1-2-6 opportunity

#1
I have been presented with an opportunity which will allow me sustained access to a "1-2-6" CSM. Any ideas or suggestions about potential experiments to conduct. I intend to insert six decks in new deck order and see what happens, as well as try to analyze its dilution. I imagine the only limits to the potential experiments would be the machine would need to be able to be returned to working order. I would be happy to try suggestions and report findings here.;)
 

AussiePlayer

Well-Known Member
#2
Fill the machine with only black cards, every time you deal a hand replace the cards dealt with red cards and see the ratio of black/red for the next few hands.
 

HockeXpert

Well-Known Member
#4
5ofclubs said:
I have been presented with an opportunity which will allow me sustained access to a "1-2-6" CSM. Any ideas or suggestions about potential experiments to conduct. I intend to insert six decks in new deck order and see what happens, as well as try to analyze its dilution. I imagine the only limits to the potential experiments would be the machine would need to be able to be returned to working order. I would be happy to try suggestions and report findings here.;)
Load the machine with one unshuffled deck so that the machine takes them in A-K order. After one complete deck has been loaded, deal cards and note the order without refeeding the machine until it runs out. Repeat until you are satisfied you have enough data to draw conclusions about the initial dilution of the cards.

Load the machine properly with between two and six unshuffled decks with the numbers 1-6 on the all the cards to represent the insertion order of the decks. Deal a round or two, record the order of the cards that are reinserted and put them in upside down, repeat over and over and record the results noting which come out upside down (reinserted cards) and right side up (original cards).

This should yield some results that we can draw conclusions about and develop better tests.
 

Gamblor

Well-Known Member
#5
Get a deck of cards. With a marker, draw a diagonal line across the side of the deck. After the shuffle, take out the whole deck and check the pattern of the line.
 

HockeXpert

Well-Known Member
#6
Gamblor said:
Get a deck of cards. With a marker, draw a diagonal line across the side of the deck. After the shuffle, take out the whole deck and check the pattern of the line.
The brilliance of simplicity!

To the OP: send pictures.
 

The Chaperone

Well-Known Member
#8
Gamblor actually has a decent idea for once.

An even better idea is to just number the cards 1-52 or 1-312. Why bother guessing at *which* Kc you are looking at when you can know precisely which one.
 

Gamblor

Well-Known Member
#9
The Chaperone said:
Gamblor actually has a decent idea for once.

An even better idea is to just number the cards 1-52 or 1-312. Why bother guessing at *which* Kc you are looking at when you can know precisely which one.
Well, guess its better than never putting forth anything remotely interesting...
 

aslan

Well-Known Member
#10
For ST, count a slug of cards and put them in upside down, then observe how they become dispersed.

For key carding, put an ace and two key cards in and observe how they come out.

Put a slug of 10 tens in upside down and observe how they disperse.

Repeat each exercise a number of times and chart the results.
 

UK-21

Well-Known Member
#11
Please also confirm . . . .

The number of trays in the wheel,

Max number of cards each tray holds,

The number of cards the output ramp holds,

Whether when cards are inserted into the output ramp, whether they are able to go to the very back, and if so how often they do (a pet theory of mine that you could prove or disprove if you have access to one).

Thanks.
 

pit15

Well-Known Member
#12
It's too bad you need to return the machine in working order.

It'd be much more interesting to have an electrical engineer take the thing apart and give the actual algorithm used to shuffle the cards.

Presumably this machine isn't going back into use at a casino right? I'd be shocked if a casino would allow an outsider to handle a machine that would actually go onto a table. It'd be far too easy for someone to rig the machine then give it back. (Not oceans 13 style where they make it give all the players blackjacks... more like making the machine not actually shuffle anything and spit out the cards in the exact order they get put in)

A good engineer CAN actually extract the algorithm for shuffling the cards without damaging the machine (though it would be much easier of course, if you could take the circuitry apart). If you can get the algo I can take a look at it to see if it's beatable. My guess though is that the shuffle is adequate enough that it's not beatable. I doubt a company like shufflemaster would rely on noone knowing their shuffle algo to protect the games. (security through obscurity vs security through strength)
 

MangoJ

Well-Known Member
#13
pit15 said:
It's too bad you need to return the machine in working order.

It'd be much more interesting to have an electrical engineer take the thing apart and give the actual algorithm used to shuffle the cards.
Even if you could do whatever you want with the machine, the shuffle algorithm will be impossible to retract if build properly.

Anyway, even if you do *know* the shuffle algorithm, will it help ?
If I would build such a device, I would also use the entropy given by the user, i.e. the microseconds between cards are pulled from the machine as an input to the algorithm. Then even if you know the algorithm, since you cannot measure microseconds at the table, there is nothing to predict.
 

pit15

Well-Known Member
#14
MangoJ said:
Even if you could do whatever you want with the machine, the shuffle algorithm will be impossible to retract if build properly.

Anyway, even if you do *know* the shuffle algorithm, will it help ?
If I would build such a device, I would also use the entropy given by the user, i.e. the microseconds between cards are pulled from the machine as an input to the algorithm. Then even if you know the algorithm, since you cannot measure microseconds at the table, there is nothing to predict.
Like I said.. I think the shuffle is probably adequate enough to not be beatable..

Of course there's going to factors like entropy from the user, but that doesn't necessarily make it unbeatable.

There's been some pretty major goofs with major things in the past, the only way to know for sure is to have all the information on it.
 

itrack

Well-Known Member
#15
The Chaperone said:
An even better idea is to just number the cards 1-52 or 1-312. Why bother guessing at *which* Kc you are looking at when you can know precisely which one.
This is what I would do if I had a machine. Placing a diagonal line on the egdge of a deck would be kind of cool to see just how well the machine actually shuffles, but I don't think it would give you any kind of usable information. By numbering each card you could make a huge spreadsheet after dealing yourself thousands of cards and try to find some sort of pattern.
 

johndoe

Well-Known Member
#16
pit15 said:
A good engineer CAN actually extract the algorithm for shuffling the cards without damaging the machine (though it would be much easier of course, if you could take the circuitry apart). If you can get the algo I can take a look at it to see if it's beatable. My guess though is that the shuffle is adequate enough that it's not beatable. I doubt a company like shufflemaster would rely on noone knowing their shuffle algo to protect the games. (security through obscurity vs security through strength)
I agree. It wouldn't be that difficult to create a system to extract the exact algorithm, even without damaging the machine. The difficulty would be letting it run long enough to generate statistically useful results.

Of course, if they aren't entirely clueless, they designed their randomization routines intelligently, with a long enough cycle time (or with sufficient external entropy) to make prediction even with algorithm knowledge useless.

But I'd love to find out if that's the case!

At the very least, you should take some detailed video and share it discreetly.
 

HockeXpert

Well-Known Member
#17
No replies from the OP, 5ofclubs and this is his one and only post. Please don't tell us this was one of those cruel, ridiculous, time-wasting, casino insider posts.:cry::laugh:
 

Gamblor

Well-Known Member
#18
HockeXpert said:
No replies from the OP, 5ofclubs and this is his one and only post. Please don't tell us this was one of those cruel, ridiculous, time-wasting, casino insider posts.:cry::laugh:
Yeah seems like a waste of time mostly, although some of us soldier on :cry: Was wishing the shuffle wouldn't look like this random shuffle, although it probably would ;)

 
Top