Can't deal with F@&*! cheating AI

[ywhtptgtfo]

Okay, so I just moved up to monarch.
I can beat prince 90% of the time, and easily, but I have trouble finishing up when I get to the industrial era. Too many other players are in better positions.

I am not sure how you can win prince 90% of the time if you don't build barracks, water mills, and workshops in the appropriate places.

By the time you get to Industrial Era, you should be very close to conquering your whole continent if that already hasn't happened. If there are any gold shrines and wonder cities on your continent, they also should've been your's. In the Monarch games I've played, losing the Liberalism race is a sign that you have not managed your economy sufficiently or have not conquered/expanded enough.

Tell me, what do you generally get for Liberalism in Prince? I think that's a good indicator of how good you are in that difficulty level.

 

[Carboniferous]

In the save posted, Cathy's raw production is approx twice yours (and your science is about equal, but her gold output is about half as much again as yours). And that's before the AI's bonuses are applied. So, it doesn't really matter how much the AI cheats.

And I would say, from a technical point of view, it does cheat, but not to a huge extent. Much of what makes people cross are game mechanics (particularly vassals) with fairly basic AI code behind them.

 

[Ataxerxes]

The AI is allowed to do that too, at least in the last patching. Take a look at what happens with FRIENDLY vassals to get the idea... That reminds me...how many times do you see an AI which is the world leader, decide to just FRIENDLY VASSAL to you for no reason at all? It does this to the other AI's ALL THE TIME.

Also, how come I can't demand an AI into a war vs someone else every few turns when it already has its hands full, yet I'm suppose to take the diplo hits when I'm already in a war and it keeps begging the same idiotic thing non-stop...

Actually, your first paragraph details one of my pet peeves - the hidden diplo modifiers from vassals where you don't know if he's really Friendly.

The second really is a diplo flaw, I agree, no question. Diplomacy is one of the things I like best in IV but it has a lot of defects, most of which are skewed against the player.

I didn't intend my post as 100% serious, as I don't automate my workers, ignore blue circles and can't even remember what the AI reccomendations were. Really, there are legitimate cheats against the player, it's true. However, the fact that I in AI stands for either idiot or insanity is all that makes the higher levels playable for the human.

 

[theskibum]

However, the fact that I in AI stands for either idiot or insanity is all that makes the higher levels playable for the human.

OTH if the I stood for Intelligence, then the AI wouldn't need all the cheats at higher levels.

If anyone has played Galatic Civilizations 2, the AI is pretty well implemented in that game. I am pretty sure it does not receive bonuses at higher levels, but rather it is programmed to do things that it will not do if you set it at a lower level. I also like that one because you can set some of the AI civs to more or less intellegence...very helpful when dealing with the Korath (think Monty for unitbuildprob plus monty's insanity, but with Shaka style empire management)

 

[The7Sins]

On one hand, I understand the furor some of you feel when someone posts an "AI cheats" thread ( I don't much care for it), but come on guys. This forum has a reputation as one of the friendliest, most helpful of all internet forums. Lets not let it degrade into the junior high-esque forum that belong to the likes of WoW.

Yes we should instead degrade this forum into a den of trolling and flames like Pojo. .................. kidding kidding.

 

[TheMeInTeam]

Another human cheat is prior knowledge of the leader personalities. As Ataxerxes mentioned if you start near shaka or another crazy AI you change how you play.

This is a total joke. AI leader personalities lead it to often sacrifice itself to create a super AI from turn 0. Not only that, but only the very extreme AIs are "safe" bets.

Civ IV AI i believe was designed to give a challenging game, but not 'play to win.' This type of AI allows for some consistency in diplomatic relations and role play in the leaders. If they all play to win you probably get the complaints you hear in Civ 5, how the AI can go from friendly to war in a turn or two.

Cop-out excuse. Civ V problem is the same as this; the AI will implode to prop up other AI at random, and gets angry over incomprehensible things that do NOT help it win. Declaring on a human (or getting mad) because the human is fighting against the empire 3x bigger than everyone else and took 2-3 cities after the AI declared first is NOT a winning strategy. It *should* turncoat the human eventually. It should NOT optimize a DIFFERENT AI's chances to win. There is a big conceptual difference here between what you think the problem in V is and what it actually is. The reality is that the AI garbage is actually very similar between the 2 games.

And obs makes another good point on AI "cheats", it can literally make demands/requests that the human often can't.

 

[babar]

Also, how come I can't demand an AI into a war vs someone else every few turns when it already has its hands full, yet I'm suppose to take the diplo hits when I'm already in a war and it keeps begging the same idiotic thing non-stop...

Yeah I hate that. Being barraged with idiotic war requests turn after turn when I'm already unlikely to survive Shaka's recent DOW. **** off you dumb AIs, can't you see I'm the one that needs help?

It would be nice if the human could go into WHEOOHRN mode, and maybe take negative diplo hit with everyone for its duration. I promise not to abuse it if it gets added in.

 

[insaneweasel]

I am not sure how you can win prince 90% of the time if you don't build barracks, water mills, and workshops in the appropriate places.

By the time you get to Industrial Era, you should be very close to conquering your whole continent if that already hasn't happened. If there are any gold shrines and wonder cities on your continent, they also should've been your's. In the Monarch games I've played, losing the Liberalism race is a sign that you have not managed your economy sufficiently or have not conquered/expanded enough.

Tell me, what do you generally get for Liberalism in Prince? I think that's a good indicator of how good you are in that difficulty level.

You would be surprised at how easy it is to win prince with a single heroic epic city, cottage spamming, and 15 rifles. I don't tend to build barracks early game unless I'm going for massive military superiority.

I usually get liberalism between 1400-1700, and get mainly steel or rifling.

Save your patronizing condescension. If someone posts a ******** thread no one's under any obligation to indulge that thread out of some sense of friendly reputation. That reputation has been earned, not granted, and if someone wants help it's easily attainable but if someone just wants to whine and rant then they shouldn't expect coddling.

I suggest you actually read my original post. The starter thread is known in advertising as a "hook" to get people to want to read the rest of my post.

The rest of what I had to say couldn't exactly be construed as "whiney." If you're going to criticize someone on a forum, don't make it a false criticism or else you end up looking like a fool.

As it stands, I had made the intro provocative in the hopes of getting real advice that I could use. I doubted that anyone would want to help me if I had a simple "Need help" post. I actually intend to do that very soon, but I hold little hope of attracting attention with that strategy. People are just too busy, especially at this time of year.

Civ V problem is the same as this; the AI will implode to prop up other AI at random, and gets angry over incomprehensible things that do NOT help it win. Declaring on a human (or getting mad) because the human is fighting against the empire 3x bigger than everyone else and took 2-3 cities after the AI declared first is NOT a winning strategy. It *should* turncoat the human eventually. It should NOT optimize a DIFFERENT AI's chances to win. There is a big conceptual difference here between what you think the problem in V is and what it actually is. The reality is that the AI garbage is actually very similar between the 2 games.

It seems (from my limited perspective) that the CiV AI is extremely inconsistent. With no religion, social policies, or tech trading, it's much easier to get an AI to hate you than to like you. In IV, you could be pretty sure that a friendly leader wouldn't turn on you if your military was a bit weak, whilst in 5 the AI can and will attack if they see an opening-no matter what your previous history with it was. The lack of consistency is what drives me crazy.

 

[peppe1]

This is a total joke. AI leader personalities lead it to often sacrifice itself to create a super AI from turn 0. Not only that, but only the very extreme AIs are "safe" bets.

If you took the best personality or best 3 and just randomly assigned them to the AI's you would probably have the more competitive gave you are after.

And maybe that is what the AI mod or an AI mod should do is collapse down how the AI plays to only the best personalities to battle against competitive players. I personally would find that to be a more boring game ... for better or worse the personalities create an interesting background to the game. If they all had the same personality or no personality then you might as well name them player 1, player 2, player 3, etc...

Cop-out excuse. Civ V problem is the same as this; the AI will implode to prop up other AI at random, and gets angry over incomprehensible things that do NOT help it win. Declaring on a human (or getting mad) because the human is fighting against the empire 3x bigger than everyone else and took 2-3 cities after the AI declared first is NOT a winning strategy. It *should* turncoat the human eventually. It should NOT optimize a DIFFERENT AI's chances to win. There is a big conceptual difference here between what you think the problem in V is and what it actually is. The reality is that the AI garbage is actually very similar between the 2 games.

And obs makes another good point on AI "cheats", it can literally make demands/requests that the human often can't.

I'll concede i don't know what specifically is wrong with civ 5's AI and its diplomacy other than 'a lot.' Civ IV BTS with mods is a fun game with a challenging AI.

On the second point above its either better AI or revdcm that adds more options to the player to AI diplomacy, so you can for example go into ignore mode with a nation just like the AI does to you. I believe there is also an option to go into or respond with 'we have too much on our hands.'

Also in these mods I have had a peace vassal select me as its protectorate. Pretty sure that is a revdcm thing, but it could be from better AI (haven't been tracking this mod's change log). So the list of things the AI can do that the player can't is dwindling...

I've been cycling through mods lately, so sorry if I don't recall exactly which ones have the features vs not.

 

[Gwynnja]

I suggest you actually read my original post. The starter thread is known in advertising as a "hook" to get people to want to read the rest of my post.

The rest of what I had to say couldn't exactly be construed as "whiney." If you're going to criticize someone on a forum, don't make it a false criticism or else you end up looking like a fool.

oh. Sorry. I guess I meant this one:

Ramses did a huge REX without COL and didn't ruin his economy.
Fredrick has cruddy land and few resources yet is the tech leader.

Catherine managed to have a two-front war with me and Mehmet.

I don't understand how they can do that.

 

[ywhtptgtfo]

You would be surprised at how easy it is to win prince with a single heroic epic city, cottage spamming, and 15 rifles. I don't tend to build barracks early game unless I'm going for massive military superiority.

I usually get liberalism between 1400-1700, and get mainly steel or rifling.

It depends on setting. I play with following settings for any game of any difficulty: Large, Marathon, 11 A.I.

Steel or Rifling are not bad deals for Liberalism. I usually beeline Scientific Methods and sling shot to Communism (instead of Biology) with my Liberalism. GS -> Golden Age + State Property are great.

I usally get a conquest/domination victory by 1500-1800 on Prince, so Liberalism at 1400 - 1700 is a bit late. I am not sure how you play in your other Prince games, but turtling until rifles does not always work.

Barracks are always something you should get if you intend to have a city to produce military units. Otherwise, you are kind of wasting your hammers by producing inferior units that die more easily. For example, a HA with Barracks and Stables have Flanking I and II or a 50% retreat chance + First strike immunity whereas an unupgraded HA has only 20% retreat chance. You can certainly win wars without these additional benefits, but this lousiness should incur a substantial hammer cost in the long run.

 

[Amao]

@insaneweasel

Here is what I learnt from DaveMcW that lead me from Noble to Monarch so far: Get your Bureaucratic Capital up with Academy asap, add library/university/observatory/oxford when they are available, and 10% from those priest training schools if you get a religion.

Another thing you want to do:
Time the workshop spamming wisely, sometime like you get chemistry, use Caste. again, this requires a lot of workers. and later use Commisium if you do need the hammers for war.

 

[TheMeInTeam]

It seems (from my limited perspective) that the CiV AI is extremely inconsistent. With no religion, social policies, or tech trading, it's much easier to get an AI to hate you than to like you. In IV, you could be pretty sure that a friendly leader wouldn't turn on you if your military was a bit weak, whilst in 5 the AI can and will attack if they see an opening-no matter what your previous history with it was. The lack of consistency is what drives me crazy.

I've run several civ V games w/o ever having them declare war. The tricks are different, but there are still tricks.

If you took the best personality or best 3 and just randomly assigned them to the AI's you would probably have the more competitive gave you are after.

Probably. The sucky thing is that the only aspect of personality variance the AI REALLY needs is expansion rate/priority. Giving one guy 12 cities while you have 3-4, as an AI, is an incredibly weak position that guarantees peace-vassaling and 0 contention to win the game. If Gandhi had more cities, maybe he wouldn't get facerolled.

 

[Loukoum427]

In IV, you could be pretty sure that a friendly leader wouldn't turn on you if your military was a bit weak, whilst in 5 the AI can and will attack if they see an opening-no matter what your previous history with it was. The lack of consistency is what drives me crazy.

I'm a little bit off-topic there, but hey, that's exactly what happened and happens in real history... To be honest, maybe you can explain that by the facts that leaders don't stay the same for thousands of years

 

[CornPlanter]

I am not sure how you can win prince 90% of the time if you don't build barracks, water mills, and workshops in the appropriate places.

I ignored water mills and workshops until I started playing Emperor. Barracks aren't really necessary @ Prince either.

I am not saying it's a good job to avoid or mess up these. I'm just saying you can easily win Prince without them. Some of the MadScientists made up rules were much harsher than that, and people could still win @ Monarch+.

 

[peppe1]

Probably. The sucky thing is that the only aspect of personality variance the AI REALLY needs is expansion rate/priority. Giving one guy 12 cities while you have 3-4, as an AI, is an incredibly weak position that guarantees peace-vassaling and 0 contention to win the game. If Gandhi had more cities, maybe he wouldn't get facerolled.

I took a quick look at the build probabilities and do see a correlation between the 'weak' civs that often get steamrolled and the civs that tend to become runaways.

Maybe better AI guys should take a look at re-balancing the unit probabilities to ensure each AI can defend itself and expand well.

Should they all be Shaka's at 40?

Edit: I was assuming workers/settlers and thus expansion come out of the build unit probability. If there is an expansion weight or something then that might be something to look at.