Cristian

Deck Shuffling Clarification - It doesn't seem random?

568 posts in this topic

On 4/14/2017 at 11:43 AM, Warp said:

I haven't noticed anything even resembling what you describe. There is no unusual amount of flushes, or straights, or even river bad beats.

 
 

Since it seems like you are unable to have a logical debate, using your own word and deflated argument, how do you measure good quality?...from you "haven't noticed anything"? Prove that there is no issue before arguing there is. 

Those who have highlighted issues on one very good post from HYCICANDR, which you have completely ignored, state the very slim chances of occurrence being spotted and repeated in succession. This in itself is the purpose of PRG. I feel from your responses that you probably have no idea how a pseudo random number generator works nor do you actually know its purpose. 

Let me tell you.

1st. it is not random (purely)

2nd. it is a massive single loop of numbers generated that will also go back to first selected. (you cannot generate the first no.) [In case, each card dealt is a loop since they do not leave the card deck static as Texas should be or any poker version]

3rd. PRG quality is based on if repetitive patterns have spotted

so...in you review to disprove, prove that it is a high-quality PRG.

I don't think you even know where to start!

Lastly, your previous post on asking how you prove it is poor quality means you have no idea when it is already been explained with valid points. If you don't get it, drop the debate. We are all here to learn from everyone's experience and I have not seen you adding to that. 

Edited by Prodigal Gil

Share this post


Link to post
Share on other sites
4 hours ago, Prodigal Gil said:

I feel from your responses that you probably have no idea how a pseudo random number generator works nor do you actually know its purpose.

In fact, I have an MSc degree in computing science, and I do programming as my payjob (and have been doing so for about 15 years, and as a hobby for longer than that). I know quite well what PRNGs are.

Cryptographically strong PRNGs are indistinguishable from true randomness (especially if the PRNG is re-seeded from time to time from a true source of randomness, but even that's not necessary for the purposes of a game like this. The main reason to re-seed in a game like this is not to increase the quality of the randomness, but to thwart people from detecting the position in the PRNG stream and thus being able to predict future values.) If you were to compare eg. card distributions produced using a cryptographically strong PRNG and distributions produced by a source of pure randomness, there would be virtually no difference. The odds and frequencies would be pretty much the same, no matter how you want to measure them. In other words, from two samples of rounds (no matter how large samples, even billions of them), one made using a PRNG and another using a true source of randomness, you wouldn't be able to tell which one used which method.

(The basic practical difference between a cryptographically strong PRNG and a true source of randomness is what I already touched upon: If you know the position of the PRNG stream, you can start predicting future values, because they are generated deterministically. However, this has no effect on the actual quality of the randomness eg. in terms of random card distributions and odds.)

Now, you could argue that PP is not using a cryptographically strong PRNG (such as the Mersenne Twister), even though it could. You would have to present an argument of why the PRNG it is using (the standard .NET random function) is not cryptographically strong, and how much poorer its quality is. Then you would need to show how the way that PP uses this PRNG leads to poor quality randomness.

Sure, I'm not happy with the method by which PP deals cards. As a programmer with quite some knowledge about these things, I would much more prefer if they used eg. the Mersenne Twister (or other cryptograhically strong PRNG), and used it to perform a simple Fisher-Yates shuffle on a simulated deck of cards, and then deal from the top. That would be much simpler programmatically (reducing the chance of bugs) than the overly and needlessly complex way they are using, and would give confidence that they are doing it right.

However, just because I don't like the method they are using doesn't somehow automatically mean that its quality is poor and not indistinguishable from true randomness. I would need to prove it before I make that accusation.

You are making the accusation based purely on conjecture and circumstantial eyewitness testimony and personal experience, not on actual measurements. I think what's happening is that you have experienced some bad beats (which always happen in poker), and you get so angry at them that you blame the program, based on nothing else than your angry feelings, rather then hard mathematical facts and measurements.

Edited by Warp
  • Like 2

Share this post


Link to post
Share on other sites
1 hour ago, Warp said:

In fact, I have an MSc degree in computing science, and I do programming as my payjob (and have been doing so for about 15 years, and as a hobby for longer than that). I know quite well what PRNGs are.

Cryptographically strong PRNGs are indistinguishable from true randomness (especially if the PRNG is re-seeded from time to time from a true source of randomness, but even that's not necessary for the purposes of a game like this. The main reason to re-seed in a game like this is not to increase the quality of the randomness, but to thwart people from detecting the position in the PRNG stream and thus being able to predict future values.) If you were to compare eg. card distributions produced using a cryptographically strong PRNG and distributions produced by a source of pure randomness, there would be virtually no difference. The odds and frequencies would be pretty much the same, no matter how you want to measure them. In other words, from two samples of rounds (no matter how large samples, even billions of them), one made using a PRNG and another using a true source of randomness, you wouldn't be able to tell which one used which method.

(The basic practical difference between a cryptographically strong PRNG and a true source of randomness is what I already touched upon: If you know the position of the PRNG stream, you can start predicting future values, because they are generated deterministically. However, this has no effect on the actual quality of the randomness eg. in terms of random card distributions and odds.)

Now, you could argue that PP is not using a cryptographically strong PRNG (such as the Mersenne Twister), even though it could. You would have to present an argument of why the PRNG it is using (the standard .NET random function) is not cryptographically strong, and how much poorer its quality is. Then you would need to show how the way that PP uses this PRNG leads to poor quality randomness.

Sure, I'm not happy with the method by which PP deals cards. As a programmer with quite some knowledge about these things, I would much more prefer if they used eg. the Mersenne Twister (or other cryptograhically strong PRNG), and used it to perform a simple Fisher-Yates shuffle on a simulated deck of cards, and then deal from the top. That would be much simpler programmatically (reducing the chance of bugs) than the overly and needlessly complex way they are using, and would give confidence that they are doing it right.

However, just because I don't like the method they are using doesn't somehow automatically mean that its quality is poor and not indistinguishable from true randomness. I would need to prove it before I make that accusation.

You are making the accusation based purely on conjecture and circumstantial eyewitness testimony and personal experience, not on actual measurements. I think what's happening is that you have experienced some bad beats (which always happen in poker), and you get so angry at them that you blame the program, based on nothing else than your angry feelings, rather then hard mathematical facts and measurements.

 
 
 
 
 
 
 

Thanks for sharing. A developer does not necessary gives you automatic credit for knowing what PRNG is as this is heavily a Mathematics specialist.  

Your responses in the previous post indicate otherwise from your list of qualification. From a Poker perspective, rather than a developer's perspective and I manage a team of developers and QA engineers in my daily job, a much simpler way to look at this rather than going around proving codes and mathematical models has, a poor PRNG will show signs of easily spotted patterns outside of the norm in its perceive randomness quality base on anecdotal evidence. I will go a far as decent poker players, semi-pro, all those who take a strong interest in poker will pick that up rather quickly. Especially with PP, it is staring you in the face. 

I would say several experienced poker players on this game has presented sufficient amount of anecdotal evidence to make such determination.

Your argument to prove it is of high quality or even so claiming acceptable, no sufficient evidence was presented. The programme got caught out and the developers claimed there were no issues with the PRNG with its quality, after finally sharing it has tweaked the entire shuffling system, which someone with your expertise seemed to have completely ignored. 

I still yet to see an argument point on the poker perspective from you, rather than sitting on your anti-debate bench. 

--------------------------

Let me ask you this since you are the 16 years expert. 

Flip a coin x amount of time in 2 sequences:

Sequence 1: HTHTHTHTHTHTHTHTHTHT

Sequence 2: HTHTTHHHTHTHHTTTTHTH

Which sequence is random? When you are able to answer the question correctly, then we can have a proper debate instead of chucking out garbage debate points because in any debate, one must share the same understanding or at least close, otherwise, there is no point in bumping head against a wall!

Edited by Prodigal Gil

Share this post


Link to post
Share on other sites

I played a 500k ring game recently.  I had one maniac player on the table 3 bet shoving over my 3x raises a few times.  So I had pocket 9's and called his KQ preflop (coin flip) and he got his K on turn.  Then two hands later he open shoves QQ and I call AA and a Q comes on the flop.

So yeah being down 1 mil sucked.  But I outplayed some other players postflop and got my stack to 1.3 mil when I left...so I only lost 300k for the game, compared to 1 mil.  My point is I didn't blame the game or say it was rigged.  Its part of poker, sometimes the better hand loses.

I hear the complaints on this thread and wonder why people still play the game if they don't like the card shuffling system.  Any form of poker is a mean game, with lots of variance.  The main fundamental is just have a big enough bankroll to take the swings and take the lumps as they go.  My bankroll is at 26.3 mil, so even a small 300k loss doesn't hurt.

Just manage your bankroll, play your best and the long term will benefit those who play well.  If this game brings one more frustration than joy then why play it? 

  • Like 1

Share this post


Link to post
Share on other sites
7 hours ago, Prodigal Gil said:

a poor PRNG will show signs of easily spotted patterns outside of the norm in its perceive randomness quality base on anecdotal evidence.

No, it won't. Your sample size is ridiculously small.

How many examples do you have? A dozen or so? That's just a laughably small sample size. It says absolutely nothing.

You are also completely ignoring the fact that these supposed "non-random" things seem to be happening to only a few people in particular, while the rest (including me) don't experience it. How exactly do you explain that? Is PP targeting those users in particular and deliberately giving them bad beats?

I have a much simpler explanation: Those particular people have got some bad beats, got angry, and are blaming the program for it.

In fact, I doubt you are experiencing any more bad beats than anybody else. You simply experienced a string of them at one point, started complaining, and are now too stubborn to back down. And, of course, now every bad beat you experience, even if it happens once for every hundred rounds you play, you will take as "evidence" of the "poor quality rng". Your stubbornness makes you biased.

Edited by Warp
  • Like 1

Share this post


Link to post
Share on other sites
22 hours ago, Prodigal Gil said:

Those who have highlighted issues on one very good post from HYCICANDR, which you have completely ignored, state the very slim chances of occurrence being spotted and repeated in succession. This in itself is the purpose of PRG. I feel from your responses that you probably have no idea how a pseudo random number generator works nor do you actually know its purpose. 

So HyricanDR got Q4 twice in a row, and then in a different hand Q4 landed on the flop...

Pardon my ignorance but how does this in any way indicate a problem with the RNG?

The purpose of the RNG is most certainly NOT to avoid patterns or repeated behaviour. It is simply to select a number, at random, between 1 and 52. If that number has already been selected in this hand, go again. Plain and simple, nothing more, nothing less. It is NOT supposed to select a number based on previous or future selections. Even if Q4 came out the last twenty times in a row, it should make no difference to the RNG how it selects the cards for the next hand. Repeat sequences are not only normal, but are expected, in a correctly operating RNG.

 

Share this post


Link to post
Share on other sites

I saw a chick get pocket aces 4 times in a row the other day.......yes!!   4....times.....in....a....row.....then quads, then a full house, then quads again...random my ass....You guys are taking advantage of a lot of older nice people that come out here to play with their pension money and whatnot, people that can't make it to the casino.You're setting them up to take their chips, some people have told me they've spent over 500$ real money to you guys. This is like a pyramid scam you guys have running and it's not right! Be careful robbing people for their hard earned real money guys....there's something called Karma out there and it bites hard......so keep making your money by lying to people.....bad business is good business nowadays.....Please post your monthly profits from everyone you have bamboozled, with your random card shuffler... Maybe post the profits right beside the random card shuffler statistics.

 Even if the pigs were called on you guys...you would just switch something out in your server room...and tell people to know their limit, play within it, or don't turn their Play station on and play the game....while you guys are sipping drinks on a beach with everyone's money....just be honest....just tell us how it really works, How does the system reallywork? it's really quite fascinating! Just stop lying.....please!   I have 40 million, I've earned, not salty about losing or bad beats, just tired of the lies you people tell us and expect us to believe and watch occur on the tables, that is all....

Share this post


Link to post
Share on other sites
14 hours ago, hustl3r said:

I saw a chick get pocket aces 4 times in a row the other day.......yes!!   4....times.....in....a....row.....then quads, then a full house, then quads again...random my ass....You guys are taking advantage of a lot of older nice people that come out here to play with their pension money and whatnot, people that can't make it to the casino.You're setting them up to take their chips, some people have told me they've spent over 500$ real money to you guys. This is like a pyramid scam you guys have running and it's not right! Be careful robbing people for their hard earned real money guys....there's something called Karma out there and it bites hard......so keep making your money by lying to people.....bad business is good business nowadays.....Please post your monthly profits from everyone you have bamboozled, with your random card shuffler... Maybe post the profits right beside the random card shuffler statistics.

 Even if the pigs were called on you guys...you would just switch something out in your server room...and tell people to know their limit, play within it, or don't turn their Play station on and play the game....while you guys are sipping drinks on a beach with everyone's money....just be honest....just tell us how it really works, How does the system reallywork? it's really quite fascinating! Just stop lying.....please!   I have 40 million, I've earned, not salty about losing or bad beats, just tired of the lies you people tell us and expect us to believe and watch occur on the tables, that is all....

WOW! I know you're just trolling but at least make it a little believable! 

Share this post


Link to post
Share on other sites
46 minutes ago, rockguy24 said:

come on Nex lock this thread already, this is getting ridiculous.

And make no sense, but in my mind all players have the same chance. With every BAD BEAT there is one GOOD BEAT

Share this post


Link to post
Share on other sites
On 16/04/2017 at 7:36 AM, Warp said:

No, it won't. Your sample size is ridiculously small.

How many examples do you have? A dozen or so? That's just a laughably small sample size. It says absolutely nothing.

You are also completely ignoring the fact that these supposed "non-random" things seem to be happening to only a few people in particular, while the rest (including me) don't experience it. How exactly do you explain that? Is PP targeting those users in particular and deliberately giving them bad beats?

I have a much simpler explanation: Those particular people have got some bad beats, got angry, and are blaming the program for it.

In fact, I doubt you are experiencing any more bad beats than anybody else. You simply experienced a string of them at one point, started complaining, and are now too stubborn to back down. And, of course, now every bad beat you experience, even if it happens once for every hundred rounds you play, you will take as "evidence" of the "poor quality rng". Your stubbornness makes you biased.

Let me just close down this ignorant argument immediately and as someone with a MSc in Computer Science should know. What has  sample sizing got to do with measuring a quality PRNG? PRNG is a algorithm meaning it follows a set of linear instruction. This also means irrelevant of sample size, it should produce exactly what it is support to do in a periodic loop. Thus, the number produce in a sequence will at some point repeat itself.

This goes back to the point that the purposes of a good quality PRNG is to produce a spread of numbers and in this case cards that meets the criteria of "subjective randomness"...look it up mate.

Uniformly distributed, uncorrelated, and do not it repeat itself within the loop. What players are seeing is blatant correlated, uneveningly distributed and repeat itself in both hold cards and board matching the hand. This is expectionally poor.

Your debate point on sample sizing  is flawed.

 

 

 

 

 

 

Edited by Prodigal Gil

Share this post


Link to post
Share on other sites
20 hours ago, rockguy24 said:

come on Nex lock this thread already, this is getting ridiculous.

That would just result in more threads of a similar nature being started every time someone has a gripe to bear. 

Share this post


Link to post
Share on other sites
6 hours ago, Prodigal Gil said:

Uniformly distributed, uncorrelated, and do not it repeat itself within the loop. What players are seeing is blatant correlated, uneveningly distributed and repeat itself in both hold cards and board matching the hand. This is expectionally poor.

'Uniformly distributed' AND Random? One does not go with the other.  

For example... here are the stats of the Euromillions Lottery over the last 58 draws. 58 Draws = 58x5 balls each draw = 290 chances. A very small sample I know but sure you don't care about sample sizes. 

As you can see, ball number 10 has landed on 12 of the 290 chances (24/1) while ball 8 has only landed on one of those 290 chances (290/1) despite both having odds of 50/1. Hardly 'uniformly distributed' are they?  Also, and this is what most novice poker players seem to have a problem understanding and what your argument seems to be based on, because ball 8 has only landed once in the last 290 occasions, does not make it any more likely to land in the next selection. Likewise, ball 10 is no less likely to land despite the fact that it has landed in 12 of 290 selections. 

There's only 169 different possible starting hands (disregarding suits) and up to 6 players per round... of course you're going to see the same hands over and over... and YES, occasionally, those hands will come one after the other and maybe, on rare occasions, three times! That's how RANDOMNESS works! It's not supposed to be uniformly distributed. If I get Q4 in this hand, I should have exactly the same chance of getting it in the next. Over a LARGE SAMPLE SIZE, the figures should be broadly distributed, but certainly NOT uniformly distributed. 

Edited by Zanatoo

Share this post


Link to post
Share on other sites
10 hours ago, Prodigal Gil said:

What has  sample sizing got to do with measuring a quality PRNG?

I have no words.

  • Like 2

Share this post


Link to post
Share on other sites

Yeah this thread has gone on a long time.  I'll give cliff notes to those who don't want to go through the whole thread.

Side A:  We have seen after 1 billion plus hands the game simulates poker results close to what one would expect from real life results.  Stuff happens so just manage your bankroll well to take the swings.  Be a big boy or girl when you get sucked out on and don't cry.

Side B:  When I don't win I blame the game for not being random enough in my eyes.  The best measure of randomness is my win rate being nearly 100%.  The game will no longer be random due to me winning almost every hand I'm favored, but hey that is the "randomness" I want. 

  • Like 3

Share this post


Link to post
Share on other sites
30 minutes ago, Squid said:

Yeah this thread has gone on a long time.  I'll give cliff notes to those who don't want to go through the whole thread.

Side A:  We have seen after 1 billion plus hands the game simulates poker results close to what one would expect from real life results.  Stuff happens so just manage your bankroll well to take the swings.  Be a big boy or girl when you get sucked out on and don't cry.

Side B:  When I don't win I blame the game for not being random enough in my eyes.  The best measure of randomness is my win rate being nearly 100%.  The game will no longer be random due to me winning almost every hand I'm favored, but hey that is the "randomness" I want. 

If I could 'like' it twice, I would!

  • Like 2

Share this post


Link to post
Share on other sites
16 hours ago, Squid said:

Yeah this thread has gone on a long time.  I'll give cliff notes to those who don't want to go through the whole thread.

Side A:  We have seen after 1 billion plus hands the game simulates poker results close to what one would expect from real life results.  Stuff happens so just manage your bankroll well to take the swings.  Be a big boy or girl when you get sucked out on and don't cry.

Side B:  When I don't win I blame the game for not being random enough in my eyes.  The best measure of randomness is my win rate being nearly 100%.  The game will no longer be random due to me winning almost every hand I'm favored, but hey that is the "randomness" I want. 

 

15 hours ago, Zanatoo said:

If I could 'like' it twice, I would!

absolutely! 

  • Like 1

Share this post


Link to post
Share on other sites
On 4/18/2017 at 4:46 AM, Zanatoo said:

That would just result in more threads of a similar nature being started every time someone has a gripe to bear. 

Yeah, pretty much. At least with this one central thread, all of the information we've presented and all the arguments around that are in a single spot.

Share this post


Link to post
Share on other sites

It's good to see this thread still kicking and screaming.

People complained about the same thing on our WSOP game, and we used an entirely different shuffling mechanic.

We have investigated these reports and came to the conclusion that there doesn't need to be any adjustment on our end.

That being said, we're not in the business of censoring threads unless they break our forum guidelines. Remember to respect your fellow peers; everyone is here for the same reason.

  • Like 3

Share this post


Link to post
Share on other sites
19 hours ago, ClearConscious said:

It's good to see this thread still kicking and screaming.

People complained about the same thing on our WSOP game, and we used an entirely different shuffling mechanic.

We have investigated these reports and came to the conclusion that there doesn't need to be any adjustment on our end.

That being said, we're not in the business of censoring threads unless they break our forum guidelines. Remember to respect your fellow peers; everyone is here for the same reason.

well fair enough I suppose.  

  • Like 1

Share this post


Link to post
Share on other sites
20 hours ago, ClearConscious said:

It's good to see this thread still kicking and screaming.

People complained about the same thing on our WSOP game, and we used an entirely different shuffling mechanic.

We have investigated these reports and came to the conclusion that there doesn't need to be any adjustment on our end.

That being said, we're not in the business of censoring threads unless they break our forum guidelines. Remember to respect your fellow peers; everyone is here for the same reason.

It is fair enough! If they play 80 Percent of her hole cards until the end, than it's normal that they get many bad beats! 👍

Share this post


Link to post
Share on other sites

Right, I play 80 percent of my hands. The arguments are getting better with every post.

Im not getting angry at all, I just think it removes value from the game.

And what is wrong with Q4 being among the dealt 15 of 52 cards 5 times in a row, 4 of those actually within holecards so 12/52? Nothing at all if it wasnt absolutely default and you have the explanation in the back of your head that they dont shuffle and reseed the rngs as seats get filled/emptied, which one would assume doesnt happen all that often in tourneys... So yeah hf playing those back to back to back hands on a bad rng sequence. 

-> I was really really honestly not surprised to get 6 8 after the 6 8 6 8 board and make trips with it. I did however lough and cry. Not because I was upset. The actual thought is simply "Of course he has the same hand with 6 6 6 weak kicker in headsup, im used to these stuck sequences from the game" and another was "I should raise big preflop, this hand will make it" "If only I had, he wouldve probably had to fold his 6 8 which isnt a good hand in general"<-  if you dont see whats wrong with that, you will not see anything anytime. I wouldnt think about raising 6 8 with real cards unless Im tryin to make pots with moves instead of cards

I am totally shocked (Not...) people who play 2 tables a day or so and have less than 200 played dont pick up on patterns and ramble about  small sameplesizes to people who play 20 one after the next resulting in 1000+ tables fast. But anyways, there wont be anything interesting following anyways so ill close the chapter for me and not bite any more baits by keyboard warriors who try to break down everything ever said anywhere with "These guys are such scrubs".

I dont play 80 percent, I dont limp with strong hands to find them cracked (Except that 6 8 of course which I sadly did limp), or any of those "Pro" suggestions. Please even if trolling, put some effort into it, nut just "l2p, git gut"

Edited by HyricanDR
  • Like 1

Share this post


Link to post
Share on other sites
On 22.4.2017 at 6:05 PM, HyricanDR said:

I am totally shocked (Not...) people who play 2 tables a day or so and have less than 200 played dont pick up on patterns and ramble about  small sameplesizes to people who play 20 one after the next resulting in 1000+ tables fast.

So you play 20 games a day, totaling over a thousand games, and thus when in one of those games a very unusual circumstance happens, it must be because the rng is bad.

You do understand that if what you claim is true, then everybody would be seeing such circumstances, not just you, who is playing dozens of games every day?

Share this post


Link to post
Share on other sites

20 games a day!!!!  Even for super accelerated play, that's 20 hours in a 24 hour day,  Plus wait time for a game to start and then add in the time to compose several multiple screen post on this forum!!

I'm impressed.

Share this post


Link to post
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now