Is AI really brain dead?

[Socratatus]

Damn. AI is important to me. they ALWAYS do this, have you noticed. In every CIv game of recent years the AI is rubbish and you have to wait 6 months before it starts being reasonable.

I am certain Firaxis has a policy with Civ games to make the AI baby-easy on release, as in next to dead, then they `improve ` it later.

 

[Socratatus]

False. In Civ 5, the AI managed to take a city from me inmy very first game. In Civ 6, I don't think they ever could.

I`ve had the AI take a city off me in Civ 5 several times- And I`m not a very good player. If the AI can`t even do this in Civ 6 then I think I will hold off. Some people like easy wins- I don`t, unless it`s easy because I did something right!

 

[Tempestfury]

I don't think it's all that bad. I mean, in my first game. Prince Difficulty. 2 of the AI were killed by one of the other AI.

 

[Kailan]

It's atrocious, just seen England with 2 cities and me with my 9 attack me with about 4 swordsman and a chariot when i have muskets and cannon, they are completely outnumbered and outgunned. This is about the 4th time they have done something like this. 1742 and i've just got tanks, 40 discoveries while 4 other civi's are 15 techs behind and this is the 3rd hardest level. The game is just so easy and feels a lot worse version of Civ 5.

 

[The_Reckoning]

It's utter garbage. I wish they'd design the game concepts with the idea firmly in mind that it should be something which is easy for an AI to cope with.

I don't think we ought to have to wait for a Kurzweilian humanity-enslaving articifical superintelligence for a good computer opponent in our games.

Just look at the staff roles at Firaxis though - 10+ people on the art team and 1 guy whose role 'includes' AI. What a joke.

 

[Sobornost]

I had read an article prelaunch that was as a whole favorable to the game, to everything that it was bringing to the table, but left as the biggest main caveat, the lament that it does not have an AI that's really worthy of it. It was some offbeat magazine though, not one of the big gaming publications of course...

I wish there was a way that we could really give >1 UPT a chance, see how the game worked with and without it, and had a chance to weigh what we liked best.

I think a big impact is just a fresh game too, the AI is bound to get better from here. A 5/10 AI might not be great but it's better than a 3/10.

 

[ThERat]

I had read an article prelaunch that was as a whole favorable to the game, to everything that it was bringing to the table, but left as the biggest main caveat, the lament that it does not have an AI that's really worthy of it. It was some offbeat magazine though, not one of the big gaming publications of course...

Well, I think the German gamestar had the same conclusion. Maybe Firaxis forgot to tell them to stand in line.

 

[jcoto]

To think I almost broke my promise of never buying a firaxis game after what they did with Beyond Earth.


Sounds like I won't buy.

Ever.

Thanks for your honest reviews.

 

[Sobornost]

It's utter garbage. I wish they'd design the game concepts with the idea firmly in mind that it should be something which is easy for an AI to cope with.

This, definitely. And I think the players themselves deserve blame here not just Firaxis. I think the playerbase as a whole dogs Firaxis to have this and that cool feature, civs, graphics issues, etc. that those things get made the priority, because they're what we playerbase buys the game over, when they're just building something the AI is never going to be able to figure out how to compete in half-decently, or investing resources in those things rather than in developing the competitive side of the game. So for a quick sexy two-minute rush of cool we demand the developer sacrifice the core game experience.

At risk of beating a dead horse, I remember reading a lot of talk of 1UPT, main points being that it felt cooler, funner, looked better, etc. And sure, that's very true. Exciting 'tactical combat' - as if you could call tactical the equivalent of playing chess with your opponent not knowing how to move any of his backline pieces. In a better world, with a supercomputer, we could make it work. But in the real world, with budget constraints for the devs and processor constraints for the players, why is it a sin to cut our losses and accept some combat mechanic that the AI can handle better (not perfectly) at the expense of some of the sexiness? AI can be taught to put together a balanced stack, and make semidecent decisions regarding when to advance it, when to retreat it, etc. It is much harder to force him to split out those pieces and move them all optimally, concentrate force on the right points, etc. There's twenty variables rather than just 2 or 4. Things the human does naturally. AI is much better with vertical than with horizontal combat (can kind of see this with ranged units in Civ 5 at least, the AI is semidecent finally at getting them to focus-fire on one target)

This goes for other mechanics too, and no civ game has been completely immune to the problem. But it certainly needs to be minimized as much as possible - the less we need to kick up the artificial difficulty hurdles to make it a challenge, the better.

 

[footslogger]

When Civ 6 was announced I loved all the new features and was only surprised to see that 1upt was being kept in the game. Now that the game has been released I am confirmed in my belief that it was a mistake to keep it in. The AI simply can't cope with it. The first time I was attacked by 3 rival civs I was briefly worried until I saw how feeble, inept and cowardly the attacking forces were. They were easily repelled by my 2 archers and 2 warriors. Now I disregard the AIs completely as they represent no threat at all, even with numerically vastly superior forces. How can you have an enjoyable, immersive strategy game when that is the case?

 

[Sascha77]

Started a new game as Japan and just discovered Cleo - who was apparently throwing a scout-orgy in one of her towns. WTH?

S.

 

[ahtiandr]

AI declared a surprised war on me and then moved his troops away from one of his cities which was closer to my border even though he has pretty good army.

 

[illusiondrmr]

Well.. So unfortunate my first post has to be a disappointment..
A game which is so much anticipated, you have the impression that its been released without been tested at all.

Im curious.. Those devs havent noticed that AI is brain dead? Not just bad, its just BRAIN DEAD for all the reasons the above other fellows mentioned.
No resistance at all in first 2 games @ emperor and immortal.. Im pretty sure deity wont change a thing..

Yes the GFX is awesome. Yes the music is awesome. But the overall outcome its just a big circled ZERO coz of this bad AI.
Therefore the gameplay is a disappointment which should be mostly taken care of, instead of everything else..

 

[Wellfooled]

Im curious.. Those devs havent noticed that AI is brain dead?

I can promise you that the Devs are acutely aware of the AI's flaws and I'd also wager a guess that they're more irritated by it than you. When you make something, you want it to be perfect. There is however only so much time in a day and there was a deadline to meet. Video game crunch is terrible. Realistically every project will need to make some kind of sacrifice and sometimes there isn't time to play test and tweak things to perfection before release. The nature of games these days allow the Devs to keep working on them after release. I'm sure we can expect patches to many of these problems soon.

 

[Frank327]

This might be cynical but it would be nice if Firaxis would open up the code for AI and sets out a contest for creating/modifying to build the best Civ AI. Let the community go to work on it asap and maybe in a year from now we have really cool AI to play around with. They did this sort of thing with open tranport tycoon and it made for some cool AI opponents that had different playstyles and everything.

For future civ games, they should make a decision early. Can we make an AI that is competent with 1UPT? If no, return to some kind of stacking system. Tactical combat isn't all that important to civ games (empire building is the core of the game, not the chess part) so there's no reason to let 1UPT destroy your AI.

 

[illusiondrmr]

Tactical combat isn't all that important to civ games (empire building is the core of the game, not the chess part) so there's no reason to let 1UPT destroy your AI.

Allow me to disagree on this mate. Of course its the chess part that makes me wanting to play civilization and not just siting back and relax building districts and wonders. I play only for dominations wins. So, since they got this victory condition included, i expect the "chess part" to be challenging.

Maybe other more peacefull people will enjoy this version of CIV more than me. Until then this game is no fun for me. ;<

 

[upi00r]

Sumeria declared a war on me with 10-15 units near my capital, they pillaged one tile and then were doing nothing / moving around until I destroyed them with my archers... When they have arrived and declared the war I had like 2 units or something so they should **** me hard. Instead they behaved like total morons :/

 

[Sascha77]

Of course its the chess part that makes me wanting to play civilization and not just siting back and relax building districts and wonders. I play only for dominations wins. So, since they got this victory condition included, i expect the "chess part" to be challenging.

I totally agree.

And even if you're not going for domination in every game: Combat is an integral part to any Civ-game, so it has to work. I mean: I usually go for other victory types (domination is just too easy against the AI), but I do like to throw in a war or two if there's a juicy target nearby. I haven't done the math but I'd hazard a guess that well over 50% of all units/buildings you can produce are military in nature. So why even include all that stuff (plus military policies and governments) if combat wasn't an important aspect of the Civ-series?

S.

 

[Xyriach]

My only real gripe with the AI is it's seemingly resistant to upgrading units. It's tactical play has improved from what I've seen with units attacking in what would be a fairly logical configuration, ranged at the back, good supply of melee to feed through to the front, good use of terrain. But as you're attacking my Modern Era City with Catapults and Warriors....pardon me if we just leisurely get the keys for the tank...

Fix that, and I would have potentially lost cities to it in my last game.

From where I'm sitting, albeit with a modest coding background, it looks like how to attack has taken precedence over what with. As if there was a checklist for combat that it's not in the right order so it's hitting the conditions for war without actually being ready for it. (Hypothesizing of course...)

But in response to the original question, of course it's brain dead, you were expecting what? Skynet and Joshua in some AI deathmatch? No 4X game I've played has AI that can reasonably take on a human without being given fairly substantial bonuses to it's ability to play the mechanics. In all honestly, I'd be worried if any developer could release a game that had human rivaling AI and often that does come down to the fact that the basics of the game are very complex and therfore extremely hard to balance correctly, but they are in fact possible to do with enough understanding of the behavior.

Some things need fixing as they're a little silly. But it's only through analyzing the behavior that you'll get a solution. Immediately pitching the current model overboard and dialing back isn't fixing, it's just rolling back. I for one would rather that we armed them with the information they need to improve upon the machine behavior, if I want to roll back to earlier versions which performed more to my liking, then it's really up to me to break out the discs.

 

[Sascha77]

My only real gripe with the AI is it's seemingly resistant to upgrading units.

The only gripe? Well.. :

From a thread in the bug reporting forum:

S.