Why did the AI need a nerf?

245

Comments

  • shteev
    shteev Posts: 2,031 Chairperson of the Boards
    Ohboy said:

    As a programmer,  you obviously know the explanation can be as simple as a tweak in priority leading to unexpected results. 
    I don't buy it. Hibernum are swamped with work as it is. They're having to take on new staff just to cope with booster crafting. Why would they tinker with code which has the opportunity to lead to unexpected results? As a non-programmer, I'm sure you can imagine all kinds of outlandish results that might happen because of that; graphical glitches, terminal crashes, bees flying out of your phone, dogs and cats living together....

    If you change code, you at least test to see if it does what you think it does. You don't just change a bunch of lines of code and send it out into the world hoping that it probably works.

    I mean, I guess Hibernum *might* be incompetent enough to do that. But why take the risk? If you haven't got time to test the new code, why not just leave it as it is?
  • WiLDRAGE
    WiLDRAGE Posts: 145 Tile Toppler
    shteev said:
    Ohboy said:

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

    If you change code, you at least test to see if it does what you think it does. You don't just change a bunch of lines of code and send it out into the world hoping that it probably works.

    I mean, I guess Hibernum *might* be incompetent enough to do that. But why take the risk? If you haven't got time to test the new code, why not just leave it as it is?
    Who has time for Regression Testing anyway?!
  • madwren
    madwren Posts: 2,259 Chairperson of the Boards
    Steeme said:

    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.


    I agree completely. It feels unnatural.

    It'd be interesting if they were experimenting with different AI levels for different tiers of play.
  • Ohboy
    Ohboy Posts: 1,766 Chairperson of the Boards
    shteev said:
    Ohboy said:

    As a programmer,  you obviously know the explanation can be as simple as a tweak in priority leading to unexpected results. 
    I don't buy it. Hibernum are swamped with work as it is. They're having to take on new staff just to cope with booster crafting. Why would they tinker with code which has the opportunity to lead to unexpected results? As a non-programmer, I'm sure you can imagine all kinds of outlandish results that might happen because of that; graphical glitches, terminal crashes, bees flying out of your phone, dogs and cats living together....

    If you change code, you at least test to see if it does what you think it does. You don't just change a bunch of lines of code and send it out into the world hoping that it probably works.

    I mean, I guess Hibernum *might* be incompetent enough to do that. But why take the risk? If you haven't got time to test the new code, why not just leave it as it is?

    You spent months trash talking about how they always do exactly this, and now you suddenly don't buy it? 
  • shteev
    shteev Posts: 2,031 Chairperson of the Boards
    Ohboy said:
    shteev said:
    Ohboy said:

    As a programmer,  you obviously know the explanation can be as simple as a tweak in priority leading to unexpected results. 
    I don't buy it. Hibernum are swamped with work as it is. They're having to take on new staff just to cope with booster crafting. Why would they tinker with code which has the opportunity to lead to unexpected results? As a non-programmer, I'm sure you can imagine all kinds of outlandish results that might happen because of that; graphical glitches, terminal crashes, bees flying out of your phone, dogs and cats living together....

    If you change code, you at least test to see if it does what you think it does. You don't just change a bunch of lines of code and send it out into the world hoping that it probably works.

    I mean, I guess Hibernum *might* be incompetent enough to do that. But why take the risk? If you haven't got time to test the new code, why not just leave it as it is?

    You spent months trash talking about how they always do exactly this, and now you suddenly don't buy it? 
    Citation, as ever, needed.
  • Ohboy
    Ohboy Posts: 1,766 Chairperson of the Boards
    Before I go to the trouble of digging. Please state for the record that you do not recall constantly complaining about the incompetence of the coding. 


  • shteev
    shteev Posts: 2,031 Chairperson of the Boards
    Ohboy said:
    Before I go to the trouble of digging. Please state for the record that you do not recall constantly complaining about the incompetence of the coding.

    I hereby state for the record that any earlier statement Ohboy pulls out of the forums will be something I stand by, AND ALSO will be able to fully justify in the context of this thread.

    Go!
  • Steeme
    Steeme Posts: 784 Critical Contributor

    I know that the majority of players love to jump on the Dev bashing train, but in reality software companies are usually dealing with a lot of legacy code written decades ago.

    We do the same thing over here.  A big portion of our base application was written years ago by monkeys.  Now we have some really competent people working on it, but stuff you didn't even know existed will break any time you try to modify it.

    I liken it to playing "Jenga".

  • shteev
    shteev Posts: 2,031 Chairperson of the Boards
    Steeme said:

    I know that the majority of players love to jump on the Dev bashing train, but in reality software companies are usually dealing with a lot of legacy code written decades ago.

    We do the same thing over here.  A big portion of our base application was written years ago by monkeys.  Now we have some really competent people working on it, but stuff you didn't even know existed will break any time you try to modify it.

    I liken it to playing "Jenga".

    And that, surely, is exactly why you don't go messing with legacy code unless it is absolutely necessary.

    So, IMO, that kinda rules out the idea that the AI for making moves was altered as a result of altering the code for the AI for making moves, as non-programmer Ohboy seems to think.

    So back to my original question: Someone give me a plausible sequence of events which would indicate this change was unintentional.
  • DumasAG
    DumasAG Posts: 719 Critical Contributor
    shteev said:
    Ohboy said:
    Before I go to the trouble of digging. Please state for the record that you do not recall constantly complaining about the incompetence of the coding.

    I hereby state for the record that any earlier statement Ohboy pulls out of the forums will be something I stand by, AND ALSO will be able to fully justify in the context of this thread.

    Go!
    That is some next-level politician stuff there
  • shteev
    shteev Posts: 2,031 Chairperson of the Boards
    DumasAG said:
    shteev said:
    Ohboy said:
    Before I go to the trouble of digging. Please state for the record that you do not recall constantly complaining about the incompetence of the coding.

    I hereby state for the record that any earlier statement Ohboy pulls out of the forums will be something I stand by, AND ALSO will be able to fully justify in the context of this thread.

    Go!
    That is some next-level politician stuff there
    I'd rather be Obama than Trump :)
  • babar3355
    babar3355 Posts: 1,128 Chairperson of the Boards

    I don't know what the optimal win rate is to keep people hooked on a game.. but I do feel like it is something less than 95%+. 

    I know in Magic Duels a win rate of around 70% is very strong, although that certainly goes up against the AI.  I also know that I am very glad the AI acts quickly in PQ rather than the slow motion moves in Duels.

    But anyway, maybe they are worried about making the AI too hard, so they don't prioritize match 5s?  Or are they so dense that they think landfalls are better?

  • Ohboy
    Ohboy Posts: 1,766 Chairperson of the Boards
    Didn't think this was going to be the hill you chose to die on, but OK. 

    Here's one week ago where you expressed a lack of surprise that the coding would be buggy 

    https://forums.d3go.com/discussion/comment/705686#Comment_705686

    Two weeks ago, same sentiment

    https://forums.d3go.com/discussion/comment/705686#Comment_705686

    Last month

    https://forums.d3go.com/discussion/comment/691852#Comment_691852

    April

    https://forums.d3go.com/discussion/comment/672985#Comment_672985





  • Steeme
    Steeme Posts: 784 Critical Contributor
    babar3355 said:

    I don't know what the optimal win rate is to keep people hooked on a game.. but I do feel like it is something less than 95%+. 

    I know in Magic Duels a win rate of around 70% is very strong, although that certainly goes up against the AI.  I also know that I am very glad the AI acts quickly in PQ rather than the slow motion moves in Duels.

    But anyway, maybe they are worried about making the AI too hard, so they don't prioritize match 5s?  Or are they so dense that they think landfalls are better?


    Well, given that JC states with absolute certainty that the AI does not have an advantage over the player in net mana gain, that means they are using some type of in-house simulation to gauge the rate at which the AI progresses in a match.

    Perhaps setting the 5-swap as a top priority lead to an imbalance in net mana gain in favour of the AI.  I mean, we still get periodic "cascade" threads.

    The solution was to set something else as the priority.  But that didn't work out as planned.

    So now they re-jigged the priorities.

    Jenga Jenga, J-J-J-Jenga

  • Corn_Noodles
    Corn_Noodles Posts: 477 Mover and Shaker
    shteev said:
    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.

    Left hand. Right hand. Ignorance.
  • shteev
    shteev Posts: 2,031 Chairperson of the Boards
    edited August 2017
    Corn Noodles said:
    shteev said:
    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.

    Left hand. Right hand. Ignorance.
    As a matter of interest, does everyone think that the changes to the free boosters and the event decks in the latest update happened accidentally as well?

    Have a look at the poll here on the forums for the new free boosters. They are universally reviled. Do you think they were trying to make a popular change, and accidentally made an unpopular one?
  • shteev
    shteev Posts: 2,031 Chairperson of the Boards
    And what about the new event objectives? The event objectives on RotGP are pretty easy, aren't they? Does that not suggest a push toward making the game easier?
  • madwren
    madwren Posts: 2,259 Chairperson of the Boards
    edited August 2017
    What I don't understand is why they would make the game easier across the board. That leads to more ties, which leads to more prize distribution, which doesn't seem their style.

    While I lean towards the thought that it was purposeful, it's just an odd strategy.
  • Gunmix25
    Gunmix25 Posts: 1,442 Chairperson of the Boards
    shteev said:
    Ohboy said:

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


    If you change code, you at least test to see if it does what you think it does. You don't just change a bunch of lines of code and send it out into the world hoping that it probably works.

    I mean, I guess Hibernum *might* be incompetent enough to do that. But why take the risk? If you haven't got time to test the new code, why not just leave it as it is?
    Do you play on console or pc games? Because Beta gameplay is exactly that. Them punching in codes and sending it out into the world (of beta testing) and see what breaks or doesn't break. I am sure that you are aware that the reasoning behind this that despite what developers can do to "test" a code... they know full well that there just isn't enough time in day to test for any and all possible glitches, bugs, interactions and what-have-yous. But with a multitude of players in beta, they can speed up the process. I've seen forums go tinykitty over beta results and glitches and scream bloody murder for whatever burr that snuck up their tinykitty and demand changes be made to X or Y. Anyways, I think that D3 should create a large random pool of players to run a dry invite only event that removes the cards after use to find issues. But that in itself is also a huge endeavor... one I sadly do not think D3 has the manpower to pull off at this time. 'Cause ya know... the not another penny movement and all.
  • hawkyh1
    hawkyh1 Posts: 780 Critical Contributor
    Steeme said:
    babar3355 said:

    I don't know what the optimal win rate is to keep people hooked on a game.. but I do feel like it is something less than 95%+. 

    I know in Magic Duels a win rate of around 70% is very strong, although that certainly goes up against the AI.  I also know that I am very glad the AI acts quickly in PQ rather than the slow motion moves in Duels.

    But anyway, maybe they are worried about making the AI too hard, so they don't prioritize match 5s?  Or are they so dense that they think landfalls are better?


    Well, given that JC states with absolute certainty that the AI does not have an advantage over the player in net mana gain, that means they are using some type of in-house simulation to gauge the rate at which the AI progresses in a match.

    Perhaps setting the 5-swap as a top priority lead to an imbalance in net mana gain in favour of the AI.  I mean, we still get periodic "cascade" threads.

    The solution was to set something else as the priority.  But that didn't work out as planned.

    So now they re-jigged the priorities.

    Jenga Jenga, J-J-J-Jenga

    I don't get how allowing the ai to always take match 5's
    would put it at an advantage against a player. I always
    take match 5's whenever I see them. if the cascades
    are a problem then it's the cascades that should be
    fixed?

    HH

Categories