1. We have added a Gift Upgrades feature that allows you to gift an account upgrade to another member, just in time for the holiday season. You can see the gift option when going to the Account Upgrades screen, or on any user profile screen.
    Dismiss Notice

[GS] How much are you willing to pay for a “great” AI?

Discussion in 'Civ6 - General Discussions' started by Engeez, Feb 5, 2019.

?

How much would you be willing to pay for a “great” AI

  1. Nothing, the AI is already great

    22 vote(s)
    17.1%
  2. Under $50

    54 vote(s)
    41.9%
  3. $50-$100

    24 vote(s)
    18.6%
  4. Over $100

    29 vote(s)
    22.5%
  1. bbbt

    bbbt Emperor

    Joined:
    Oct 21, 2013
    Messages:
    1,965
    Yes, I'm aware of different ways AI works generally. I'm saying I don't think people are asking for a 'perfect' AI for the most part - that's sort of a strawman in these threads (yes, there's always someone who pops up with something ridiculous like "they should just make the AI to be better on deity instead of giving it bonuses" but that's not the majority). I think most just an AI that can be somewhat competent while actually roleplaying - ie.repairs it's damaged walls instead of leaving them unrepaired for 2 centuries after an attack, shoots units attacking it's cities instead of just sitting, takes cities instead of attacking them until halfway and then sailing off, actually prioritizes the steps of the Science Victory when it hits the correct technological era first instead of just researching unrelated techs, etc.
     
  2. Duuk

    Duuk Doom-Sayer Supporter

    Joined:
    Dec 16, 2005
    Messages:
    1,912
    Gender:
    Male
    Location:
    Detroit, Michigan, USA
    See, that's perfectly reasonable. I wish there was a way to separate the "OMG FIX THE AI" threads into "Can you fix these specific AI issues, which I feel are actually broken?" discussions and automatically filter out the spiral those discussions always turn into. Because I completely agree the AI does dumb stuff (I remember going back and forth with @Victoria ((I think)) where we were discussing why the AI seemed to get distracted and not take cities and we were ALMOST THIS CLOSE to being able to replicate the behaviour with about 95% certainty.)

    There is a difference between "The AI should be smarter!" and "why does the AI start a science victory path and then ignore the next steps?, it should either not start it or, once started, make it a priority." You're not wrong.
     
  3. AriochIV

    AriochIV Analyst

    Joined:
    Jul 25, 2006
    Messages:
    5,618
    Location:
    San Jose, California
    Nothing. I shouldn't have to pay extra for AI that can play the game.
     
  4. TheMeInTeam

    TheMeInTeam Top Logic

    Joined:
    Jan 26, 2008
    Messages:
    25,318
    "Good" or "great" AI means different things to different people.

    We need a good design framework first or the question posited by the OP is untenable.

    In Civ 6 this is an oxymoron. It's impossible.

    It isn't reasonable. The request for an AI that can beat top level human players w/o bonuses is more reasonable and self-consistent, and that's not exactly a reasonable request in its own right. But at least it isn't instantly self-contradictory.
     
    Last edited: Feb 5, 2019
    AmazonQueen and ChocolateShake like this.
  5. ChocolateShake

    ChocolateShake Prince

    Joined:
    Dec 28, 2016
    Messages:
    525
    Gender:
    Male
    Yup, to me, that would be far more interesting and fun.

    I've formally studied AI and computer science, and in the end building a great AI in my opinion comes down to giving it a great set of algorithms to use. I think discussing algorithms would be helpful and constructive, and would be beneficial to modders too. Whether those are feasible to implement is a different story.

    And I agree with your earlier statement about Stellaris. There isn't a single 4x game out there that doesn't have a faction of fans complaining about the AI. I wanted to buy Endless Space 2 for a while but never did because of the repeated complaints about the AI. I did finally pick it up but I haven't played it yet. I've played GalCiv 3 and there the AI makes bad decisions too, but I still had fun with it. I plan on playing Stellaris too but there again I know I'll see the AI do dumb stuff.

    It's prime time for a company to decide to make a name for itself on a true hardcore AI, but I don't think any big company would be jumping at that chance. There just isn't enough incentive to really invest in an AI in single player games. It'll require a lot of resources, code optimization, etc. for something that will only please what I'd say would be a minority and wouldn't provide a good return on investment. I think it would only make sense for indie developers at the moment. The others will be content to add new features, and most players are happy with that as long as any glaring weaknesses are fixed.
     
    Trav'ling Canuck likes this.
  6. TheMeInTeam

    TheMeInTeam Top Logic

    Joined:
    Jan 26, 2008
    Messages:
    25,318
    I suspect more of how good or bad an AI is comes down to design choices interacting with it rather than just raw AI skill. Similar to Firaxis, Pdox gives its AI no chance for example. Even if their AI devs were top-tier amazing the game's design intentionally hamstrings it and further exacerbates that by changing the game rules often.

    So too for Civ 6 and its predecessors. At least we don't have peace vassals anymore.
     
    jdevo and Duuk like this.
  7. Duuk

    Duuk Doom-Sayer Supporter

    Joined:
    Dec 16, 2005
    Messages:
    1,912
    Gender:
    Male
    Location:
    Detroit, Michigan, USA
    You just hit the nail on the head. "Hardcore AI mode" is the 2020's version of what we used to call in my misspent youth "grognard games". The old "Empires in Arms", 10,000 piece, takes 4 hours just to set up the game, takes a month of weekends to play, the rulebook is 200 pages long, nobody has ever really played the game, games.

    And the problem is there are like 10 people that would actually care enough to pay extra for it. All of those companies that made those games went under. All of them. Hasbro owns like 2/3 of them these days, and all of the games are out of print. You can still get them on eBay of course, and shockingly most of them are close to original list price because NOBODY WANTS THEM.

    It's kind of like how people used to howl and whine on the WoW forums that twinks weren't really about beating up on lesser, ungeared players in PvP battles, so Blizzard gave them their own battlegrounds to fight each other... and none of them queued for it.

    The loud minority of people (say... the Fanatics) howl that they want the Deep Blue AI that can beat a player as competently as another player can... but 20,000 people are playing Civ6 at any given time, and only a few hundred are posting on civfanatics or in the steam forums or on reddit. Less than half of civ6 owners have EVER won a victory at ANY difficulty level. Don't quote me but I believe it was something like 3% had won on Deity, and I'm willing to bet that a good chunk of those (mine included) were turn 1 Rome score victories to get the achievement (Yup. Cheevo whoring that day. Guilty.) So... in reality... I feel fairly confident when I say that for most people the AI is "good enough".
     
    ChocolateShake likes this.
  8. Futumch

    Futumch Calm as a Coma

    Joined:
    Dec 27, 2001
    Messages:
    416
    Location:
    142 E 42 S
    Would you pay for a manual transmission car that only has the bottom two gears working? Hell no, I would want all the gears working properly, and I would not be paying extra for something that should have been good right from the get-go.
     
  9. Onin

    Onin Chieftain

    Joined:
    Jan 22, 2019
    Messages:
    50
    Balancing a good AI is hard and making the game still enjoyable at the same time is even harder. At the end of the day the game is less about the AI and more about the player using their resources and game mechanics wisely enough to gain an advantage over an AI that will progress at a moderate rate that generally varies based on difficulty levels. That is about it and yes the player expects to win. The game designers want to make sure that if a player wants Petra they have to focus on that decision and get it quick or they won't get it in time. So an AI that builds it in a stupid spot really isn't a major problem.

    The purpose is less about the AI's efficiency but the player's. I think for the most part this is fine with the exception of war. War is still the one area where the AI really struggles and the player can often leverage this weakness relatively easily. Other than that I feel like the game works pretty well. I think there are ways to play that are more or less challenging with regards to how you will beat the AI and if people want a decent challenge they can find it. The majority of players will also find a level of difficulty that works for them.
     
  10. Archon_Wing

    Archon_Wing Vote for me or die

    Joined:
    Apr 3, 2005
    Messages:
    4,669
    Gender:
    Male
    $0

    Modders will do the job if Firaxis can't. It hss happened before and most likely will again.

    I mean we already have users that can make the ai use planes..... Why should I pay to have it do what it should be doing in the first place?
     
    679x and Duuk like this.
  11. AmazonQueen

    AmazonQueen Virago

    Joined:
    Nov 25, 2007
    Messages:
    4,307
    Location:
    Climbing Kero Fin
    I still have some of those games. A small group of people bought them. We maybe played them once or twice(probably solo because getting opponents for them was hell, if you actually got an opponent what were the chances he'd stick around for a whole game). And yes if you look at the boardgames being produced today they are better games whilst being just as challanging as the old-style SPI,AH or GDW games. They just lack the flavour of a game that had a rule for every eventuality, however unlikely.
    So how does that actually affect modern computer games? We have a range of games being made. Total War, Civilisation or Paradox to a certain extent its what you are interested in.
     
    ChocolateShake likes this.
  12. TheMeInTeam

    TheMeInTeam Top Logic

    Joined:
    Jan 26, 2008
    Messages:
    25,318
    If this were strictly or even predominantly an AI issue, we should anticipate that PvP has roughly equal or at least similar win rates per each victory condition.

    Is that the result we observe in practice in PvP games?
     
  13. Onin

    Onin Chieftain

    Joined:
    Jan 22, 2019
    Messages:
    50
    I don't see why you would think that would be true and taking advantage of AI at war applies to all victory types, not just domination.
     
  14. 679x

    679x Warlord

    Joined:
    Mar 30, 2017
    Messages:
    219
    Gender:
    Male
    I'm not a great Civ player but I do value having a 'good' AI. By 'good', I mean an AI that is able to understand and take into account the game's mechanics - such as loyalty when forward-settling, or effectively laying siege and conquering a city - and an AI that gets a little smarter/better at higher difficulties, rather than merely having silly head starts but still being just as dumb.

    However, I do want to see the AI making proper tactical decisions, and multiplayer will never satisfy that for me because A: I want to take my time on turns and sometimes take a bit to mull over which religious beliefs I want or where I want to put my districts or whatever, and B: I don't like sitting at the computer stuck in a game of Civ for hours straight, which is what would happen in multiplayer unless you coordinate with friends, and none of my IRL friends play civ.

    So, ultimately, I don't really want to have to pay for that if modders can do it eventually anyway once they have access to the proper tools, but I also kind of wish that they had streamlined the AI a little bit right when the game came out.
     
  15. Wizard-Bob

    Wizard-Bob Always remember to Find Your FUN!

    Joined:
    Sep 7, 2010
    Messages:
    170
    Gender:
    Male
    Location:
    Florida
    Arioch, I *really* love your site and all you contribute here. Thanks so much for that!!!

    This comment strikes me as wrong - only because, to me, the AI CAN INDEED play the game. Maybe not as well as others would like, I'll grant you that. But at my low level of play, man those barbs can pose a challenge! And barbs are AI, correct?
     
  16. pgm123

    pgm123 Emperor

    Joined:
    May 21, 2017
    Messages:
    1,168
    Gender:
    Male
    Isn't PvP "player vs. player"?
     
  17. TheMeInTeam

    TheMeInTeam Top Logic

    Joined:
    Jan 26, 2008
    Messages:
    25,318
    AI also doesn't attack to prevent victory. War is a central part of the game's actual incentives, regardless of victory condition. Any victory condition would see an attempt at intercepting it if opponents are trying, which necessarily means war in Civ 6.

    Also, this is a matter of degree. The AI is broadly terrible at peaceful empire management, too. Its bonuses on high difficultly levels offset more of its deficiencies, but it's still pretty bad.

    Sort of. They use AI, but are not "the AI" in the sense of an AI taking over a direct competitor for a victory condition. As such they are more so a mechanic with an AI script, rather than an AI script managing a competitor in game terms.
     
  18. AriochIV

    AriochIV Analyst

    Joined:
    Jul 25, 2006
    Messages:
    5,618
    Location:
    San Jose, California
    What I'm referring to here is the several important features (such as air units) which the AI is completely incapable of using. I also find the game interesting and challenging (at the appropriate difficulty level), and I understand from experience that creating challenging game AI is difficult, but the concept of being asked to pay extra for something that should be part of the base game offends me.

    It's a moot point, because I've never heard of a AAA game developer offering a paid module that only improves the AI. But I think there's a reason they don't do that -- and it's probably because many customers (like myself) would find it absolutely outrageous. It would be tantamount to asking for extra money to fix bugs.
     
    Trav'ling Canuck likes this.
  19. pgm123

    pgm123 Emperor

    Joined:
    May 21, 2017
    Messages:
    1,168
    Gender:
    Male
    Is the AI really that bad at empire management? In my experience, they put down districts, build wonders, collect great people, spread their religion, etc. It's not perfect, but it's functional. The main problem, to me, seems to be that the AI gets ADD when going for the space victory and rarely wins it. Part of me wonders if this is a catch-up mechanic for worse players to make sure the AI doesn't really beat them. It's similar to how the AI looks to make peace after conquering a city.
     
  20. TheMeInTeam

    TheMeInTeam Top Logic

    Joined:
    Jan 26, 2008
    Messages:
    25,318
    The difference between someone like Civtrader winning peaceful deity games and AI even on deity is enormous by turn count. > 100 turns, despite the large bonus differential in AI's favor.

    You can make a case that this is nevertheless a smaller deficiency than its war capability, but it's pretty darned bad. Players half as good as Civtrader outperform it trivially.
     

Share This Page