Shuffle prevention bug

[tomaszavenger]

Let's say I'm searching my deck to get a card. If I accidentally shuffle before I conceal my deck, when I press the "Reveal to owner: None and Shuffle", it conceals all the cards, but does not shuffle; this means I know the order of my cards. The "none and shuffle" should probably be active while the deck is visible. Ideally you should prevent the press of "none and shuffle" when the cards are face down.

Just my two cents.

[Remag]

On the note of the shuffler I would appreciate it and think it would be more effective if we could have the shuffler do it's thing two or three times when it is commanded to do so (all at once) as I have noticed a pattern of the loss of a true of mix of the cards since the last change was made. I have  also surmised this by asking players and from some knowledge of curve charts that give the percentage chances of cards being drawn over a linear course of draws.

Let it be known also that I chose to wait at least 50 games since to see if what I was observing (from my decks and others) occasionally was a recurring pattern and indeed there is a marked difference imho from the way the shuffler used to mix and how it does so now.

What I am talking about seeing more frequently than before, for example, is long(er) land pockets or waht is know as land screw but more frequently I have been seeing and hearing about 3 out of 4 copies of a card being placed right next to each other and on one occasion for me so far 4 out of 4 so far.   

Now, one might consider that a freak of probability or just luck but that is not likely when I have heard it (and witnessed it) from other players.

Not that the shuffler is very bad, in fact many new members as of late have shared with me how they find Lackeys shuffler better than others.  Just to say I think it could mix more efficiently.   

[Trevor]

Let's say I'm searching my deck to get a card. If I accidentally shuffle before I conceal my deck, when I press the "Reveal to owner: None and Shuffle", it conceals all the cards, but does not shuffle; this means I know the order of my cards. The "none and shuffle" should probably be active while the deck is visible. Ideally you should prevent the press of "none and shuffle" when the cards are face down.

Just my two cents.

Fixed.

[Trevor]

On the note of the shuffler I would appreciate it and think it would be more effective if we could have the shuffler do it's thing two or three times when it is commanded to do so (all at once) as I have noticed a pattern of the loss of a true of mix of the cards since the last change was made. I have  also surmised this by asking players and from some knowledge of curve charts that give the percentage chances of cards being drawn over a linear course of draws.

Let it be known also that I chose to wait at least 50 games since to see if what I was observing (from my decks and others) occasionally was a recurring pattern and indeed there is a marked difference imho from the way the shuffler used to mix and how it does so now.

What I am talking about seeing more frequently than before, for example, is long(er) land pockets or waht is know as land screw but more frequently I have been seeing and hearing about 3 out of 4 copies of a card being placed right next to each other and on one occasion for me so far 4 out of 4 so far.   

Now, one might consider that a freak of probability or just luck but that is not likely when I have heard it (and witnessed it) from other players.

Not that the shuffler is very bad, in fact many new members as of late have shared with me how they find Lackeys shuffler better than others.  Just to say I think it could mix more efficiently.   

Currently the shuffle algorithm is perfect, and it has been thoroughly tested. Multiple shuffles are redundant. This issue has been brought up many times, so I made a suite of tests to test it experimentally. If you want to generate some data to confirm your suspicion, I'd be happy to look at it. Perhaps your real cards never get shuffled well and you aren't familiar with the real randomness of the computerized shuffle.

No matter how perfect a shuffle is, people will always complain it is not random when they are unlucky. At this point, I'm only going to investigate it further if people present me with statistical data. There are many ways to do that, and I have made tools to help me do that. One thing you could do is something like this: Make a deck of twenty cards, each with a different name ascending alphabetically, like Awesomeness, Bash, Coolness, and so on. Shuffle it. Each shuffle, each card has a 50% chance to be in the top half as it does in the bottom half of the deck. You could track a specific card, and repeat this a billion times. The more times you repeat it, the more the average will approach 50%. On second thought, I don't think you can feasibly test this without the test routines I made. I can reenable them and tell you how to use them if you actually want to support your suspicions.

Without the test functions, if you can find any repeatable pattern, let me know. I don't suggest you have multiple cards with the same name in the test deck. The shuffler algorithm doesn't differentiate same-name cards with different name cards in any way, but it can confuse you in which version of that card you are seeing. If you have 20 copies of a card in a 60 card deck, it is very likely you will get SOME of that card in your opening hand, but you may think you are seeing the same one over and over and over, but in reality you are seeing a different card that just has the same name.

As a side note, there is no true randomness in the universe, so no computer can actually make a truly random number generator. But in this context, "random" means "with no discernible order."

[Alastair]

The random library is far from perfect. No computerized "random" effect is in itself truly random because they're based on static algorithms. Likewise it can't have been thoroughly tested by a single person (or even a small group of people). Trevor could overhaul the random lib time and time again and there will still be some people that it just seems to hate and some that seemingly always get the perfect draw. Such is the nature of a computer based randomizer.

So while I do have to disagree with Trevor's position that the shuffler is perfect, I'm of the opinion that it's just about as good as it is going to get.

[Remag]

Okay Trevor and Alastair, I am good with that and that was quite insightful as well as reassuring.

Especially regarding the algorithms. So, as the algorithms function currently, would or could they tend to favor a certain deck type, for example, like one that contains cards that have cards requiring the deck to be shuffled more often? If you please Alastair, what do you mean about Trevor being able to overhaul the lib again? Would that alter kind of randomness which would in some way favor, as in another example of above, a deck that has more singletons than four ofs? Ty

By the way, perhaps the shuffler does hate me as much as Laconic Spray man over there

[Remag]

Congradulations Sei, you have officially become a troll. Hope you find a better day job.

[Trevor]

The random library is far from perfect. No computerized "random" effect is in itself truly random because they're based on static algorithms. Likewise it can't have been thoroughly tested by a single person (or even a small group of people). Trevor could overhaul the random lib time and time again and there will still be some people that it just seems to hate and some that seemingly always get the perfect draw. Such is the nature of a computer based randomizer.

So while I do have to disagree with Trevor's position that the shuffler is perfect, I'm of the opinion that it's just about as good as it is going to get.

No one can discern any pattern, so it is perfectly random by the definition I gave. Of course you could predict things if you knew the internal workings of the algorthms used, but it's absurd to think anyone would actually be doing that. For the purposes that I use the shuffler, it is perfectly random, by the definition I gave.

To reiterate: random = no discernible order or pattern.

[Trevor]

LOL... I am just toying with you. It seems I succeeded at getting under your skin!

I believe that is the definition of a troll.

[Alastair]

If you please Alastair, what do you mean about Trevor being able to overhaul the lib again?

I didn't say he was going to. I said he could. Meaning he could swap to a different library and it'd achieve pretty much the same results.

No one can discern any pattern, so it is perfectly random by the definition I gave. Of course you could predict things if you knew the internal workings of the algorthms used, but it's absurd to think anyone would actually be doing that. For the purposes that I use the shuffler, it is perfectly random, by the definition I gave.
To reiterate: random = no discernible order or pattern.

I have yet to see a machine code randomizer that's actually random. It'll always favor one set of numbers over another, and the math library Lackey is using definately falls into that category. The debate of algorithm generated randomness has been going on for years and years. Which isn't really the point though, you could spend the next ten years trying to get the perfect "random" shuffler and you'd still be basically where we are now. Lackey's "random" isn't perfect, but it's the best we're going to get.

[Trevor]

If you please Alastair, what do you mean about Trevor being able to overhaul the lib again?

I didn't say he was going to. I said he could. Meaning he could swap to a different library and it'd achieve pretty much the same results.

No one can discern any pattern, so it is perfectly random by the definition I gave. Of course you could predict things if you knew the internal workings of the algorthms used, but it's absurd to think anyone would actually be doing that. For the purposes that I use the shuffler, it is perfectly random, by the definition I gave.
To reiterate: random = no discernible order or pattern.

I have yet to see a machine code randomizer that's actually random. It'll always favor one set of numbers over another, and the math library Lackey is using definately falls into that category. The debate of algorithm generated randomness has been going on for years and years. Which isn't really the point though, you could spend the next ten years trying to get the perfect "random" shuffler and you'd still be basically where we are now. Lackey's "random" isn't perfect, but it's the best we're going to get.

It is perfect depending on how you define "random".

[Remag]

Agreed that there is no true randomness in this universe as we know it, which is intruiging and comforting to say the least.

Please do not construe my curiosity as any kind of complaint toward Lackey or your work Trevor. Admittedly, I was frustrated but that is gone with the wind now as I can thank you all for being helpful in the comprehending of certain methodologies, realities and strategies (already lol).

Alastair, I didn't ask if he was going to, nor implying that you were saying that. I know you said he could. Saying "he could swap to a different library and it'd achieve pretty much the same results" is a half answer that would seem to indicate then that there would be a difference, quite possibly favoring something in one way or another. If you can, please provide more detail in your answer(s) to the questions as I asked them sir.   

Not saying why because this could just be an accident but it would seem that communication regarding the algorithm has gone selective and defensively cryptic since I asked ...

"...as the algorithms function currently, would or could they tend to favor a certain deck type, for example, like one that contains cards that have cards requiring the deck to be shuffled more often? If you please Alastair, what do you mean about Trevor being able to overhaul the lib again? Would that alter kind of randomness which would in some way favor, as in another example of above, a deck that has more singletons than four ofs?"

Please do not take these question as anything but from curiosity and a thirst to learn.

No one can discern any pattern, so it is perfectly random by the definition I gave. Of course you could predict things if you knew the internal workings of the algorthms used, but it's absurd to think anyone would actually be doing that.

I certainly did discern a pattern, although not pertaining directly to your algorithm. There are patterns all around us (and abound...) in the micro and macro time space scope of things. It is the pattern that was provided for me to see that caused me to post in this thread in the off chance that my words might help but I know better now that you have things in order here Trevor. All cool. 

However, not that I initially thought of it but, why would it be "absurd" to think that someone might benefit or revel in a kind of (even if petty) power in some way? Very often, a lot of people do things that do not follow common sense or substantial reason. In that particular case, they would not be inviting ridicule as no one in particular would know and it would seem there could in fact be reason, exercised in a logical common sense way. 

Anyway, no complaint here.  After all, the dog barks and the caravan moves on, correct? Pride also taketh away Sei (only to leave the Demonic). I pray you are healed as surely as you will be forgiven and may you know a deeper love. Pardon is another matter though as, yeah, we all got a lot to learn...    

By the way, the only reason I chose to play a less than 5 min match (that I threw because I know orientated your brain is too needing to win) with you, Sei, was because, to put it bluntly (what the heck) I thought you might have gotten past being a prick. There is also the fact that you were blocking me from hosting the day earlier. Thought I might try to put you out from under the bridge but Alas, not quite yet it would seem as you only did that so that you could in...

Poor taste, hijack one of my youngest rogue test decks and attempt in your insincere way to cover up what is obviously another fail in getting some rise (?) out of me through sarcastic ridicule. Wow.

Just leave the bad half of pride outside and life gets better unfettered. Or you could just waste your time proving to 3 people how tough an interwebz Level 20 Troll Fighter you are. Flip a coin and see if I care about that any further.  I did high school once already Sei, you can relive it on your own if that's what turns your crank. 

If that is the case, then just leave me out of it.

[Trevor]

Remag, you did not discern any real pattern. What you think you see is an illusion, a trick human brains play on their owners. We tend to see patterns in randomness. It's called pareidolia. What you won't find is any patterns that you can actually use to predict the future. You may "see" a pattern when you draw a hand, but you can't say something like your next hand will have X card, or even that your next hand will have a disproportionately high or low chance to have a certain pattern. When you draw a hand, you WILL likely think you see a pattern, but that would be the case in a lot of random samplings, and people also have a tendency to forget about the times when "patterns" don't happen, thus leading to a statistical bias and the belief that you think things aren't functioning as they would if they were random.

My point about using a deck of one-ofs was so you aren't confused. In reality, the algorithm doesn't consider card names at all, so any error in the shuffler wouldn't be based on card names. The shuffler doesn't consider the card type or anything about the card. It just randomizes the whole deck, ignorant of what the cards are. That's what a random shuffle would do.

"start learning how to play Magic instead of blaming the shuffler all the time!"
Blaming the shuffler can be interpreted as blaming randomness. The best player with the best deck can lose to a mediocre player with a mediocre deck. Randomness does play a part in most games, and almost all CCGs. Blaming the shuffler is not the same thing as saying the shuffler is not perfect.

[Alex_Steinkohl]

There's a random a random factor in almost any game, even chess.
To eliminate that would require you to be able to read your opponents mind!

[Trevor]

There's a random a random factor in almost any game, even chess.
To eliminate that would require you to be able to read your opponents mind!

Chess is deterministic. That's one of the draws, since you can guess long chains.
If I do this, he can do this, and then I would do this, and then he would do this, and so on....