So, after having played after around 10 games of Civ 6, my conclusion is . . .

[Orup The Great]

. . . my conclusion is: a REALLY great group of concepts-most of them top notch-but all of it, for me, spoiled by-you guessed it: a REALLY underwhellming A.I.

I basically skipped through the Civ V era (couldn't stand the game, for reasons too numerous to mention, and my feelings on the game are not at all relevant now). I regarded it as a huge failure when matched up against all the other iterations of Civ (I've played since Civ I. Civ IV was my fave.) What I really loved about Civ IV, amongst other things, was how well it had gathered together an ever-increasingly complex set of systems, like Religion, World Diplomacy, Cultural mechanics, Espionage, etc., and presented them all into an engaging package. And yet, as satisfyingly complex as I found Civ IV, I remember still wanting a game that was able to model even more aspects of the world.

After my high hopes for Civ V were dashed, I was really disillusioned and almost gave up on Civilization entirely. After reading up on Civ VI late last year I decided to give it a try. I did find some oversimplifications and simple "just for the sake of making it a game" additions distracting or just down-right irritating ("Quests" come to mind; another would be the ability to aquire territory through "buying" (so, who are we paying for this territory? God?)), but after playing VI a couple of times, I must admit: I was sold. These stupid systems didn't become any more sensical to me, but rather everything else peculiar to the game was so awesome that my gripes about the SillySystems just didn't matter. Policies: awesome concept. Revamped Tech tree: awesome. Civics tree: double-awesome! Districts: great. Bonuses: very interesting, super-detailed and promising many, many hours of being entangled within them, learning them, so . . . I'll call it: awesome. New Leaders with Agendas: great. Ranged Units: love them. The list goes on. Civ VI shows so much promise that it becomes triply sad to have to acknowledge that the one big comeuppance about this game, by all accounts, is as pervasive as it is unsolvable: completely dumb A.I. in warmaking.

One thing I'll say about EVERY iteration of Civilization up to IV: half of the fun of playing any of those Civ games was building your empire up from nothing. And the other half was the thrill of actually feeling imperiled. I can remember a IV game where I had left the northward reaches of my borders sparsely guarded a few too many turns until one turn a massive invasion of Cossacks poured into exactly the border zone I had ignored. The A.I. had been watching what I was-and was not doing, it had taken note of it, and made me pay for it. I spent the next several turns expelling them from my land and then the next several turns after that repairing all that they had destroyed. All of it, thrilling.

I have yet to feel, in any way, imperiled by the presence of any A.I. opponent. My experiences of this game matches those of many, many other players whose accounts I have read: once you reach Medieval times, if you've achieved superiority over the A.I., you've won; A.I. doesn't attack in any meaningful, tactically competent way.

I feel that the problem lies in the tactical requirements put upon A.I. as a result of 1UPT. There were similar complaints laid at the feet of Civ V, and of course, 1UPT is one of the things that V and VI have in common. And, of course, the one thing that all the other Civ iterations pre-V had in common was multiple units per square. I think that multiple units per is so, so much easier for A.I. to handle that makes it easier for them to imperil the human player, which is key to injecting a sense of thrill and interest into the game.

Many have said that the tactical shortcoming of the game are not going to change because they are basically hard-coded into the game, and are inaccessible to modders. Given this assumption, I gotta say, the whole thing is a little depressing to me. Civ VI is the first time in my more than 20 years of playing this game where I regularly am too bored to finish, by mid-game. I have finished games more often than I have just simply stopped them-but I've had to push myself, more out of a desire to see what would happen, see if things will get better (which they haven't).

This is a pathetic state of affairs to me: there are things that I really, really want to experience (like all the systems I lauded earlier) but they're joined at the hip to a combat system that makes things a complete yawner . I have tried the RealStrategy mod. Thus far I can't see a huge difference in the way the A.I. plays, nor in the amount of fun I'm having, but I'll still try it more. But, thus far, it's looking like my Civ VI experience is turning out to be a disappointing bummer (Sigh)

 

[Trigan Emperor]

I have yet to feel, in any way, imperiled by the presence of any A.I. opponent.

A better AI has been top of my Civ wishlist since at least Civ IV.

The Tough A.I. mod can make the AI imperil the human player by their presence, but it does so by increasing AI Combat Strength bonuses, rather than making them smarter.

 

[Sostratus]

We do not know the current situation but if they follow past games (and their statements during release about modding) once they do release the .dll etc then you will probably start seeing some really good AI mods.

This may include units per tile concerns- either making 1UPT AI better or creating multiple unit per tile systems for the AI. (Iirc we can do multiple units per tile via mod, the AI just doesn’t know how to handle it.)

I mean look at handheld games from 15-20 years ago, like the Advance Wars series. 1UPT but pretty good AI- can handle random and player made maps, different leader abilities, on equal footing starts, etc. I think people see 1UPT as implemented in Civ, but overly discount the methodology because of it.

 

[Trigan Emperor]

I mean look at handheld games from 15-20 years ago, like the Advance Wars series. 1UPT but pretty good AI- can handle random and player made maps, different leader abilities, on equal footing starts, etc. I think people see 1UPT as implemented in Civ, but overly discount the methodology because of it.

Advance Wars is a good reference. I only played Advance Wars: Dual Strike, but it was a really fun tactical game.

ETA: Here's a better video of an Advance Wars battle.

 

[Victoria]

This is a pathetic state of affairs to me

Been discussed to death in a zillion threads. In essence spreading units out rather than having doomstacks makes the game more popular but significantly damages the AI ability to play as does the profit line. At the end of the day they are a business not a charity and if/when the dll is released things can be tweaked but you will never get to the level of doomstacks (which I hated)

 

[fantsu]

Its just not 1upt, in civ 6 the AI feeds its units on purpose, the AI is a many steps back from even the original civilization. Its just lazy and bad programming from people who dont care.

But its still possible to enjoy this game a lot, it just takes an attitude change from the player.

 

[Tech Osen]

Its just not 1upt, in civ 6 the AI feeds its units on purpose, the AI is a many steps back from even the original civilization. Its just lazy and bad programming from people who dont care.

But its still possible to enjoy this game a lot, it just takes an attitude change from the player.

At one point you have to either accept the state of the AI for what it is: bad. Or move on to another game. Like Victoria said, it's beating a dead horse now. To me it feels like they put in some features with multiplayer in mind, features the AI has a very hard time with. No idea how popular the game is for multiplayer but my guess is it's small compared to people who just do singleplayer.
I don't expect it to change.

 

[Socrates-]

Totally agree with the compliments towards the new & improved systems in CIV6. Sad to hear you're not able to enjoy playing the game because of the AI. Since there's a small chance the AI will be improved the solution for a more even playing field might be to handicap the player.

I've seen ton of in-game challenges/handicaps pass through these forums (One City Challenge, no campus SV, Peaceful domination) maybe have a look at those for a better (more challenging) playing experience?

Aside from the challenges I think roleplaying civ's is also a good way to enjoy your games. Not pursuing the most efficient route to victory but going for a more themed approach does add some flavor to your play.

Last tip: No-rerolling starting positions or save-scumming!

 

[Cerilis]

this Thread warrants only one response

 

[Casworon]

We could move conversation towards what we can do as players to make it more challenging.

The Tough A.I. mod recommended above is one example of what player's can do.

Another is a series of mods on the steam workshop that change the way difficulty is managed in game. Rather than giving the A.I. % higher resource buffs on the higher difficulties it imposes a negative % resource modifier on the player instead. I find this actually makes the game harder than giving the AI a bonus.

It all comes doing to making it harder for the player to start snowballing. The further into the game you push the point of snowballing the harder the game is seen to be. Also spending more of the game in the period where decisions matter rather than playing autopilot once you know you've won and no one can challenge you.

The other alternative is encouraging AI snowballing so the player has a massive strong opponent to fight in the late game. However even if you use local multiplayer with only A.I opponents and change the individual difficulty's of the A.I. Diety A.I. still seem to struggle against Emperor A.I. opponents the moment walls appear and loyalty issues make them lose the cities they do capture during warfare

 

[Tech Osen]

It all comes doing to making it harder for the player to start snowballing. The further into the game you push the point of snowballing the harder the game is seen to be.

I usually refrain from early conquest and try to stick to cities I settled myself. Trying to quickly expand early on without falling too hard behind that I become easy pickings for my neighbors is a decent challenge playing on Emperor.
The other day I played Eleanor's France and while it's fun to take over 2/3 of neighboring ally's cities it once again showed how bad snowballing can get. Leading easily in science whilst focusing hard on culture and GW/GA/GMs....

 

[Civrinn]

I'm a bit confused by this post. In regards to AI firaxis is simply being cheap. For the people who parrot "but the AI can't be complex because 1UP", no you are wrong. Civ 5 VP mod exists therefore this is false.

This though:
"I basically skipped through the Civ V era (couldn't stand the game, for reasons too numerous to mention, and my feelings on the game are not at all relevant now). I regarded it as a huge failure when matched up against all the other iterations of Civ (I've played since Civ I. Civ IV was my fave.) What I really loved about Civ IV, amongst other things, was how well it had gathered together an ever-increasingly complex set of systems, like Religion, World Diplomacy, Cultural mechanics, Espionage, etc., and presented them all into an engaging package. And yet, as satisfyingly complex as I found Civ IV, I remember still wanting a game that was able to model even more aspects of the world"

All of those things were in 5. In fact did you play 5?

Regardless this is truly beating a dead horse. Civ 6 AI is bad and firaxis won't fix it. The end.

 

[SammyKhalifa]

THE AI HAS NEVER BEEN GOOD

Go back and play the rightly acclaimed Alpha Centauri and tell me that the AI is any better than it is now, with much simpler rules. You know what though? It was still fun.

Dead horse and all one of my pet peeves is this notion that somehow the games in the past had this tremendous intelligence and they just let it go. It seems like nostalgia.

 

[Gedemon]

I'm a bit confused by this post. In regards to AI firaxis is simply being cheap. For the people who parrot "but the AI can't be complex because 1UP", no you are wrong. Civ 5 VP mod exists therefore this is false.

This though:
"I basically skipped through the Civ V era (couldn't stand the game, for reasons too numerous to mention, and my feelings on the game are not at all relevant now). I regarded it as a huge failure when matched up against all the other iterations of Civ (I've played since Civ I. Civ IV was my fave.) What I really loved about Civ IV, amongst other things, was how well it had gathered together an ever-increasingly complex set of systems, like Religion, World Diplomacy, Cultural mechanics, Espionage, etc., and presented them all into an engaging package. And yet, as satisfyingly complex as I found Civ IV, I remember still wanting a game that was able to model even more aspects of the world"

All of those things were in 5. In fact did you play 5?

Regardless this is truly beating a dead horse. Civ 6 AI is bad and firaxis won't fix it. The end.

the correct sentence is "the AI seems bad because of the chosen rules for 1UPT and because developers time is neither free or infinite"

if you look at the code for example, civ5 tactical AI is (a bit) more complex than civ4, yet civ4 AI seemed to perform better.

 

[jasper]

People dont care about ai as much as they think. People want civs, mechanics and features. Dlcs dont contain improved ais; they contain the things listed above .

 

[Civrinn]

The argument isn't whether it's easier to code because 1UPT exists, the argument is "guys it's impossible to code the AI properly because 1UPT so just be happy with the AI not making aircraft or whatever the new nonsense is (I'm hoping by now that's at least resolved).

Also if their time and money are limited, don't you think a strategy game should devote it's time and money into improving the strategy portion instead of flashy graphics?

 

[Civrinn]

People dont care about ai as much as they think. People want civs, mechanics and features. Dlcs dont contain improved ais; they contain the things listed above .

Interesting I'm glad you speak for everyone.

 

[Orup The Great]

I'm a bit confused by this post. In regards to AI firaxis is simply being cheap. For the people who parrot "but the AI can't be complex because 1UP", no you are wrong. Civ 5 VP mod exists therefore this is false.

The reason why there is such a thing as Vox Populi is because VERY shortly after they released Vanilla V, they also released the dll for the game-something that they had promised to do with VI but, 4 years later, they still haven't released the dll code (this is a good article on the subject: https://www.pcgamesn.com/civilization-6/dll-source-release-modding-community) Without access to the dll, mods like VP would've been impossible.

What I really loved about Civ IV, amongst other things, was how well it had gathered together an ever-increasingly complex set of systems, like Religion, World Diplomacy, Cultural mechanics, Espionage, etc., and presented them all into an engaging package. And yet, as satisfyingly complex as I found Civ IV, I remember still wanting a game that was able to model even more aspects of the world"

All of those things were in 5. In fact did you play 5?

I played-and got irretrievably annoyed with-Vanilla V, giving it multiple chances to win me over, until it finally lost me. From what I remember, the only of these concepts that was included with Vanilla was a Cultural dynamic, and I remember being dissatisfied and unimpressed with it's implementation.

I must admit however, that in my thirst for another challenging playing experience throughout all of this, I have done research on VP and: it sounds like something I might be interested in. In particular, the way that it mitigated V's hated nation-wide happiness dynamic (and how it prevented large civs) sounds really promising. I'm thinking of giving it a whirl.

There are a few thread respondants-you, Civrinn included-that bemoan what you consider to be the existence of a thread that, according to you, is "beating a dead horse". Why then spend the energy on posting within the thread? What bothers you about somebody else opining on a matter that, according to you, no longer interests you (rhetorical question)? Save your energies for threads that you feel worthy of your energies is my suggestion. And I can continue to benefit from the opinions and information-sharing that occurs in the thread, from people that feel it worthy of their time to simply share opinions and information. And to those, I say: thank you, because I am learning more valuable information from the opinions and facts that you share.

 

[Gedemon]

The argument isn't whether it's easier to code because 1UPT exists, the argument is "guys it's impossible to code the AI properly because 1UPT so just be happy with the AI not making aircraft or whatever the new nonsense is (I'm hoping by now that's at least resolved).

That's not my argument, yet mine is also answering your "simply being cheap" point.

Also if their time and money are limited, don't you think a strategy game should devote it's time and money into improving the strategy portion instead of flashy graphics?

I agree, that's why Firaxis and 2K would never take you or me to be their next lead designer for civ7: we'll make a game that take longer to develop, and sell less.

Yeah, sure, I can imagine it would be a good strategic game, old school challenging.

But the targeted audience doesn't want good old school, it wants instant gratification, short playing sessions, shiny graphics, complex mechanisms over good AI.

And here you got me really parroting, because there is only one viable solution for everyone to be happy: release the source code, please. And earlier than 1 year ago would be nice, tomorrow or next hour is too late.

 

[Civrinn]

Check out VP. You will probably like it. For a while they were locking AI posts because it's a carbon copy of what you have said. Yes AI is bad. Hence why people are saying it's beating a dead horse.