I actually think a site with real cards has marketability and could attract customers if done right. I know too many people who tried online poker and then gave up after losing because "they could just tell" it wasn't realistic or nonsense like that. But for such a site to succeed they should back off the accusations and implications against the other sites. Just deal real cards and shut up about it. The more you explain why and how you're legit the more you confuse and the more some people will think you aren't legit because all that 8x2000 stuff sounds weird.

Real cards. Shut up. Deal. Easy game.

Such a site that was run halfway decently and not supervised and marketed by ******s would actually have a chance...although not a huge one simply because its a tough market to crack.

Making an hour video to babble about your super amazing formula does not qualify as non-******ed marketing.

There is absolutely a niche market for it. But execution is key.

this site is a huuuge fail. so much publicity and only 50 people online at traffic jam ;p

maybe ppl dont like to play at rigged sites. no competition and stuff ;<

Hey you!!!! It's a 6 month "soft launch". OK?!

Are you the deckmaster? I am the matrixkeeper.

Any rakeback? Any games above micros? I'd consider having a session vs the riggies.

"Why we created Real Deal Poker" - So your quad Kings don't get Burned on the river. *




Feel that Burn! You silly RNG players.





* Joke - he was pointing out that this situation would be auditable, of course.

oh my god the stupidity is just too much to comprehend

You don't need to know the starting deck. The system takes something that is random and then using a finite number of secret but static algorithms transforms the original random ordering into 2,000 new orderings that are NOT random. Since these orderings are related to each other with enough data even if you don't know the master from which they are transformed you can still identify which transformation you are being dealt using limited info (hole cards and flop) and then from that and data from other tables predict your turn and river. Having multiple master decks just increases the amount of processing capability you need and lowers the certainty in your predications but not enough to matter since even the limited information is enough for someone to exploit.

It was this.

59 minutes and 21 seconds of it.

This boggled my mind as well, I tried to process it a few times couldn't make sense of it either.

http://www.cigital.com/papers/downlo...r_gambling.php

That was the first thing that came to mind when he mentioned using time.


Then he mentioned that the system was specifically designed to prevent a permutation of a master deck to be dealt to a table a second time. It is clear that this system introduces a bullseye at where and how to attack it.
I'd like to see the group that did the article I quoted take this on, of course they would need to have enough traffic to get on enough tables which isn't likely going to happen, not with that slobbering drunk out making promotional videos anyway.

I agree. Many players will never be winners but rather than think they need to improve their game they blame it on stars being rigged. That never happens on Tuesday nights at my buddies house is a common topic I hear.

When I say Im a winner they say because stars likes me. LOL

To their credit, they are getting a ton of attention (even here) while being nothing more than one of hundreds of skins in a network that has a couple dozen players.

Think Red Fox poker gets as much attention for instance?

LOL.... so the future of internet poker will be everyone must be on a webcam, you will have a live feed screen where u see a real person shuffle and deal, someone else looks at your cards and selects them and they pop up on your screen. You then see the action... you can look at the person to acts face and see your opponets real time reaction to the flop, turn, river, your bet, your sneeze, you sticking them the finger, your girl walking by naked after getting out of the shower... etc..etc

meanwhile, in casinos they will have tables with electronic screens and use and RNG...

interesting...

I don't even think that would be very successful because of what spade mentions here.

You'd have to gather a lot of information from other tables to even possibly figure out all the cards and their order. Then you'd need to somehow know which decks being dealt to the different tables are related. I don't know how you'd be able to figure that out.

So, basically you'd need to data mine thousands of tables (lol that they'll ever get the traffic to run that many) simultaneously and somehow know when they're using decks from the same matrix. Then use time to determine which permutation they're using at which table and you'd get a very small subset of known cards on each table. Then maybe with all that you'd know the cards being dealt to a couple tables.

You'd have to have one heck of an operation going on to take advantage of that. Plus you might know all the cards and end up in a situation where you're in the seat that gets 72o and you know another player has AA and is going to hit a set on the flop. Not much good it does you to know you have no chance of winning the pot.

I think it would be smarter just to focus on hacking them and knowing the cards that way

DrMichHead in no way condones hacking.

Ok, after reviewing the rambling video a couple times here's my understanding of how it works.

1. The mechanical shufflers are providing a new scanned deck to the system about every 16 seconds. The matrix system stores the most recent 8 decks only. I assume this is first-in first-out but it wasn't specified. These are called the Master Deck or MD.

2. Only the most recently created MD can be used when a deal is called for by a table. The others are just processing deals already in progress until they expire.

3. The Matrix contains 2000 preset algorithms for reshuffling the deck to give up to 2000 permutations of the physical deck. From everything said in the video (and in the patent application) it appears these do not change. These are called S1 to S2000. They could be anything, from swap all even and odd cards, to reverse the deck, to random mappings created when they designed it, but whatever they are, they are preset and fixed.

4. The first Sx algorithm that gets used by any new MD is determined by the timestamp on the MD. Since we are creating about 8 MDs every 2 minutes, apparently that interval is broken down into 2000 parts for choosing the first Sx algorithm to be used for that deck, so about every 60-millisecond interval would have to increment to a new S value. The exact way timestamps are mapped to S values wasn't specified. But there is a mapping and I can only assume it will evenly distribute particular Sx algorithms over time, but that was certainly not apparent from the explanation. And since the physical shufflers are on periodic intervals and not random, I question how they ensure an even distribution across the Sx values, and not clumping according to the physical shuffle intervals.

5. So we have 8 decks stored in the Matrix continuously, and each one has a particular Sx algorithm queued up to be used when a deal is called for by a table. Once that Sx actually gets used, that MD proceeds sequentially to use Sx+1, Sx+2, etc. until that MD is pushed out of the system by 8 new ones coming in.

6. When a table calls for a deal, the most recent MD will run the Sx algorithm that is queued up for it and reshuffle the deck and send it to the table to be cut and played. The next table that calls for a deal with get a deck from the same MD if it is still the newest one, but will use the Sx+1 algorithm to reshuffle the deck this time. The next request would use Sx+2, etc. If a newer MD has come in since the last deal, then that newer one is used instead, using whatever Sx is queued up for that new MD according to its timestamp as explained in #4. Only the newest MD can be used to fill requests for deals from a table.

7. One exception to #6 is that if a table calls for a deal while the same MD used for that table last is still the active one, that same MD can't be used again on that table. How this gets resolved wasn't specified, but I assume that deal request must wait for the next new MD to get created. Since these happen about every 16 seconds, there might be a delay waiting for a deck if a hand completed in less than 16 seconds. The video said the newest MD is always the active one, so I don't see any alternative here other than the table must wait for a new MD. The video was clear that they only store 8 and only the newest one can fill table requests for a deck.

--------------
So if 8 tables are active, we can provide a deal to all of them every 16 seconds. If 800 tables are active we can still provide a new deal to all of them every 16 seconds, and we would use 800 of the Sx algorithms for the current MD. If over 2000 tables are active, then the Sx algorithms are used more than once, up to 8 times each for 16000 tables. But the 8 would all come from different MDs.

This sound right?

I'm not sure if this is exploitable or predictable, but it does raise some security questions since the order of the next 2 minutes of deals is always known in advance (we're storing 8 decks, and the S transformations are preset). That's plenty of time for a lot of stuff to happen, particularly with low traffic. One of their claims is that the system prevents the possibility of insider cheating. I don't see how that's true, but I'm also not suggesting it is likely either. I just don't see it being ruled out by this.

Oh yeah I forgot to LOL about this section of the presentation. It was rather hilarious that he was making up everyone else's hand and talking about how great quads are and how he'd "expect" to win with them.

I also liked his explanation of how "every" RNG site works. With the cards getting chosen multiple times and having to get a new RNG to until it finds an unused card. How the hell can he know that's how RNG sites work? First off I thought PS just creates a randomly ordered deck and deals it out. So, there's no looking for an unused card every time one is needed to be dealt. They could also just put all the cards in a list and for the first card generate a random number between 0 and 51 and choose a card. Then just remove that card from the list and generate a random number from 0 to 50 for the next card chosen. If it hits the exact same number you'd still get a new card because the old card is no longer in the list. *&^@#$

It's like this guy has no ability to think of other possible alternatives. Once he's come up with an idea in his head about how something works that's all there is. No other possibility could exist. I'm going to go out on a limb and guess that he's a losing player.

Wow, that complicated algorithm seems much simpler than just using a real RNG. And much more effective. And the "Rigg"Tards to whom this dumpster fire of a site is marketed are sure to love and trust this.

If you've ever played in a casino you'll know that bad players blame the dealer all the time. It's not a situation that's unique to online. The only difference is nobody can throw their cards at the RNG...

So, what do you guys think the audit process is like? Let's say the quads over quads hand happened on RDP. KK boy requests an audit.

Do they just have the "third party" that they chose send you an email that says, "All is well we confirmed that you lost fairly."? That seems less than reassuring. The dude said he will not reveal the algorithms that spit out the different permutations. So, what could they possibly give the user to reassure him?

1. video of shuffled cards
2. time stamp
3. ???
4. cards that were dealt to you
5. busto

Would that make anyone feel better? Much less the riggies.

This doesn't seem very well designed to me.

Why not use a FIFO list for the 8 decks and select one on the basis of time at which it is selected (3 least significant bits) and the Sx index using another 11 bits? Obviously they would need to use the precision timer rather than the main system clock as only that would have adequate resolution to provide the number of bits required but using that isn't a problem. I would actually use more decks and a smaller number of Sx algorithms.

In reality, of course, a properly designed software RNG + event time based re-seeding is perfectly adequate for poker. Even a hardware RNG is overkill and only used to satisfy those who have a poor grasp of RNG's.

If I understand correctly you have to PAY RDP $50 to audit the hand.
I'm pretty damn sure that after losing with quads having to pay $50 (or any amount) isn't going to make you feel any better at all.

A hardware RNG is going to provide an extra layer of security to prevent 'tampering'. If there is no capability to 'flash' the program once it is built, tested and verified it would be far more difficult to modify it than it would a software based system.

Using an event time based re-seeding with a software based RNG has already proven to have the potential for disaster.

Gene was pretty misleading about RNG's and tried to lead the people who do not know better that there would be an audit trail available for any (reputable) site.
No matter what data the site uses to 'seed' the RNG the seed values are recorded with the hand number along with all other data, the resulting deck, player actions etc. The deck actually would not even need to be recorded in the DB as by supplying the seed values that were recorded the exact same deck would always be generated, of course storing the resulting deck along with the seed values adds another layer of confidence.

That's a spurious consideration. If you are going to tamper you're going to tamper after the RNG, whatever flavour, hardware or software.

Again, spurious. Any method of generating random numbers has the potential for disaster if it is not done competently. In fact, hardware RNG's actually need software to overcome certain inherent limitations.