Jump to content

well that deal generator


Recommended Posts

This thing that deal holdings on BBO. While it locates all the possible deals that can be made, some number with 28 zeros. Can it produce 2 identical holdings before it have wandered around and emtyed all possible deals. Curious for some reason I dont know. Guess I want to impress some on my bridge-partner sigh B-)
Link to comment
Share on other sites

I don't understand your question. If you are asking:

 

"Can the hand generator on BBO repeat a hand before having every possible hand comes up?"

 

I can say (even without knowing a thing about how it is being precisely implemented) that the answer is 100% yes. Not only *can* it happen, it *will* happen with 100% certainty (well, not 100%, but it is better odds than 1:(number of particles in the universe)^(number of particles in the universe) by a lot! It's not even close.)

 

I hope this answered your question.

  • Upvote 1
Link to comment
Share on other sites

I don't understand your question. If you are asking:

 

"Can the hand generator on BBO repeat a hand before having every possible hand comes up?"

 

I can say (even without knowing a thing about how it is being precisely implemented) that the answer is 100% yes. Not only *can* it happen, it *will* happen with 100% certainty (well, not 100%, but it is better odds than 1:(number of particles in the universe)^(number of particles in the universe) by a lot! It's not even close.)

 

I hope this answered your question.

Tx Sir I will trust you, unless ......lol

Link to comment
Share on other sites

This thing that deal holdings on BBO. While it locates all the possible deals that can be made, some number with 28 zeros. Can it produce 2 identical holdings before it have wandered around and emtyed all possible deals. Curious for some reason I dont know. Guess I want to impress some on my bridge-partner sigh B-)

 

Are you referring to hands randomly generated?

Link to comment
Share on other sites

Not sure what the algorithm is, but if it wouldn't generate at least 1 hand twice before having generated all possible deals, then it would be a poor (pseudo)randomizer. I'm confident Fred and his crew are smarter than that. ;)

 

A better question would be if it's possible to generate all hands before the world ends (and I don't mean 2012)...

Link to comment
Share on other sites

A better question would be if it's possible to generate all hands before the world ends (and I don't mean 2012)...

 

Well, here's a list of likely ways the world will end. I'll pick a "sure thing" one in that the sun's gonna give up in about 5 billion years.

 

 

We'd need to be producing 336,123,795,000 hands a second to succeed (if we have no repeats). Good luck!

Link to comment
Share on other sites

Well, here's a list of likely ways the world will end. I'll pick a "sure thing" one in that the sun's gonna give up in about 5 billion years.

 

 

We'd need to be producing 336,123,795,000 hands a second to succeed (if we have no repeats). Good luck!

It is for me totally unbelivable,..guess its true but nevertheless I dont belive it,...just 52 cards.... Think this is intresting for those who read this and not before been thinking about these shocking total sums,..when trying to promote bridge I often talk about that 28 zero number. Oh well

Link to comment
Share on other sites

I expect the answer is yes, since I expect a garden variety pseudorandom number generator is being used.

 

It's actually an interesting programming problem to achieve the opposite effect, sampling without replacement -- one way to do it is with things called quasirandom numbers rather than pseudorandom numbers -- and if the number of hands were merely billions or trillions, it would be quite useful to be able to get a 'fast' Monte Carlo estimate for how common something was, that turned into an enumeration if allowed to run for awhile.

Link to comment
Share on other sites

Free: Couldn't disagree more. Would you trust a computerized set of dice that *never* rolled double 6's 10 times in a row? Sacrificing ACTUAL randomness for APPARENT randomness is one of the worse traits a RNG could have.

This is not a matter of trust, it's just facts. You either choose a true RNG which doesn't contain any flaws if it's truely random, or a pseudo RNG which obviously has some (minor) flaws. But I'd rather use a PRNG that rolls double 6's maximum 9 times in a row but has great characteristics otherwise, than a PRNG that needs to roll every other combination first before it can roll double 6's again (whatever other characteristics it has).

Link to comment
Share on other sites

Well, here's a list of likely ways the world will end. I'll pick a "sure thing" one in that the sun's gonna give up in about 5 billion years.

 

We'd need to be producing 336,123,795,000 hands a second to succeed (if we have no repeats). Good luck!

Will be easy for quantum computers. Hopefully we'll invent one within 5 billion years. :D

Link to comment
Share on other sites

Will be easy for quantum computers. Hopefully we'll invent one within 5 billion years. :D

 

Why would it be easy for quantum computers? AFAIK they can only search through lists and factor better than normal computers...this exhaust would actually be slower on one.

Link to comment
Share on other sites

We'd need to be producing 336,123,795,000 hands a second to succeed (if we have no repeats). Good luck!

 

We are trying to get some new players tomorrow at the university. I'll be sure to bring up this point :)

 

Can't claim that the game is boring if it takes that long to go through all the deals.

Link to comment
Share on other sites

In this context, "hands" and "deals" are very different things.

 

The number of "hands" is simply 52 choose 13, which is 6.35 x1011. If we generate 1000 per second with no duplication, we can get them all in about 20 years.

 

The number of "deals" is immensely larger, and I am not quickly sure how to figure it up.

Link to comment
Share on other sites

In this context, "hands" and "deals" are very different things.

 

The number of "hands" is simply 52 choose 13, which is 6.35 x1011. If we generate 1000 per second with no duplication, we can get them all in about 20 years.

 

The number of "deals" is immensely larger, and I am not quickly sure how to figure it up.

 

A good point, I was discussing "deals".

Link to comment
Share on other sites

The number of deals is easy: permute the 52 cards, then factor out rearrangements within a hand and rotations.

Not sure I understood the discussion on PRNGs. I don't think anyone's hand-writing heuristics to generate randomness.

 

Why not just choose 13 for N, choose 13 for S, and choose 13 for E?

 

52C13 * 39C13 * 26C13

Link to comment
Share on other sites

FWIW, most of the standard random number generators in common use have dead zones.

 

There are certain outputs that do not get generated.

 

This is very, very true. I don't know anything about BBO's RNG, though.

Link to comment
Share on other sites

Why not just choose 13 for N, choose 13 for S, and choose 13 for E?

 

52C13 * 39C13 * 26C13

 

You have to then remove the rotational symmetries, but that's just a /4 at the end...Unless you count the same hand but rotated as different.

 

There are also dealer and vulnerability issues, but I didn't consider those.

Link to comment
Share on other sites

wyman, I believe that comes out to the same thing.

PRNG people, quite simply, a PRNG that has dead zones is not a PRNG by the cryptographic definition. I suspect we may be discussing different things?

 

re: comment 1. Yeah, but my way is easier to compute :)

 

re 2: I think this is semantics. You can use, say, a 32-bit PRNG to generate "pseudorandom bridge deals" but certainly you won't see all of the deals. That's all whoever meant by "dead zones" as I read it.

Link to comment
Share on other sites

How is it easier? Permute 52 cards: 52! divide by internal order of each hand: /4*13!. Rotations and dealer positions cancel each other out, so just add vulnerability and you're set.

Your way: C(13, 52) = 52!/(13! * 39!) * C(13, 39) = 39!/(13! * 26!) * C(13, 26) = 26! / (13! * 13!) and then you just notice some things cancel each other out and end with the same expression.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...