It's also worth noting that even if dice sites had some cutting edge techniques that broke sha2 and were able to generate collisions, still their ability to undetectably cheat is still insanely limited. That's because generating a single collision isn't enough, but you need to generate one that would give all previous games, which is an incomprehensibly hard problem that gets harder exponential with the amount of games played.

In reality, if there's any weakening of sha256 you can be pretty confident it will be used *against* dice sites (and/or the bitcoin network) long before dice sites are able to feasibly use it against players.

No, you can very easily overcome the house edge and people regularly do in practice  

It's most simple to analyze if you just consider if there's one other person playing.

 But the rule of thumb to remember, if you can average 1% bonuses, you're neither losing or making money due to the bonus scheme. If you can average 2% bonuses, you're guaranteed to be overcoming the house edge (but most of the time, you need less than that. e.g. if you're playing at 2x, you only need an average bonus of 1.5%)

Bustabit is a bit unique in the sense that there is actually a small aspect of skill (bonus hunting) so practicing definitely can help. So if you want to try play against real people and not worry about losing anything then bustaclam might be a good option

Looking at the charts:

https://dicesites.com/satoshidice

On Jan 11 they ostensibly lost 326 BTC, so that's probably where the drop comes from

Yeah, that way I can test withdrawals/deposits too. But there would also be an onsite faucet, so it's easy to acquire anyway (and there's also a whole bunch of testnet faucets) that give out tens of test coins so they're pretty easy to get

Yeah, actually that's what I've been working on today. I'm planning on having a permanent test.bustabit.com which will operate a complete clone of the site (but in total isolation) but with bitcoin testcoins, instead of real bitcoin. But it'll allow depositing/withdrawing/investing, so that it'll not only will it be a good place for users to screw around with "fake" money, but give me a good way to do live-deploys and test buggy code, get feedback etc.

There's actually a few pretty cool people on bustabit who have never deposited and unquestionably make the place a better and more interesting place. But most of them are interested in the maths/game concept/bonus hunting strategy stuff, rather than being focused on becoming rich off free money, or what not. Actually one of them was a mathematician who helped me solve (read: did it for me) a maths problem I had been struggling with for the last 6 months 

So i definitely don't mind if people have no intention of ever depositing, I've just had a bad experience with the sort of people who are professional fauceters 

I numb it with alcohol 

But more seriously, I've started keeping the hot wallet extremely full so I don't need to have the think about it when someone wins a couple hundred bitcoin


Although actually I think the swings now in bitcoin price are causing a lot more networth variance for me.

Not a bad way to make 10.8 BTC:

https://www.bustabit.com/game/3553128

Nah, it was just a play on "stunna"  


Again, they don't owe you any evidence. And I'm not sure what exactly you think happens in a hack? A hack doesn't leave some cryptographically verifiable evidence of a hack.

If you choose to believe it was a publicity stunt, then fine. But if you're going to call someone out for lying with zero evidence of your own, you just look like a douche (which I guess explains why you're using a new account).

I've done extensive simulation on the topic and studied the maths, it's seriously not a viable business. There's no way you can stay in business taking unbounded risk for 0 expected profit. There's a lot of psuedo-science around thinking that "greedy" strategies are good for the house, but it's simply not true. (It's just equivalent to playing at a high payout)

However, if you do it pvp way, then it's totally possible. Then you could take no risk, and match gamblers with gamblers. That could totally work.


But another way to look at it, even with a house edge running a casino is an extremely risky business. In the calendar month of October, I lost around ~402 BTC bankrolling bustabit, where the weighted average house edge during that month was around ~0.7%.  

No sane person is going to take those sort of risks for no expected profit  

https://thebitcoinstrip.com/leaderboard.html


 

I'm really not sure what more evidence you could possibly expect? They've provided a lot more evidence that it was a real heist than you provided it's fake (i.e. none) and posted a pretty detailed write-up of exactly what happened.

If you choose to believe it was a publicity stunt .. well then go for it. The truth is the only money was lost was their own, they really owe you nothing.


If you compare it to bitfinnex who lost a huge amount of other peoples money and has provided way less information (and not even said if they've found and fixed the issue) I think the PD team has handled it stunningly well.

Yeah, I'm rather against deposit bonuses. I've really yet to see any that aren't traps for players, and they'd be better off playing without them. I feel like the best (and most honest) thing I can do, is keep the house edge as low as economically possible.

I do plan on bringing back the faucet however, but in a limited form. The way I've coded it so far (which is subject to change) is you'll get 5 claims per ip address. Once an ip address has claimed a faucet more than 5 times, it's forever locked (for that ip address)

The primary point of this, is it should give most new players a chance to try the game without depositing (assuming they're not using a shared ip that's already burnt the faucet) but it'll not be a big target for abuse or professional fauceters 

Another possibility would be to let it hit the db (that way you can use other APIs, as it would have a proper bet id and what not) but have a daily job that deletes any 0 bets older than a day. It shouldn't really affect the audibility of anything, as those bets wouldn't have affected balance and you don't have to deal with the long term storage of it either.

Unfortunately not, we now require all users to deposit to play

In the past we had a faucet, but unfortunately it attracted a lot of the wrong sort of people (abusers and professional beggars) so I removed it, which had the unfortunate side effect of making it annoying for new players to try the game (wait for a deposit). So in the near future (~1 month), I'll add a small faucet (with a limited amount of claims per ip address) to assist with trying the game

It's pretty difficult to say. If I had to guess, I'd say bustabit isn't affected too much as at least I certainly haven't noticed any changes in volumes, although since bitcoin is worth more now, our volume denominated in USD would have gone up a lot.

What really makes a huge difference, is that as in bustabit everyone is playing the same game, if the recent history (especially last 8 or so hours?) has had a lot of good game results, we have a lot of activity (people recycling their winnings, etc) and if we have bad games activity goes way down (e.g. a lot of people bust or rage quit). So it's pretty easy to correlate activity with the recent history of game results, but bitcoin prices, not noticeably so.

No, not really. Someone knowing how you pick passwords would have got it in less than 10 tries, or perhaps 26 if they wanted to guess every letter. You can't lock accounts, or ban ips for a handful of tries or it would suffer from a huge false-positive problem and/or be a huge vulnerability (easy to lock other peoples accounts).


This will be my last reply, as it's becoming a waste of time. But honestly, you should start taking some responsibility. It seems you've learnt two good lessons:
a) Don't post an active password in a public spot
b) Use a password manager to pick good password.  A single letter, and a common suffix is insanely guessable.


I hope you don't let this lesson go to waste 

I was actually the one who originally asked you to share your password (after you changed it) so we could see if it was a secure password or not (like you claimed).



Your whole thread is about problems in PrimeDice, while in reality it's simply a case of you using a trivially guessable password (to anyone who looked up how you pick passwords on a password leak site).

Comments that are edited after a threshold (5 minutes I think) look like this:

https://imgur.com/a/BOWYt

(that's my post, for testing)

And you can hover over it, to see the edit time.

However, Stunna's was never edited (at least after the threshold)