Unpopular opinion: people that blame AI don't play on deity

[Absolution]

Generally I disagree.
I won't get into details because I'm no expert on Civ6, but even a look back on Civ4 could show some difference.

Personally, when I wish the AI to be improved, it is mostly not because I wan't the AI to win more often - It is because I want them to play a more interesting game.
To have a more consistent strategy during a game, to take more realistic decisions, to have some human-like or historically accurate ambitions and tendencies.
A major problem is the diplomatic system, which blocks any possibility for practical statesmenship and downgrades it to a childish game of good manners and meaningless decisions.

 

[acluewithout]

It must be purposeful to make the game easier vs. the AI. I mean simply coding the AI to build a single ram before the wave of catapults it conjures, attaching the ram to an infantry unit that is itself coded to take the tortoise promotion first (+10 vs. ranged), then sending the infantry to charge a given city, then once positioned next to a city wall have all units within range attack, and all units that are close move towards the city. How hard can it be?

I don’t get why people keep asserting that FXS are deliberately coding the AI to be bad. It’s just not a credible assertion.

It’s clear FXS have been trying to improve the AI, and with some success. There are also much better ways to make the game easier rather than deliberately borking player controlled players.

It’s true FXS don’t code the AI to bee-line victory - instead, the AI is more coded to create interesting games although even that still encapsulates some attempt at victory. But to say that extends to making the computer players deliberately throw the game is just not... credible.

Honestly. These discussions about the AI are getting very silly. It’s either “this game needs Deep Blue or AlphaGo” or “FXS are evil overlords deliberately dumbing down the game”. Both are silly lines of discussion.

It’s clear FXS are devoting resources to the AI. This is improving the AI, but not enough for many people. There are a few potential causes. More than one may be at play and many are related. First, it’s really hard to code the AI for Civ (because it’s hard to program, AI for strategy games, or because the rules are still changing, or because particular mechanics are hard for the AI to use eg 1upt). Second, while FXS have devoted resources to making the AI better, it’s just not enough. Third, there’s some fundamental problem, such as using Branching Tree decision making (or whatever it’s called) or the AI is timing out.

 

[BarbarianHunter]

I don’t get why people keep asserting that FXS are deliberately coding the AI to be bad. It’s just not a credible assertion.

I disagree with the fundamental premise of your argument that says my statement is not credible. Generally, I note that the AI often has a significant number of units about the front in a given war. I also generally find that they have essentially no hope of conquering a city due to the human player's defenses. If they were to bring rams instead of catapults with their armies, my belief is they would be more successful. My hypothesis as to why they don't: Most of the people that buy and play the game are not CivFanatics and if they were to get stomped, they'd opt out and become disillusioned with the franchise.

Edit: I hardly ever build walls, and generally the AI still can't conquer my cities. I'm just saying that FXS are probably at least tacitly incentivized to brush this problem under the table. Whether they realize this or not, I would argue it is part of their decision making process.

 

[Pietato]

The idea is for the AI to get better as the difficulty level increases, instead of giving it stupid bonuses. The AI would be the same at King.

 

[acluewithout]

I disagree with the fundamental premise of your argument that says my statement is not credible. Generally, I note that the AI often has a significant number of units about the front in a given war. I also generally find that they have essentially no hope of conquering a city due to the human player's defenses. If they were to bring rams instead of catapults with their armies, my belief is they would be more successful. My hypothesis as to why they don't: Most of the people that buy and play the game are not CivFanatics and if they were to get stomped, they'd opt out and become disillusioned with the franchise.

Edit: I hardly ever build walls, and generally the AI still can't conquer my cities. I'm just saying that FXS are probably at least tacitly incentivized to brush this problem under the table. Whether they realize this or not, I would argue it is part of their decision making process.

Your argument is that FXS have deliberately made the AI incompetent so as to not discourage weaker players. But there are more reasonable explanations, that don’t require assumed bad faith by FXS, which are therefore more credible explanations. For example: the AI is incompetent because FXS can’t work out how to make it good; or it’s hard to make the AI good; or FXS haven’t devoted sufficient resources to it; or they haven’t got to it yet.

Deliberately making the AI incompetent to assist weaker players is also very backwards. Lots of games “cheat” to help the players - giving extra hit points the player doesn’t see they have, making opponent AIs miss, giving the player the benefit of a few extra pixels landing space in a platformer or making their hit box smaller than you’d expect. But they usually do so in a way that the player isn’t aware that’s happening. If your theory is right, FXS have tried to assist weaker players but only in the most obvious (and in the nose) way possible.

And then there are the bad reviews on Steam and on these forums because of the AI. Why would FXS want those?

And there are the myriad of other ways FXS could assist weaker players that wouldn’t be so obvious or on the nose. Why wouldn’t FXS use things like eg player buffs or AI debuffs instead?

And then there’s the fact FXS have been improving the AI over time. Why do that if FXS deliberately want the AI to be incompetent? Why improve it at all?

You’re welcome to your views. But I don’t think it’s credible to assume bad faith by FXS when there is no concrete evidence of bad faith here, no real substantive history of bad faith, and there are other more reasonable explanations for the AI’s problems.

 

[Elhoim]

Personally, when I wish the AI to be improved, it is mostly not because I wan't the AI to win more often - It is because I want them to play a more interesting game.

That's the main point, I believe. Currently, the problem with the AI is mostly regarding the use of units, making combat way less interesting, and rendering big parts of the game useless. For example, the issue with air units. The AI not really using them renders the whole air defense aspect meaningless, with lots of units and promotions choices that are not really used.

Also very glaring stuff like not fixing improvements when they have workers around, that also detracts from the experience.

You can cover with higher bonuses the deficiencies in building selection, city and district placement, but the use of units is something that it's harder to fix, it's the most glaring and has the biggest effect on making a game interesting or not.

 

[BarbarianHunter]

Your argument is that FXS have deliberately made the AI incompetent so as to not discourage weaker players. But there are more reasonable explanations, that don’t require assumed bad faith by FXS, which are therefore more credible explanations. For example: the AI is incompetent because FXS can’t work out how to make it good; or it’s hard to make the AI good; or FXS haven’t devoted sufficient resources to it; or they haven’t got to it yet.

I think we essentially agree on this matter. I know something on how decisions are made in corporations, and there most probably are resources that could be devoted to the matter. But as a relatively small percentage on the player base (and revenue stream) feel the AI is incompetent, once this reaches finance the mantra becomes "if it ain't broke for 95% of the player base, don't fix it." I don't think it is bad faith as you describe. It's just a corporation in a capitalist system offering a viable and fun product for the vast majority of the customers while maximizing profits. JMO.

 

[Victoria]

I don’t get why people keep asserting that FXS are deliberately coding the AI to be bad. It’s just not a credible assertion.

When I was young toy grabber machines came out where you had to get a toy with the grabber. My dad explained how they are programmed to only grab once they have made enough money and I was shocked. There was a program about that and slot machines a few years ago, blatant money making machines.

If anyone thinks online poker is programmed fairly and does not manipulate the player, I have some bad news.
Even candy crush is highly manipulative to give the user a ‘better’ experience and encourage them to pay.
In the case of Firaxis it may be as simple as do not build rockets until T250 or There is less ‘encouragement’ to build cities beyond certain numbers.

I feel it is credible.

 

[FearSunn]

I think we essentially agree on this matter. I know something on how decisions are made in corporations, and there most probably are resources that could be devoted to the matter. But as a relatively small percentage on the player base (and revenue stream) feel the AI is incompetent, once this reaches finance the mantra becomes "if it ain't broke for 95% of the player base, don't fix it." I don't think it is bad faith as you describe. It's just a corporation in a capitalist system offering a viable and fun product for the vast majority of the customers while maximizing profits. JMO.

But we have real symphony orchestra instead!

 

[BarbarianHunter]

But we have real symphony orchestra instead!

I'm not exactly sure what you mean by real symphony orchestra, so I ran a search.

Spoiler :

2nd from the top I found a reference to whacko. So I queried for the exact definition. And (coincidentally) 2nd from the top I found something that made sense relative to the AI's units wandering about (but never doing much).

Spoiler :

In order to arrive @ this conclusion, unfortunately one must be a bit mad oneself . Am I right though?

 

[FearSunn]

Search google for "civilization 6 symphony orchestra" and you will get the right answer.

 

[Archon_Wing]

I don’t get why people keep asserting that FXS are deliberately coding the AI to be bad. It’s just not a credible assertion.

I know, I think that is just giving them too much credit lol.

It just makes more sense that since most players do not go past diff 2....

 

[cvb]

When I was young ...

AFAIR Friedrich Nietzsche commented, that money which one is defrauded of, is _best_ invested money, because it goes directly 100% into experience ... no teacher or guide will ever reach that.

In the case of Firaxis it may be as simple as do not build rockets until T250 or There is less ‘encouragement’ to build cities beyond certain numbers.

We are doing infinite loops ... the VAST majority of players are "lamebrain". Civ provides for (nearly) EVERYONE to beat eg. Alexander the Great. What more could you wish for as publisher of such a fantastic game?

For us few, which think, "it is just too easy", finally the only thing, which helps, will be to mod it ourself. Means to throw sand in the machinery of the human player confronted by AIplayers not build rockets until T250 ...
For example, revert ONLY FOR THE HUMAN PLAYER to the yields of the older civ versions: grasslands 2-0-0, plains 1-1-0; grasslands hills 1-1-0, plains hills 0-2-0; tundra 1-0-0; tundra hills 0-1-0; i.e. hills give +1 production, but take away -1 food.
(No more issue of HIGH food of grasslands hills mines: Together with Sugar/Citrus/Spice ... a city in the grasslands never needs farms. Because grasslands hills mines already deliver enough food.)

PS. I'm not a masochist. Just like "fair" play.

 

[Victoria]

PS. I'm not a masochist. Just like "fair" play.

Well, the AI is growth crazy which means they slot amenity cards, which they still do.
Once you know this, it is just not fair... bring on the mods!

 

[The googles do nothing]

Based on the new features in GS Firaxis intentionally wants players to look at civ 6 as not a 'war game'. Environmental Disasters, dams and canals, Canada and Mâori, and World congress don't really add anything to the domination game. They are all intended for peaceful play styles. The GDR is the only domination feature and in the Inca Gamplay the said it's intended to speed up the end of a domination game.

 

[cvb]

bring on the mods!

The human brain is very adaptable (especially after decades of civ-play ). If I began to play on civ6vanilla release, i would be probably already be burned out (BEFORE the game is ready with the release of GS).
I'll start AFTER the game is ready .

Well, the AI is growth crazy which means they slot amenity cards, which they still do.
Once you know this, it is just not fair...

PLEASE, explain, "AI is growth crazy which means they slot amenity cards, which they still do" - I don't understand, please, explain.

Based on the new features in GS Firaxis intentionally wants players to look at civ 6 as not a 'war game'. [...] They are all intended for peaceful play styles.

"The new features in GS" provide sand in the machinery of the human player. We need this for balance as long as AI isn't as good as it should be.

We have to balance it even more, away from 'front-loading' towards 'continuous boni' ...

 

[Victoria]

PLEASE, explain, "AI is growth crazy which means they slot amenity cards, which they still do" - I don't understand, please, explain.

In the log files you can see what cards a civ slots.
If you look at the AI civs at T70-100 their cities are huge, they push pop to the limit,
To counteract this, instead of going coliseum or entertainment districts they slot the cards that give amenities.

This means as soon as you start slotting other cards you start catching up.

I have an egyptian king gamegoing that is at T100, shall I see what they have slotted?

 

[Talcove]

Google spent $10 billion on AlphaGo, if you give Firaxis that money they will may up with a fantastic AI. Unfortunately not every PC have a fast GPU for neural networks, and therefore only the highest-level computer owners can enjoy that AI.

Do neural nets and game AI use the graphics processing unit over the central processing unit? I would have thought that the GPU was just for the visuals.

 

[Kamamura]

Unpopular opinion - huge cheating bonuses are no substitute for an actual AI.

Also, Nvidia RTX cards now contain enough paralellized computing power for reasonable AI. The only constraint is the devs thinking they can get away without one.

 

[cvb]

In the log files you can see what cards a civ slots.
If you look at the AI civs at T70-100 their cities are huge, they push pop to the limit,

sounds good

To counteract this, instead of going coliseum or entertainment districts they slot the cards that give amenities.

I suppose, that's the Civ1 equivalent to simply turn discontent citizens into neutral (not working, but eating) Elvises instead of building Temples, Cathedrals or even Colosseums ... I wouldn't be surprised, if such is just not implemented. (not as Behaviour tree, not at all)
Hope we can see positive signals towards DLL sources soon. "Soon!"