Opinions on the "AI"

[Aristos]

Also you need to remove your memory, the AI has no memory of something it saw last turn.

That is true. "Memory" is very hard to simulate/program in any software, let alone a game like civ. IIRC, Soren gave civ4's AI "pseudo-memory" by letting it see some tiles in, either partially or all the time (don't remember). Gazebo and Ilteroi programmed some type of pseudo-memory into Vox Populi's AI, but I don't know the details on how they did it. Paradox claims to have done something similar in the latest patches for EU4, and judging by the AI's behavior in that game, it is working well enough to not force me as a player to "remove my memory".

So, it's not impossible, but very hard to do. There are more pressing issues with this AI before any attempt of "pseudo-memory" can and should be done.

 

[Victoria]

it's not impossible, but very hard to do.

yep, and how much do you use memory in a game?
The trouble is once you know the fallback tricks you can catch anything including scouts sometimes.
In fact the trouble is once you know the weaknesses its sometimes hard not to use them to your advantage, luckily I tend to play "speed civ" because of a lack of time.
I loose lots on deity due to it... and just plain CBA with micromanagement

 

[cvb]

the players here are like in the top 1-10% of Civilization players. There are hundreds of thousands of people playing the game! We are talking within a small sample size of people who are really invested in the game, but do not represent everyone. Firaxis has to take into consideration the many players who do not visit online message boards but are still paying customers.

I can tell you with absolute certainty that this isn't the only place where people go on about the AI. Anywhere that people express their views on the game ends up having plenty of complaints towards that. Sure, most people don't express their opinions on the game, but claiming that acknowledging the terrible AI is limited to the top players is flatout wrong.

If there aren't serious issues with the AI, we wouldn't that much discuss about it and trying to find ways to improve it, would we? Much of what you say is correct, but I suppose, the numbers are not so overwhelming as you interpret them.

Suggestion: Just go to youTube and search for 'civilization 6 gameplay part 1', sort by "upload date" not by "view count" or "rating", pick one, fastforward for actual game start and watch the first 10 minutes ...
Just a dozen or 2 random examples will reveal to you, how the bulk of peoples, who think they have something to show or even something to teach, really plays ... (it is hard to imagine, but there must be people playing even worse, "struggling with the game" by their own assessment)

I don't like the "casual player strength" at all, but it cannot be overlooked (at least not by the company).

 

[ashendashin]

If there aren't serious issues with the AI, we wouldn't that much discuss about it and trying to find ways to improve it, would we? Much of what you say is correct, but I suppose, the numbers are not so overwhelming as you interpret them.

Suggestion: Just go to youTube and search for 'civilization 6 gameplay part 1', sort by "upload date" not by "view count" or "rating", pick one, fastforward for actual game start and watch the first 10 minutes ...
Just a dozen or 2 random examples will reveal to you, how the bulk of peoples, who think they have something to show or even something to teach, really plays ... (it is hard to imagine, but there must be people playing even worse, "struggling with the game" by their own assessment)

I don't like the "casual player strength" at all, but it cannot be overlooked (at least not by the company).

I'm plainly aware of the popularity Civ6 has on youtube. I'm pretty sure that's a consequence of it being new and most youtubers moving on to it. Civ lets plays have been popular for a while. Marketing, I think, has plenty of effect on that front. Moreso than the quality of the game, which is left to those who leave reviews to reflect. The bulk of peoples will always be taken by the pretty colors and popular youtube/streaming personalities, but my point has always been that ignoring problems will get nowhere good.

Edit:
And, once again, just because most casual players don't play the game properly, that doesn't mean a better AI absolutely has to leave them in the dust. This isn't an issue of raising the difficulty across the board. A better AI could easily lose many of the bonuses currently in place. Think that any form of competence will leave those casuals in the dust, even on an even field? Handicap the AI. That's an option you know. Firaxis/2K/Take2 looks at them and sees easy money, I acknowledge that much. I don't understand how so many people here defend such business practices.

 

[cvb]

ignoring problems will get nowhere good.

100%. Intentionally ignoring problems I would consider on the same level as trolling. I see nobody on this forum, who wholeheartedly says "All is fine".

On the other hand: "strong" words won't help us either. Confucius said "Who speaks load in strong words will be heard, but forgotten." I believe a cooperative style will gain more than a confrontative one.
In the videos I perceive an Ed Beach, who sees and understands much more than he is allowed to comment on and "the programmer", who is under immense pressure timewise ...

just because most casual players don't play the game properly, that doesn't mean a better AI absolutely has to leave them in the dust. This isn't an issue of raising the difficulty across the board. A better AI could easily lose many of the bonuses currently in place. Think that any form of competence will leave those casuals in the dust, even on an even field? Handicap the AI. [...] Firaxis/2K/Take2 looks at them and sees easy money, I acknowledge that much. I don't understand how so many people here defend such business practices.

Of course it would be a dream to have a competent "AI" and dumb it down as necessary ... (well, some players who now play Immortal or Deity would not like it to find themselves playing on Warlord or Prince but ok). But honestly, it would significantly cost more than can be earned more because of it.
You can demand what you want, no sane company will on the long term deliver parts, which are relatively loss-making. They know their numbers well - if there would be the possibility to make a profit with "good AI", I AM sure, they would deliver a superb AI.

I think, the best option is to give the community access to the sources and let them mod.
Btw, you cannot compare the results 1:1, because when I modify, it is a piece of love - I don't count the hours ... and hell, no: there is no specification book, product requirements document ... nobody is going to tell me, what I do or not!

 

[ashendashin]

100%. Intentionally ignoring problems I would consider on the same level as trolling. I see nobody on this forum, who wholeheartedly says "All is fine".

On the other hand: "strong" words won't help us either. Confucius said "Who speaks load in strong words will be heard, but forgotten." I believe a cooperative style will gain more than a confrontative one.
In the videos I perceive an Ed Beach, who sees and understands much more than he is allowed to comment on and "the programmer", who is under immense pressure timewise ...
Of course it would be a dream to have a competent "AI" and dumb it down as necessary ... (well, some players who now play Immortal or Deity would not like it to find themselves playing on Warlord or Prince but ok). But honestly, it would significantly cost more than can be earned more because of it.
You can demand what you want, no sane company will on the long term deliver parts, which are relatively loss-making. They know their numbers well - if there would be the possibility to make a profit with "good AI", I AM sure, they would deliver a superb AI.

I think, the best option is to give the community access to the sources and let them mod.
Btw, you cannot compare the results 1:1, because when I modify, it is a piece of love - I don't count the hours ... and hell, no: there is no specification book, product requirements document ... nobody is going to tell me, what I do or not!

All is apparently fine for the majority(which I put to question), so leaving it to modders is the best option, it seems. How about we get a confirmation on that happening? Things are limited right now. Firaxis has practically given a "maybe" on that changing.
Seriously, this is a strategy game. A competent AI would surely give them a noticeable return. For something better than the current, it wouldn't even require much investment besides actually deciding to deal with it. They're not some poor indie company with no backing so as to get a free pass on not being interested in improving their game as much as possible. If they do have too many company limitations and genuinely want their game to have a lasting impact, then I hope they do eventually make it possible for modders to freely make the AI capable in all aspects of the game. So long as we have no confirmation, I will not place trust in their competence until either their priorities change or we get some solid developments in the modding scene.

Strong words? Man, you haven't even seen what the people who actually worked on Civ5s DLL have said.

 

[cvb]

Well, so we can agree, that the glass is half full / half empty (but not completely empty, "broken" or completely full, "fine") depending from which edge one is looking, shall we?

 

[Victoria]

@ashendashin The entire community, all forums are littered with these threads with much abuse for a Firaxis.
It does not stop the way they run their gravy train, and gravy train it must be.
There is loads of people who visit these forums who are very decent nice practical modders and they just say let’s wait until the dll is released.
Anything they do before then is a bonus and would be much appreciated but I doubt they will change their spots significantly.
If the game was unplayable then no-one would be playing it so it’s a playable game and that is what works for 2K.

There was a poll a while back something to the effect of what do you want in the new version and I voted for fixing the things that were broken but I was in the minority, it was all new DLC’s and new functionality.

How can you fight against that? Firaxis is in fact delivering what the majority want and this is not just a gaming thing, it’s in the computer business as well, new stuff sell’s.

Fight all you want, you are in the minority and there will always be crusaders for doing the right thing so keep it up, I just find I get a sore head after a while bashing my head against a wall.

 

[ashendashin]

Well, so we can agree, that the glass is half full / half empty (but not completely empty, "broken" or completely full, "fine") depending from which edge one is looking, shall we?

Yes. So let's move beyond the personal offense topic and acknowledge that things can and should be better for both modders and Firaxis when it comes to the AI.

 

[ashendashin]

How can you fight against that? Firaxis is in fact delivering what the majority want and this is not just a gaming thing, it’s in the computer business as well, new stuff sell’s.

I'd say a decent AI would be a totally new feature to come from Firaxis. Not to mention that they have a huge team that's surely capable of multi-tasking. Forgive me for believing that a video game company should have some passion towards quality.

 

[cvb]

edit: #127

How about we get a confirmation on that happening?

The amount of hate / opposition ("I feel betrayed") makes it difficult to make statements, too easily it may return as "broken promise". So it seems better for them to remain silent.

Because of this I advocate for a more cooperative dialog between company and community.

 

[Victoria]

Forgive me for believing that a video game company should have some passion towards quality.

One key thing is that artwork requirement are HUGE now, not just more detailed images but shadows etc. Wonders being built. The artwork team will be ginormous and to keep employing them they need new work. It is just they way things have gone, they may want quality but the budget is not theirs.

 

[ashendashin]

edit:
The amount of hate / opposition ("I feel betrayed") makes it difficult to make statements, too easily it may return as "broken promise". So it seems better for them to remain silent.

Because of this I advocate for a more cooperative dialog between company and community.

You think it hasn't been tried? It was done back with Civ4. Firaxis has almost completely locked up to DLL modders since then.

 

[Archon_Wing]

"as an absoloutely barebones AI that usually is unpleasant to play against."
"AI is just unbearable, both diplomatically and in warfare."
"in 20 or so years diplomacy has not advanced much at all. its still complete garbage. as is the AI in general."
"How can the AI be so bad - WTF???"
"The AI got worse, don't bother."
"Game is not bad if you are new to 4X but otherwise the IA is so bad that after two games there is absolutely no challenge."

And these were from Steam reviews from just this month under "most helpful reviews in the last 30 days". This means this group is already filtered because they had to be upvoted by people that thought the same way.

Sure, not everyone writes reviews and reviews are always skewed negative. I also know many of the reviews are from illiterate people that thinks Rise and Fall should come with the Deluxe. But go find a review out there, out of the 28,000, that says the AI is not bad.

Yea, it may be true the overall skill level is low as most players don't beat the game on Prince. However, if you go by that and there are still complaints from casuals, well....

 

[MarigoldRan]

Excellent opportunity for some other company like Google to make their own version of Civ. Nothing kicks a company into action then getting one-upped.

Google's planning to go into gaming, btw.

 

[Unconquered Sun]

Yea, it may be true the overall skill level is low as most players don't beat the game on Prince. However, if you go by that and there are still complaints from casuals, well....

The point of having so many Difficulty settings is that all players can find one where they are challenged enough to not be bored, per the Flow game design fundamental concept



Civ 6's problem isn't Chieftain being too hard on the casual players. It's Deity being way, way too easy.

 

[Aristos]

How can you fight against that? Firaxis is in fact delivering what the majority want and this is not just a gaming thing, it’s in the computer business as well, new stuff sell’s.

This a thousand times! Hence my "blame the Audience" argument, buried somewhere in this thread before the fireworks started... we are converging, Vicky, we are.

edit: #127
The amount of hate / opposition ("I feel betrayed") makes it difficult to make statements, too easily it may return as "broken promise". So it seems better for them to remain silent.

Because of this I advocate for a more cooperative dialog between company and community.

How when there is never an answer? Even for bug reports for which we did a few (many) during Chao's CQUI times, we never got any answer at all, not one.. in fact, some of those nasty bugs appear to be still around, like the strange city list behavior in the deal screen...

Dialog implies, from the Latin root, a two-way communication... in my experience, the attempts came only from a portion of the community (after Soren's times, that is), without any response from them.

 

[Bitterman]

I have a friend that has absolutely (I repeat, absolutely) no clue about strategy games, being the closest thing for him Age of Empires, if something, a game in wich he was terribly bad.

Well, even him can easily beat the unmodded AI on king, just by making military and stomping everyone. He does not understand unit movement. He does not understand culture. He does not understand science (he knows that the more campuses, the faster it goes though). And of course he has no idea about faith or religion. The same goes for many core mechanics of the game, like spies, tourism, citizens managment etc... and even so he is capable of not only winning games on king, but winning them by a far margin.

For me, this is nothing but an indicator of how poor the AI is. It is not about the top players or something like that. Top players don't even bother playing against vanilla AI I pressume. Normal people that have no idea of how the game works are able to beat it on higher diffculties. This needs to be adressed. It is ridicolous that someone who does not even know how to use the simplest of mechanics is able to get victories in a (supposedly) strategy game. As for my friend, I'm sure that he will not be playing this more than a month. He is a super casual player, and will most likely abandon this the moment it loses its novelty.

As for the problem itself, I'm in for giving the AI some work on science and military. Those two things are the ones wich really make the difference between a top civ and just a regular civ. Even if you are the best player in the world, it's quite difficult to mess with someone who has more and more advanced units than you. It also adds that feeling of being threatened, of having to be careful with your neighbour and having to maintain good relationship with him/her if you don't wanna get invaded. Right now that simply does not exist.

 

[darko82]

I have a friend that has absolutely (I repeat, absolutely) no clue about strategy games, being the closest thing for him Age of Empires, if something, a game in wich he was terribly bad.

Well, even him can easily beat the unmodded AI on king, just by making military and stomping everyone. He does not understand unit movement. He does not understand culture. He does not understand science (he knows that the more campuses, the faster it goes though). And of course he has no idea about faith or religion. The same goes for many core mechanics of the game, like spies, tourism, citizens managment etc... and even so he is capable of not only winning games on king, but winning them by a far margin.

For me, this is nothing but an indicator of how poor the AI is. It is not about the top players or something like that. Top players don't even bother playing against vanilla AI I pressume. Normal people that have no idea of how the game works are able to beat it on higher diffculties. This needs to be adressed. It is ridicolous that someone who does not even know how to use the simplest of mechanics is able to get victories in a (supposedly) strategy game. As for my friend, I'm sure that he will not be playing this more than a month. He is a super casual player, and will most likely abandon this the moment it loses its novelty.

As for the problem itself, I'm in for giving the AI some work on science and military. Those two things are the ones wich really make the difference between a top civ and just a regular civ. Even if you are the best player in the world, it's quite difficult to mess with someone who has more and more advanced units than you. It also adds that feeling of being threatened, of having to be careful with your neighbour and having to maintain good relationship with him/her if you don't wanna get invaded. Right now that simply does not exist.

It seems Firaxis does not care. Or they cannot deal with the problem. They do nothing about it. But I don't think they are as stupid as the AI

I will not buy anything from Firaxis if the challenge and the AI will not be good enough. I lost my money only once. Do you also want to lose your money? In Civ IV, the AI was stupid, but I was afraid of being attacked by the AI. In Civ V, VP made the AI challenging. Otherwise, I would not have played Civ V because of the AI. Firaxis cannot compete with some modders in this respect, it seems.

They do not think consumer/gamer-wise; they think business-wise, even if the product is a crap in other categories - like the AI. Marketing and selling is the most important thing for them, not the quality of the whole product. That's my opinion.

 

[Aristos]

They do not think consumer/gamer-wise; they think business-wise, even if the product is a crap in other categories - like the AI. Marketing and selling is the most important thing for them, not the quality of the whole product. That's my opinion.

Clearly. The question is why. My personal answer to that question is, as I will say it a thousand times, because Audience. And if I am right, then it's hard to blame FXS because they are a private company whose ONLY purpose is to make money. It's the audience, or market, of any given product (including political products), that in the end decides the quality of the end product. In other words, we have only ourselves to blame.