It is currently 14 Aug 2018, 09:09
   
Text Size

How rigged is the shuffler?

New decks and cards for Stainless Games' release

Moderators: Xander9009, CCGHQ Admins

How rigged is the shuffler?

Postby addict insane » 24 Jul 2015, 01:52

I've seen it being discussed, but only vaguely, so I wanted to ask the question.

Exactly how rigged is it? I'm getting mana flooded with my 16-18 land decks.

By the way, I'm talking about dotp 2014.

Also, I'm getting roughly the same amount of lands whether I play a deck with 24 lands or with 18: Does the land count actually count for this game, pun intended? This game seems designed for playing with 24 land decks, I'm getting just about the same win rate also with a standard deck than with a modern powerhouse like affinity.
addict insane
 
Posts: 133
Joined: 02 Mar 2015, 22:20
Has thanked: 13 times
Been thanked: 1 time

Re: How rigged is the shuffler?

Postby Chakan » 24 Jul 2015, 02:48

This is a good question I'd like to know more about as well. I usually make 62 card decks with 22 land, and more often than not I get flooded with land. I always suspected that the shuffling system was a bit biased but never really thought much of it until you mentioned it.
User avatar
Chakan
 
Posts: 163
Joined: 07 Jun 2014, 23:08
Has thanked: 11 times
Been thanked: 8 times

Re: How rigged is the shuffler?

Postby RiiakShiNal » 24 Jul 2015, 10:50

Towards the player the shuffle should be purely random.

For the AI though there is shuffling bias based on the difficulty the player has selected and how the deck is configured. When creating a deck using the Deck Builder you will notice a column listed as "Bias" this determines the level of difficulty required to prevent the card from being shuffled towards the bottom of the deck. If the bias is met then the card is supposed to get shuffled as normal, if not met then the card is supposed to be shuffled towards the bottom.

Though it is possible they also biased the shuffler based on land cards, we don't really have any solid information on whether they did or not.
Problems with base game cards: DotP 2014 Core Fixes
Need a deck builder: DotP 2014 Deck Builder
Problems Modding: DotP 2014 Frequent Modding Mistakes
RiiakShiNal
Programmer
 
Posts: 2098
Joined: 16 May 2011, 21:37
Has thanked: 73 times
Been thanked: 462 times

Re: How rigged is the shuffler?

Postby addict insane » 27 Jul 2015, 02:37

RiiakShiNal wrote:Towards the player the shuffle should be purely random.

For the AI though there is shuffling bias based on the difficulty the player has selected and how the deck is configured. When creating a deck using the Deck Builder you will notice a column listed as "Bias" this determines the level of difficulty required to prevent the card from being shuffled towards the bottom of the deck. If the bias is met then the card is supposed to get shuffled as normal, if not met then the card is supposed to be shuffled towards the bottom.

Though it is possible they also biased the shuffler based on land cards, we don't really have any solid information on whether they did or not.
Thank you. As always, very informative and well put.
addict insane
 
Posts: 133
Joined: 02 Mar 2015, 22:20
Has thanked: 13 times
Been thanked: 1 time

Re: How rigged is the shuffler?

Postby Shadowcran » 31 Aug 2015, 23:50

Listen, I've done more research on the faulty shuffler than anyone and this goes back to 2013 version.

It "clumps" draws, especially land BUT there are workarounds, 5 to be exact. IF this helps in your configuring, let me know.

1. Scry. Cards with scry help break up the clumps a little together with
2. Draw-Every deck I've built has to have adequate draw for Duels
3. Ramp- and Deck thinning with it helps break up the land clumps a little.
4. Shuffling. After making a fresh deck, go into vs AI, mulligan 3 times, restart, then repeat. Then play the deck as is. You'll note this breaks up the clumps some.
5. This one is seen by many as a possible "cheat"..Deck conditioning. If you go into deck builder and remove 2 or 3 cards, then add them back, they'll have a 70.2% chance of showing up in opening hand or near it.

This bad shuffling is prevalent in every digital card game if you'll note. THis tells me they're all either using the same RNG to determine shuffles or nearly the same.

This isn't a debate about it, but what I've noted from playing every version, including Hearthstone and other digital card games. The same thing happened in Magic: The Gathering: Tactics and could be fixed by the same 5 ways. This problem is real, and a person who can come up with a better RNG for this could probably write their own golden ticket to $$$.

I've a longer discussion about this saved on multiple sites. Look up this search term, with Shadowcran as author: The case for problems with the Random Number Generator(RNG)in video games. Copy and paste that and you'll find all of them.

Ignore the debates as they're just others saying "Uh Uh" but with sarcasm. The problem is real, now let's work together and fix it.

**Most who treated my original works with scorn and sarcasm are now arguing that there IS a problem.
Shadowcran
 
Posts: 38
Joined: 17 Apr 2011, 20:11
Has thanked: 0 time
Been thanked: 0 time

Re: How rigged is the shuffler?

Postby addict insane » 07 Sep 2015, 15:17

Shadowcran wrote:Listen, I've done more research on the faulty shuffler than anyone and this goes back to 2013 version.

It "clumps" draws, especially land BUT there are workarounds, 5 to be exact. IF this helps in your configuring, let me know.

1. Scry. Cards with scry help break up the clumps a little together with
2. Draw-Every deck I've built has to have adequate draw for Duels
3. Ramp- and Deck thinning with it helps break up the land clumps a little.
4. Shuffling. After making a fresh deck, go into vs AI, mulligan 3 times, restart, then repeat. Then play the deck as is. You'll note this breaks up the clumps some.
5. This one is seen by many as a possible "cheat"..Deck conditioning. If you go into deck builder and remove 2 or 3 cards, then add them back, they'll have a 70.2% chance of showing up in opening hand or near it.

This bad shuffling is prevalent in every digital card game if you'll note. THis tells me they're all either using the same RNG to determine shuffles or nearly the same.

This isn't a debate about it, but what I've noted from playing every version, including Hearthstone and other digital card games. The same thing happened in Magic: The Gathering: Tactics and could be fixed by the same 5 ways. This problem is real, and a person who can come up with a better RNG for this could probably write their own golden ticket to $$$.

I've a longer discussion about this saved on multiple sites. Look up this search term, with Shadowcran as author: The case for problems with the Random Number Generator(RNG)in video games. Copy and paste that and you'll find all of them.

Ignore the debates as they're just others saying "Uh Uh" but with sarcasm. The problem is real, now let's work together and fix it.

**Most who treated my original works with scorn and sarcasm are now arguing that there IS a problem.
I read your thread on nogoblinsallowed and am not surprised at the amount of skepticism and hate you get from other players. I don't know why, but gaming communities seem to be filled with that. I'm also surprised that you tried the 80 lands - 20 spells, 50 lands - 10 spells experiments because I can't believe I haven't tried it myself.

The worst scenario that I have was from playing a Living End deck (which does a lot of drawing) and drawing eleven lands in six turns. The deck had Nineteen lands. I literally drew 58% of lands in six turns.

I don't care how skeptic people are about this, the drawing in dotp 2014 is just stupid unreal. However, I think you are wasting time and energy trying to convince others.
addict insane
 
Posts: 133
Joined: 02 Mar 2015, 22:20
Has thanked: 13 times
Been thanked: 1 time

Re: How rigged is the shuffler?

Postby Xander9009 » 07 Sep 2015, 18:27

We could fix that ourselves.

Each land would have to have an "initial shuffle" function which it calls when the game starts. It would obviously have to leave the hand alone, but the deck itself could be randomized. Make sure only one land runs the function (by setting a value that the function checks before running). Make it iterate through all player's decks.

Code: Select all
for i=0,Player:Library_Count()-1 do
  local RandomIndex = math.Random(0,Player:Library_Count()-1-i)
  local Card = Player:Library_GetNth(RandomIndex)
  if Card ~= nil then
    Card:PutOnBottomOfLibrary()
  end
end
And if needed, you can provide different seeds for math.Random using math.RandomSeed(SomeSeed). I tried os.time(), but we can't use os, unfortunately. I didn't try importing it, though. It may just need imported, but I doubt it. It's possible io would work, which means that we could just make it read the size of some file which changes sizes a lot (the CW's size changes every day). Add the player's index to the seed for each player so each player uses a different seed.

Thoughts?
_______________________________
Community Wad - Community Wad Website - How to Help and Report Bugs
Discord: discord.gg/4AXvHzW
User avatar
Xander9009
Programmer
 
Posts: 2818
Joined: 29 Jun 2013, 07:44
Location: Indiana, United States
Has thanked: 101 times
Been thanked: 420 times

Re: How rigged is the shuffler?

Postby addict insane » 07 Sep 2015, 21:02

This coming from a complete ignorant, but your solution sounds both difficult to implement and like it would slow the game down.

I would love it, though, if you told me I was wrong on both accounts.

I hadn't noticed until Shadowcran mentioned it but I've been recently forced to play decks with either many scry lands or deep into the library manipulation, card drawing territory (Brainstorm, Ponder, Dig Through Time, Eidolon of Blossoms, Dark Confidant, and so on). I Have lost terribly with my pet Merfolk deck when drawing seven lands and two Aether Vials. Way more than once.

I have no idea what will become of my Abzan Constelation deck now that you instructed the A.I. to kill my Eidolon of Blossoms on sight.

Anyways, enough with the rant. Your idea sounds great if you think you can pull it off.
addict insane
 
Posts: 133
Joined: 02 Mar 2015, 22:20
Has thanked: 13 times
Been thanked: 1 time

Re: How rigged is the shuffler?

Postby Xander9009 » 07 Sep 2015, 21:33

Hey, the Eidolon of Blossoms was your idea. haha Should I lower the score a bit? As I said, it's all guesswork, so I may have set it too high.

As for the function, there are already MANY cards that run something at the beginning of the game. They can slow down the game, but only a moment of lag when you first accept your hand. Once that moment of lag is over, for the rest of the game, the function is ignored, so the game plays without being slowed down.

As for the difficulty, it could potentially be moderately complicated, but it would probably be really simple. The code I posted is already 90% of the work. The only read difficulty would be adding it to all lands. But if it's identical on all lands and requires no managers (which it doesn't), then I could run through and add it to all lands VERY quickly and easily with a program I could write in 15 minutes or less. The most likely thing to be complicated and/or difficult is getting a good seed. To be fair, we could likely use ANYTHING as a seed without it even changing between duels and it would probably be enough.

However, there might be a simpler way. Shadowcran, how knowledgeable are you on this topic? I ask because rather than sorting through each card to pick a random one, it would be a lot faster to just call the shuffle function multiple times. If the shuffle function on its own can properly randomize the decks after a sufficient number of iterations, then that may be a better way to go. Possibly choosing a random number between 1 and 10, and then shuffling each deck 10+RandomNumber times.

But that would all depend on me testing this very thoroughly. I understand you're probably sick of having people be skeptical, but if I'm literally going to modify every single land in the game, I need to be 100% certain it's necessary. That probably means testing it myself a few times before starting. Then I'd make a test file for some else who has experienced the problem to test a few times and see if it fixes it. If anyone would be up for that, let me know.
_______________________________
Community Wad - Community Wad Website - How to Help and Report Bugs
Discord: discord.gg/4AXvHzW
User avatar
Xander9009
Programmer
 
Posts: 2818
Joined: 29 Jun 2013, 07:44
Location: Indiana, United States
Has thanked: 101 times
Been thanked: 420 times

Re: How rigged is the shuffler?

Postby Shadowcran » 12 Sep 2015, 07:12

addict insane wrote:
Shadowcran wrote:Listen, I've done more research on the faulty shuffler than anyone and this goes back to 2013 version.

It "clumps" draws, especially land BUT there are workarounds, 5 to be exact. IF this helps in your configuring, let me know.

1. Scry. Cards with scry help break up the clumps a little together with
2. Draw-Every deck I've built has to have adequate draw for Duels
3. Ramp- and Deck thinning with it helps break up the land clumps a little.
4. Shuffling. After making a fresh deck, go into vs AI, mulligan 3 times, restart, then repeat. Then play the deck as is. You'll note this breaks up the clumps some.
5. This one is seen by many as a possible "cheat"..Deck conditioning. If you go into deck builder and remove 2 or 3 cards, then add them back, they'll have a 70.2% chance of showing up in opening hand or near it.

This bad shuffling is prevalent in every digital card game if you'll note. THis tells me they're all either using the same RNG to determine shuffles or nearly the same.

This isn't a debate about it, but what I've noted from playing every version, including Hearthstone and other digital card games. The same thing happened in Magic: The Gathering: Tactics and could be fixed by the same 5 ways. This problem is real, and a person who can come up with a better RNG for this could probably write their own golden ticket to $$$.

I've a longer discussion about this saved on multiple sites. Look up this search term, with Shadowcran as author: The case for problems with the Random Number Generator(RNG)in video games. Copy and paste that and you'll find all of them.

Ignore the debates as they're just others saying "Uh Uh" but with sarcasm. The problem is real, now let's work together and fix it.

**Most who treated my original works with scorn and sarcasm are now arguing that there IS a problem.
I read your thread on nogoblinsallowed and am not surprised at the amount of skepticism and hate you get from other players. I don't know why, but gaming communities seem to be filled with that. I'm also surprised that you tried the 80 lands - 20 spells, 50 lands - 10 spells experiments because I can't believe I haven't tried it myself.

The worst scenario that I have was from playing a Living End deck (which does a lot of drawing) and drawing eleven lands in six turns. The deck had Nineteen lands. I literally drew 58% of lands in six turns.

I don't care how skeptic people are about this, the drawing in dotp 2014 is just stupid unreal. However, I think you are wasting time and energy trying to convince others.
I agree. Except for here, at Slightlymagic, I encounter outright hate for even mentioning problems. It's a community problem as cheaters and hackers(not modders, I don't mean modders)get supporters and apologists and those trying to stop them get ostracized and hate.

I also get this hate for daring to suggest...gasp...larger than 60 decks, something I wouldn't do in PAPER magic. I've noticed the problems with the shuffler for years now so I've adapted around the problem. I've even been very successful with it. Everyone must chalk up my success to "luck" as although I've ranked #1 for years now, everyone treats my ideas as garbage and instead use mathematical formulaic builds.

In truth, I'm tired of trying to convince people and getting this treatment.
Shadowcran
 
Posts: 38
Joined: 17 Apr 2011, 20:11
Has thanked: 0 time
Been thanked: 0 time

Re: How rigged is the shuffler?

Postby Shadowcran » 12 Sep 2015, 07:15

Xander9009 wrote:Hey, the Eidolon of Blossoms was your idea. haha Should I lower the score a bit? As I said, it's all guesswork, so I may have set it too high.

As for the function, there are already MANY cards that run something at the beginning of the game. They can slow down the game, but only a moment of lag when you first accept your hand. Once that moment of lag is over, for the rest of the game, the function is ignored, so the game plays without being slowed down.

As for the difficulty, it could potentially be moderately complicated, but it would probably be really simple. The code I posted is already 90% of the work. The only read difficulty would be adding it to all lands. But if it's identical on all lands and requires no managers (which it doesn't), then I could run through and add it to all lands VERY quickly and easily with a program I could write in 15 minutes or less. The most likely thing to be complicated and/or difficult is getting a good seed. To be fair, we could likely use ANYTHING as a seed without it even changing between duels and it would probably be enough.

However, there might be a simpler way. Shadowcran, how knowledgeable are you on this topic? I ask because rather than sorting through each card to pick a random one, it would be a lot faster to just call the shuffle function multiple times. If the shuffle function on its own can properly randomize the decks after a sufficient number of iterations, then that may be a better way to go. Possibly choosing a random number between 1 and 10, and then shuffling each deck 10+RandomNumber times.

But that would all depend on me testing this very thoroughly. I understand you're probably sick of having people be skeptical, but if I'm literally going to modify every single land in the game, I need to be 100% certain it's necessary. That probably means testing it myself a few times before starting. Then I'd make a test file for some else who has experienced the problem to test a few times and see if it fixes it. If anyone would be up for that, let me know.
I don't know how to manipulate code or anything computer related first of all, BUT:

I've more free time than probably any 5 other players combined. I'm disabled and have time to experiment and playtest.

I first noticed the shuffling problem in Magic The Gathering: Tactics. (If you modders took this over and recreated it, I'd probably die happy). There, the mana is auto provided so "standard deck size" was 40 cards. The shuffling of only spells was still problematic as they would still clump. I experimented constantly and found ways around it. We didn't have scry, but did possess draw and ramp so I built my decks around that at first. Then I tried larger decks up to 75 cards. I could overcome the game timer this way due to constant practice and actually ran out of cards many times due to my speed(but you don't lose in Tactics when you run out of cards).

The result? I won constantly. Had more gold than anyone else and 5x every card as well as more tournament wins. I usually gave cards and gold away to new players as I had no use for it once SOE pulled all of the developers.

Then I came to DoTP and noticed the same problems with the shuffler. I've also tried other card games and saw the same exact problem. My conclusion was everyone used the same RNG for their games. I adapted accordingly.

What I've determined is the shuffler does not change drawing odds. Let's say you have 20 lands in a 60 card deck and have drawn 18 lands in a row. Think the odds of drawing the last 2 lands next is 2/42 or 1/21? No, It's still 33.3%. There's the problem. That's why you get clumping. The shuffler is not mixing the deck properly as it doesn't calculate the changing odds. With a larger deck(no more than 70 in DoTP)it actually draws more consistently due to it's faulty program.

This is why deck conditioning actually makes a 70.2% chance of drawing the cards you alter in the deck builder in either opening hand or first few draws.

Since this site doesn't notify me when I've been answered, contact me at Shadowcran@aol.com. I've had this address since 2003 so I simply haven't changed it despite aol being a joke now.
Shadowcran
 
Posts: 38
Joined: 17 Apr 2011, 20:11
Has thanked: 0 time
Been thanked: 0 time

Re: How rigged is the shuffler?

Postby Shadowcran » 12 Sep 2015, 07:41

2 other things I didn't mention earlier is Healing and defense which collectively equals "stalling".

These mechanics give your deck(especially a larger one) time to break the clumping yourself and kill through sheer attrition if your combos don't arrive. You're stalling the match til the terrible shuffle favors you.

But you can't just use any old healing spell. It has to have a dual purpose like Solemn Offering which kills enchants AND heals you or a creature that heals as it enters the battlefield. Spells like ANgel's Mercy just don't cut it.

Defense is a concept alien to most players I've encountered. They don't see it or refuse to see it. When they can't aggro past your defenses, they are giving you more time(turns)to build for wins. They'll call it stalling and that's just what it is. Creatures heavier on the toughness, walls that do more than just sit there, and spells that delay make up for a proper enough defense. You can't really get this with ones of 5 mana or over, like Siege Mastodon, but one like Joruvai Murk Lurker or Graveyard Marauder more than fit the bill.

These are the methods I do to overcome the terrible shuffler.
Shadowcran
 
Posts: 38
Joined: 17 Apr 2011, 20:11
Has thanked: 0 time
Been thanked: 0 time


Return to Magic: The Gathering - Duels of the Planeswalkers

Who is online

Users browsing this forum: No registered users and 1 guest


Who is online

In total there is 1 user online :: 0 registered, 0 hidden and 1 guest (based on users active over the past 10 minutes)
Most users ever online was 279 on 11 Jul 2013, 22:03

Users browsing this forum: No registered users and 1 guest

Login Form