Bitcoin Forum
July 31, 2024, 06:17:11 PM *
News: Help 1Dq create 15th anniversary forum artwork.
 
  Home Help Search Login Register More  
  Show Posts
Pages: « 1 ... 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 [148] 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 ... 573 »
2941  Bitcoin / Legal / Re: 4Grinz Bitcoin Casino Official Thread on: July 25, 2015, 07:04:26 AM
maybe it is time that we as a crypto gambling community start to engrave our shared visions of industry-wide minimum standards into stone to help new operators set up their business with those requirements.

at least I feel like almost everyone who has been in crpyto gambling for a while would appreciate stuff like signed cold wallets, verifiable provably fair methods etc.

A Crypto Gaming Commission Cheesy 

We could sell certificates maybe?

"Certified Provably Fair"?

Then the people who think that appeal to authority proves something could proudly display their certification for all to see.  Roll Eyes
2942  Economy / Gambling / Re: Why You Should NEVER Deposit To Primedice on: July 25, 2015, 06:56:44 AM
I think there are alot of these threads, but too many are by butthurt losers. I'm not butthurt, in total, I am 5+ BTC in profit on Primedice, but I still hate it. I think a thread must be made, because many noobies are coming in thinking that Primedice is the best dice site that exists. Here are my reasons why you should NEVER EVER play on Primedice.

1. The MAX button has been proven rigged.

Yes, yes it's true. Not just an accusation. It's been proven rigged after lots of investigation. Are normal rolls rigged? No of course not. But clicking max triggers a code that makes you lose a ridiculous more percentage than normal. If you absolutely insist on playing on Primedice, NEVER click the max button.

Maybe include your proof. The site is provably fair so it should be easy. Until then I don't think it's worth reading the rest of your post.
2943  Bitcoin / Legal / Re: 4Grinz Bitcoin Casino Official Thread on: July 25, 2015, 06:54:06 AM
though it would be nice, if they could prove the almost 900 btc max jackpot announced on their site  Wink

It's probably something like 100 jackpots of 9 BTC each, like that other old-style casino that was advertising 60k jackpots. The advertised amount is the total of all the jackpots, not an actual amount they have available to win.
2944  Economy / Gambling / Re: ♛ BitCasino.IO - 700+ games / 63,900+ BTC Jackpot / MICROGAMING, NETENT, BETSOFT on: July 25, 2015, 02:07:18 AM
They claim they are provably fair casino https://bitcasino.io/bitcoin-gambling/provably-fair-bitcoin-casino/  Huh Huh Huh

They seem pretty clueless. From that page you linked:

Quote
How do I know if I’m playing at a provably fair Bitcoin casino?

Finding out whether the casino you’re playing at is accredited is very easy. In fact, all it really takes is 2 steps. Firstly, make sure you’re playing at a Bitcoin casino, as only this type of casino is supported by provably fair gambling methods and verification software.

What nonsense! Accepting Bitcoin and being provably fair are two unrelated things. You can be provably fair and only accept dollars, DOGEs, or anything else.

Also, if you want a laugh, check out this thread in which a another casino with a similar bunch of games demonstrates their complete misunderstand of what provable fairness is...
2945  Alternate cryptocurrencies / Announcements (Altcoins) / Re: [ANN][CLAM] CLAMs, Proof-Of-Chain, Proof-Of-Working-Stake on: July 24, 2015, 11:53:29 PM
Is this talk about a new coin distributed to Clam addresses?

Yes. But there's no reason to believe it will happen.
2946  Bitcoin / Legal / Re: 4Grinz Bitcoin Casino Official Thread on: July 24, 2015, 11:12:08 PM
Now, I do not want to be lynched here as you are well respected on this forum but one could say you do the same provably fair or not: prey on kids. Maybe that's why in my case you decided to not side with the player?

I'm not siding with the casino. I think the guy who is speaking for them is either badly misinformed or deliberately deceptive. But I don't think it's fair to jump from that to "any therefore he's a scammer". For some reason he is unwilling or unable to understand how the idea of provable fairness could help his business, and that is concerning.

It's like the difference between a casino saying:

  "we have 1000 BTC put aside to pay out jackpots - look, here's the address, here's a signature proving we control that address"

and a casino saying:

  "we have 1000 BTC put aside to pay out jackpots - we don't want to show you where, but look, here's a certificate saying we have it"

Which one would you rather play at?

Does 4grinz publish their cold storage address by the way? I don't think anyone asked about that yet.
2947  Alternate cryptocurrencies / Announcements (Altcoins) / Re: [ANN][CLAM] CLAMs, Proof-Of-Chain, Proof-Of-Working-Stake on: July 24, 2015, 08:41:06 PM
Oh, is bitaddress.org yours too or did you rip their design only? Tongue I used bitaddress quite some time already.

No, bitaddress.org isn't mine. It's open-source, so I forked it and made a CLAM version. I also used a small bit from https://bitcoinpaperwallet.com/bitcoinpaperwallet/generate-wallet.html (which itself is based on bitaddress.org) to allow skipping of the entropy collection.

Could this be the first CLAM paper wallet in the world?



He used photoshop to replace the low-resolution image produced by clamaddress.org with a higher resolution version. I wonder how best to allow clamaddress.org to produce print-ready images itself, while still fitting on the screen. Any ideas?
2948  Alternate cryptocurrencies / Announcements (Altcoins) / Re: [ANN][CLAM] CLAMS, Proof-Of-Chain, Proof-Of-Pearl, Recent Mandatory Update on: July 24, 2015, 04:24:01 PM
Here is some info from my debug.log file that might help. If you want the whole thing I can get that for you too.

Code:
$ cat debug.log | grep ORPHAN > debug_orphans
$ cat debug_orphans | curl -F 'clbin=<-' https://clbin.com
https://clbin.com/FTBHj

I tried syncing from scratch. It worked flawlessly. I only saw 414 orphans in total, so never hit the 750 limit:

$ grep ORPHAN debug.log | wc -l
414

I guess it depends which peer you're downloading from.
2949  Bitcoin / Legal / Re: 4Grinz Bitcoin Casino Official Thread on: July 24, 2015, 06:26:30 AM
Provably fair systems are open source,

Most provably fair sites are closed source. Some aren't. The openness of the source isn't related to whether the events are generated in a provably fair way or not.

which puts players and their computers at great risk -

Is that why Linux is so much less secure that Windows? Oh, wait.

especially today.

What happened today?

If it's not the best implementation, or even a great implementation, then why are we arguing this?

Try to keep up. You said that it was impossible to make a provably fair slots game, so I proved you wrong by showing you a working example of one. You said it many times, even after I posted the link to that one.

The process is disturbing. It pulls data from a player's browser and computer each and every spin. In also puts the burden on the player. 

Sigh. It was showing you (and mostly everyone else) that you were wrong. Their choice of provably fair system is poor, but that isn't what we were discussing. You claimed it was impossible so I showed you that it isn't.

If you choose not to play on our site, then vote with your web wallet.

I already did.

If you wish to disparage us because we don't use a dice system to measure the fairness of slots, then expect a swift response.

"dice system"? What are you talking about. Provable fairness is unrelated to dice. It applies to any single-player game of chance, including slots, as I have already pointed out many times. You are really bad at arguing.

If you'd like to discuss scam sites, then please do so on another thread. This is our official thread meant to answer our player's questions, update them on contest results, and make promotional announcements.

We have gone out of our way to explain that Provably Fair systems were used during initial testing, and they were ineffective when applied to a slots scenario, and for several security reasons, we do not subject our software and platform to systems that put our security and that of our players at risk.

You still don't seem to understand the point of provable fairness. The point is that the casino can prove *to the player* that each and every game event was fairly determined. What you seem to have is some figure in authority which you claim has proved to themselves that your games are fair. That proves nothing to the player, since you could be lying about them having looked at your games, and they could be lying about your games being fair. You could also have changed your games since they looked at them, or change them just for your biggest bettors.

Provable fairness is something that happens while players play, to satisfy the players. It isn't a process you do in secret with some company who gives you a stamp of approval. The difference between these two is the word "proof". (Look it up).

I think the point has been hammered home enough times by now that if you don't understand then you never will, so I guess I'll leave it at that.
2950  Bitcoin / Legal / Re: 4Grinz Bitcoin Casino Official Thread on: July 24, 2015, 05:39:07 AM
Why are you just insistent on posting relentless garbage? Don't you want your site to be fair to your user base?

I think it's important to be careful with the language. There's nothing to suggest that his site isn't fair. The issue is that we can't know whether it is or not unless he proves it to us.

This is what I would like to see more of here. People stepping in and proving these non-provably fair bitcoin casinos are scammers.

Again, we haven't proved any such thing. At worst we've proved that he either doesn't understand provable fairness, or wants to pretend that he doesn't for some reason. Neither is a crime.

The thing is when these kind of casinos want you to win you will win big. When they want you to lose? It can go on for MONTHS or until you are homeless. At least with provably fair you know it's bad luck and bad luck only. The trick is to know when to hit & run these corrupt casinos. Though, many have tried and failed.

It's very hard to tell the difference between a straight house and a rigged one. I know Just-Dice isn't rigged, but we have such runs of good and bad luck that it looks as if it is at times. We had a 6 week period where most big players won their big bets. An outsider looking in could easily think we were doing it on purpose, setting them up for a fall. Then we had a week where the actual site profit was around 9 times the expected site profit for that week. "Variance is a bitch" is the common saying that expresses this phenomenon. It's quite possible that the long winning and losing streaks you see at the non-provably-fair casinos are legit. There's just no way of knowing for sure.

These non-provably fair bitcoin casinos are the most greediest and corrupt online casinos i have ever seen.

I'm uncomfortable with the conclusion that because a site doesn't offer provable fairness it means they have something to hide. It's definitely a possibility, but it seems unfair to assume it to be the case.
2951  Bitcoin / Legal / Re: 4Grinz Bitcoin Casino Official Thread on: July 24, 2015, 05:31:22 AM
Now if you would like to talk about the PF stall method I'm all ears.

I'm unfamiliar with the term. Are you referring to this in Trevor Xavier's piece on reddit?

Quote
Stalling

The easiest way to compromise a dealt hand is to redo it. After receiving the client seed, if the server does not like the outcome of a particular deal, it can stop responding to the client. Generally, this results in a hang: the roulette wheel will spin indefinitely, cards will not be dealt, and so on. When the user eventually gives up and reloads the page, it will produce a new client seed. The player sees that no bitcoins have been lost, claims no harm, and proceeds with a new deal, giving the house a new round to work with.

If so, that's one of the many problems with old-school provably fairness that is not a problem with the modern systems. If you reload at Just-Dice, PrimeDice, PRCdice, or any other reputable modern provably fair site you will have the same client and server server seeds as before you reloaded. The aim of provable fairness is to make it impossible for the casino to *undetectably* cheat the player. To that end, this "stall method" no longer 'works'.

Otherwise you and Canton can go back to making printed holiday bitcoin wallets.

I don't know who Cantor Becker is. I put clamaddress.org up this morning and immediately received a bug report that it wouldn't work on iOS. I found that a fork of bitaddress.org by Cantor Becker had already fixed the problem so I took his fix, added it to my site, and added his name to the copyright notice, since I had used some of his copyrighted open source code. I very much doubt he is in any way involved with Just-Dice or CLAMs.

Read bitZino's evaluation of Provably Fair, and you decide. https://www.reddit.com/r/Bitcoin/comments/1frm4x/provably_fair_by_bitzino_not_provable_with/

Provably Fair is better than anything out there. False.

What makes you think that evaluation is "bitzino's"? It isn't. It clearly isn't.

It's not clear to me whether you are playing dumb here or not.

You guys are going a little far with this no?

As an outsider looking in I would like to point a "probably unfair" team up of associated board members on this particular thread.
Its fascinating to me that
Dooglas
tryphe
EBK1000
webcris
TigerTatas

I found this thread when someone at Just-Dice drew my attention to it. After having made my first reply, I posted a link to it in the chat there, because I found some of what you were saying astonishing. That's presumably why so many people from the Just-Dice community posted replies here. No conspiracy, just a bunch of people who understand how provable fairness works trying to help you understand. Whether you want to or not. Wink

All coincidentally seem to also be working on the clam client together and obviously have strong dice affiliate associations if not tier 2 direct affiliates

I don't know what "tier 2 direct affiliates" means. Sounds like some kind of unpleasant marketing term to me.

tryphe and Tiger hang out on the JD chat sometimes. I think I've seen a webchris there a couple times. I don't know who EBK is.
2952  Bitcoin / Legal / Re: 4Grinz Bitcoin Casino Official Thread on: July 24, 2015, 03:28:55 AM
Fascinating and relevant interview with Bitcoin Gambling and Betting Expert, James Canning.

That sentence, no verb.

Now, now, that's not very professional. "This is wrong," is hardly proof that provably fair systems are applied to slots, and a casino that runs on providers using tested algorithms is the point. Isn't it? Seems you were unable to find the resources necessary to back up your original claim that slots are Provably Fair. Stuff happens. Have a nice day, and good luck finding that provably fair slot casino in the sky.

I just provided an existence proof.
2953  Bitcoin / Legal / Re: 4Grinz Bitcoin Casino Official Thread on: July 24, 2015, 02:52:29 AM
While provably fair systems apply to dice scenarios, as has been so painstakingly noted by dooglus, it does not apply to a slots scenarios. Provably Fair was never developed, nor intended for slots. Moreover, we don't host dice games at 4Grinz.com. But thank you for that eloquent and detailed breakdown. We hope that dice players find this useful, however, we doubt they will be searching the 4Grinz.com Official Thread for advice on games we don't actually host on our platform.  

I think you just totally missed my point, for I was saying the exact opposite.

There is no fundamental difference between 'dice' and 'slot' games. In both games you somehow pick an outcome and the player wins or loses varying amounts. The problem is how you demonstrate to the player than you picked the outcome you picked because you had to, rather than because you wanted to. THAT is what provable fairness is all about, and it applies equally to all single-player games of chance.

You may have overlooked the paragraph that clearly states that 4Grinz is Provably Fair through GL and TST testing, that its providers are tested, and that we continue to measure algorithms daily.

There is a world of difference between an appeal to authority and actual proof.

You are asking us to trust you, and to trust the people who tested your site, and to trust that the version they tested is the version you actually run.

Provably fair sites don't ask you to trust anything. They say "if you want to check whether your game was fair, follow these steps", and provide a list of steps that you can follow to be 100% certain that your game was fair.

Do you understand the difference?

Our point is that Provably Fair is not a system that can be applied in a slot scenario. Anyone that makes such a claim is not only misguided, but misinformed.

I don't think you said what you meant to say there, but I agree that you are not only misguided, but misinformed when you say that slots games cannot be provably fair.

Check https://bitcoinvideocasino.com/slots for example. It is a provably fair slot game:

Quote
PLAYING SLOTS AT BITCOIN VIDEO CASINO IS PROVABLY FAIR

Play with confidence! We guarantee that every Slots game you play at Bitcoin Video Casino is completely fair. You will be given a completely random spin (random number) every game, and we can provably demonstrate that we have in no way manipulated the outcome of the spin.

Instead of just taking our word for it, we make available all of the information you need to verify that our game is fair.

The reason that we can guarantee this is that your web browser supplies a random number that we must incorporate into the random number generator in a provably consistent way. Our server starts by choosing a private server seed for the random number generator that your computer doesn't know. Your computer then sends us a random number that we must incorporate into the seeding of that random number generator.

To manually verify a game that has been played, press the green verify button in the MY GAMES table after a game has been played.

It's not a great implementation since it uses a new pair of seeds for every spin and puts too much burden on the player if he wants to verify each roll, but it does work, and serves as a sufficient example to prove your claim wrong.
2954  Bitcoin / Legal / Re: 4Grinz Bitcoin Casino Official Thread on: July 24, 2015, 01:31:41 AM
Another link about Provably Fair with some great code examples. https://www.reddit.com/r/Bitcoin/comments/1frm4x/provably_fair_by_bitzino_not_provable_with/

Bitzino did a lot of pioneering work with provable fairness, but their system does have some weaknesses.

The technology has moved on since then, addressing the concerns raised in that reddit post.
2955  Bitcoin / Legal / Re: 4Grinz Bitcoin Casino Official Thread on: July 24, 2015, 01:29:03 AM
we’re going to accept your challenge in disproving the all-encompassing necessity of Provably Fair

I don't think anyone is claiming it is necessary to prove that your games are fair. But when it's free to do so why wouldn't you, unless you want to be able to cheat when luck goes against you?

The Provably Fair algorithm in laymen’s terms is an application great for calculating the randomness of dice rolls - those little cubes with only six sides. Its hash identifier and algorithm is easy for the player to wrap his/her head around because there are only two dice, or six-squared possibilities, equaling 36.

People use provable fairness in all kinds of games, and in all kinds of ways. To talk about "THE provably fair algorithm" makes it sound like you think there is only one of them.

If you have a slot with 5 reels, and 50 symbols of each reel, that's 50^5 = 312500000 = 2^28.2 different combinations. sha512 produces 512 unpredictable output bits. You only need 28.2 bits of entropy for the 5*50 slot.

A hash is a long stream of letters and numbers that is a mathematical problem with a solution that is otherwise very difficult to find, but once the solution is presented, it’s easier to reverse the solution to define an actual outcome. The solution to a hash is a piece of data that applies to the roll. With dice, these outcomes happen more often, giving the player more data to compare, thus proving each roll is random and not rigged. When the same system is applied to greater variables, such as card decks or slots, no provable results have been obtained.

You seem very confused. I don't know where to start with that paragraph. A hash function is a function which takes an input string of arbitrary length and returns a output number within a fixed range. There's no way of reversing the function to get from the output number to the input string uniquely, and with a good hash function it's impossible to find any suitable input that hashes to a given output. You kind of used those words, but in a way that makes it sound like you don't really understand the concept.

If you think there aren't any provably fair card or slot games out there, you haven't researched the market you're in very well.

Applying the provably fair system to a deck of 52 cards is nearly impossible, and casino card games are something we at 4Grinz know a lot about.

Not at all. You can shuffle a deck using a seeded pseudo-random number generator. There are only 2^226 ways of shuffling a deck of 52 cards. A 512 bit hash has plenty of entropy.

In fact, you’re probably not going to find a team in the Bitcoin casino industry with more knowledge of card games that recognizes the importance of random shuffles and deals. We’ve grown through the evolution of software and hardware companies claiming to guarantee the perfect random shuffle going back to Shuffle Master, Delphi 4, and STL random shuffle methods in live and online casino settings.  

"guarantee" is a funny word. It's a bit like "prove" only without the actual proof aspect. If you guarantee perfect randomness but offer no proof of it, how is anyone ever able to check the randomness to call you on your guarantee?

For Provably Fair to perform in only single-deck games, it must calculate exactly 52 x 51 x 50 x 49 x 48, and so on ……x 3 x 2 to come up with 80,658,175,170,943,878,571,660,636,856,403,766,975,289,505,440,883,277,824,000,000,000,000 possibilities and then attach a hash to outcomes that can be reversed.

Forget about reversing hashes. It's meant to be impossible. Think of a hash as a one-way function from input to output. That number you came up with is the same as the 2^226 approximation I mentioned above. Only 226 bits of entropy required.

The question becomes, “How do we or Provably Fair create a hash generator that a player or operator can utilize and apply?” We can't, and neither can Provably Fair.

What is a hash generator? Just use a standard hash function. Like sha512.

While the way in which a deck is shuffled is paramount, there are much more important considerations, like who is behind the software, who is managing the casino, whether players are able to withdraw in a timely manner, how revenue is being stored (jackpots, casino revenue, and player funds), backend security protecting everyone from hackers and potential viruses, and the use or possible abuse of players’ personal information.  

These things are all important, true. Casinos can be provably fair and still just run away with everyone's money. It has happened before. Provable fairness isn't a silver bullet. It's simply a way (the only way I know of) for a casino to prove that its games aren't rigged.

Flashback to 2008, when three large online poker brands were caught committing fraud. This happened while pro and amateur players were utilizing every possible variation of provably fair software recommended by licensers, created by developers, and sold by a dozen or more commercial providers.

None of those poker sites were in any way provably fair. In the case of AB/UB, potripper was using a backdoor that allowed him to see everyone's hole cards. Even if the game was provably fair that could still have happened. The same with Full Tilt stealing player funds. No provably fair algorithm would have prevented that (although regular proof of solvency proofs could have brought it to light sooner, but that's an argument for a different post).
 
In the end, it did nothing to protect players and served as a false flag. While so much focus was being placed on the integrity of a deck, other, more sinister, offenses like viewing hole and river cards were being committed. With today’s technology, such behavior is now easy to identify and there are several companies that offer related software.

There was no provable fairness in those cases, and so it was no false flag. Provable fairness is a simple way of letting the player know that you aren't cheating him games of chance. It doesn't solve all problems, but it solves some that need solving, and which no other technique is known to solve.

A card counter makes calculations based on outcomes. Outcomes are determined by the cards you can see with the deduction of those you can’t see. The randomness of the shuffle is only monitored by Provably Fair, and doesn’t actually ensure the integrity of the deck or decks, but only that the order of the cards differ substantially for each hand.

I think you really need to try to understand what provable fairness is, and how it works before you comment further on it. You are way off base here.

This is great, but how do you attach a hash to so many possible outcomes to provide comparisons establishing fairness? We invite anyone to square and cube the above number to determine randomness in dealing blackjack and poker, and challenge them to provide tangible results.

When you play with 6 decks, you don't use all 6 decks. You shuffle after every hand, and so never deal more than the top say 52 cards from the 6 deck shoe. So long as the top 52 cards are randomly selected from the full 6 decks, the game plays exactly the same as if you had shuffled the whole shoe. If you need more than 512 bits of entropy because for some reason you need to shuffle the whole 6 decks, you can simply use multiple hashes. Use sha512(serverseed:clientseed:0) and sha512(serverseed:clientseed:1), etc. Re. tangible results, be careful what you ask for. It sounds like you're asking for someone to implement the games you offer in a provably fair manner, while refusing to do so yourself.

Provably Fair doesn’t claim a fair outcome when applied to one or more decks. Rather, it proves the randomness. While random shuffles are important to online and brick-and-mortar card rooms, what’s happening on the backend is what should concern players. Security has less to do with randomness and more with outcome.

That's exactly what provable fairness promises. And the randomness determines the outcome. The backend is a black box. Provable fairness publishes the inputs and outputs of that black box as well as the algorithm it runs, such that the player can reproduce the steps it takes and verify that it didn't cheat him. Without that, the player has to simply trust that your dealer hit his 21 against the player's 20 by luck and not due to foul play.

The benefit of a provably fair system is its third-party verification and auditing for cards and dice. Again, the shuffle and roll are important and shouldn’t be discounted, but recently, Provably Fair’s major flaw has been it’s susceptibility to unscrupulous players or competitors who make baseless claims against service operators by implying they cheat.

Previous you said it didn't work for cards. Now you say it does?

I've not seen anyone making these baseless claims. Do you have a reference? I see people making baseless claims that "primedice cheat because I lost 10 50% bets in a row". In response the site can tell them to verify their rolls. If they weren't provably fair the player would have no way to distinguish between a run of bad luck and rigged casino.

Most players only familiar with the phrase “provably fair” will take such a claim for its word, believing a casino is creating a larger risk, when Provably Fair was never designed for slots or that purpose.

Most players don't really care about fairness I think. They think of gambling as a way of passing the time and expect to lose. I suspect that the percentage of rolls that are ever actually checked is very low. But the fact that it is possible to check any roll keeps a provably fair casino honest, because they know there's a chance that any cheating will be detected.

While Provably Fair is effective in dice and single-deck blackjack, it’s not an effective tool when applied to multiple decks, shoes, and certainly not slots.

I disagree.

Worse yet, the Provably Fair system and its open-source software is responsible for exposing several operators to hacking attempts. This is why Provably Fair is applied separate from our platform and in conjunction with other testing.

Crappy coders write crappy code. I recommend testing before deployment. Maybe ask someone who knows what they're doing to help. What does "applied separate from our platform" mean? Are your games provably fair or not? Nothing needs to be open source. You don't need to publish any of the code that you run on your server. Simply describe the algorithm.

If your head hasn’t already exploded from all of this information, we challenge you to take the Provably Fair system a step further by applying the hash algorithm to slots. It’s being attempted, but the outcomes mean nothing. Because the hash will most likely never attach itself to the same outcome produced by (oh, let’s say) 36 lines determined by a wheel with (oh, let’s say) 50 identifiers, times (oh, let’s say) five or more wheels on any given spin, giving you a 0% Provably Fair rate, and merely the appearance of fairness. Finally, we challenge anyone to find a central source behind Provably Fair. Go ahead, Google it.

36 lines? You don't spin the lines, you spin the reels. You have 5 reels which each have 50 different positions. That's just 5^50 end positions. Then you check the 36 lines from there. What does it mean for a hash to "attach itself"? What is a "provably fair rate"? What do you mean by "central source"? Does the source of an algorithm matter? I get a strange feeling reading your words. It's like you're using all the right words but in the wrong order. Like you've read a buzzword dictionary but don't understand any of them.

So what is the real solution for players and casino managers hosting more than just dice and blackjack games? Real science, inspections, visual tools, and supervision.

You seem to be trying hard to justify not proving the fairness of your games to your customers. Why is that?

To put all concerns to rest, we’d like to state for the record, 4Grinz does employ Provably Fair, just not directly (for the above security reasons). 4Grinz is powered by CoinGaming.io and scrutinized by Gaming Labs International (GL), and Provably Fair has never been our end-all for establishing fairness - it’s merely an element. We layer our testing to ensure safety, reasonable outcomes, and everyone’s protection from malware and hackers.

What does "not directly" mean? Either you prove the fairness to your players, or you don't.

Fairness in iGaming and slots has also been established by the big-brand software and product providers like Betsoft, Ezugi, Gamearts, Endorphina, Takisto, and Play-N-Go. Just like IGT solicits trusted developers to provide similar software to brick-and-mortar casinos, our online developers are scrutinized by hundreds of casinos and security experts and pass rigorous testing before each of their games can be licensed.

There are many developers out there, but none as closely monitored as the brands and their products that we offer our players. Coingaming.io provides 4Grinz with additional layers of security, and then 4Grinz security experts ensure all standards are being met. More importantly, our outcomes are compared with all of Coingaming.io’s partner sites and their outcomes daily, and visa versa.

In addition, 4Grinz does the following:
1. Research each and every provider on our site and establish clear lines of communication with immediate access to security reports that flag uncommon outcomes.  
2. Powered by Coingaming, 4Grinz enjoys another layer of management, scrutiny, and access to algorithms that display real-time statistical data 24 hours a day, seven days a week, on demand and at a glance. At anytime, this data can be shared and compared with the software provider with whom we also have direct contact. 4Grinz, Coingaming.io, and the casino game providers are able to review and collect data from each and every play, a series of plays, while comparing several games over short or long periods of time, and then analyze that data to make a determination. Coingaming.io is one of the largest bitcoin gaming platforms having processed a huge sample size of more than 100 million bets (transactions) on its platform. This affords its team the ability to consistently and accurately analyze the expected house edge from the gaming software providers with actual results to ensure the validity of house percentages.
3. Gaming Labs International (GLI) is our preferred testing lab. Why? Because GLI acquired Technical Systems Testing (TST), established in 1993, and its technology - including Provably Fair. We can assure you, GLI, which has owned TST since the online poker meltdown of 2010, is the world's most experienced gaming test labs and internationally recognized testing facilities offering a full range of testing and consulting services to both the iGaming and land-based markets. It doesn’t get any better than that. Finally, when assessing the total value of a service provider, we prefer to talk to a real person.

That's all very nice, but isn't relevant to the question of why you don't prove the fairness of your games to your customers.

What does that mean for the player? It means that we know and trust the providers, that the providers are tested, that our standards are tested, and that all of this is monitored in ways like never before. We also communicate with other casinos and providers using the same GL Standard games. This allows us to compare outcomes, address possible concerns, and move forward as a gaming community to guarantee GLI performs at the highest standards.

We invite one to visit an actual GLI facility before coming forth as an expert on iGaming fairness and security. Their labs are strategically positioned around the world, with bases in London, Macau, Manila, Netherlands, Vancouver, and Italy. Each lab works closely with industry operators, software suppliers, and manufacturers. Its jurisdictional regulators verify compliance, ensuring GLI upholds superior technical and new world industry standards. Their website is comprehensive and very educational for one wishing to stay up with iGaming security, safety, and fairness issues.

Best part, they are real people.

This humble author and gaming expert has witnessed several developers and testers challenge each other mentally and physically over the years, with a few of those challenges resulting in more than one concussion. Do we discourage such behavior? No. We just get better insurance, because developers and testers challenging each other at escalated levels is exactly what we want for the protection of our company and our players.

So you trust the game providers, and we have to trust you.

That seems like at least two points of failure.

With provable fairness no trust is required. That's why we call it "provable" not "trusted".
2956  Alternate cryptocurrencies / Announcements (Altcoins) / Re: [ANN][CLAM] CLAMs, Proof-Of-Chain, Proof-Of-Working-Stake on: July 23, 2015, 07:00:47 PM
I forked bitaddress.org and changed it to work with CLAM addresses:
  https://github.com/dooglus/bitaddress.org

I need the graphics updated if anyone has the time and skills required.

It's up at:

  https://clamaddress.org

I made some crappy 'art' for the logo and wallets. I'm happy to accept better versions.

2957  Alternate cryptocurrencies / Announcements (Altcoins) / Re: [ANN][CLAM] CLAMS, Proof-Of-Chain, Proof-Of-Pearl, Recent Mandatory Update on: July 23, 2015, 04:17:46 PM
Syncing a virgin db, and clamd seems to be getting "stuck" on orphan blocks...

ERROR: ProcessBlock() : duplicate proof-of-stake (COutPoint(31c8d98ed4, 1), 1409328810) for orphan block ab81478d442bbc067c091b86ddd81d8c415b7fba4cea652a78f7549a515b70a8
ProcessBlock: ORPHAN BLOCK 751, prev=ab81478d442bbc067c091b86ddd81d8c415b7fba4cea652a78f7549a515b70a8
ERROR: ProcessBlock() : duplicate proof-of-stake (COutPoint(4b4c0aacd4, 2), 1409328811) for orphan block e9f85e1f3d10894e42aa0982b22e1a1a74d3c22e1847adc2b801fdbd70fea9f6
ProcessBlock: ORPHAN BLOCK 751, prev=e9f85e1f3d10894e42aa0982b22e1a1a74d3c22e1847adc2b801fdbd70fea9f6
ERROR: ProcessBlock() : duplicate proof-of-stake (COutPoint(df96fd5846, 1), 1409328935) for orphan block 331a1a344def2ff02b76726b37a5ad1cc03af2f9c85b084ba31de073a9cf09f0
ProcessBlock: ORPHAN BLOCK 751, prev=331a1a344def2ff02b76726b37a5ad1cc03af2f9c85b084ba31de073a9cf09f0
ERROR: ProcessBlock() : duplicate proof-of-stake (COutPoint(d93414f7d2, 1), 1409328928) for orphan block f09b4a9d0f0d1d060b256805248696fc38ecf9ff26c06119d653af9b3bb7a656
ProcessBlock: ORPHAN BLOCK 751, prev=f09b4a9d0f0d1d060b256805248696fc38ecf9ff26c06119d653af9b3bb7a656
ERROR: ProcessBlock() : duplicate proof-of-stake (COutPoint(3f2c6d206d, 1), 1409329004) for orphan block 6795a64418eb9b18880e4d7ec125552be3ff73d8f9de7f825ebc311d276d02ec
ProcessBlock: ORPHAN BLOCK 751, prev=6795a64418eb9b18880e4d7ec125552be3ff73d8f9de7f825ebc311d276d02ec
ERROR: ProcessBlock() : duplicate proof-of-stake (COutPoint(7b54b4e3b6, 1), 1409329082) for orphan block a9ad274839def4b39c0d2e8a44fa14f9d15a7665f37f202d1f9ab80edf56fc85
ProcessBlock: ORPHAN BLOCK 751, prev=e8be94c3aa50abd60dbe64dde7b76672b5a97c8d4ec50da9afa9b70dc10f3550
ERROR: ProcessBlock() : duplicate proof-of-stake (COutPoint(770d08c5c8, 1), 1409335177) for orphan block ebb678b09629ae0657500521b73aafd4fa66f441458a14c70380c0ca05eccfbb


At this point it's just a bunch of "ORPHAN BLOCK 751" messages, although the block hashes for each seem to be different. It doesn't seem to be accepting new blocks. Is the client hitting some sort of limit? I have to stop and restart it in order for it to accept more network blocks. Once it hits 751 orphan blocks it gets stuck again. This has happened about 5 times now.

The phrase "ORPHAN BLOCK 751" currently appears 51327 times in my debug log. Now on 6th restart.

Guys I'm getting this same problem. ORPHAN BLOCK 751 and gets stuck. I am running the latest clamd built from the source code on an AWS micro instance (ubuntu).

What version of the source code did you build from?
What block number did it get stuck on?

As a workaround, see https://bitcointalk.org/index.php?topic=623147.msg9772191#msg9772191 for a bootstrap.dat file that will get you all the blocks. I'll see if I can reproduce and hopefully fix this orphan problem.
2958  Economy / Scam Accusations / Re: FastMiner has trojen do not try it on: July 23, 2015, 07:22:49 AM
hellow all there mates ...I am here to warn you all that fast miner software has trojen malware in it..so please do'not try it at any cost..I suggested my friends and they got affected luckyly I am safe.. https://i.imgur.com/pu89mW4.png here is the proof..and if you don't belive me review their codes yoy will find yourself..please I request modrators to take strick actions on these because there are some open threads till now about the fastminer and peoples are downloading it till now and even paying 0.75btc - 2.5btc for this ..

That screenshot looks familiar. I made it, and posted it about a week ago to warn people about this scam site. But a forum moderator deleted my warning at the same time as he deleted the post I was warning about.

I'm glad you got to see it and spread the word before my warning was censored.

2959  Economy / Scam Accusations / Re: SatoshiDice - Need investigation into these bugs. [SEVERE] on: July 23, 2015, 07:14:29 AM
Okay, but why is the gap between round initiated and completed around 50 minutes?

I think that's the start and end time of the "streak". SatoshiDice lists bets as streaks, not individual bets.
2960  Alternate cryptocurrencies / Service Announcements (Altcoins) / Re: Scrypt.CC | Scrypt Cloud Mining on: July 23, 2015, 06:16:12 AM
The most stupid thing i ever read. You make complaint to any other police or some like that they just laugh on you because what? because you become most stupid people they ever met.

They never ask you or force you to invest into their website or wallet, every people decide they own path.

If I offer to sell you something, we agree a price, you send the money but I never send the item, I didn't force you to send the money, but you have still committed fraud. You "decided your own path" in trusting that I would send you the thing you paid for. Does that mean I'm not guilty of a crime when I decided to rip you off?

This is the same. S.CC customers have agreed that they are paying for the company to mine on their behalf, but the company aren't doing what they agreed to. Instead they're deleting customer support tickets when people find their funds are unavailable.

So no, they didn't force anyone to do anything but that doesn't mean they are innocent.
Pages: « 1 ... 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 [148] 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 ... 573 »
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!