Observations from a new Civ 5 player

Forget about that. It has no priority.
All resources go towards CivWorld and that game is in a horrid state.

That game should have never been created. People I have talked too say it is a total waste of time and resources. In any case, firaxis loves it still when hardcore traditional civ gamers, who have played for years like me, buy dlc. Since we by dlc, maybe they should fix the game, and give us our moneys worth for once.

nokmirt:

No Civ AI ever has, you know. The CiV AI is predictable, it just looks crazy when you don't know anything about how they behave (and refuse to learn).

The problem is that the CIV AI does the same thing over and over. Each new game is really no different than the one before it. BTW we all know too well how it behaves, why do you think everyone keeps telling firaxis to give it an overhaul. The AI is too predictable your right, but that is our main concern. We want an air of unpredictability, a surprise now and then to make the game more interesting.
 
We want an air of unpredictability, a surprise now and then to make the game more interesting.

No, we don't want that! The AI is predictable because it's weak and typically unfriendly for no reason. Making it less predictable will not make it better, but will just make it even more crazy/"stupid".
 
nokmirt:

Top reasons why "everyone" tells Firaxis to fix the AI:

1. Deity-only players can't compete with AI production bonuses, leading to military estimation mistaches. The effect is that Deity AIs are more aggressive than they are a lower diff settings where the AI is supposed to function well.

2. Everyone just repeats what the Deity players say, because they think that it has to be right.

3. Players don't have insight into their own actions, mistaking aggressive and jerk maneuvers for neutral ones, and getting angry when the AI does something similar to them, purely by accident.
 
The AI quite frankly needs to learn to play tactically. It's general strategy is to mass produce one unit type and then just blindly march forward.

It's using stacking logic in a 1upt environment, basically.

If it had a more sophisticated understanding of tactics and unit deployment it would be fine, however, I for one wouldn't volunteer to program it! :lol:

The thing is, the AI is never going to get to a level where it presents the human player with a regular thrashing. Whilst I'm sure it's possible, to have AI of that level would render the game unplayable for a large number of people who simply couldn't compete against it. That's not to say it doesn't need a fair bit of tinkering.

To be honest, I think a large amount of the issue in these discussions is the players misunderstanding of what is meant by AI.
 
The AI quite frankly needs to learn to play tactically. It's general strategy is to mass produce one unit type and then just blindly march forward.

It's using stacking logic in a 1upt environment, basically.

If it had a more sophisticated understanding of tactics and unit deployment it would be fine, however, I for one wouldn't volunteer to program it! :lol:

The thing is, the AI is never going to get to a level where it presents the human player with a regular thrashing. Whilst I'm sure it's possible, to have AI of that level would render the game unplayable for a large number of people who simply couldn't compete against it. That's not to say it doesn't need a fair bit of tinkering.

To be honest, I think a large amount of the issue in these discussions is the players misunderstanding of what is meant by AI.

I disagree... i think that AI can be much better, the problem here is not impossibility, but will to do, the AI the way it is now can be much improved with resource and time, but it will not be so becouse instead of evoulting this AI they will began working on another game (maybe even a new version of civ) with diferent mechanics, diferent resources, and they will have to reprogram the AI to consider it and not improve (at least not as much as we would like)... the problem here is not get the AI to be a good human equal, but at least to make it satisfatory at playing.
If at least they released the source so we could try to improve it (will not be as much as a company could do of course, but...)

As for making it unplayable for the normal player, well... for that are the dificult levels, you just have to make the ai make stupid things on lower one... much easier than make the good AI
 
As for making it unplayable for the normal player, well... for that are the dificult levels, you just have to make the ai make stupid things on lower one... much easier than make the good AI

This, is much harder for a game like Civ than for something like an FPS. In an FPS you can make the AI have a scaling margin of error particularly, miss chance. You couldn't do that in Civ so you end up with the scaling of production speeds and research bonuses.

These factors are currently used to make up for the AIs tactical and strategic inadequecies, if you eliminated that fault, I believe it would be extremely difficult to get the AI to scale it's tactical ability in a functional manner.

The overriding problem is, that to achieve the level of AI we would like to see, they would need to introduce an entirely new system. You can see just from the way that the AI approaches combat that it doesn't understand it at all, let alone properly. So it approaches combat as a numerical issue rather than a tactical one.

My point here, is that if they did introduce that level of AI capability, getting the difficulty scaling right would be virtually impossible because the range of tactics available don't require any system you can scale as far as I can see.

It's either smart, or it's not.
 
Becomedeath:

I've actually met with an Alex that marched forward with a combination of Hoplites and Companion Cavalry, and when that didn't work, he marched forward with Knights and Catapults.

Likewise, I had a Hiawatha AI go against me with a combo of Mohawks and Trebuchets, and a China AI go against me with Cho Ko Nu and Warriors/Pikemen (because it didn't have Iron).

In all of these cases, the AI would have been better had it actually gone for the one unit type strategy, as many human players would have done in the same place.

The way to create a brutal AI for a game is to determine ALL the optimal ways to play the game and then program the AI to execute the "combo" that's most likely to win from a play position. That said, Civ has never been designed as a MP game in the first place. The very conceptual foundations of Civ design assume that it is an assymetric single player game that's pretending that it's an MP game.
 
This, is much harder for a game like Civ than for something like an FPS. In an FPS you can make the AI have a scaling margin of error particularly, miss chance. You couldn't do that in Civ so you end up with the scaling of production speeds and research bonuses.

These factors are currently used to make up for the AIs tactical and strategic inadequecies, if you eliminated that fault, I believe it would be extremely difficult to get the AI to scale it's tactical ability in a functional manner.

The overriding problem is, that to achieve the level of AI we would like to see, they would need to introduce an entirely new system. You can see just from the way that the AI approaches combat that it doesn't understand it at all, let alone properly. So it approaches combat as a numerical issue rather than a tactical one.

My point here, is that if they did introduce that level of AI capability, getting the difficulty scaling right would be virtually impossible because the range of tactics available don't require any system you can scale as far as I can see.

It's either smart, or it's not.

That is done in chess and in that case the same thing can be done here, the AI just have to do bad tactical decisions sometimes... if the AI can evaluate what is a good tactical decision he can know what is a bad by exclusion... so on lower levels it have only to choose the bad or not so good ones sometimes, like letting that crucial unit alone unprotected, or atack that ubber protected choke point (where it will actuallt lose), that protected city, let the crucial resource unprotected, etc,etc,etc...

And on even lower levels the human player can have bonus like the AI have today on dificult levels.
 
nokmirt:

Top reasons why "everyone" tells Firaxis to fix the AI:

1. Deity-only players can't compete with AI production bonuses, leading to military estimation mistaches. The effect is that Deity AIs are more aggressive than they are a lower diff settings where the AI is supposed to function well.

2. Everyone just repeats what the Deity players say, because they think that it has to be right.

3. Players don't have insight into their own actions, mistaking aggressive and jerk maneuvers for neutral ones, and getting angry when the AI does something similar to them, purely by accident.

Wrong again APOC! There are enough players who know what I am talking about. In any case, it had nothing whatsoever to do with diety players, not sure why you would come to that conclusion?

No, we don't want that! The AI is predictable because it's weak and typically unfriendly for no reason. Making it less predictable will not make it better, but will just make it even more crazy/"stupid".

Nothing could make it worse than it is now. And, yes sir we do want that! Otherwise I wouldn't have said it, do you think I like to repeat myself for my health. A game is no fun, when you already know what to expect. All I am saying is that the AI should try different strategies, strategically and through diplomacy, so it does not do the exact same thing every game. That way every game is different from the one before it. Who wants to play the same old thing with the same old outcome. What would you get from it? NOTHING BUT BOREDOM!
 
Nothing could make it worse than it is now.

Emotive dramatisation and very little else. This sort of sums it up.

Go back to 1.0 and say that again.

Civ 5 has come a long long way since it's release, and quite frankly I'm getting bored of the utopia crew expecting the impossible.

If I've said it once, I've said it a thousand times, there are millions of Civ 5 players who love it. Being part of the outspoken few who want to claim disaster only means you're part of a group who appear to being playing a game you don't enjoy. Call the police, get them to get that guy with the gun to your head away from you.

The AI in Civ 5 is far from perfect, but if you're not willing to accept that then you have options. 1. Wait for it to become better as Friaxis patch it. 2. Do something else.

Mechanically, programming requirements for something as complex as AI in a game as complex as Civ 5 isn't something that's ever going to be right. I strongly suggest you either accept it, or go back to your console.

You're not playing on a supercomputer and the game isn't made by MIT, accept it's limitations.
 
If I've said it once, I've said it a thousand times, there are millions of Civ 5 players who love it.
Hardly. They can be measured in the thousands, and that's it.
The AI in Civ 5 is far from perfect, but if you're not willing to accept that then you have options. 1. Wait for it to become better as Friaxis patch it. 2. Do something else.
Option 1 seems to be no option, as the AI in Civ5 doesn't show improvements.
It may have changed, as I will admit, but it is still dump as bread.

I strongly suggest you either accept it, or go back to your console.

You're not playing on a supercomputer and the game isn't made by MIT, accept it's limitations.
Why should anybody accept the weak performance?
First of all, we were promised a good AI. Firaxis has not delivered.
Second, if we accept the AI's weaknesses as given, then we have to face the fact that the core principles of the game expose this weakness even more than in previous versions. Which in turn means that Firaxis has made some major faults in their design.

Pointing out the desastrous result can only lead to an improvement.
Trying to find excuses will only be an incentive to Firaxis to release even more weak versions. Do you really want this?
 
Hardly. They can be measured in the thousands, and that's it.

Option 1 seems to be no option, as the AI in Civ5 doesn't show improvements.
It may have changed, as I will admit, but it is still dump as bread.


Why should anybody accept the weak performance?
First of all, we were promised a good AI. Firaxis has not delivered.
Second, if we accept the AI's weaknesses as given, then we have to face the fact that the core principles of the game expose this weakness even more than in previous versions. Which in turn means that Firaxis has made some major faults in their design.

Pointing out the desastrous result can only lead to an improvement.
Trying to find excuses will only be an incentive to Firaxis to release even more weak versions. Do you really want this?

Thank God I am not the only CiV player who sees reason.
 
Emotive dramatisation and very little else. This sort of sums it up.

Go back to 1.0 and say that again.

Civ 5 has come a long long way since it's release, and quite frankly I'm getting bored of the utopia crew expecting the impossible.

If I've said it once, I've said it a thousand times, there are millions of Civ 5 players who love it. Being part of the outspoken few who want to claim disaster only means you're part of a group who appear to being playing a game you don't enjoy. Call the police, get them to get that guy with the gun to your head away from you.

The AI in Civ 5 is far from perfect, but if you're not willing to accept that then you have options. 1. Wait for it to become better as Friaxis patch it. 2. Do something else.

Mechanically, programming requirements for something as complex as AI in a game as complex as Civ 5 isn't something that's ever going to be right. I strongly suggest you either accept it, or go back to your console.

You're not playing on a supercomputer and the game isn't made by MIT, accept it's limitations.

I love the game, just feel it should have been fixed by now. BTW I do not expect it to be perfect, just much better. I do realize this is only the vanilla version, it will improve. Like others I am concerned with spending money on a game that is made better, one small bit at a time.
 
Back
Top Bottom