Opinions on the "AI"

[FearSunn]

I think it would help if the AI got free units in certain situations (both offensive and defensive), instead of just at the start of the game. The trick would be to give the AI units which reflect what it should have been building in the first place, had it been actually planning ahead, so that the bonus units don't feel like a boondoggle.

Just for the record. This is already implemented in More Strategy (and with lesser extend in Adaptive Difficulty) mods. AI tracks human military strenght and prepares itself in advance for possible human offence. Also AI gets religious units if it has its own religion founded and if human is buying and actively using his own religious units. With this domination and religious victories became a true challenge. The only one victory not "solved" yet remains cultural. Just giving culture is not enough and there is no specific unit or building type to facilitate this kind of victory.

EDIT:
http://steamcommunity.com/sharedfiles/filedetails/?id=1226610353

 

[MarigoldRan]

I Think of the Game as Sim City

With the current AI Opposition, that's basically what Civ 6 is.

 

[darko82]

It's a shame Firaxis served such poor AI to the players and fans. I think it was offensive to demand so much money for this game.

The AI is a hoax. As a consumer and player, I personally feel deceived by the AI and Firaxis.

And I will repeat once again:

As long as it is accepted by the players, it will carry over to the next Civ game, sadly.

The AI cannot play the game it is designed for. A stupid dog would beat it.

It reminds me a bit of Might and Magic Heroes VII, which was a fiasco, too. It did not sell well because fans could not accept the design and very poor AI.

Civ VI was already bundled within a year since the release. It's not a coinsidence.

Some players were turned off by the artstyle, the others by the AI.

Firaxis aims at new, young players. Easy buck. They don't aim to invest time and money to create a legendary and challenging strategy game ever created.

Dissapointing. I fell and lost my money only once. I like many new features in Civ VI.

But the AI is a deal breaker.

 

[FearSunn]

It's a shame Firaxis served such poor AI to the players and fans. I think it was offensive to demand so much money for this game.

The AI is a hoax. As a consumer and player, I personally feel deceived by the AI and Firaxis.

I fully agree. As a programmer I bought that game already knowing it is broken and I will have fun for just fixing it.
For all other consumers not savvy in lua and sql this is unfair at least.

 

[cvb]

I fully agree. As a programmer I bought that game already knowing it is broken and I will have fun for just fixing it.

I'd used other words along 'already knowing it has serious flaws and I will have fun for just making a good thing better' ... but words are nothing but smoke and mirrors.
Besides from me liking to be positive and thinking in solutions instead of problems etc. etc.

Just reminded me of that bit some years ago:

Don't worry, probably it is just the next level of gaming experience ...
playing CIV: build an empire, rule the world, stand the test of time ... (under given rules)
modding CIV: create the world, make the rules, be GOD ... (watching the AIPlayers build their empires)

In civ1 I used dice to vary the paths through the tech tree and built wonders and used several HouseRules. So it was a challenge again ... in civ2 Brian Reynolds was my hero and gave me "a lot I needed for modding my version" including the AI initially using an AWACS (Airborne Warning and Control System) as throwaway bomber ... civ3, civ4 ... again & again ALL repeats ... also always new disappointed newbies threatening NEVER BUY AGAIN ... whow ... you are vociferous, but the majority here is simply bored about such "news" ... ok, going back under my stone and will skip the next hundred invitations to answer to any statement deemed unique or "new".

 

[Victoria]

I bought that game already knowing .....

I bought the game already knowing it would be half full with what I like.
I didn't get what I wanted but I got what I needed.
One of those things was that the AI could move and shoot archers in the same turn.

I used to get upset that only one city each turn would shoot at me, then looking deeper it seems there is timeouts involved and that sort of made me happier.
You can have a smarter AI that takes 10 minutes to take a turn or you can have a game you can play in a night or two.

That does not mean they cannot improve the game and make it more 'playable', it just means there will always be compromises. Lately I have been seeing more AI archers in wooded hills not moving from these nasty positions, seems like baby steps are happening.

 

[Aristos]

You can have a smarter AI that takes 10 minutes to take a turn or you can have a game you can play in a night or two.

If you take the Vox Populi AI, developed by a history teacher that learned to program C++ for fun, as proof of anything, it would probably mark the above as a fallacy, don't you think?

And NO, Vox Populi turns do not take 10 minutes... in fact, they are faster than vanilla, because the "amateur" devs also optimized a lot of what they call "spaghetti code", referring to the overall quality of the coding in vanilla Civ 5.

That speaks volumes about the argument of Quality being the issue here, don't you think?

 

[Victoria]

That speaks volumes about the argument of Quality being the issue here, don't you think?

I got too bored of civ V to play Vox. I heard it was good but I was burnt out.
Quality is a symptom not an issue. It’s investment.
The fact of amateur dev’s plural as opposed to Firaxis’ Single AI person speaks volumes.

 

[ashendashin]

I got too bored of civ V to play Vox. I heard it was good but I was burnt out.
Quality is a symptom not an issue. It’s investment.
The fact of amateur dev’s plural as opposed to Firaxis’ Single AI person speaks volumes.

Vox is like a totally new game. I put over a thousand hours into it after getting burnt out of vanilla.
Most of VP was coded and is maintained by Gazebo. It was built onto Whoward's DLL years ago and had some overall contributions from the community, but as far as I know, the AI work can be almost entirely credited to Gazebo and the tactical AI programmer, Ilteroi.
So it took DLL work from Whoward, then for G to realize that the AI can get overhauled with a proper DLL, and for Ilteroi to get to work on a new tactical AI. The Community Patch is the core that those two did most of the massive improvements on, and Vox Populi is a package of 6 components that expanded towards a community overhaul.
Meanwhile Firaxis has around 200 employees from what I read on wikipedia. Dunno how many of those are active devs, but some priorities obviously need straightening out.

 

[Victoria]

Well I stand corrected then.
I would say in the defence of Firaxis that its a hell of a lot of design work and artwork and non AI programming that needs doing, that their AI guy says that moddable code can never be coded as cleverly as less moddable code. But primarily they are focused on getting a working game out and milking it. DLC release requirements would all have been planned out way in advance, release dates and everything and as each date gets closer they run out of time... why? Because thats the way company projects work. Code still not working right? Just bang it out, well put an emergency patch out the week after.

Vox is like a totally new game. I put over a thousand hours into it after getting burnt out of vanilla.

I only played BNW but thrashed it severely, took me a long time to come back to the game after Civ VI annoyed the hell out of me. If Vox still has the National college thats enough for me not to play it.

 

[Aristos]

The fact of amateur dev’s plural as opposed to Firaxis’ Single AI person speaks volumes.

I'm sorry, but this one sounds to me like another fallacy. Firaxis' single AI person (if true) is supposed to be a full time, experienced AI programmer, correct? And he is being paid for it, full time. Gazebo is a history teacher that picked up C++ as a hobby, and programmed Vox Populi in his spare time. Such an inversely proportional imbalance between background, time and result is what really speaks volumes.

A smart developer (lead designer, manager, call it whatever you want) would have kept people like Gazebo and Ilteroi very close, if not directly working on the new AI, at least beta testing it permanently. But Firaxis went the other way, completely ignoring them and even allegedly treating VP as the "enemy" (according to some reports of people claiming that they have been banned from 2K forums for just mentioning Vox Populi)... how is that smart?

 

[Victoria]

I'm sorry, but this one sounds to me like another fallacy.

... well what you are saying is the Firaxis developers are morons then, I just think thats a bit of a moronic opinion in itself

 

[Aristos]

I only played BNW but thrashed it severely, took me a long time to come back to the game after Civ VI annoyed the hell out of me. If Vox still has the National college thats enough for me not to play it.

You will hate VP as a beginner for many reasons, but none of them will be the overwhelming sensation that there is only ONE way to play it (your NC argument). You will hate VP as a newbie for the many ways the improved AI will trash you (as opposed to you trashing the game for boring linearity).

 

[Victoria]

You will hate VP as a newbie

You will hate VP as a beginner

I will... I will ... really? is a game all about being thrashed or pushed to your limit?
I work damn hard, my mind has to work through quite complex stuff ande my job is ever changing... last thing I want to do when I come home is have to work for my enjoyment
I appreciate everyones view as long as they realise that life is not about one size fits all, jeez
I may play it and love it straight away but I would be turning down difficulty for pleasure.

 

[Aristos]

I will... I will ... really? is a game all about being thrashed or pushed to your limit?
I work damn hard, my mind has to work through quite complex stuff ande my job is ever changing... last thing I want to do when I come home is have to work for my enjoyment

Well, then, maybe you should reconsider your participation in a thread that is all about the lack of challenge from the AI in civ 6 (ergo, the lack of an AI that makes us "work" harder to achieve something in the game)? I can understand your point, nothing wrong with that, but nothing wrong with the "opposite" view either, the one that some of us (many?) see as fun ("working" harder for victory)...

 

[Victoria]

So you are saying I cannot have an opinion?

 

[Aristos]

So you are saying I cannot have an opinion?

Where do I say that? I get your opinion, you don't want a harder AI because for you it equates to "hard work" and you had enough of that during your day (who doesn't?). Is that interpretation correct? If yes, then why insist on it inside a thread that is clearly dedicated to analyzing the lack of challenge from an AI that you don't want to be more challenging? I don't get it.

But enough of this. This is my last personal answer. Let's keep rolling on how to hopefully press FXS to consider making this AI more challenging, and/or releasing the dll source sooner so that we can get our hands on it and make it ourselves.

 

[ashendashin]

I will... I will ... really? is a game all about being thrashed or pushed to your limit?
I work damn hard, my mind has to work through quite complex stuff ande my job is ever changing... last thing I want to do when I come home is have to work for my enjoyment
I appreciate everyones view as long as they realise that life is not about one size fits all, jeez
I may play it and love it straight away but I would be turning down difficulty for pleasure.

Eh, you might get frustrated at first. Some people do. I loved it when I started as there were so many things to learn and I was more than willing to drop the vanilla experience. Typically all you've gotta do is play the appropriate difficulty level to avoid too much frustration. Or you can just throw yourself at it like it's vanilla and get frustrated when things don't work like you're used to.
You certainly won't be thrashing it though since whatever massive imbalances you're used to aren't a thing in it.

 

[Victoria]

If yes, then why insist on it inside a thread that is clearly dedicated to analyzing the lack of challenge from an AI that you don't want to be more challenging? I don't get it.

Because someone was bragging they could win every game without save scumming which is just BS. Oh... that was you!
I'm with all those people in the thread that say make your own challenge if the default ones are to hard. Play a one city challenge on deity for example.
My last entry also

. I loved it when I started as there were so many things to learn and I was more than willing to drop the vanilla experience

Thanks!, I might give it a go, I'm still ripping VI to bits though, its the mechanic in me.

 

[Infixo]

Ok, now that we established that everyone can play what they want to play in a way they want to play it (yey ), let’s get to business.

Getting the source code for the Civ6 dll will ofc open new possibilities but it will NOT be as huge brakthrough as it was with Civ5. Entire AI logic in Civ5 has been coded in the dll (literally everything but flavors) and any improvement on this front required changes to the dll. Civ6 uses a different model: a big chunk of AI is actually programmed via behavior trees, lists of AiFavoredItems, hundreds of modifiers and quite a lot of various parameters scattered around. This means that many changes and improvements can be done right now.

I would say that number of things that cannot be changed or tweaked is actually much smaller that things that can be changed. Things we cannot change, from my observations, are mainly:
- tactical behavior of specific units
- building / unit valuations and their impact on build decisions
- assessment of strategic situation

The list may not be long but these are pretty important things, unfortunately. That is why imho the community can make an AI better but not as good as it should be.