Deck RNG

Options
DBJones
DBJones Posts: 803 Critical Contributor
in MtGPQ General Discussion
Previously, I had seen on these forums that the cards you draw in a game are produced by randomly shuffling 4 copies of each card in your deck, and that seemed reasonable from what I saw. However, since the update this doesn't feel true anymore. The fact that they said this update fixed the HUF issue, which was related to the 40 card decks from what I've heard, makes me think they may well have changed the way this works. Anyone have concrete evidence one way or another?
«1

Comments

  • MDsupa
    MDsupa Posts: 124 Tile Toppler
    Options
    Yes, I'm noticing alot more bad card draws.  I would get 3-4 copies of the sames 4-cards over and over.  Deck RNG draw has been bad.  on top of the New Cascading and AI now prioritize 5-match on top of their Clairvoyance skyfall.   Someone wanted a tougher game.. you got it.
  • DBJones
    DBJones Posts: 803 Critical Contributor
    Options
    Looks like the decks are still created in the same way at least. I just cycled my way through 80 cards in a deck, and found four copies of each card in both the first and second 40. That makes the odds of a different method pretty low. As to the order of the cards, and initial hands, I haven't done nearly enough for any trends to mean anything, and I haven't even noticed any.
  • wereotter
    wereotter Posts: 2,064 Chairperson of the Boards
    Options
    I’ve been observing a lot more highly improbable AI plays. Incidents of the AI able to play the same creature three turns in a row starting on the first or second turn.

    The odds of drawing the same card three out of four cards out of 40 seem too low to see this as often as I have this update. 
  • Stormbringer0
    Stormbringer0 Posts: 190 Tile Toppler
    Options
    wereotter said:
    I’ve been observing a lot more highly improbable AI plays. Incidents of the AI able to play the same creature three turns in a row starting on the first or second turn.

    The odds of drawing the same card three out of four cards out of 40 seem too low to see this as often as I have this update. 
    On the other hand, I cycled through 20 cards in my librarys in search for Dissenter's Deliverence (for just one support on the battlefield) and didn't find it. Those situations may some just as random as any oher scenario, but for it shows that something is wrong with which cards we draw.
  • rafalele
    rafalele Posts: 876 Critical Contributor
    Options
    Something has been changed , I also get draws of 3/4 copies of the same card, more than before the update.
  • Coilbox
    Coilbox Posts: 202 Tile Toppler
    Options
    DBJones said:
    Previously, I had seen on these forums that the cards you draw in a game are produced by randomly shuffling 4 copies of each card in your deck, and that seemed reasonable from what I saw. However, since the update this doesn't feel true anymore. The fact that they said this update fixed the HUF issue, which was related to the 40 card decks from what I've heard, makes me think they may well have changed the way this works. Anyone have concrete evidence one way or another?
    Grab an umbrealla mate, it's gonna rain on you.
  • Volrak
    Volrak Posts: 732 Critical Contributor
    Options
    DBJones said:
    Looks like the decks are still created in the same way at least. I just cycled my way through 80 cards in a deck, and found four copies of each card in both the first and second 40. That makes the odds of a different method pretty low. As to the order of the cards, and initial hands, I haven't done nearly enough for any trends to mean anything, and I haven't even noticed any.
    In June last year the 40 card deck mechanic was temporarily not working for some reason: https://forums.d3go.com/discussion/65639/decks-are-not-40-cards-4x-of-each-card/p1.  So it wouldn't be the strangest thing if it happened again.

    But what you've done above is the perfect way to check; it can be done without too much effort any time it looks like it might be off.
  • bk1234
    bk1234 Posts: 2,924 Chairperson of the Boards
    edited February 2018
    Options
    I've always said the 40 card deck is a figment of our imagination. Cast Uncage the Menangerie on turn 1 for proof of my assertion. Or HUF with AotSS and another AotSS in hand. 

    I've also drawn more than 4 copies of numerous other cards on their own in the first 3 - 5 turns -- and seen AI do the same. Any draw deck will show you there is no such thing as the 40 card deck. 

    It's an RNG game with an RNG deck -- and I've seen nothing to convince me otherwise. (Though I welcome proof to the contrary.)
  • Volrak
    Volrak Posts: 732 Critical Contributor
    Options
    bken1234 said:
    I've always said the 40 card deck is a figment of our imagination. Cast Uncage the Menangerie on turn 1 for proof of my assertion.
    The way it's been explained is that an initial set of 40 cards is formed, and any time any fetch or draw effect looks for a card that can't be found there (but which exists in your deck), another 40 are added to the bottom.  Uncage is still interesting - if for example you have just one creature card in your deck, and you uncage 6 copies into your hand, then you should expect to draw no creatures normally until you've gone through at least 40 cards.

    Most times anyone's bothered to count their cards in a game, the 40 card description has held up, but there are cases where it hasn't.  Your mileage may vary.
  • bk1234
    bk1234 Posts: 2,924 Chairperson of the Boards
    edited February 2018
    Options
    I had a Koth cast 5 Inferno Jets in 2 turns on Tuesday. It was in The first 10 rounds. He wasn’t using any draw or fetch mechanism like HUF. The first two cast on Round 3 (I had GI out) — I Cast Nyx in round 4 and GI came back. 

    Over the next 2 turns he filled and held 3 cards. On Round 7 he cast 3 Inferno Jets and hit me with 2x damage with the last one. 

    So that’s 5 of one card in one deck. He went on to cast the same card a few more times.

    This is noteworthy for me because 1. it was the red node so I lost a bonus and 2. He held 3 IJ to get past my GI — I’m having a blast observing smarter Greg - So I did double check the battle log after to confirm this was the the case and I didn’t miss HUF or any other fetch of draw mechanism. 
  • DBJones
    DBJones Posts: 803 Critical Contributor
    Options
    As I said a bit earlier, I cycled my way through 80 cards, and got two sets of the 40 card deck that has 4 of each card. While not conclusive, and there could be unexpected things that mess it up besides fetching, the odds of getting the described 40 card deck twice in a row from a purely RNG deck are pretty low. However, the AI might have different rules, did anyone ever see Greg have that issue where HUF didn't fetch as many cards as it should? Or any other glitches where the game failed to create another forty card deck when it was needed?
  • Kinesia
    Kinesia Posts: 1,621 Chairperson of the Boards
    Options
    I think the 40 card deck is the _basic_ setup as DNJones demonstrates, but there is SOMETHING that messes it up, maybe more than 1 thing, and we currently have zero idea what. At all, just wild guesses.

  • ILikePancakes
    ILikePancakes Posts: 101 Tile Toppler
    edited February 2018
    Options
    Just played a match where the ai cast 5 gideons intervention  (at least that's what I thought) checked the battle log after I got my **** kicked to see if it was 4 or 5 and the log showed only one being cast. Huh?

    Note: these were cast on separate turns each time.
  • Kinesia
    Kinesia Posts: 1,621 Chairperson of the Boards
    Options
    Did they have Starfield? (Although it should show Starfield triggering and then GI coming back into play, but the log is faulty with some things.)
  • ILikePancakes
    ILikePancakes Posts: 101 Tile Toppler
    edited February 2018
    Options
    Yes, they did. Thank you, had to read what that card does. Makes sense.

    What a pain in the...
  • Kinesia
    Kinesia Posts: 1,621 Chairperson of the Boards
    Options
    If the battlelog didn't show it I'd make a bug report for that, it really needs to.

    With the new Across Ixalan event I was having a hard time working out what was happening and it turns out to be similar, the "node" effects are not recorded in the battlelog, so it looks like the AI is just cheating and buffing its creatures for no reason.
  • Sirchombli
    Sirchombli Posts: 322 Mover and Shaker
    Options
    I did fall victim to 5 lay claims last night. I don't think they drew past 40, but things didn't start getting weird til their 6th turn, and 3 happened off of a single cascade. Definitely weren't any day's undoings going on. It's hard to guage if anything is fishy right now because I'm still adjusting to  cascadapalooza2k18. Feels bad, but lots feels bad when you watch Greg cast 70 mana worth of cards from one cascade . On the upside, Greg's targeting is wonkier than ever, so I survived 5 cast outs. 
  • wereotter
    wereotter Posts: 2,064 Chairperson of the Boards
    Options
    Had Raider’s Wake in play during a recent match and creatures consistently in play to trigger raid. Watched the AI cast three copies of Legion Conquistador.

    This is vital as it proves the decks aren’t 40-card decks. Every time you cast a Legion Conquistador, you tutor up the next one into hand. However it is the last card in hand, meaning my Raider’s Wake should force a discard of the tutored version. This means the AI had to draw three copies in addition to the three tutored up for a total of 6 before running through a 40-card deck. So either the AI drew key cards needed outside the normal peremiters, the decks are just RNG picking any of the 10 possible cards, or the AI can rearrange its hand after casting cards. 

Categories