Try watching your own site for more than 5 minutes. It happens pretty regularly.

This one was interesting - the game ran when there was only one participant. At least he won, I guess:



Neither screenshot is fake. Both these things really happened. Your inability to accept feedback and react to it is appalling. I'm not even asking for bug bounties. This is all free!

(You're welcome )

Heheh.

I just saw a game end with "Participants(0), winner: Zeb" but I didn't manage to screenshot it in time.

As Creative says, run the clam client with the -rescan command line argument.

How did you import your wallet? I wonder why it didn't rescan automatically for you. It should do.

Which version of the client did you use?

I use the 'watchlist' feature extensively.

Sometimes a thread shows up on the watchlist for me but when I look at the only new post is by a user I have on my ignore list.

Would it be possible to have threads only show up on the watchlist if someone I don't ignore has posted to them?

I just watched a game where the "lukcy" winner wasn't even one of the two people who were playing:



I guess you have bigger problems than provable fairness.

You said that already. Do you have any actual arguments? Saying "false" seven times doesn't count.

No, you haven't.

Look at your 7 falses, and see if you can justify them.

You asked me to prove your site wasn't provably fair. I broke it down into 7 steps so you could understand. You just wrote "false" next to each of them, when I'm sure even you could understand at least some of them if you tried your hardest.

You're clearly trolling, not me.

Unfortunately, lots of places do it because they know that if you can't withdraw you will probably be tempted to play some more, but it is borderline scammy behavior. Any reputable casino should allow you to withdraw your winnings whenever you want them.

Obviously nobody is going to keep their whole bankroll online, available for instant withdrawal, but there's no reason not to keep enough available online to service reasonable withdrawals.

At Just-Dice for instance we have a bankroll worth around $1.5M but only have coins worth something like $40k available for instant withdrawal. If you win more than $40k's worth you'll need to wait for up to 12 hours for the hot wallet to be refilled.

I am encouraging him to make it better, but he refuses to understand what is wrong with it.

His site has approximately zero volume. It isn't in competition with Just-Dice at all. I have no problem with competition. My problem is with deception. OP is lying about his site being provably fair. If he thinks he's able to get away with that without me pointing it out, he's mistaken.

OK, let me stop you there.

If you don't think it is possible to generate a transaction without broadcasting it then you really have no idea how Bitcoin works.

You never heard of offline wallets? Armory? Paper wallets?

This alone proves to me that you either don't have a clue what you are talking about or are deliberately being deceptive.

I'll leave you appropriate feedback as soon as I finish this post.


So you're saying each time someone bets, you have 10 seconds to figure out which of your pre-generated transactions would make you win? That's about a thousand times longer than you need.


Magic, luck, or cheating. Any of those 3 would work.


I don't think you are a super computer. I think you're probably either a child, an idiot, or a scammer. It's hard to say which. Possibly all three I guess.

You know exactly when the game is ending: 10 seconds after the last bet.
You know which transaction will let you win: the one which selects a ticket you bought.
You know how to push a transaction to blockchain.info in less than a second and pick it.


Try making some arguments against me. Writing "false" after my seven valid statements just makes you look silly.


Wow, thanks for that. You're a lot more patient with this guy than I am. You actually made valid arguments supporting all seven of my statements, even though he probably didn't even read them.


True that.


Well, it has to be accepted by blockchain.info. I think they reject transactions with too-low fees. I'm not sure exactly what their rules are though, so maybe you could find a way of cheating for free. Either way if the 0.0001 BTC fee is the only thing stopping people from cheating I hardly think we can call it 'provably fair'.


I know you're joking, but do you think mr. raffle is going to have any idea?


How about a dance off? The one with the freshest moves is right.


I wanted to make sure he understood how his site wasn't provably fair and give him a chance to fix it before leaving appropriate feedback.

I don't want him to be able to claim that I left it just because I ran a competing site.

But as you can see above, he's just trolling. He won't respond to any technical points, either because he can't or because he is willfully ignoring them.

So now that it's clear that he doesn't want to fix his site I'll leave him the feedback and move on.


I'm not sure even that is strong enough.

I can't wait until they invent super-duper impregnable I-can't-believe-it's-not-rigged undeniable provable fair. With extra timestamps proving we picked a random transaction within 2 seconds of the end of the game to decide the winner. For fuck sake.

No, that probably means you are typing a BTC address when you should be typing a CLAM address.

The /dig command on JD will tell you the CLAM address for any given BTC address. That's the one you need to check with validateaddress in the CLAM client:

I type /dig then my address and see:

    11:50:01 INFO: CLAM address [x9KpGgpB] corresponding to BTC address [122BNoyh] was funded with 4.60545574 CLAM ...

I click the [x9...] link and it shows me the full address. I use that full address with 'validateaddress' and see:

  "isvalid" : true,
  "ismine" : false

ie. it isn't imported into my current CLAM wallet, but is a valid CLAM address.

One of:

1. the keys don't really have CLAMs in them; are you sure they do? can you give an example address?
2. the import didn't work; can you try 'validateaddress' in the CLAM debug console for the corresponding CLAM address and check 'ismine':true?
3. the blockchain isn't really synced; check 'getblockcount' in the CLAM debug console

For step 2, '/dig <address>' in the Just-Dice trollbox will tell you the corresponding CLAM address.

Maybe there's a 4th thing that may have gone wrong, but I can't think of it.

So first you were saying it was IMPOSSIBLE for you to use pregenerated hashes:


And now you are saying that it is possible so long as you are willing to pay a few dust fees?


It takes less than a second to get a transaction onto blockchain.info, so you know which second it will be there - it's the same second as you sent it to them.

So it seems this attack has gone from "impossible, 100% false" to "anyone can do it for 0.0001 BTC". Hmmm.


Which bit is impossible? Affording the fee? Or sending a transaction at the right time?


OK. One more time:

The system is not provably fair because the site can affect the outcome of the game after the tickets have been bought by broadcasting pregenerated transactions at the right time, and deliberately selecting that transaction as the one that affects the selection of the winning ticket.


You want to see proof of what, exactly? Here are the main points:

1) you can generate transactions without broadcasting them
2) you can check the txid of those transactions before you broadcast them
3) once you know the number of satoshis bet, you can figure out which ticket would win for each transaction
4) so you can find out which of your 100 pregenerated transactions would make you win if it was picked
5) it takes less than a second to get a transaction picked up by blockchain.info
6) you have no way of proving which was the last transaction you saw on blockchain.info after the 30 seconds is up
7) you can pick any transaction that was shown on blockchain.info in the relevant second

Can you tell me which of those 7 points you need proof of? They all seem self-evident to me.

I have no reason to make any site look good. My interest is in making sure that sites which claim to be provably fair actually are. And yours isn't.

Typo:

"if you do not use a unique passwords"

Care to explain why you couldn't use pre-generated hashes? You would just need to broadcast them shortly before the 30 second timeout.


I understand it just fine. It isn't provably fair.

Maybe, though some think he had cryptorchidism.

The first I heard of your site was when you turned up on the pevpot thread making accusations that they were cheating their players.

I checked out your site and found that it isn't provably fair, whereas pevpot *is*. We went over this in detail on the pevpot thread.

I have tried numerous times on that thread to get you to understand the problem, but both you and the "developer" you used fail to grasp the concept.

I can't tell whether you're leaving the site how it is because you're incapable of understanding what's wrong with it or because you like having the option of cheating your players. I guess it doesn't matter which of the two it is - either way it would be a mistake for anyone to trust your site.

Please either fix the site so that it is provably fair OR stop claiming that it is provably fair. Come on, do the right thing here.

There's one draw every 1000 blocks. Since there are 6 blocks per hour and 24 hours per day, that means there should be one draw every 6.944 days:

  >>> 1000.0 / 24 / 6
  6.9444444

If you don't want to have your coins tied up for too long, you can wait until near the end to buy your tickets. But you get a better share of the "early bird bonus" if you buy your tickets early.

I answered every one of your comments, and this is all you have for a response?

I think it stands as a brilliant concise demonstration of why nobody should employ you in this field.

1. you're rude, resorting to ad hominem attacks rather than making valid points
2. you haven't researched the field you are attempting to enter
3. you're technically incompetent
4. you can't keep up in a technical discussion
5. you have no interest in understanding where you are wrong

My point exactly. You can't check the time a txid was generated.

You are using unconfirmed transactions. They don't have timestamps.

blockchain.info assigns timestamps to the txids they happen to notice, but that's just in their own database.


Yes, just like that. Which part do you think you couldn't achieve reliably?


No, but so long as my points are valid why would that matter?


Again with the insults? I clearly know exactly what I'm talking about.


No, let's assume hashes are irreversible. Re-read what I wrote - I propose a 'generate and test' scheme, not a 'reverse all the hashes scheme'. I said:

  "I would generate the number between 1 and 10000 for each of the transactions I have pre-generated
   using that number of satoshis. About 10% of them would give a number in the range 1000-2000, so
   on average I would only have to test 10 transactions to find a 'winning' one"

Get it?


Again, read what I wrote. At no point did I suggest reversing any hashes.


No.


I don't know where this came from. It seems entirely unrelated to anything we're talking about.


I would have to be crazy to hire you. You have shown yourself to be incapable of understand basic arguments. You seem to have convinced this btc-raffle guy that you have provided him with a provably fair game when you have done anything but. And when people try to point of where you went wrong you do your best not to understand them. Mr btc-raffle should be asking for his money back. Whatever he paid you was too much.