AI not conquering each other kills this game for me

[kaspergm]

I know people don’t like these kind of solutions ...

Yeah, that would be a very unsatisfactory solution imo.

 

[TreVor_ish]

I remember devs saying there will be bigger support for modding.....has anyone seen MODDING section in any of the updates during NFP?
And Yeah....AI improvements in those updates....can You be any more cryptic....

 

[plus]

Civ V had that weird overpunishing wide approach. However, it is also clear to me that civ VI is designed so the game turns peaceful from middle ages till end game.

VI has now that weird plain DF penalty, that means that if the player takes more than a civ in the game, the whole WC, competitions, emergencies, and favor mechanics become pointless.

VI is designed so domination victory is the hardest to get, and the least satisfactory one, due to the fact that forces the domination player to micromanage hell exponentially with the ammount of cities, and offers no domination alternative to take cities.

VI is designed to not even consider late wars a thing. As so, AI domination victory is imposible making domination the only victory type that the AI is uncapable of achieving. Furthermore, the AI even lacks the knowledge and will to try to achieve it. This is then the first Civ game in 30 years where the AI does not challenge the player to domination, and where big scale wars are imposible. So basically is the first civ game ever where conquering the world is not supposed to be a goal.

It is true that civ VI rewards having more cities, and therefore makes the game easier the more you war, but is also true that the game is also designed to make the game more pointless and boring the more you war. Which is the worst way posible to discourage the player. It is true that is not the first game of the series that is guilty of this, but it may be the first that is intentionally designed to be this way. With repetition and shallowness being not a side effect in late game, but a core design element.

So I think, quite the opposite is true, Civ VI not only discourages war, but is designed to remove domimation and war from the late game entirely.

I would also argue that religious victory, culture victory and diplomatic victory are so pointless and badly implemented and designed that should not even be in the game. While i will also admit that science victory is the best it has ever been.

The game is as a result a game of civilization where global conflict does not exist, domination is non-fuctional, war is broken and diplomacy is pointless. As such, civ VI may be a fun game in many ways, and certainly is one of the most enjoyable games of the series in city building, multiplayer, scenarios and variety. But as a civ game, I think is time to be honest and say it... Civilization VI is probably the worst civilization game of the series to the point it does not work as a civilization game at all. Personally it is the first civilization game (and I played them all) that seriously dissapointed me, and as is also by far the most expensive game of the series. So I think it may be the last civilization game I ever buy it they decide to keep this philosophy.

I agree with you.thsts why i only play civ5,none of the problems mentioned are there.i have full scale wars,civs that are decimated city states that actually last and not run over by ai,i tried pkaying 6 after release but realized its more a city builder than civ.and for city building there are other games..

 

[m_mus]

Proposal: Hard'mode'
Just add another gamemode that lets the AI game to win, to conquer each other and dogpile the player in case he or she is close to winning.
Call it 'hardmode' and make it optional, so that everyone can play according to his pleasure.

 

[kaspergm]

Personally, I feel there's a distinction between the AI being capable of achieving a victory and the AI being conscious about a victory being a thing (in as much as it makes sense to use the word conscious in the context about an AI).

What I mean is: The AI should be programmed to follow different game focuses/specializations - what is currently called victory conditions. For instance, if an AI decides to play science focused, it should maximize its science input, try to get the space techs unlocked asap., build space ports and space projects, and grab great people boosting science and space projects. If the AI decides to play conquest focus, it should focus on military, it should be aggressive, and it should be ruthless and opportunistic in eliminating another civ, if it deems it will benefit itself (not just randomly accepting peace offers after the minimum 10 turns etc).

However, I don't like the aspect of the game where the AI is aware that the human is "winning". I disliked the "they are afraid that you are winning" diplomatic modifiers in Civ5, and I would hate for the AI to randomly start DoW'ing me or nuking me just because I had my space ports up and running, if we were otherwise on good diplomatic status. That just kills immersion for me. However, many of these elements can be implemented in a different framing that I would be less negative towards. I wouldn't mind an AI being jealous of my wealth or prosperity if I started pulling too far ahead, and if that spawned them to turn negative towards me and in time DoW me to claim some of my wealth for their own, that would definitely be meaningful to me. Even more so if sharing my wealth with them would actually eliminate this negativity.

Similarly, I wouldn't object to the AI ganging up against the player in the world congress if the players starts monopolizing the vote to his own benefit - again, Civ5 world congress were worlds ahead of Civ6 in this regard, where they would actually be diplomatic modifiers for voting with or against the AI, where in Civ6 we're just left with a random mess where you have to guess or know from the hard-coding how the AI is going to vote. This would also make emergencies and world congress much more meaningful in the game, because it would allow you to have propositions that would benefit civs falling behind - for instance, tech sharing, research is cheaper for civs when another civ has already researched the tech. Voting for or against such a suggestion would allow you to either keep your advantage - at the cost of negative diplomatic relationships - or give up (some of) your advantage, but at the benefit of a diplomatic boost.

 

[Krajzen]

Personally I'd honestly revolutionize the entire way 'victory' works in this game. I don't think there is another strategy game in which there is such tension between the gameplay and victory conditions. On one hand Civ (sometimes) wants to be immersive, on another hand it has those wildly arcade victory conditions akin to the sport competition. Once again, the victory conditions don't really make sense withing the world of historical empires. What exactly does it mean for historical culture to 'win'? Did America get a standing ovation and title of the best country from UN because of the Moon landing or its few decades of popcultural dominance? Did Jews got a special statue of them built from gold to mark their religious victory across the globe? Don't even ask me about the diplomatic victory.

Even besides that, victory conditions contribute to the problem of terrible endgame, because the way they are done is inherently benefitting linear growth of runaway elite. There is nothing allowing any sort of dramatic change in those races from antiquity to space. By the renaissance era you can basically divide factions on the camp of 'obstacle to victory' and 'certainly not an obstacle to victory'. Factions from the second camp never get into the first.

Paradox games have no victory conditions at all, and thus can invest fully into immersion and lack of derailing mechanics; all countries in eu4, human included, have their own realpolitik. Humankind is instead going for a victory system based on 'Fame', namely spectacular and risky actions done at every age of the game and summarized at the end. Both seem to be better solutions to me than the civ method.

 

[TreVor_ish]

Personally I'd honestly revolutionize the entire way 'victory' works in this game. I don't think there is another strategy game in which there is such tension between the gameplay and victory conditions. On one hand Civ (sometimes) wants to be immersive, on another hand it has those wildly arcade victory conditions akin to the sport competition. Once again, the victory conditions don't really make sense withing the world of historical empires. What exactly does it mean for historical culture to 'win'? Did America get a standing ovation and title of the best country from UN because of the Moon landing or its few decades of popcultural dominance? Did Jews got a special statue of them built from gold to mark their religious victory across the globe? Don't even ask me about the diplomatic victory.

Even besides that, victory conditions contribute to the problem of terrible endgame, because the way they are done is inherently benefitting linear growth of runaway elite. There is nothing allowing any sort of dramatic change in those races from antiquity to space. By the renaissance era you can basically divide factions on the camp of 'obstacle to victory' and 'certainly not an obstacle to victory'. Factions from the second camp never get into the first.

Paradox games have no victory conditions at all, and thus can invest fully into immersion and lack of derailing mechanics; all countries in eu4, human included, have their own realpolitik. Humankind is instead going for a victory system based on 'Fame', namely spectacular and risky actions done at every age of the game and summarized at the end. Both seem to be better solutions to me than the civ method.

While I agree with You on most points, CIV VI does have a score victory, which is somewhat like fame in Humankind....also u can remove any type of victory from Your game or every one of them - leaving score and even changing number of turns when it's triggered. So it is quite agile in that regard. The problem is even if removed certain one, AI will still try to win using programmed path, which bring us to issue that is being repeated at least 100 times a day....let us mod AI - with lua, or release dll....

 

[Krajzen]

While I agree with You on most points, CIV VI does have a score victory, which is somewhat like fame in Humankind....also u can remove any type of victory from Your game or every one of them - leaving score and even changing number of turns when it's triggered. So it is quite agile in that regard. The problem is even if removed certain one, AI will still try to win using programmed path, which bring us to issue that is being repeated at least 100 times a day....let us mod AI - with lua, or release dll....

No, because Fame is a refined system of much more than 'just another way to name score count.' You do get era 'stars' per specific actions in every era separately and they benefit risky actions; iirc they specifically said something like 'it may pay off to be spectacular in one era even if you collapse if the next one' - there is more dynamism in this sentence than in the entire civ6 combined, with its linear snowballing slog until inevitable victory. Civ6 score victory is just was static and boring as each other victory type, because it's just the linear score counter frm which you can easily deduce endgame victors halfway through the game. At least on paper Humankind more aims at the approach of 'Mongols went really crazy in one era, and even though they collapsed later they are still one of the greatest empires of human history'.

Of course God knows how it will work in practice, but here at least devs acknowledge the issue of static mid to late game and are trying some radical ideas to deal with that.

 

[TreVor_ish]

No, because Fame is a refined system of much more than 'just another way to name score count.' You do get era 'stars' per specific actions in every era separately and they benefit risky actions; iirc they specifically said something like 'it may pay off to be spectacular in one era even if you collapse if the next one' - there is more dynamism in this sentence than in the entire civ6 combined, with its linear snowballing slog until inevitable victory. Civ6 score victory is just was static and boring as each other victory type, because it's just the linear score counter frm which you can easily deduce endgame victors halfway through the game. At least on paper Humankind more aims at the approach of 'Mongols went really crazy in one era, and even though they collapsed later they are still one of the greatest empires of human history'.

Of course God knows how it will work in practice, but here at least devs acknowledge the issue of static mid to late game and are trying some radical ideas to deal with that.

Yes, the mid/Iate game in CIV VI is sometimes extremely static, specially when (in most cases actually) u know it's already a win. Adding new completely detached mechanisms only makes it worse.
Also looking forward to Humankind, and Amplitude does communicate with community much more frequent.

 

[oSiyeza]

I know people don’t like these kind of solutions but if the problem is truly the walls why not remove the combat penalty for attacking cities with walls for the AI on deity and immortal? I don’t think that would be too hard for us players to deal with and since the AI they attack will also benefit I’d call it balanced.

Also I STRONGLY agree with changing siege to support class units and allowing them to be stacked.
As wonderful as 1UP is for us players, I think it’s crippling.

I hope we eventually get access to the DLL for mod teams to work on the AI because every patch that there is “AI improvement” in the notes I feel it’s just adjusting numbers on the tables which is something we can do ourselves. I did notice some small improvement the first patch or so of NFP but the last few patches have seemed to make things worse. For anyone who isn’t aware, you can get the AI to trade many of their cities away again now.

That will make the game more challenging on inmortal and deity, true. But that will not sove the problem. I dont play on those difficulties, and the problem to me is not that the game is too easy, is that it becomes boring and repetitive and unsatisfactory and you find yourself pushed out of the world of the game because the AI is badly designed, and the mechanics are shallow and repetitive.

The solution is just make consistent, inmersive, deep and fun game systems that can be handled by a good AI, and to design the AI along with the game not as an afterthought. It is a matter of money resources and dedication to the game not as a product, but as a game, as an idea. Fxs had plenty of time to do this, and even to correct it, or at least try. However, Civ VI will the testament on how an awesome foundation, talented people and incredibe new ideas ideas are not enough to make a good game, if the design is not managed properly.

 

[agonistes]

Can't say I notice this. I usually end up never meeting 1-2 civs because they died early.

Me too.

 

[Archon_Wing]

Me too.

What are your game settings?

I dunno. I usually play pretty aggressive and work with other AIs to take down other AIs. I'm aware the AI is a big potato, but if you help them, then things become much more dynamic. Now, if you're passive and leave the AI to its own devices, it's a bit slow.

 

[agonistes]

What are your game settings?

I dunno. I usually play pretty aggressive and work with other AIs to take down other AIs. I'm aware the AI is a big potato, but if you help them, then things become much more dynamic. Now, if you're passive and leave the AI to its own devices, it's a bit slow.

Settings vary. Normally I play at 7x marathon, gigantic map.

But now and again I'll play standard huge just so I can see some fancy new late game stuff without having to wait thousands of turns.

 

[Xur]

I dont have a problem with strong walls that acts as a buffer. My problem is that the walls can shoot. The AI brings 2 siege units on a good day, so they are fairly easy to stop. The AI itself even attacks siege units over anything else first. Remove the ranged attack from walls, and much of the problem is gone... I think.

 

[Time Tested]

I dont have a problem with strong walls that acts as a buffer. My problem is that the walls can shoot. The AI brings 2 siege units on a good day, so they are fairly easy to stop. The AI itself even attacks siege units over anything else first. Remove the ranged attack from walls, and much of the problem is gone... I think.

Maybe for the player, but the AI will still bring a bunch of non-siege units to try and take a city and just attack until they die. It's not the ranged attack that gets them, it's attacking as an underdog repeatedly.

I feel like the AI ignores walls in their attack strategy. They just treat a city like a normal enemy.

 

[Archon_Wing]

Could not find Robert the Bruce until near the end just because Sweden destroyed him. I end up liberating the capital. in this picture but he would just die again soon. Also I tried to spare Jadwiga one city for a culture victory but then Sweden just came in and razed the city. Yikes.

 

[jasper]

What a strange false dichotomy. Each civ should simply care for itself like in real world, not form some hivemind against human. It has worked like this in previous games and in other strategy games.

What a strange misunderstanding. Civs dont just care for themselves in the real world. They very much form alliances and pay attention to who is top dog in the region. They definitely team up against someone when its to their advantage.

Even board games irl, youd absolutely team up or at least consider teaming up against the guy about to win.

 

[Krajzen]

What a strange misunderstanding. Civs dont just care for themselves in the real world. They very much form alliances and pay attention to who is top dog in the region. They definitely team up against someone when its to their advantage.

Even board games irl, youd absolutely team up or at least consider teaming up against the guy about to win.

That's exactly what I said, or at least meant.

 

[Foufa131]

I know it's a little late but the " no walls " mod helped me a little. now ai is better at taking cities.

 

[Abaxial]

There was an interesting experiment conducted by Marbozir - he created a scenario where one AI had a giant death robot (and some uranium) from turn one. Did the AI use this huge advantage to roll up all the other civs? No, because not being sentient it had no way of thinking that it had a huge advantage. It just followed its estabished routines - build new settler - escort settler with strongest unit available - and so on.