PokerCoin - Blockchain based poker :p
I've talked to some devs about this. They said it would bloat the blockchain quite a bit and cost a lot in transaction fees. Along with that, you are left with unconfirmed transactions unless you wait 10 minutes between the the flop, river, etc.Naw i don't mean use the bitcoin blockchain, i mean create our own blockchain that is intended specifically for the kind of transactions we would be doing. Nice, fast, and reallly super long term storage isn't that necessary - a game will only last a few hours at the most.
We need PokerCoin - Blockchain based poker :p
I've talked to some devs about this. They said it would bloat the blockchain quite a bit and cost a lot in transaction fees. Along with that, you are left with unconfirmed transactions unless you wait 10 minutes between the the flop, river, etc.
We need PokerCoin - Blockchain based poker :p
Thought about this a bit. One way you could do it is to have an event that happens 50% of the time be part of your transaction.
Not sure what event has a 50% chance of happening other than a trusted oracle that signs transactions 50% of the time.
Basically for rock, paper, scissors, you would make a transaction with two inputs, the two betters and have it spent to one users output if an oracle signs it within 24 hours. If an oracle doesn't sign it, then the other user gets it.
Edit: Look at this post, he has a much better algorithm than me: https://bitcointalksearch.org/topic/fair-coin-toss-with-no-extortion-and-no-need-to-trust-a-third-party-277048
Not sure what event has a 50% chance of happening other than a trusted oracle that signs transactions 50% of the time.
Basically for rock, paper, scissors, you would make a transaction with two inputs, the two betters and have it spent to one users output if an oracle signs it within 24 hours. If an oracle doesn't sign it, then the other user gets it.
Edit: Look at this post, he has a much better algorithm than me: https://bitcointalksearch.org/topic/fair-coin-toss-with-no-extortion-and-no-need-to-trust-a-third-party-277048
sounds good in theory, but you still have to see your own hand, which means the other side has to give you their part of the key, which also means you can see their hand.
Seems like it could be gamed pretty easy. If a user loses rock paper scissors, they move their coins to another address before the winner broadcasts his winning transaction.
rock, paper, sissors is suitable for a simple form of gambling, but the ties slow things down.
It is better if each side chooses a number. If the number sum to even, then person A wins. If they sum to odd, then person B wins.
Maybe person A will try and say his number slightly late, so that he can listen to B and cheat.
To avoid this, both parties should write down the number they chose, and then show each other.
or equivalently, they trade hashes of the number they chose first.
This is a very simple case of http://en.wikipedia.org/wiki/Distributed_computing distributed computing.
I have a plan on how I can use this to build a trustless distributed poker system.
It is better if each side chooses a number. If the number sum to even, then person A wins. If they sum to odd, then person B wins.
Maybe person A will try and say his number slightly late, so that he can listen to B and cheat.
To avoid this, both parties should write down the number they chose, and then show each other.
or equivalently, they trade hashes of the number they chose first.
This is a very simple case of http://en.wikipedia.org/wiki/Distributed_computing distributed computing.
I have a plan on how I can use this to build a trustless distributed poker system.
Jump to: