Does anyone else agree that the AI is the highest priority?

[noto2]

I've pondered the idea of making Civ more MP friendly, but this runs into many problems. I play Dota and enjoy it, but the fact that I can only play it online gives it certain limitations. I can't play Dota if I only have 45 minutes to spare. I can't play it if I'm expecting a call, or anticipating the need to get up and do something in the near future. I can't save a game and pick it up later. Dota works because most games only last about 45 minutes, on average, or less.

I don't see how Civ could be enjoyable as a game that only lasts an hour per game. So I don't see any way to make it practical as a MP experience for most people. If everyone had to clear their schedule for 8-10 hours to play a game it would be something that most of us could barely do once per week, if that.

Given this, we are stuck with Civ being a mostly SP experience. We will be playing against the computer. To me, the AI is the single biggest reason I get bored of the game, it is the biggest negative factor to replay value. Once I learn the way the AI behaves and become familiar with it, the game becomes incredibly boring.

I'd rather play a game with a simpler design and poorer graphics but outstanding AI, than a game with wonderful features and graphics but a terrible AI. The latter case is something I might thoroughly enjoy a few times and then put down forever.

Am I alone in this? If I could just tell Firaxis the number 1 most important thing to me in a Civ game, I'd say the AI, hands down. Ramping up difficulty by giving the AI bonuses to production ruins the balance of certain strategies and ruins the balance of the game itself. I really hope Civ6 has an AI that is truly better than what Firaxis normally produces. To me, it is absolutely the highest priority for a fun game.

 

[Ryika]

It's certainly not "highest" priority for me, but it's a lot higher than it probably was for the devs when they created CiV.

In general I assume it will probably be better than the CiV-AI was, just because they now already have experience with what doesn't work in an environment with tactical combat (just having the AI loop through all its units one after the other for example). I hope they don't make the decision to ignore improving the AI, although from a business point of view it may just make sense. The majority of the Civ Community seems to be of the more casual brand and doesn't seem to care that much about higher difficulties and AI capabilities (Of course that's where Civfanatics is different from the overall Community).

Expecting an AI that doesn't need "AI bonuses" however seems very far-fetched to me. That would require the AI to be able to adapt like a human player does, which just isn't going to happen. I hope they fix the major problems that made it so the AI required extremely high bonuses, but I have no doubt that going up in difficulty will still require unique bonuses for the AI, especially in Civ where the environment is different every time and the AI can't just make up their stupidity with an insane amount of APM.

And I don't think that's a problem either, AIs aren't "opponents", AIs are puzzles to solve.

 

[Oerdin]

In Civ: Call to Power 2 the AI was terrible but this didn't really matter because the multiplayer function was great so everyone just joined in online games. Sadly, in Civ 5 the AI was terrible but the game was always terrible at MP so there was no saving it.

 

[MadDjinn]

the most important thing is features, mechanics and balance.

if the game doesn't have working features and mechanics, and isn't well balanced, then it doesn't matter if the AI is good or MP works.

 

[kaltorak]

And I don't think that's a problem either, AIs aren't "opponents", AIs are puzzles to solve.

Yup, I don't know if I would call the puzzles, but definitely don't regard them as equals. Singleplayer games are not an 8 player game, where the AI is supposed to act like a player, it's you versus 7 AI leaders playing their roles and acting with their personality

 

[stealth_nsk]

SP is more important than MP for Civilization. However, the term "Good AI" is not straightforward and could be quite subjective. Jon Shafer had pretty good article about it, but unfortunately he don't maintain his blog anymore and the hosting seems to be expired.

(Do you still have, the article anywhere, Jon? )

 

[Stilgar08]

From what I heard Ed Beach stated, AI is still one of the biggest challenges to his team, which is not very encouraging to be honest... So I won't be surprised if AI is better than in CIV but still lacks far behind expectations in CivVI.
I expect from the secret "hidden agenda" AI-leaders are supposed to have that more surprises come up during a game but that will reduce in the later ages.
We'll have to see if AI can use the diplo-system and the combat (please let this be the case).

AI-warfare was always a flaw in the civ-series and the player could always play out it's strategical and tactical thinking advantage. I doubt this is going to change a lot, but nevertheless the supporting units and corps/army mechanics could help a bit.

MP is not interesting for me when playing Civ at all. I'd rather have them refrain from the whole MP idea and instead focussing on SP-experience!

 

[ddd123]

Not at all?

If you have any knowledge of video games you should know that a good AI is totally impossible, so either you are a casual and then you shouldnt worry about AI, or you are a pro that seek challenges and then there cant be any with an AI, your only chance is playing with other humans.

Ai is pointless anyway and it was clear all civ 4/5 where to beat the hardest you just had to abuse in every possible way trades, city states and all this that i dont really know why ppl even enjoyed in the first place.

 

[dexters]

the most important thing is features, mechanics and balance.

if the game doesn't have working features and mechanics, and isn't well balanced, then it doesn't matter if the AI is good or MP works.

ding ding ding

 

[ddd123]

the most important thing is features, mechanics and balance.

if the game doesn't have working features and mechanics, and isn't well balanced, then it doesn't matter if the AI is good or MP works.

I dont agree, considering how many options there are its easy to tweak balance, like when the zulus were banned, or the no barb, no city state options and so on.

In the end MP is superior even in this cause many things can be fixed, ofc not everything, if the game is in a pathetic state there is nothing we can do about it, but even if there isnt great balance the online can play around it pretty well.

 

[darko82]

Yes, it is one of the most important features. Without good AI, there is no challenge. It also makes the game a lot more interesting. Why would I need to build walls etc. if the AI could not even attack me? Or it attacks me even though it's a lot weaker and gets wiped etc.

 

[inthesomeday]

I don't think people are really getting what noto says here, I agree completely that MP is better and more enjoyable and I play it pretty exclusively for CiV but I don't think noto is disagreeing. Rather, the point made in the OP is that it's impossible to make MP both -fully satisfactory, IE a full game with complexity and enjoyability, as well as -completable in a short period of time. So, in those times where you have only 1 hour but you wanna play some Civ, you have to play SP, and then it becomes necessary to make better AI.
I agree with most of what noto says here. The reason I only ever play MP is because the AI is so bad. If the AI was better, I'd still play mostly MP just for personal enjoyment, but it would be nice to then be able to have SP as an alternative option when necessary.

 

[Ryika]

Economic Left/Right: -10.0
Social Libertarian/Authoritarian: -8.26

It's OffTopic but as a side note before the actual post: Jesus Christ, if those are real then those are the most extreme results I've ever seen.

I don't think people are really getting what noto says here, I agree completely that MP is better and more enjoyable and I play it pretty exclusively for CiV but I don't think noto is disagreeing. Rather, the point made in the OP is that it's impossible to make MP both -fully satisfactory, IE a full game with complexity and enjoyability, as well as -completable in a short period of time. So, in those times where you have only 1 hour but you wanna play some Civ, you have to play SP, and then it becomes necessary to make better AI.
I agree with most of what noto says here. The reason I only ever play MP is because the AI is so bad. If the AI was better, I'd still play mostly MP just for personal enjoyment, but it would be nice to then be able to have SP as an alternative option when necessary.

That's looking at it entirely from a multiplayer perspective though. The AI will never* be able to "replace" or even come close to being comparable to real players, their design will never* be able to represent what a real player would do and for somebody who plays the game mostly in multiplayer the AI will never* be able to live up to what people want from them. That's just an impossible standard that the AI will never* be able to live up to.

Playing MP and playing against AIs is bound to be two completely different experiences. There's tons of obvious AI-stupidity that should be stomped out, but that won't make it a replacement for MP.

*(And by 'never' I actually don't mean 'never' but rather 'not within the foreseeable future' )

 

[anandus]

SP is more important than MP for Civilization.

But is it? I think it depends. Many people have played a lot more multiplayer Civ than singleplayer.
In my 2000 hours of Civ 5 I might've played 10% singleplayer?

Although I must admit in general Civ 5 was more played in singleplayer than multiplayer, but that's also in large part because multiplayer wasn't stable for a lot of people.

That said, single- and multiplayer in a discussion about AI is actually besides the point, as you can also have AI-players in a multiplayer game.

For instance, I often play a 10-player game with 4 friends and 5 AI.

 

[stealth_nsk]

The funny thing is - there's no definition of "good AI".

From gameplay perspective for a pure single-player games, AI is just a way to present game challenges for player, like puzzle or minigame. It doesn't matter how the challenge is presented.

For immersion, however, the AI needs to imitate some character so player can personalize the opponent. But the immersion is purely subjective thing. There are some more or less common immersion breakers, like stupid tactical mistakes, but it's not enough to define what's objectively the good AI is.

The more hell appears once game allows both SP and MP. In this case AI needs to follow a lot of rules designed for human players. At this point we have another set of immersion breakers. For example, if AI has better research and production, that's looked like acceptable bonus by most players; if AI units are stronger than human ones, many players think it's cheating; if AI units appear from nowhere, that's considered plain cheating by nearly everyone.

On top of this, there's another level of madness - while players expect AI to play by same rules as human players and don't accept clear AI mistakes, they expect AI to "roleplay" - the thing which human not do and is often a clear mistake.

Once you look at all this, it becomes clear what good AI is just impossible, due to level of subjectiveness in this term. There are some guidelines to make AI acceptable by majority of players, but that's titanic work. When Ed says AI is the biggest challenge for them, I'm optimistic as this means they understand what they are working on

EDIT:

But is it? I think it depends. Many people have played a lot more multiplayer Civ than singleplayer.
In my 2000 hours of Civ 5 I might've played 10% singleplayer?

I don't think you're majority.

 

[dexters]

Good AI for me is just an AI that doesn't cheat*; can use the 'systems' in the game and is constrained by said systems, and an AI that reads the Civ map the way a human player can.

*The word Cheat is subjective as the AI obviously gets massive bonuses on harder difficulties in every Civ game in recent memory. This may include production bonuses or bending the rules of some game systems by getting higher or lower tresholds. That kind of 'cheating' is OK. Blatant cheating like summoning units out of the thin air, to defend a city is not.

The reading game map part is a huge deal for me. I think we've long past the point of the AI mindlessly rolling an RNG each turn to decide who to attack next. Irregardless or how good or bad the game systems are, an AI that is stuck to doing that is just a lazy AI.

The AI needs to see and understand its own core interests. And see things like the slow erosion of his or her city-state influence, or the creeping borders of another Civ and react to it. AI needs to not only be looking at the score table to see who is their next threat, but understand the relative growth in threat for each. Is #1 still as big a threat if what was formerly #6 is now #3 and quickly closing the gap? Maybe ask #1 to attack #3 instead of going ahead with plans to attack #1. AI needs to remember (not just add a + or - modifier) when bad things are done to them and this can be directly bad things (losing cities/tiles) or bad things done to them geopolitically.

The problem with +/- modifiers without any other AI systems behind it is what I'd like to call laundering reputation. Like money laundering by putting bad money into a casino or bank then withdrawing it as 'good currency' ; A reputation system based solely on +/- modifiers as seen in Civ3/4 means players can do some egregious things and if the AI can't understand what has happened to it and merely applies the max penalty the game tells it it can apply, that modifier becomes just as good/bad as any other modifier in the mix. So you get people gaming the system by calculating that they can take the temporary hit to reputation then recoup some of it back by a free tech gift, or some bags of gold later on. Even though they just ravaged the AI by stealing their best lands, countered all their national interests, and left them with a rump state.

I have faith Ed and his team will do the right thing with Civ6, but to cycle back to my first post, good AI or MP is meaningless if the systems aren't balanced or features aren't there. So I vote for a good core game first. Worry about the AI later.

 

[Nixalo]

The AI is important but it doesn't have to be "good". It just needs to understand the game. It will never out tactics a human. But it could at least understand how to build a city, use an archer, or who their allies should be.

The features, mechanics, balance, and options are more important.

 

[stealth_nsk]

The AI is important but it doesn't have to be "good". It just needs to understand the game. It will never out tactics a human. But it could at least understand how to build a city, use an archer, or who their allies should be

That's the point. AI can't understand anything, it doesn't have a cognition.

 

[dexters]

That's the point. AI can't understand anything, it doesn't have a cognition.

'understand' in the context of AI in the game is purely a semantic issue. If they can react properly to a situation, it is understanding.

How the programmer gets the AI to do that is a detail.

 

[stealth_nsk]

'understand' in the context of AI in the game is purely a semantic issue. If they can react properly to a situation, it is understanding.

But the "properly" is subjective