[GS] New patch: AI walled city attack experiment results

[Gorbles]

Civ 5 being a mess on release is no longer relevant when comparing V and VI. Both had two expansions so we have to compare the wonderfully-improved post-BNW civ V with still-the-Same-as-vanilla GS-civ VI.

I'm assuming that you're referring to the AI here (difficult, considering the reference to CiV was the whole game, but certainly "wonderfully-improved" applies to both games), but uh. Are you forgetting the AI on release? We emphasise the existing faults (and some things that have persisted since vanilla - though I haven't seen many / any G&S-specific AI-using-aircraft reports), but to claim there haven't been solid improvements is . . . well, just wrong, really.

This normally gets countered with "well the current state is something I can't stand", and that's perfectly fair. But not the same argument.

 

[Siptah]

I'm assuming that you're referring to the AI here (difficult, considering the reference to CiV was the whole game, but certainly "wonderfully-improved" applies to both games), but uh. Are you forgetting the AI on release? We emphasise the existing faults (and some things that have persisted since vanilla - though I haven't seen many / any G&S-specific AI-using-aircraft reports), but to claim there haven't been solid improvements is . . . well, just wrong, really.

This normally gets countered with "well the current state is something I can't stand", and that's perfectly fair. But not the same argument.

I wasn‘t taking about the AI specifically, but the games in general. So it is actually off-topic. And yes, many things in civ VI improved as well, but did the expansions bring really game-changing improvements? I don‘t think so.

 

[teakbois]

I wasn‘t taking about the AI specifically, but the games in general. So it is actually off-topic. And yes, many things in civ VI improved as well, but did the expansions bring really game-changing improvements? I don‘t think so.

A lot of this is due to Civ VI not needing many game changing improvements. Almost everything in the Civ V expansions was already in Civ VI, and much of it better than Civ V.

It does have plenty of game enhancing improvements though. Small things like Grievances and strategic resource changes make a big difference in the game. And they've done a good job with making floods and eruptions be good risk vs reward. The game is more interesting in the middle section now.

I think global warming and future era are a bit disappointing. Endgame is as much of a slog as ever.

Im hoping for a 3rd expansion that doesn't add new systems but does for science, diplo, and/or religious victories what BNW did for culture. And make global warming more relevant.

 

[Siptah]

A lot of this is due to Civ VI not needing many game changing improvements. Almost everything in the Civ V expansions was already in Civ VI, and much of it better than Civ V.

It does have plenty of game enhancing improvements though. Small things like Grievances and strategic resource changes make a big difference in the game. And they've done a good job with making floods and eruptions be good risk vs reward. The game is more interesting in the middle section now.

I think global warming and future era are a bit disappointing. Endgame is as much of a slog as ever.

Im hoping for a 3rd expansion that doesn't add new systems but does for science, diplo, and/or religious victories what BNW did for culture. And make global warming more relevant.

Oh don‘t get me wrong, I like civ VI and liked it more than V in vanilla already. Still, the expansions would have had much more potential imho and are a major disappointment for me. I hope for such a 3rd expansion as well, yet I doubt we‘ll get one like that.

 

[kaspergm]

I believe they (Firaxis) try to replicate the Ideologies with the mechanics available in Civ6. In Civ5 different Ideologies caused many wars and that was the whole idea and it worked.
So, in Civ6, they have to use different governments situation. There is even a casus beli "Ideological War" but I've never seen it used.

I can see the reasoning for this, if the AI is actually programmed to follow the government choice of its friends. But even with that, it feels very halfbaked. Civ5 actively encouraged you to follow ideologies based on pressure and diplomacy, weighed up against the benefits of being first adopters. In Civ6, there's no such mechanism, or if there is, it's completely hidden.

 

[kaspergm]

I think global warming and future era are a bit disappointing. Endgame is as much of a slog as ever.

Im hoping for a 3rd expansion that doesn't add new systems but does for science, diplo, and/or religious victories what BNW did for culture. And make global warming more relevant.

I must admit I really liked the seasteads and windfarms, they added a whole new level of territory development which gave me something to do in the late game, that I really enjoyed.

 

[Seek]

Also, for better or worse, the new diplomatic modifier for different governments is kind of ridiculous in late game. Modifier for different government was -66 (WTH?) in Modern age and -88 (seriously) in Future age? Modifier for same government was +40. That's basically saying: If we have same government, we like you, if we have different government, we will dow. you.

This would be fine if the modern and future governments had mechanics which gave a reason for such animosity (ie, trade bonuses for like govs, bonuses for denouncing civs in other govs, etc), but since they don't it feels gamey as heck. Arg, another missed opportunity.

Anyway, don't want to derail the thread, carry on everyone.

 

[Lord Shadow]

You can try to explain it away any way you like, but the fact remains that Civ VI's AI is incapable of using its dozens of units to take a single target, which is the ONLY military target on the entire map. There is no excuse for this. And OP is playing on a Duel map, which is not bigger than Advance Wars ones. The unit variety is also actually much greater in Advance Wars - most of Civ VI's units are simply upgrades of other units(which are treated identically by the AI), while all units in AW are unique. The AI is predictable in those games, but that's because it takes smart options instead of moving what seems to be completely randomly.

Feel free to not respond, I'll take that as an admission.

Moderator Action: Please do not troll. leif
Please read the forum rules: http://forums.civfanatics.com/showthread.php?t=422889

There's still no comparison. Advance Wars is a pure wargame where peace isn't even in the picture. The AI has but a single opponent, a single objective, and most campaign missions are quite scripted.

You won't see it performing complex feats even Civ AI has trouble with, like old school amphibious invasions (with actual transports) or making effective use of aircraft carriers unless it's been explicitly scripted to do so in the specific context of unequivocally fixed missions and maps. Hell, even combat is simple since it's always unit-on-unit, with no more complication than terrain in some cases. Furthermore, on many missions the AI begins play with a fully formed army composition and a solid foothold, and the challenge relies more on bringing down that wall than withstanding coordinated, unscripted attacks by said force.

Beyond all that, the AI in Advance Wars doesn't have to trouble itself with research, developing cities, expanding, evangelizing, improving the land, diplomacy, etc. etc. in addition to waging war. Civ AI's Achilles' heel tends to be prioritization in the face of having so many things to juggle, which is why it's not uncommon to witness its indecision.

The first series of tests throws the AI into a war it didn't want nor prepare for, and it's expected to conquer a city it may not even want. It is caught off-guard by the declaration, and it unsurprisingly flounders if one takes a moment to consider the context regardless of the subjective goal the human player had in mind for the experiment.

The second series of tests, perhaps more representative of the AI's reality, produces much more promising results. It shows a more decisive AI mainly due to the fact it seemed to want the conflict and had prepared for it.