Quote from: espringe on August 26, 2014, 02:55:04 AM

But yeah, there's definitely a real danger of someone putting a significant dent in my bankroll. But hey, I guess I'm just a gambler at heart. Grin

At least this time the odds are with me

Good luck with that. I just wanted to make you aware of the risk you were taking if you weren't already.

You're the third instance of someone limiting the maximum bet (and not just the maximum profit per bet) I've seen recently.

PrimeDice effectively does that, by limiting the maximum payout per bet (including the return of your stake) to 40 BTC. So if you want to bet at 98%, 1.01x, you're limited to betting around 39.6 BTC to win 0.4 I think.

Then dice.ninja launched, and they limit the max bet to 5 times the max profit (which is 0.5% of their bankroll). Apparently the same guys own dark-dice or some such, which lost a lot of coins to someone betting high stake, high chance of winning, low payout bets over and over and getting lucky.

I wonder if there's a good reason to limit the max bet beyond just limiting the maximum profit per bet. The Kelly criterion tells me no, but when I see it happening in three different places maybe there's a reason for it.

Quote from: espringe on August 26, 2014, 03:58:52 AM

Quote from: dooglus on August 25, 2014, 05:05:53 PM

So you have a billion-long chain of sha256 hashes.

Hmm. I just tried it, and was able to create a 0.2% undetectable house advantage for me, for the first year of operation.

Given this, I'm not sure if I should bother continuing down this route?

Well, this is still better than the current situation, which is that you get to create whatever house advantage you like, whenever you like. Don't you?

The problem could be solved using the server-seed, client-seed technique of other sites. Only instead of per player and per randomize, it only happens once.

Create your chain of a billion sha256 hashes, publish the billionth hash, then (and only then) have someone trusted pick a client seed which will be combined with each of your billion hashes in turn to pick the outcome of each game. That basically makes you commit to your hash chain before you are able to tell whether it is a "good" or "bad" one.