Why did the AI need a nerf?

babar3355
babar3355 Posts: 1,128 Chairperson of the Boards
in MtGPQ General Discussion

This was an unannounced but obvious change in the recent update.  The AI, who most considered totally inept pre-update, has now been further downgraded.

Although Greg (our name for the AI) has never been the brightest bulb when it came to gem matching, it has recently started avoiding match 5's more frequently and even avoided breaking single shield supports.   Greg has classically taken many forms of match 5's, but screwed up the L shaped version a good amount of the time.  Now they will miss the straight line match 5's and not even for a higher priority target like an activate or support gem. Greg seems to prefer to not break player supports at this point either. 

I assume the reason for dumbing down Greg even further is the continued desire to increase your retention rate on new players who download the game? 

Spoiler: It wont work. Cascades will still happen that will cause people to rage. 

What might work?  At the end of each chapter of story mode, give players a big box  premium pack.  Also, up the drop rates.  The best way to create the addiction you are looking for is to have those shaking gems come back with a nice bomb (or even a perceived one to newer players).

I will yet again chastise the company for taking a brilliant model of a collectible card game and refusing to let the players collect the cards.  We won't get board if we have all the cards!  There is plenty of options for new gameplay content and we all know Ixalan is right around the corner. 

But I digress... please don't make the game any easier.  Greg needed a buff to competence if anything. We shouldn't have tons of perfect scores on the first HoR and RotGP. It makes the game even more boring.

«1345

Comments

  • Corn_Noodles
    Corn_Noodles Posts: 477 Mover and Shaker
    I noticed this as well. It might not have been on purpose.
  • Marrhault
    Marrhault Posts: 23 Just Dropped In
    I've seen something similar, but when i did notice it Greg appeared to favor greatly Mana gains.  Even if the 5 would have given another roll the 4 match would yield more Mana.
  • speakupaskanswer
    speakupaskanswer Posts: 306 Mover and Shaker
    I noticed avoiding support shields that would have helped him (why not her?) too. That's a new one!
    But why are we even talking about this when the AI is still incapable of using the most powerful mechanic the game has at the moment? So, whenever you run into a cycling deck, you know it will waste mana on Drake Havens that will never be triggered?
  • babar3355
    babar3355 Posts: 1,128 Chairperson of the Boards
    speakupaskanswer said:
    I noticed avoiding support shields that would have helped him (why not her?) too. That's a new one!
    But why are we even talking about this when the AI is still incapable of using the most powerful mechanic the game has at the moment? So, whenever you run into a cycling deck, you know it will waste mana on Drake Havens that will never be triggered?

    If we can only focus on one problem at a time then I guess we should still be focusing on the plethora of bugs and server issues.  But can't we multi-task?

    But to your point.. if the AI learns to cycle then I will immediately quit the game.  It's broken enough that I can build a 100 HP stack of drake havens.  It's unthinkable that the AI can do it against me.

  • speakupaskanswer
    speakupaskanswer Posts: 306 Mover and Shaker
    I'm not saying it should learn to use it but when discussing the AI that's what comes to mind first. It doesn't have the ability to use a key mechanic! It's the main reason why Solemnity is such an absurd card. So, if they couldn't even manage that (or make cycling balanced), why should we expect the AI to make reasonable gem swap decisions. Your question is valid, though, why make it even worse than before?
  • shteev
    shteev Posts: 2,031 Chairperson of the Boards
    speakupaskanswer said:
    I'm not saying it should learn to use it but when discussing the AI that's what comes to mind first. It doesn't have the ability to use a key mechanic! It's the main reason why Solemnity is such an absurd card. So, if they couldn't even manage that (or make cycling balanced), why should we expect the AI to make reasonable gem swap decisions. Your question is valid, though, why make it even worse than before?
    It's not so much asking for effort to be put in to making the AI better... it's more a case of why effort was put into doing things that nobody asked for (like making the AI worse, or cocking up the event decks), instead of things that people HAVE been asking for.

    My own personal theory is that they've been talking to a whale who's had some troubles with the difficulty level in the game, and since he pays their salaries, they leapt into action to comply with his request to make it easier.
  • Mburn7
    Mburn7 Posts: 3,427 Chairperson of the Boards
    Its funny, this happens every update.  They fix all AI bugs, and then when they release new cards and mechanics it breaks again.  It'll gradually be fixed, don't worry.  Just in time to break with the next set.  For now, enjoy the free match 5s and general ineptitude
  • fiirst
    fiirst Posts: 438 Mover and Shaker
    edited August 2017
    my AI (sorin) match 5 white gems as L shape. i think it depend pw color too.
  • Steeme
    Steeme Posts: 784 Critical Contributor

    While I'm enjoying the fact that my supports last more than a turn or two, it feels dirty.  The AI not taking the swap to destroy my Desert's Hold and instead taking a 3-swap of it's secondary color literally feels like it is intentionally "going easy" on me to let me win.

  • Doomstat
    Doomstat Posts: 115 Tile Toppler
    I know this is way off base, but what if this was intentional? Lessening the AI difficulty rating could have been a strategic maneuver by the development team to try to tempt players away from Cycle. Knowing that the AI won't prioritize your supports could make you try other decks that might have a chance. 

    I have moved away from Cycle decks to try Anointed Procession decks. Sandwurm Convergence and other token generating spells/supports have revitalized my enjoyment and enthusiasm for playing in more than just the coalition events.

    Note, I'm not saying they had the foresight of this happening. I'm saying that maybe they realized that this might be a good thing for moving the meta from Cycle to something better for the game. We all love a challenge, but it's nice to not auto against cycle played by the AI.
  • Pcell777
    Pcell777 Posts: 51 Match Maker
    It bothers me more, now, that AI deliberately avoids 5-match gems than them having crazy cascades and extra turns.

     I felt disrespected.
  • tfg76
    tfg76 Posts: 258 Mover and Shaker
    The weird thing is that gem optimization seems like the simplest thing to implement! To be honest, the simplistic AI is one of the things that most hurts this game in my opinion. Most decent players can expect a win ratio of 95-99% against random decks, and that makes the leaderboards in events more about the tertiary objectives than anything else. Let's at least not make it even worse!
  • shteev
    shteev Posts: 2,031 Chairperson of the Boards
    Doomstat said:
    I know this is way off base, but what if this was intentional?
    Of course this was intentional. Someone give me a plausible sequence of events which would indicate this change was unintentional.
  • Ohboy
    Ohboy Posts: 1,766 Chairperson of the Boards
    1) let's improve the ai
    2) tries to maximize mana gain
    3) priority now shifted to landfall->mana gain->everything else in whatever order, including match 5 and special gems

    All yours big guy. 

    As a programmer,  you obviously know the explanation can be as simple as a tweak in priority leading to unexpected results. 
  • babar3355
    babar3355 Posts: 1,128 Chairperson of the Boards
    Ohboy said:
    1) let's improve the ai
    2) tries to maximize mana gain
    3) priority now shifted to landfall->mana gain->everything else in whatever order, including match 5 and special gems

    All yours big guy. 

    As a programmer,  you obviously know the explanation can be as simple as a tweak in priority leading to unexpected results. 

    Sure, but why would you ever have anything higher priority than a match 5?  Lets just stick that at the top of the priority chain for the rest of time, and then shuffle the other stuff around.  Seems pretty obvious doesn't it?

    And I'm honestly asking, as I do not have a coding background.

  • DumasAG
    DumasAG Posts: 719 Critical Contributor
    Oh good, I thought I was just going crazy. So this was a real decision made by real people. Seems like a waste of time and a detriment to the game. 
  • Ohboy
    Ohboy Posts: 1,766 Chairperson of the Boards
    edited August 2017
    babar3355 said:
    Ohboy said:
    1) let's improve the ai
    2) tries to maximize mana gain
    3) priority now shifted to landfall->mana gain->everything else in whatever order, including match 5 and special gems

    All yours big guy. 

    As a programmer,  you obviously know the explanation can be as simple as a tweak in priority leading to unexpected results. 

    Sure, but why would you ever have anything higher priority than a match 5?  Lets just stick that at the top of the priority chain for the rest of time, and then shuffle the other stuff around.  Seems pretty obvious doesn't it?

    And I'm honestly asking, as I do not have a coding background.


    Theoretically you shouldn't. But it was never the top priority in this game, ever. 

    It used to roughly go

    1) landfall 
    2) match 5
    3) special gems
    4) mana maximisation

    Sometimes 3 and 4 swapped places. As far as I could tell, the game has a defence and aggro mode, so sometimes priorities gets shifted around during the different modes. The game always starts off defensive and ends up aggro. I never narrowed down the trigger that made it shift gears. 

    It just seems that the current state is the result of someone putting 4 permanently at the top. 

    If the ai was optimized, we would be hearing a lot more complaints about it. I think they deliberately crippled it before. Whether this move is a further cripple or an accident isn't clear. 
  • WiLDRAGE
    WiLDRAGE Posts: 145 Tile Toppler
    edited August 2017
    If this were me doing it, the priority system would be:
    1. Match 5
    2. Use PW ability (Koth/Chandra break Match-5s all the time)
    3. Highest Mana Match (calculating Activated gems but not Cascades)
    4. Match 4
    5. Match 3 to lower/break Opponent Support
    6. Match 3 of Opponent Highest Mana Match
    7. Match 3
    Other than making the AI better and checking if off-colour Cascades give more mana than a straight up match at Step 3, I don't see why priority should be anything else.

    Edit: Add Embalm/Eternalize/Aftermatch for and against in there somewhere.  Probably between 4 and 5.
  • Ohboy
    Ohboy Posts: 1,766 Chairperson of the Boards
    There's actually a bit of nuance to play with. 

    In defensive mode, the ai used to occasionally take a suboptimal match to totally screw your options over. I think it still does it. 
  • Laeuftbeidir
    Laeuftbeidir Posts: 1,841 Chairperson of the Boards
    tfg76 said:
    The weird thing is that gem optimization seems like the simplest thing to implement! To be honest, the simplistic AI is one of the things that most hurts this game in my opinion. Most decent players can expect a win ratio of 95-99% against random decks, and that makes the leaderboards in events more about the tertiary objectives than anything else. Let's at least not make it even worse!
    I entered the wrong event deck by accident in TotP and managed a win with Kioras Beginner Deck. Just to prove your point :D

Categories